Skip to content

Instantly share code, notes, and snippets.

@HDCharles

HDCharles/gist:c0d418e1307d9f5248b359b2ffa25427

Created March 20, 2025 18:58
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save HDCharles/c0d418e1307d9f5248b359b2ffa25427 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save HDCharles/c0d418e1307d9f5248b359b2ffa25427 to your computer and use it in GitHub Desktop.
Download ZIP
tl parse
Raw
gistfile1.txt
V0320 11:46:30.704000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0, "has_payload": "6dda4945313dbc76cddf217f3df965aa"}
{
"name": "dynamo",
"ts": 1742496390704270.5,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:30.705000 18434 site-packages/torch/_logging/structured.py:27] {"str": ["/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/convert_frame.py", 0]}
V0320 11:46:30.705000 18434 site-packages/torch/_logging/structured.py:27] {"str": ["/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py", 1]}
V0320 11:46:30.706000 18434 site-packages/torch/_logging/structured.py:27] {"str": ["/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/nn/modules/module.py", 2]}
V0320 11:46:30.706000 18434 site-packages/torch/_logging/structured.py:27] {"str": ["/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py", 3]}
V0320 11:46:30.706000 18434 site-packages/torch/_dynamo/convert_frame.py:1025] {"dynamo_start": {"stack": [{"line": 118, "name": "<module>", "filename": 1, "loc": "moe_c(input1)"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 1762, "name": "_call_impl", "filename": 2, "loc": "return forward_call(*args, **kwargs)"}, {"line": 586, "name": "_fn", "filename": 3, "loc": "return fn(*args, **kwargs)"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 24, "name": "forward", "filename": 1}]}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.706000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0, "has_payload": "b436bcade896acb7d14f10a3f2067e71"}
{
"name": "entire_frame_compile",
"ts": 1742496390706748.2,
"args": {
"fn_name": "_compile.compile_inner",
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:30.710000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 0, "describer_id": 0, "size": 48}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.710000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 0, "ndim": 3, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [1, 6, 4], "is_leaf": true, "stride": [24, 4, 1], "storage": 0, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fbce5593430>)", "describer_id": 0}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.711000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 0, "id": 0, "source": "L['x']"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.717000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 1, "describer_id": 0, "size": 64}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.718000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 2, "ndim": 2, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [8, 4], "is_leaf": true, "requires_grad": true, "is_parameter": true, "stride": [4, 1], "storage": 1, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Parameter object at 0x7fbce5592300>)", "describer_id": 0}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.718000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 0, "id": 2, "source": "L['self']._modules['gate']._parameters['weight']"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:30.745000 18434 site-packages/torch/_dynamo/symbolic_convert.py:512] {"artifact": {"name": "dynamo_graph_break_reason", "encoding": "string"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 0, "has_payload": "731bddafe90368d45f590b3d83b841c9"}
Graph break in user code at /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:74
Graph Break Reason: Dynamic slicing with Tensor arguments
Explanation: Creating slices with Tensor arguments is not supported. e.g. `l[:x]`, where `x` is a 1-element tensor.
Hint: It may be possible to write Dynamo tracing rules for this code. Please report an issue to PyTorch if you encounter this graph break often and it is causing performance issues.
Developer debug context: SliceVariable start: TensorVariable(), stop: TensorVariable(), step: ConstantVariable(NoneType: None)
User code traceback:
File "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py", line 32, in forward
out = self.cond_ffn(x, expert_indices, expert_weights, self.num_activated_experts)
File "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py", line 74, in forward
token_indices_per_expert = [ordered_tokens[cum_tokens_per_expert[expert]:cum_tokens_per_expert[expert+1]] for expert in expert_list]
Traceback (most recent call last):
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 794, in wrapper
return inner_fn(self, inst)
^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 2753, in CALL
self._call(inst)
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 2747, in _call
self.call_function(fn, args, kwargs)
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 1050, in call_function
self.push(fn.call_function(self, args, kwargs)) # type: ignore[arg-type]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/variables/lazy.py", line 201, in realize_and_forward
return getattr(self.realize(), name)(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/variables/nn_module.py", line 952, in call_function
return variables.UserFunctionVariable(fn, source=source).call_function(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/variables/functions.py", line 405, in call_function
return super().call_function(tx, args, kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/variables/functions.py", line 186, in call_function
return tx.inline_user_function_return(self, [*self.self_args(), *args], kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 1067, in inline_user_function_return
return InliningInstructionTranslator.inline_call(self, fn, args, kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 3526, in inline_call
return tracer.inline_call_()
^^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 3705, in inline_call_
self.run()
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 1216, in run
while self.step():
^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 1126, in step
self.dispatch_table[inst.opcode](self, inst)
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/symbolic_convert.py", line 2234, in BUILD_SLICE
self.push(SliceVariable(items))
^^^^^^^^^^^^^^^^^^^^
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/variables/lists.py", line 975, in __init__
unimplemented_v2(
File "/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/torch/_dynamo/exc.py", line 519, in unimplemented_v2
raise Unsupported(msg)
torch._dynamo.exc.Unsupported: Dynamic slicing with Tensor arguments
Explanation: Creating slices with Tensor arguments is not supported. e.g. `l[:x]`, where `x` is a 1-element tensor.
Hint: It may be possible to write Dynamo tracing rules for this code. Please report an issue to PyTorch if you encounter this graph break often and it is causing performance issues.
Developer debug context: SliceVariable start: TensorVariable(), stop: TensorVariable(), step: ConstantVariable(NoneType: None)
V0320 11:46:30.746000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 0, "describer_id": 1, "size": 48}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:30.746000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 0, "ndim": 3, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [1, 6, 4], "is_leaf": true, "stride": [24, 4, 1], "storage": 0, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fbce5593430>)", "describer_id": 1}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:30.747000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 1, "id": 0, "source": "L['x']"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:30.749000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 1, "describer_id": 1, "size": 64}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:30.749000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 1, "ndim": 2, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [8, 4], "is_leaf": true, "requires_grad": true, "is_parameter": true, "stride": [4, 1], "storage": 1, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Parameter object at 0x7fbce5592300>)", "describer_id": 1}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:30.749000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 1, "id": 1, "source": "L['self']._modules['gate']._parameters['weight']"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:30.755000 18434 site-packages/torch/_dynamo/output_graph.py:1411] {"dynamo_output_graph": {"sizes": {"l_x_": [1, 6, 4], "l_self_modules_gate_parameters_weight_": [8, 4], "x": [6, 4], "scores": [6, 8], "expert_weights": [6, 8], "expert_weights_1": [6, 2], "expert_indices": [6, 2], "sum_1": [6, 1], "expert_weights_2": [6, 2]}}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "11ba0199cd7be0af4686a1e52138e5c6"}
class GraphModule(torch.nn.Module):
def forward(self, L_x_: "bf16[1, 6, 4][24, 4, 1]cuda:0", L_self_modules_gate_parameters_weight_: "bf16[8, 4][4, 1]cuda:0"):
l_x_ = L_x_
l_self_modules_gate_parameters_weight_ = L_self_modules_gate_parameters_weight_
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:27 in forward, code: x = x.view(-1, self.dim) # x: [T, D]
x: "bf16[6, 4][4, 1]cuda:0" = l_x_.view(-1, 4); l_x_ = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:28 in forward, code: scores = self.gate(x) # [T, E]
scores: "bf16[6, 8][8, 1]cuda:0" = torch._C._nn.linear(x, l_self_modules_gate_parameters_weight_, None); l_self_modules_gate_parameters_weight_ = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:29 in forward, code: expert_weights = F.softmax(scores, dim=-1)
expert_weights: "bf16[6, 8][8, 1]cuda:0" = torch.nn.functional.softmax(scores, dim = -1); scores = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:30 in forward, code: expert_weights, expert_indices = torch.topk(expert_weights, self.num_activated_experts, dim=-1) # [T, A], [T, A]
topk = torch.topk(expert_weights, 2, dim = -1); expert_weights = None
expert_weights_1: "bf16[6, 2][2, 1]cuda:0" = topk[0]
expert_indices: "i64[6, 2][2, 1]cuda:0" = topk[1]; topk = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:31 in forward, code: expert_weights /= expert_weights.sum(dim=-1, keepdim=True) # [T, A]
sum_1: "bf16[6, 1][1, 1]cuda:0" = expert_weights_1.sum(dim = -1, keepdim = True)
expert_weights_1 /= sum_1; expert_weights_2: "bf16[6, 2][2, 1]cuda:0" = expert_weights_1; expert_weights_1 = sum_1 = None
return (x, expert_indices, expert_weights_2)
V0320 11:46:30.755000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "6452acd59c133984d45419ec2dd05913"}
{
"name": "backend_compile",
"ts": 1742496390755824.8,
"args": {
"fn_name": "OutputGraph.call_user_compiler",
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:30.756000 18434 site-packages/torch/_inductor/compile_fx.py:1965] {"inductor_pre_grad_graph": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "b4dfd4f2af5a8aa284093f4419abbbf4"}
class GraphModule(torch.nn.Module):
def forward(self, L_x_: "bf16[1, 6, 4][24, 4, 1]cuda:0", L_self_modules_gate_parameters_weight_: "bf16[8, 4][4, 1]cuda:0"):
l_x_ = L_x_
l_self_modules_gate_parameters_weight_ = L_self_modules_gate_parameters_weight_
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:27 in forward, code: x = x.view(-1, self.dim) # x: [T, D]
x: "bf16[6, 4][4, 1]cuda:0" = l_x_.view(-1, 4); l_x_ = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:28 in forward, code: scores = self.gate(x) # [T, E]
scores: "bf16[6, 8][8, 1]cuda:0" = torch._C._nn.linear(x, l_self_modules_gate_parameters_weight_, None); l_self_modules_gate_parameters_weight_ = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:29 in forward, code: expert_weights = F.softmax(scores, dim=-1)
expert_weights: "bf16[6, 8][8, 1]cuda:0" = torch.nn.functional.softmax(scores, dim = -1); scores = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:30 in forward, code: expert_weights, expert_indices = torch.topk(expert_weights, self.num_activated_experts, dim=-1) # [T, A], [T, A]
topk = torch.topk(expert_weights, 2, dim = -1); expert_weights = None
expert_weights_1: "bf16[6, 2][2, 1]cuda:0" = topk[0]
expert_indices: "i64[6, 2][2, 1]cuda:0" = topk[1]; topk = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:31 in forward, code: expert_weights /= expert_weights.sum(dim=-1, keepdim=True) # [T, A]
sum_1: "bf16[6, 1][1, 1]cuda:0" = expert_weights_1.sum(dim = -1, keepdim = True)
expert_weights_1 /= sum_1; expert_weights_2: "bf16[6, 2][2, 1]cuda:0" = expert_weights_1; expert_weights_1 = sum_1 = None
return (x, expert_indices, expert_weights_2)
# graph id: 140438521437872
V0320 11:46:30.756000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "af0940862c43496ff0574a817f39661c"}
{
"name": "_recursive_pre_grad_passes",
"ts": 1742496390756867.2,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:30.759000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "7cf3be4490835c66ad576beb30b56aea"}
{
"name": "_recursive_pre_grad_passes",
"ts": 1742496390759648.5,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:30.761000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "1eccfaa40d3c32b9480f4c74dbe9a705"}
{
"name": "inductor_codecache_torch_key",
"ts": 1742496390760971.5,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:30.832000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "ebd4e6ef70a11e817241573bbde313a3"}
{
"name": "inductor_codecache_torch_key",
"ts": 1742496390832129.8,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.040000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "63c7dbcf237f395e444e642ea25cdb42"}
{
"name": "create_aot_dispatcher_function",
"ts": 1742496391040483.5,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.063000 18434 site-packages/torch/_functorch/_aot_autograd/dispatch_and_compile_graph.py:214] {"artifact": {"name": "aot_forward_graph_fw_metadata", "encoding": "string"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "d093dbc5e2c9cbfbc83ed56a0ae1f13d"}
ViewAndMutationMeta(input_info=[InputAliasInfo(is_leaf=True,
mutates_data=False,
mutates_metadata=False,
mutations_hidden_from_autograd=True,
mutations_under_no_grad_or_inference_mode=False,
mutation_inductor_storage_resize=False,
mutates_storage_metadata=False,
requires_grad=False,
keep_input_mutations=True),
InputAliasInfo(is_leaf=True,
mutates_data=False,
mutates_metadata=False,
mutations_hidden_from_autograd=True,
mutations_under_no_grad_or_inference_mode=False,
mutation_inductor_storage_resize=False,
mutates_storage_metadata=False,
requires_grad=True,
keep_input_mutations=True)],
output_info=[OutputAliasInfo(output_type=<OutputType.alias_of_input: 2>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=0,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=<torch._functorch._aot_autograd.functional_utils.FunctionalTensorMetadataEq object at 0x7fba640dee10>),
OutputAliasInfo(output_type=<OutputType.non_alias: 1>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=None,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=None),
OutputAliasInfo(output_type=<OutputType.non_alias: 1>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=None,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=None)],
num_intermediate_bases=0,
keep_input_mutations=True,
traced_tangents=[],
subclass_inp_meta=[PlainTensorMeta(unwrapped_idx=0,
memory_format=None),
PlainTensorMeta(unwrapped_idx=1,
memory_format=None)],
subclass_fw_graph_out_meta=[PlainTensorMeta(unwrapped_idx=0,
memory_format=None),
PlainTensorMeta(unwrapped_idx=1,
memory_format=None),
PlainTensorMeta(unwrapped_idx=2,
memory_format=None)],
subclass_tangent_meta=[],
is_train=False,
traced_tangent_metas=None,
num_symints_saved_for_bw=None,
grad_enabled_mutation=None,
deterministic=None,
static_input_indices=[1],
tokens={},
indices_of_inputs_that_requires_grad_with_mutations_in_bw=[],
bw_donated_idxs=None,
num_backward_tokens=0,
num_graphsafe_rng_states=0,
graphsafe_rng_state_index=None)
V0320 11:46:31.063000 18434 site-packages/torch/_functorch/_aot_autograd/dispatch_and_compile_graph.py:232] {"aot_inference_graph": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "d4325981c3a68bcc45a1d876d261a76e"}
class <lambda>(torch.nn.Module):
def forward(self, arg0_1: "bf16[1, 6, 4][24, 4, 1]cuda:0", arg1_1: "bf16[8, 4][4, 1]cuda:0"):
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:27 in forward, code: x = x.view(-1, self.dim) # x: [T, D]
view: "bf16[6, 4][4, 1]cuda:0" = torch.ops.aten.view.default(arg0_1, [-1, 4]); arg0_1 = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:28 in forward, code: scores = self.gate(x) # [T, E]
permute: "bf16[4, 8][1, 4]cuda:0" = torch.ops.aten.permute.default(arg1_1, [1, 0]); arg1_1 = None
mm: "bf16[6, 8][8, 1]cuda:0" = torch.ops.aten.mm.default(view, permute); permute = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:29 in forward, code: expert_weights = F.softmax(scores, dim=-1)
convert_element_type_2: "f32[6, 8][8, 1]cuda:0" = torch.ops.prims.convert_element_type.default(mm, torch.float32); mm = None
amax: "f32[6, 1][1, 1]cuda:0" = torch.ops.aten.amax.default(convert_element_type_2, [-1], True)
sub: "f32[6, 8][8, 1]cuda:0" = torch.ops.aten.sub.Tensor(convert_element_type_2, amax); convert_element_type_2 = amax = None
exp: "f32[6, 8][8, 1]cuda:0" = torch.ops.aten.exp.default(sub); sub = None
sum_1: "f32[6, 1][1, 1]cuda:0" = torch.ops.aten.sum.dim_IntList(exp, [-1], True)
div: "f32[6, 8][8, 1]cuda:0" = torch.ops.aten.div.Tensor(exp, sum_1); exp = sum_1 = None
convert_element_type_3: "bf16[6, 8][8, 1]cuda:0" = torch.ops.prims.convert_element_type.default(div, torch.bfloat16); div = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:30 in forward, code: expert_weights, expert_indices = torch.topk(expert_weights, self.num_activated_experts, dim=-1) # [T, A], [T, A]
topk = torch.ops.aten.topk.default(convert_element_type_3, 2); convert_element_type_3 = None
getitem: "bf16[6, 2][2, 1]cuda:0" = topk[0]
getitem_1: "i64[6, 2][2, 1]cuda:0" = topk[1]; topk = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:31 in forward, code: expert_weights /= expert_weights.sum(dim=-1, keepdim=True) # [T, A]
sum_2: "bf16[6, 1][1, 1]cuda:0" = torch.ops.aten.sum.dim_IntList(getitem, [-1], True)
div_1: "bf16[6, 2][2, 1]cuda:0" = torch.ops.aten.div.Tensor(getitem, sum_2); getitem = sum_2 = None
return (view, getitem_1, div_1)
V0320 11:46:31.064000 18434 site-packages/torch/_functorch/_aot_autograd/jit_compile_runtime_wrappers.py:185] {"artifact": {"name": "torch._functorch.config", "encoding": "string"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "64d79258b64ad8a5b3d76727b5c240ae"}
{
"TYPE_CHECKING": false,
"functionalize_rng_ops": false,
"fake_tensor_allow_meta": true,
"debug_assert": false,
"debug_partitioner": false,
"decompose_custom_triton_ops": true,
"static_weight_shapes": true,
"cse": true,
"enable_autograd_cache": true,
"enable_remote_autograd_cache": null,
"view_replay_for_aliased_outputs": true,
"max_dist_from_bw": 1000,
"ban_recompute_used_far_apart": true,
"ban_recompute_long_fusible_chains": true,
"ban_recompute_materialized_backward": true,
"ban_recompute_not_in_allowlist": true,
"ban_recompute_reductions": true,
"recompute_views": false,
"activation_memory_budget": 1.0,
"activation_memory_budget_runtime_estimator": "flops",
"activation_memory_budget_solver": "dp",
"visualize_memory_budget_pareto": false,
"aggressive_recomputation": false,
"fake_tensor_allow_unsafe_data_ptr_access": true,
"unlift_effect_tokens": true,
"fake_tensor_crossref": false,
"fake_tensor_propagate_real_tensors": false,
"donated_buffer": true,
"torch_compile_graph_format": "svg",
"generate_fake_kernels_from_real_mismatches": false,
"graphsafe_rng_functionalization": true,
"strict_autograd_cache": false,
"disable_guess_zero_tangent_for_mutated_input_subclass": false
}
V0320 11:46:31.065000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "139225ee4ce7c8dd9273d2a21e37135b"}
{
"name": "compile_fx.<locals>.fw_compiler_base",
"ts": 1742496391065284.2,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.065000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "9ddceaae212b3bd39c870ede18caf14f"}
{
"name": "_recursive_joint_graph_passes",
"ts": 1742496391065760.0,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.259000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "02d5398e6b75e42b332ccf26a1a2cad6"}
{
"name": "pad_mm_benchmark",
"ts": 1742496391259792.8,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.260000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "b922cefea85aa08d5adc9f4b4f21eddb"}
{
"name": "pad_mm_benchmark_get_do_bench",
"ts": 1742496391260846.0,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.261000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "cea0d54fb59827056aee17acc5462a84"}
{
"name": "pad_mm_benchmark_get_do_bench",
"ts": 1742496391261398.2,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.332000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "b2b799d07dfd071c65f014a86da6334d"}
{
"name": "pad_mm_benchmark",
"ts": 1742496391331900.0,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.333000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "bbaa17363e8ba8a90c98bb09787a098b"}
{
"name": "_recursive_joint_graph_passes",
"ts": 1742496391333732.8,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.334000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "ee2a42c4630ce3e07fafca54167fc9d6"}
{
"name": "inductor_compile",
"ts": 1742496391334192.0,
"args": {
"fn_name": "compile_fx_inner",
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.339000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "3b1f370b4417d07c84a085485dafaec7"}
{
"name": "TritonBundler.read_and_emit",
"ts": 1742496391339944.2,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.340000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "a78be66965519f228c1844749ab829e5"}
{
"name": "TritonBundler.read_and_emit",
"ts": 1742496391340414.2,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.340000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "9277611f3d735eb824eb358ba14f2153"}
{
"name": "PyCodeCache.load_by_key_path",
"ts": 1742496391340724.0,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.345000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "228e525b385a9a3c116dc89067e12961"}
{
"name": "async_compile.precompile",
"ts": 1742496391345332.8,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.357000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "c3acb609621f73d31d8c36dc456e3f39"}
{
"name": "async_compile.precompile",
"ts": 1742496391357463.5,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.358000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "08c9746c5fc618d30465340d52e4e06f"}
{
"name": "async_compile.precompile",
"ts": 1742496391358166.2,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.362000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "8a30c5d9322b7f42d6b2ea6d762f1a2a"}
{
"name": "async_compile.precompile",
"ts": 1742496391362764.8,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.363000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "e5532ec494e71430702476d737e8d8ad"}
{
"name": "async_compile.wait",
"ts": 1742496391363124.8,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.363000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "24d504b80859cc828720713b492b51fe"}
{
"name": "async_compile.wait",
"ts": 1742496391363628.8,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.364000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "c0c81d79d41392404e318f098795c3b3"}
{
"name": "PyCodeCache.load_by_key_path",
"ts": 1742496391363972.5,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.364000 18434 site-packages/torch/_inductor/codecache.py:1094] {"inductor_output_code": {"filename": "/tmp/torchinductor_cdhernandez/fv/cfv6ao6d4pq4zvnvsj2vljwky2ikmcygekpu7xsbu5r3rt33d5ae.py"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "a1eb4b622ede9afb0f8aee405654ee5c"}
# AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from cmath import nanj
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import (
grid,
split_scan_grid,
grid_combo_kernels,
start_graph,
end_graph,
cooperative_reduction_grid,
)
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
empty_strided_p2p = torch._C._distributed_c10d._SymmetricMemory.empty_strided_p2p
# kernel path: /tmp/torchinductor_cdhernandez/c6/cc6e2f2wpsexjwpnpswzrkc7z7tbgplcvftlu4wg4pdwub6c2wsg.py
# Topologically Sorted Source Nodes: [expert_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# expert_weights => amax, convert_element_type_2, convert_element_type_3, div, exp, sub, sum_1
# Graph fragment:
# %convert_element_type_2 : [num_users=2] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%mm, torch.float32), kwargs = {})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%convert_element_type_2, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%convert_element_type_2, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
# %convert_element_type_3 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%div, torch.bfloat16), kwargs = {})
triton_per_fused__softmax_0 = async_compile.triton('triton_per_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, DeviceProperties
triton_helpers.set_driver_to_gpu()
@triton_heuristics.persistent_reduction(
size_hints={'x': 8, 'r0_': 8},
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {'in_out_ptr0': '*bf16', 'xnumel': 'i32', 'r0_numel': 'i32'}, 'device': DeviceProperties(type='cuda', index=0, multi_processor_count=132, cc=90, major=9, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, warp_size=32), 'constants': {}, 'configs': [AttrsDescriptor.from_dict({'arg_properties': {'tt.divisibility': (0,), 'tt.equal_to': ()}, 'cls': 'AttrsDescriptor'})]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__softmax_0', 'mutated_arg_names': ['in_out_ptr0'], 'optimize_mem': True, 'no_x_dim': False, 'num_load': 1, 'num_reduction': 2, 'backend_hash': 'CE326918D8FE98067798167ABD26A2E4EDFD110D9ECD4380441C64512F6D164E', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__softmax_0(in_out_ptr0, xnumel, r0_numel, XBLOCK : tl.constexpr):
xnumel = 6
r0_numel = 8
R0_BLOCK: tl.constexpr = 8
rnumel = r0_numel
RBLOCK: tl.constexpr = R0_BLOCK
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
r0_index = tl.arange(0, R0_BLOCK)[None, :]
r0_offset = 0
r0_mask = tl.full([XBLOCK, R0_BLOCK], True, tl.int1)
roffset = r0_offset
rindex = r0_index
r0_1 = r0_index
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (r0_1 + 8*x0), xmask, other=0.0).to(tl.float32)
tmp1 = tmp0.to(tl.float32)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, R0_BLOCK])
tmp4 = tl.where(xmask, tmp2, float("-inf"))
tmp5 = triton_helpers.max2(tmp4, 1)[:, None]
tmp6 = tmp1 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tl.broadcast_to(tmp7, [XBLOCK, R0_BLOCK])
tmp10 = tl.where(xmask, tmp8, 0)
tmp11 = tl.sum(tmp10, 1)[:, None]
tmp12 = tmp7 / tmp11
tmp13 = tmp12.to(tl.float32)
tl.store(in_out_ptr0 + (r0_1 + 8*x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: /tmp/torchinductor_cdhernandez/rv/crvip4xdyk7nmm26bsfefmqcvz7pxdmepfqg6axnhvee2qetk6sm.py
# Topologically Sorted Source Nodes: [sum_1, expert_weights_2], Original ATen: [aten.sum, aten.div]
# Source node to ATen node mapping:
# expert_weights_2 => div_1
# sum_1 => sum_2
# Graph fragment:
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%getitem, [-1], True), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%getitem, %sum_2), kwargs = {})
triton_poi_fused_div_sum_1 = async_compile.triton('triton_poi_fused_div_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, DeviceProperties
triton_helpers.set_driver_to_gpu()
@triton_heuristics.pointwise(
size_hints={'x': 16},
filename=__file__,
triton_meta={'signature': {'in_ptr0': '*bf16', 'out_ptr0': '*bf16', 'xnumel': 'i32'}, 'device': DeviceProperties(type='cuda', index=0, multi_processor_count=132, cc=90, major=9, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, warp_size=32), 'constants': {}, 'configs': [AttrsDescriptor.from_dict({'arg_properties': {'tt.divisibility': (0, 1), 'tt.equal_to': ()}, 'cls': 'AttrsDescriptor'})]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_sum_1', 'mutated_arg_names': [], 'optimize_mem': True, 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'CE326918D8FE98067798167ABD26A2E4EDFD110D9ECD4380441C64512F6D164E', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_sum_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 2
tmp0 = tl.load(in_ptr0 + (x2), xmask).to(tl.float32)
tmp1 = tl.load(in_ptr0 + (2*x1), xmask, eviction_policy='evict_last').to(tl.float32)
tmp2 = tl.load(in_ptr0 + (1 + 2*x1), xmask, eviction_policy='evict_last').to(tl.float32)
tmp3 = tmp1 + tmp2
tmp4 = tmp0 / tmp3
tl.store(out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (1, 6, 4), (24, 4, 1))
assert_size_stride(arg1_1, (8, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((6, 8), (8, 1), torch.bfloat16)
# Topologically Sorted Source Nodes: [scores], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(arg0_1, (6, 4), (4, 1), 0), reinterpret_tensor(arg1_1, (4, 8), (1, 4), 0), out=buf0)
del arg1_1
buf3 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [expert_weights], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_per_fused__softmax_0.run(buf3, 6, 8, grid=grid(6), stream=stream0)
# Topologically Sorted Source Nodes: [expert_weights, topk], Original ATen: [aten._softmax, aten.topk]
buf4 = torch.ops.aten.topk.default(buf3, 2)
del buf3
buf5 = buf4[0]
assert_size_stride(buf5, (6, 2), (2, 1))
buf6 = buf4[1]
assert_size_stride(buf6, (6, 2), (2, 1))
del buf4
buf7 = empty_strided_cuda((6, 2), (2, 1), torch.bfloat16)
# Topologically Sorted Source Nodes: [sum_1, expert_weights_2], Original ATen: [aten.sum, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_div_sum_1.run(buf5, buf7, 12, grid=grid(12), stream=stream0)
del buf5
return (reinterpret_tensor(arg0_1, (6, 4), (4, 1), 0), buf6, buf7, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((1, 6, 4), (24, 4, 1), device='cuda:0', dtype=torch.bfloat16)
arg1_1 = rand_strided((8, 4), (4, 1), device='cuda:0', dtype=torch.bfloat16)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
V0320 11:46:31.365000 18434 site-packages/torch/_dynamo/utils.py:1821] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "4f81fc7f126c6ed9e20a1d010b82a8ff"}
{
"name": "fx_graph_cache_hit",
"ts": 1742496391335333.2,
"args": {
"triton_bundler_meta": "TritonBundlerMetadata(cached_kernel_names=[])",
"key": "ffonikg3i6odlzyyko7li37lboivis5j2xfma222ff7zwgeiqx6f",
"components": [
"[7mm2sfcyb7ag7il5aa5xcoxu4ci4mxpmu5zbgn5ngyscgjnsx4i] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1, arg1_1):\n view = torch.ops.aten.view.default(arg0_1, [-1, 4]); arg0_1 = None\n permute = torch.ops.aten.permute.default(arg1_1, [1, 0]); arg1_1 = None\n mm = torch.ops.aten.mm.default(view, permute); permute = None\n convert_element_type_2 = torch.ops.prims.convert_element_type.default(mm, torch.float32); mm = None\n amax = torch.ops.aten.amax.default(convert_element_type_2, [-1], True)\n sub = torch.ops.aten.sub.Tensor(convert_element_type_2, amax); convert_element_type_2 = amax = None\n exp = torch.ops.aten.exp.default(sub); sub = None\n sum_1 = torch.ops.aten.sum.dim_IntList(exp, [-1], True)\n div = torch.ops.aten.div.Tensor(exp, sum_1); exp = sum_1 = None\n convert_element_type_3 = torch.ops.prims.convert_element_type.default(div, torch.bfloat16); div = None\n topk = torch.ops.aten.topk.default(convert_element_type_3, 2); convert_element_type_3 = None\n getitem = topk[0]\n getitem_1 = topk[1]; topk = None\n sum_2 = torch.ops.aten.sum.dim_IntList(getitem, [-1], True)\n div_1 = torch.ops.aten.div.Tensor(getitem, sum_2); getitem = sum_2 = None\n return (view, getitem_1, div_1)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[qxar3fbb6cbebfcmf4v7wtn7r7d6kbjfoxp7xzr6kuf6rewbefy] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([1, 6, 4]), stride=(24, 4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[4pipazss7m3tbrpbeuptmx2ccqordlcix2mmbf7rfc62mqdffei] example_inputs[1]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([8, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=64, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
"[dmwmlxgi7iucpywnwyvxk44cggr2qactwqrdd5h3tamispdqeyz] fx_kwargs[static_input_idxs]: [1]",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_event_time": 1742496391364786949,
"cache_state": "hit",
"time_saved_ns": 297465102,
"compile_id": "0/0"
},
"ph": "i",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0,
"s": "p"
}
V0320 11:46:31.365000 18434 site-packages/torch/_inductor/compile_fx.py:827] {"artifact": {"name": "fx_graph_cache_hit", "encoding": "json"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "e972c1b3c0f30234679b995ef5b54b5f"}
{"triton_bundler_meta": "TritonBundlerMetadata(cached_kernel_names=[])", "key": "ffonikg3i6odlzyyko7li37lboivis5j2xfma222ff7zwgeiqx6f", "components": ["[7mm2sfcyb7ag7il5aa5xcoxu4ci4mxpmu5zbgn5ngyscgjnsx4i] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1, arg1_1):\n view = torch.ops.aten.view.default(arg0_1, [-1, 4]); arg0_1 = None\n permute = torch.ops.aten.permute.default(arg1_1, [1, 0]); arg1_1 = None\n mm = torch.ops.aten.mm.default(view, permute); permute = None\n convert_element_type_2 = torch.ops.prims.convert_element_type.default(mm, torch.float32); mm = None\n amax = torch.ops.aten.amax.default(convert_element_type_2, [-1], True)\n sub = torch.ops.aten.sub.Tensor(convert_element_type_2, amax); convert_element_type_2 = amax = None\n exp = torch.ops.aten.exp.default(sub); sub = None\n sum_1 = torch.ops.aten.sum.dim_IntList(exp, [-1], True)\n div = torch.ops.aten.div.Tensor(exp, sum_1); exp = sum_1 = None\n convert_element_type_3 = torch.ops.prims.convert_element_type.default(div, torch.bfloat16); div = None\n topk = torch.ops.aten.topk.default(convert_element_type_3, 2); convert_element_type_3 = None\n getitem = topk[0]\n getitem_1 = topk[1]; topk = None\n sum_2 = torch.ops.aten.sum.dim_IntList(getitem, [-1], True)\n div_1 = torch.ops.aten.div.Tensor(getitem, sum_2); getitem = sum_2 = None\n return (view, getitem_1, div_1)\n \n# To see more debug info, please use `graph_module.print_readable()`", "[qxar3fbb6cbebfcmf4v7wtn7r7d6kbjfoxp7xzr6kuf6rewbefy] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([1, 6, 4]), stride=(24, 4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[4pipazss7m3tbrpbeuptmx2ccqordlcix2mmbf7rfc62mqdffei] example_inputs[1]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([8, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=64, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ", "[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False", "[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None", "[dmwmlxgi7iucpywnwyvxk44cggr2qactwqrdd5h3tamispdqeyz] fx_kwargs[static_input_idxs]: [1]", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0", "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)", "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)", "[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>", "[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}", "[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}", "[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False", "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True", "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}", "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False", "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True", "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default", "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300", "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False", "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP", "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True", "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False", "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0", "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0", "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ", "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT", "[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False", "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4", "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1", "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128", "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False", "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None", "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096", "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False", "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128", "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True", "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None", "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None", "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []", "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']", "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False", "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp", "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton", "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host", "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda", "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021", "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"], "cache_event_time": 1742496391364786949, "cache_state": "hit", "time_saved_ns": 297465102, "compile_id": "0/0"}
V0320 11:46:31.368000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "73ceaf533d10429882ffcbd3636cff96"}
{
"name": "inductor_compile",
"ts": 1742496391367857.5,
"args": {
"fn_name": "compile_fx_inner",
"compile_id": "0/0",
"is_backward": false,
"cached_kernel_names": [],
"cache_state": "hit",
"cache_event_time": 1742496391335333336,
"key": "ffonikg3i6odlzyyko7li37lboivis5j2xfma222ff7zwgeiqx6f",
"components": [
"[7mm2sfcyb7ag7il5aa5xcoxu4ci4mxpmu5zbgn5ngyscgjnsx4i] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1, arg1_1):\n view = torch.ops.aten.view.default(arg0_1, [-1, 4]); arg0_1 = None\n permute = torch.ops.aten.permute.default(arg1_1, [1, 0]); arg1_1 = None\n mm = torch.ops.aten.mm.default(view, permute); permute = None\n convert_element_type_2 = torch.ops.prims.convert_element_type.default(mm, torch.float32); mm = None\n amax = torch.ops.aten.amax.default(convert_element_type_2, [-1], True)\n sub = torch.ops.aten.sub.Tensor(convert_element_type_2, amax); convert_element_type_2 = amax = None\n exp = torch.ops.aten.exp.default(sub); sub = None\n sum_1 = torch.ops.aten.sum.dim_IntList(exp, [-1], True)\n div = torch.ops.aten.div.Tensor(exp, sum_1); exp = sum_1 = None\n convert_element_type_3 = torch.ops.prims.convert_element_type.default(div, torch.bfloat16); div = None\n topk = torch.ops.aten.topk.default(convert_element_type_3, 2); convert_element_type_3 = None\n getitem = topk[0]\n getitem_1 = topk[1]; topk = None\n sum_2 = torch.ops.aten.sum.dim_IntList(getitem, [-1], True)\n div_1 = torch.ops.aten.div.Tensor(getitem, sum_2); getitem = sum_2 = None\n return (view, getitem_1, div_1)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[qxar3fbb6cbebfcmf4v7wtn7r7d6kbjfoxp7xzr6kuf6rewbefy] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([1, 6, 4]), stride=(24, 4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[4pipazss7m3tbrpbeuptmx2ccqordlcix2mmbf7rfc62mqdffei] example_inputs[1]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([8, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=64, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
"[dmwmlxgi7iucpywnwyvxk44cggr2qactwqrdd5h3tamispdqeyz] fx_kwargs[static_input_idxs]: [1]",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_bypass_reason": null,
"remote_cache_enabled": false,
"local_cache_enabled": true
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.368000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "cbc9a940b40031fa28a32b31cce13695"}
{
"name": "compile_fx.<locals>.fw_compiler_base",
"ts": 1742496391368795.0,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.371000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "daca98c30c4c13348b7e83fd0f1e7a11"}
{
"name": "create_aot_dispatcher_function",
"ts": 1742496391371120.2,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.371000 18434 site-packages/torch/_dynamo/utils.py:1821] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "b1254a0d42edbe9c97be6f94bdca82cf"}
{
"name": "autograd_cache_miss",
"ts": 1742496391040462.0,
"args": {
"key": "adxlfdccj5ayvstlbwcclfh3fhyirlxqnwzm4tpud5c3slnoxtuv",
"cache_state": "miss",
"components": [
"[grknehpvthkvpyli527v2gv7a2bns6jdzqdq3gkpuxsbhnvuv2m] aot_config: (0, True, False, False, False, [LocalSource(local_name='x', is_input=True, is_derefed_cell_contents=False), DictGetItemSource(base=UnspecializedBuiltinNNModuleSource(base=UnspecializedParamBufferSource(base=DictGetItemSource(base=UnspecializedNNModuleSource(base=AttrSource(base=LocalSource(local_name='self', is_input=True, is_derefed_cell_contents=False), member='_modules')), index='gate'), member='_parameters')), index='weight')], True, False)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False",
"[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>",
"[qdeea6rqsd6dqmumcynajmwho5ywmyykchgc3ipwukdnbd77gvn] gm: GraphModule()\n\n\n\ndef forward(self, L_x_ : torch.Tensor, L_self_modules_gate_parameters_weight_ : torch.nn.parameter.Parameter):\n l_x_ = L_x_\n l_self_modules_gate_parameters_weight_ = L_self_modules_gate_parameters_weight_\n x = l_x_.view(-1, 4); l_x_ = None\n scores = torch._C._nn.linear(x, l_self_modules_gate_parameters_weight_, None); l_self_modules_gate_parameters_weight_ = None\n expert_weights = torch.nn.functional.softmax(scores, dim = -1); scores = None\n topk = torch.topk(expert_weights, 2, dim = -1); expert_weights = None\n expert_weights_1 = topk[0]\n expert_indices = topk[1]; topk = None\n sum_1 = expert_weights_1.sum(dim = -1, keepdim = True)\n expert_weights_1 /= sum_1; expert_weights_2 = expert_weights_1; expert_weights_1 = sum_1 = None\n return (x, expert_indices, expert_weights_2)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[qxar3fbb6cbebfcmf4v7wtn7r7d6kbjfoxp7xzr6kuf6rewbefy] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([1, 6, 4]), stride=(24, 4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[phtbnuvssvvwlgi7rvholonmhqlieu6msxincv5u3o2uax7okez] example_inputs[1]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([8, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=True, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"compile_id": "0/0"
},
"ph": "i",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0,
"s": "p"
}
V0320 11:46:31.372000 18434 site-packages/torch/_functorch/_aot_autograd/autograd_cache.py:804] {"artifact": {"name": "aotautograd_cache_miss", "encoding": "json"}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "a02e322727c5a5e016477dce48933443"}
{"key": "adxlfdccj5ayvstlbwcclfh3fhyirlxqnwzm4tpud5c3slnoxtuv", "cache_state": "miss", "components": ["[grknehpvthkvpyli527v2gv7a2bns6jdzqdq3gkpuxsbhnvuv2m] aot_config: (0, True, False, False, False, [LocalSource(local_name='x', is_input=True, is_derefed_cell_contents=False), DictGetItemSource(base=UnspecializedBuiltinNNModuleSource(base=UnspecializedParamBufferSource(base=DictGetItemSource(base=UnspecializedNNModuleSource(base=AttrSource(base=LocalSource(local_name='self', is_input=True, is_derefed_cell_contents=False), member='_modules')), index='gate'), member='_parameters')), index='weight')], True, False)", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False", "[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>", "[qdeea6rqsd6dqmumcynajmwho5ywmyykchgc3ipwukdnbd77gvn] gm: GraphModule()\n\n\n\ndef forward(self, L_x_ : torch.Tensor, L_self_modules_gate_parameters_weight_ : torch.nn.parameter.Parameter):\n l_x_ = L_x_\n l_self_modules_gate_parameters_weight_ = L_self_modules_gate_parameters_weight_\n x = l_x_.view(-1, 4); l_x_ = None\n scores = torch._C._nn.linear(x, l_self_modules_gate_parameters_weight_, None); l_self_modules_gate_parameters_weight_ = None\n expert_weights = torch.nn.functional.softmax(scores, dim = -1); scores = None\n topk = torch.topk(expert_weights, 2, dim = -1); expert_weights = None\n expert_weights_1 = topk[0]\n expert_indices = topk[1]; topk = None\n sum_1 = expert_weights_1.sum(dim = -1, keepdim = True)\n expert_weights_1 /= sum_1; expert_weights_2 = expert_weights_1; expert_weights_1 = sum_1 = None\n return (x, expert_indices, expert_weights_2)\n \n# To see more debug info, please use `graph_module.print_readable()`", "[qxar3fbb6cbebfcmf4v7wtn7r7d6kbjfoxp7xzr6kuf6rewbefy] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([1, 6, 4]), stride=(24, 4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[phtbnuvssvvwlgi7rvholonmhqlieu6msxincv5u3o2uax7okez] example_inputs[1]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([8, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=True, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ", "[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)", "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)", "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)", "[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>", "[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}", "[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}", "[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False", "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True", "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}", "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False", "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True", "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default", "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300", "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False", "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP", "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True", "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False", "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0", "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0", "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ", "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT", "[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False", "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4", "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1", "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128", "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False", "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None", "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096", "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False", "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128", "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True", "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None", "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None", "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []", "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']", "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False", "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp", "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton", "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host", "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda", "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021", "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"], "compile_id": "0/0"}
V0320 11:46:31.372000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "e337c7e213408c414d20296a0e62305a"}
{
"name": "backend_compile",
"ts": 1742496391372671.2,
"args": {
"fn_name": "OutputGraph.call_user_compiler",
"compile_id": "0/0",
"requires_subclass_dispatch": false,
"dispatch_mode": "inference",
"cache_state": "miss",
"cache_event_time": 1742496391040462053,
"key": "adxlfdccj5ayvstlbwcclfh3fhyirlxqnwzm4tpud5c3slnoxtuv",
"components": [
"[grknehpvthkvpyli527v2gv7a2bns6jdzqdq3gkpuxsbhnvuv2m] aot_config: (0, True, False, False, False, [LocalSource(local_name='x', is_input=True, is_derefed_cell_contents=False), DictGetItemSource(base=UnspecializedBuiltinNNModuleSource(base=UnspecializedParamBufferSource(base=DictGetItemSource(base=UnspecializedNNModuleSource(base=AttrSource(base=LocalSource(local_name='self', is_input=True, is_derefed_cell_contents=False), member='_modules')), index='gate'), member='_parameters')), index='weight')], True, False)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False",
"[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>",
"[qdeea6rqsd6dqmumcynajmwho5ywmyykchgc3ipwukdnbd77gvn] gm: GraphModule()\n\n\n\ndef forward(self, L_x_ : torch.Tensor, L_self_modules_gate_parameters_weight_ : torch.nn.parameter.Parameter):\n l_x_ = L_x_\n l_self_modules_gate_parameters_weight_ = L_self_modules_gate_parameters_weight_\n x = l_x_.view(-1, 4); l_x_ = None\n scores = torch._C._nn.linear(x, l_self_modules_gate_parameters_weight_, None); l_self_modules_gate_parameters_weight_ = None\n expert_weights = torch.nn.functional.softmax(scores, dim = -1); scores = None\n topk = torch.topk(expert_weights, 2, dim = -1); expert_weights = None\n expert_weights_1 = topk[0]\n expert_indices = topk[1]; topk = None\n sum_1 = expert_weights_1.sum(dim = -1, keepdim = True)\n expert_weights_1 /= sum_1; expert_weights_2 = expert_weights_1; expert_weights_1 = sum_1 = None\n return (x, expert_indices, expert_weights_2)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[qxar3fbb6cbebfcmf4v7wtn7r7d6kbjfoxp7xzr6kuf6rewbefy] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([1, 6, 4]), stride=(24, 4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[phtbnuvssvvwlgi7rvholonmhqlieu6msxincv5u3o2uax7okez] example_inputs[1]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([8, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=True, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_bypass_reason": null,
"remote_cache_enabled": false,
"local_cache_enabled": true
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.379000 18434 site-packages/torch/_dynamo/guards.py:2537] {"dynamo_cpp_guards_str": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "8f1e3b1bcb628b703f29caca7ce393d6"}
TREE_GUARD_MANAGER:
+- RootGuardManager
| +- DEFAULT_DEVICE: utils_device.CURRENT_DEVICE == None # _dynamo/output_graph.py:519 in init_ambient_guards
| +- GLOBAL_STATE: ___check_global_state()
| +- TORCH_FUNCTION_MODE_STACK: ___check_torch_function_mode_stack()
| +- GuardManager: source=L['x'], accessed_by=FrameLocalsGuardAccessor(key='x', framelocals_idx=1)
| | +- TENSOR_MATCH: check_tensor(L['x'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.bfloat16, device=0, requires_grad=False, size=[1, 6, 4], stride=[24, 4, 1])
| | +- NO_HASATTR: hasattr(L['x'], '_dynamo_dynamic_indices') == False
| | +- GuardManager: source=L['x'].view, accessed_by=GetAttrGuardAccessor(view)
| +- GuardManager: source=L['self'], accessed_by=FrameLocalsGuardAccessor(key='self', framelocals_idx=0)
| | +- TYPE_MATCH: ___check_type_id(L['self'], 97578784)
| | +- GuardManager: source=L['self'].__dict__, accessed_by=GetGenericDictGuardAccessor
| | | +- GuardManager: source=L['self'].dim, accessed_by=DictGetItemGuardAccessor('dim')
| | | | +- EQUALS_MATCH: L['self'].dim == 4
| | | +- GuardManager: source=L['self']._modules, accessed_by=DictGetItemGuardAccessor('_modules')
| | | | +- TYPE_MATCH: ___check_type_id(L['self']._modules, 9693856)
| | | | +- GuardManager: source=L['self']._modules['gate'], accessed_by=DictGetItemGuardAccessor('gate')
| | | | | +- TYPE_MATCH: ___check_type_id(L['self']._modules['gate'], 87672272)
| | | | | +- GuardManager: source=L['self']._modules['gate'].__dict__, accessed_by=GetGenericDictGuardAccessor
| | | | | | +- DICT_CONTAINS: not ___dict_contains('forward', L['self']._modules['gate'].__dict__)
| | | | | | +- GuardManager: source=L['self']._modules['gate']._parameters, accessed_by=DictGetItemGuardAccessor('_parameters')
| | | | | | | +- TYPE_MATCH: ___check_type_id(L['self']._modules['gate']._parameters, 9693856)
| | | | | | | +- GuardManager: source=L['self']._modules['gate']._parameters['bias'], accessed_by=DictGetItemGuardAccessor('bias')
| | | | | | | | +- ID_MATCH: ___check_obj_id(L['self']._modules['gate']._parameters['bias'], 9677920)
| | | | | | | +- GuardManager: source=L['self']._modules['gate']._parameters['weight'], accessed_by=DictGetItemGuardAccessor('weight')
| | | | | | | | +- TENSOR_MATCH: check_tensor(L['self']._modules['gate']._parameters['weight'], Parameter, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.bfloat16, device=0, requires_grad=True, size=[8, 4], stride=[4, 1])
| | | | +- GuardManager: source=L['self']._modules['cond_ffn'], accessed_by=DictGetItemGuardAccessor('cond_ffn')
| | | | | +- TYPE_MATCH: ___check_type_id(L['self']._modules['cond_ffn'], 11695168)
| | | +- GuardManager: source=L['self']._parameters, accessed_by=DictGetItemGuardAccessor('_parameters')
| | | | +- TYPE_MATCH: ___check_type_id(L['self']._parameters, 9693856)
| | | +- GuardManager: source=L['self'].num_activated_experts, accessed_by=DictGetItemGuardAccessor('num_activated_experts')
| | | | +- EQUALS_MATCH: L['self'].num_activated_experts == 2
| +- GuardManager: source=G, accessed_by=GlobalsGuardAccessor
| | +- GuardManager: source=G['torch'], accessed_by=DictGetItemGuardAccessor('torch')
| | | +- ID_MATCH: ___check_obj_id(G['torch'], 140455470383408)
| | | +- GuardManager: source=G['torch'].topk, accessed_by=GetAttrGuardAccessor(topk)
| | | | +- ID_MATCH: ___check_obj_id(G['torch'].topk, 140455466204304)
| | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_linear'], accessed_by=DictGetItemGuardAccessor('__import_torch_dot_nn_dot_modules_dot_linear')
| | | +- ID_MATCH: ___check_obj_id(G['__import_torch_dot_nn_dot_modules_dot_linear'], 140452002807568)
| | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_linear'].F, accessed_by=GetAttrGuardAccessor(F)
| | | | +- ID_MATCH: ___check_obj_id(G['__import_torch_dot_nn_dot_modules_dot_linear'].F, 140452002807728)
| | | | +- OBJECT_ALIASING: G['__import_torch_dot_nn_dot_modules_dot_linear'].F is G['F']
| | | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_linear'].F.linear, accessed_by=GetAttrGuardAccessor(linear)
| | | | | +- ID_MATCH: ___check_obj_id(G['__import_torch_dot_nn_dot_modules_dot_linear'].F.linear, 140453682498432)
| | | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_linear'].F.softmax, accessed_by=GetAttrGuardAccessor(softmax)
| | | | | +- ID_MATCH: ___check_obj_id(G['__import_torch_dot_nn_dot_modules_dot_linear'].F.softmax, 140449307253536)
| | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_module'], accessed_by=DictGetItemGuardAccessor('__import_torch_dot_nn_dot_modules_dot_module')
| | | +- ID_MATCH: ___check_obj_id(G['__import_torch_dot_nn_dot_modules_dot_module'], 140452866578352)
| | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_module']._global_forward_hooks, accessed_by=GetAttrGuardAccessor(_global_forward_hooks)
| | | | +- TYPE_MATCH: ___check_type_id(G['__import_torch_dot_nn_dot_modules_dot_module']._global_forward_hooks, 9685856)
| | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_module']._global_backward_hooks, accessed_by=GetAttrGuardAccessor(_global_backward_hooks)
| | | | +- TYPE_MATCH: ___check_type_id(G['__import_torch_dot_nn_dot_modules_dot_module']._global_backward_hooks, 9685856)
| | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_module']._global_forward_pre_hooks, accessed_by=GetAttrGuardAccessor(_global_forward_pre_hooks)
| | | | +- TYPE_MATCH: ___check_type_id(G['__import_torch_dot_nn_dot_modules_dot_module']._global_forward_pre_hooks, 9685856)
| | | +- GuardManager: source=G['__import_torch_dot_nn_dot_modules_dot_module']._global_backward_pre_hooks, accessed_by=GetAttrGuardAccessor(_global_backward_pre_hooks)
| | | | +- TYPE_MATCH: ___check_type_id(G['__import_torch_dot_nn_dot_modules_dot_module']._global_backward_pre_hooks, 9685856)
| | +- GuardManager: source=G['F'], accessed_by=DictGetItemGuardAccessor('F')
| | | +- OBJECT_ALIASING: G['__import_torch_dot_nn_dot_modules_dot_linear'].F is G['F']
Guard latency = 0.93 us
V0320 11:46:31.380000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "9c729b89ce1690af62b49b22592c895f"}
{
"name": "entire_frame_compile",
"ts": 1742496391380370.2,
"args": {
"fn_name": "_compile.compile_inner",
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.380000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "e280150ced9c12cbca4c5d21852ea24a"}
{
"name": "gc",
"ts": 1742496391380700.8,
"args": {
"compile_id": "0/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.381000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "91a5275f358c0e56e07d2c743bb98350"}
{
"name": "gc",
"ts": 1742496391381516.0,
"args": {
"compile_id": "0/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.383000 18434 site-packages/torch/_dynamo/utils.py:1483] {"compilation_metrics": {"compile_id": "0/0", "frame_key": "1", "co_name": "forward", "co_filename": "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py", "co_firstlineno": 24, "cache_size": 0, "accumulated_cache_size": 0, "guard_count": 34, "shape_env_guard_count": 0, "graph_op_count": 8, "graph_node_count": 11, "graph_input_count": 2, "start_time": 1742496390.704986, "entire_frame_compile_time_s": 0.673622, "backend_compile_time_s": 0.616846, "inductor_compile_time_s": 0.033665, "code_gen_time_s": null, "fail_type": null, "fail_reason": null, "fail_user_frame_filename": null, "fail_user_frame_lineno": null, "non_compliant_ops": [], "compliant_custom_ops": [], "restart_reasons": ["Dynamic slicing with Tensor arguments\n Explanation: Creating slices with Tensor arguments is not supported. e.g. `l[:x]`, where `x` is a 1-element tensor.\n Hint: It may be possible to write Dynamo tracing rules for this code. Please report an issue to PyTorch if you encounter this graph break often and it is causing performance issues.\n\n Developer debug context: SliceVariable start: TensorVariable(), stop: TensorVariable(), step: ConstantVariable(NoneType: None)\n"], "dynamo_time_before_restart_s": 0.038427114486694336, "has_guarded_code": true, "remote_cache_time_saved_s": 0.297465, "structured_logging_overhead_s": 0.024049, "config_suppress_errors": false, "config_inline_inbuilt_nn_modules": true, "specialize_float": false, "dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": true, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}", "is_forward": true, "num_triton_bundles": null, "remote_fx_graph_cache_get_time_ms": null, "remote_fx_graph_cache_put_time_ms": null, "start_time_us": 1742496390704986, "duration_us": 676871, "dynamo_cumulative_compile_time_us": 673622, "aot_autograd_cumulative_compile_time_us": 616846, "inductor_cumulative_compile_time_us": 33665, "inductor_code_gen_cumulative_compile_time_us": null, "triton_compile_time_us": 17232, "runtime_cudagraphify_time_us": null, "runtime_triton_autotune_time_us": null, "dynamo_compile_time_before_restart_us": 38427, "cuda_synchronize_time_us": null, "distributed_ephemeral_timeout_us": 297465, "structured_logging_overhead_us": 24049, "remote_fx_graph_cache_get_time_us": null, "remote_fx_graph_cache_put_time_us": null, "backward_cumulative_compile_time_us": null, "end_time_us": 1742496391381857, "pre_grad_pass_time_us": 2781, "post_grad_pass_time_us": null, "joint_graph_pass_time_us": 267972, "log_format_version": 3, "inductor_config": "{\"TYPE_CHECKING\": false, \"_cache_config_ignore_prefix\": [\"trace\", \"cuda.cutlass_dir\", \"worker_start_method\", \"compile_threads\", \"post_grad_custom_post_pass\", \"post_grad_custom_pre_pass\", \"always_complex_memory_overlap_TESTING_ONLY\"], \"_collective.auto_select\": false, \"_collective.one_shot_all_reduce_threshold_bytes\": 131072, \"_fuse_ddp_bucket_size\": 25, \"_fuse_ddp_communication\": false, \"_fuse_ddp_communication_passes\": [\"fuse_ddp_with_concat_op\", \"schedule_comm_wait\"], \"_micro_pipeline_tp\": false, \"_pre_fusion_custom_pass\": null, \"_profile_var\": \"\", \"_raise_error_for_testing\": false, \"_save_config_ignore\": [\"trace.upload_tar\", \"joint_custom_pre_pass\", \"joint_custom_post_pass\", \"pre_grad_custom_pass\", \"aot_inductor.repro_level\", \"aot_inductor.dump_aoti_minifier\"], \"add_pre_grad_passes\": null, \"aggressive_fusion\": false, \"allow_buffer_reuse\": true, \"always_complex_memory_overlap_TESTING_ONLY\": false, \"always_keep_tensor_constants\": false, \"annotate_training\": false, \"aot_inductor.allow_stack_allocation\": false, \"aot_inductor.compile_wrapper_with_O0\": false, \"aot_inductor.debug_compile\": false, \"aot_inductor.debug_intermediate_value_printer\": \"0\", \"aot_inductor.dump_aoti_minifier\": false, \"aot_inductor.filtered_kernel_names\": null, \"aot_inductor.force_mmap_weights\": false, \"aot_inductor.metadata\": {}, \"aot_inductor.output_path\": \"\", \"aot_inductor.package\": false, \"aot_inductor.package_constants_in_so\": true, \"aot_inductor.package_cpp_only\": false, \"aot_inductor.presets\": {}, \"aot_inductor.raise_error_on_ignored_optimization\": true, \"aot_inductor.repro_level\": 2, \"aot_inductor.serialized_in_spec\": \"\", \"aot_inductor.serialized_out_spec\": \"\", \"aot_inductor.use_minimal_arrayref_interface\": false, \"aot_inductor.use_runtime_constant_folding\": false, \"assert_indirect_indexing\": true, \"assume_aligned_inputs\": false, \"autoheuristic_collect\": \"\", \"autoheuristic_log_path\": \"DEFAULT\", \"autoheuristic_use\": \"mixed_mm\", \"autotune_fallback_to_aten\": true, \"autotune_in_subproc\": false, \"autotune_local_cache\": true, \"autotune_multi_device\": false, \"autotune_num_choices_displayed\": 10, \"autotune_remote_cache\": null, \"b2b_gemm_pass\": false, \"batch_fusion\": true, \"benchmark_combo_kernel\": false, \"benchmark_epilogue_fusion\": true, \"benchmark_fusion\": false, \"benchmark_harness\": true, \"benchmark_kernel\": false, \"bundle_triton_into_fx_graph_cache\": true, \"bundled_autotune_remote_cache\": null, \"bw_outputs_user_visible\": true, \"can_inplace_pad_graph_input\": false, \"check_stack_no_cycles_TESTING_ONLY\": false, \"combo_kernel_allow_mixed_sizes\": 1, \"combo_kernel_foreach_dynamic_shapes\": false, \"combo_kernels\": false, \"combo_kernels_autotune\": 1, \"comment_origin\": false, \"compile_threads\": 32, \"comprehensive_padding\": true, \"compute_all_bounds\": false, \"constant_and_index_propagation\": true, \"conv_1x1_as_mm\": false, \"coordinate_descent_check_all_directions\": false, \"coordinate_descent_search_radius\": 1, \"coordinate_descent_tuning\": false, \"cpp.cxx\": [null, \"g++\"], \"cpp.descriptive_names\": \"original_aten\", \"cpp.dynamic_threads\": false, \"cpp.enable_concat_linear\": false, \"cpp.enable_floating_point_contract_flag\": \"off\", \"cpp.enable_grouped_gemm_template\": false, \"cpp.enable_kernel_profile\": false, \"cpp.enable_loop_tail_vec\": true, \"cpp.enable_tiling_heuristics\": true, \"cpp.enable_unsafe_math_opt_flag\": false, \"cpp.fallback_scatter_reduce_sum\": true, \"cpp.gemm_cache_blocking\": null, \"cpp.gemm_max_k_slices\": 1, \"cpp.gemm_thread_factors\": null, \"cpp.inject_log1p_bug_TESTING_ONLY\": null, \"cpp.inject_relu_bug_TESTING_ONLY\": null, \"cpp.max_horizontal_fusion_size\": 16, \"cpp.min_chunk_size\": 4096, \"cpp.no_redundant_loops\": true, \"cpp.simdlen\": null, \"cpp.threads\": -1, \"cpp.vec_isa_ok\": null, \"cpp.weight_prepack\": true, \"cpp_wrapper\": false, \"cpu_backend\": \"cpp\", \"cuda.arch\": null, \"cuda.compile_opt_level\": \"-O1\", \"cuda.cuda_cxx\": null, \"cuda.cutlass_backend_min_gemm_size\": 1, \"cuda.cutlass_dir\": \"/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/third_party/cutlass\", \"cuda.cutlass_instantiation_level\": \"0\", \"cuda.cutlass_max_profiling_configs\": null, \"cuda.cutlass_max_profiling_swizzle_options\": [1, 2, 4], \"cuda.cutlass_op_allowlist_regex\": null, \"cuda.cutlass_op_denylist_regex\": null, \"cuda.enable_cuda_lto\": false, \"cuda.enable_debug_info\": false, \"cuda.enable_ptxas_info\": false, \"cuda.generate_test_runner\": false, \"cuda.use_fast_math\": false, \"cuda.version\": null, \"cuda_backend\": \"triton\", \"custom_op_default_layout_constraint\": \"needs_fixed_stride_order\", \"dce\": false, \"debug\": false, \"debug_fusion\": false, \"debug_index_asserts\": false, \"debug_ir_traceback\": false, \"decompose_mem_bound_mm\": false, \"developer_warnings\": true, \"disable_cpp_codegen\": false, \"disable_padding_cpu\": true, \"disable_progress\": true, \"dynamic_scale_rblock\": true, \"efficient_conv_bn_eval_fx_passes\": false, \"emulate_precision_casts\": false, \"enable_auto_functionalized_v2\": true, \"enable_linear_binary_folding\": false, \"enabled_metric_tables\": \"\", \"epilogue_fusion\": true, \"epilogue_fusion_first\": false, \"estimate_op_runtime\": \"default\", \"external_matmul\": [], \"fallback_random\": false, \"force_disable_caches\": false, \"force_fuse_int_mm_with_mul\": false, \"force_layout_optimization\": false, \"force_pointwise_cat\": false, \"force_same_precision\": false, \"force_shape_pad\": false, \"freezing\": false, \"freezing_discard_parameters\": false, \"fx_graph_cache\": true, \"fx_graph_remote_cache\": null, \"fx_passes_numeric_check\": {\"num_iterations\": 1, \"pre_grad\": false, \"precision\": 0.0001, \"requires_optimizer\": true}, \"generate_intermediate_hooks\": false, \"global_cache_dir\": null, \"graph_partition\": false, \"group_fusion\": false, \"halide.asserts\": false, \"halide.cpu_target\": \"host\", \"halide.debug\": false, \"halide.gpu_target\": \"host-cuda\", \"halide.scan_kernels\": false, \"halide.scheduler_cpu\": \"Adams2019\", \"halide.scheduler_cuda\": \"Anderson2021\", \"implicit_fallbacks\": true, \"inplace_buffers\": true, \"inplace_padding\": true, \"inter_node_bw\": 25, \"intra_node_bw\": 300, \"is_nightly_or_source\": true, \"is_predispatch\": false, \"joint_custom_post_pass\": null, \"joint_custom_pre_pass\": null, \"joint_graph_constant_folding\": true, \"keep_output_stride\": true, \"kernel_name_max_ops\": 10, \"layout_opt_default\": \"1\", \"layout_optimization\": true, \"loop_ordering_after_fusion\": false, \"max_autotune\": false, \"max_autotune_conv_backends\": \"ATEN,TRITON\", \"max_autotune_gemm\": false, \"max_autotune_gemm_backends\": \"ATEN,TRITON,CPP\", \"max_autotune_gemm_search_space\": \"DEFAULT\", \"max_autotune_pointwise\": false, \"max_autotune_subproc_graceful_timeout_seconds\": 1.0, \"max_autotune_subproc_result_timeout_seconds\": 60.0, \"max_autotune_subproc_terminate_timeout_seconds\": 2.0, \"max_epilogue_benchmarked_choices\": 1, \"max_fusion_size\": 64, \"max_pointwise_cat_inputs\": 8, \"memory_planning\": false, \"memory_pool\": \"intermediates\", \"mixed_mm_choice\": \"heuristic\", \"nan_asserts\": false, \"optimize_scatter_upon_const_tensor\": true, \"pad_channels_last\": false, \"pad_outputs\": false, \"padding_alignment_bytes\": 128, \"padding_stride_threshold\": 1024, \"pattern_matcher\": true, \"permute_fusion\": false, \"pick_loop_orders\": true, \"post_grad_custom_post_pass\": null, \"post_grad_custom_pre_pass\": null, \"post_grad_fusion_options\": {}, \"pre_grad_custom_pass\": null, \"pre_grad_fusion_options\": {}, \"profile_bandwidth\": false, \"profile_bandwidth_output\": null, \"profile_bandwidth_regex\": \"\", \"profile_bandwidth_with_do_bench_using_profiling\": false, \"profiler_mark_wrapper_call\": false, \"prologue_fusion\": true, \"realize_acc_reads_threshold\": 8, \"realize_opcount_threshold\": 30, \"realize_reads_threshold\": 4, \"remove_pre_grad_passes\": null, \"reorder_for_compute_comm_overlap\": false, \"reorder_for_compute_comm_overlap_passes\": [\"reorder_compute_for_overlap\", \"sink_waits\", \"raise_comms\"], \"reorder_for_locality\": true, \"reorder_for_peak_memory\": true, \"rocm.arch\": [], \"rocm.ck_dir\": null, \"rocm.ck_supported_arch\": [\"gfx90a\", \"gfx942\"], \"rocm.compile_opt_level\": \"-O2\", \"rocm.flush_denormals\": true, \"rocm.generate_test_runner\": false, \"rocm.is_debug\": false, \"rocm.n_max_profiling_configs\": null, \"rocm.print_kernel_resource_usage\": false, \"rocm.rocm_home\": null, \"rocm.save_temps\": false, \"rocm.use_fast_math\": true, \"rocm.use_preselected_instances\": false, \"save_args\": false, \"scalar_asserts\": true, \"score_fusion_memory_threshold\": 10, \"search_autotune_cache\": false, \"shape_padding\": true, \"size_asserts\": true, \"sleep_sec_TESTING_ONLY\": null, \"split_cat_fx_passes\": true, \"split_reductions\": true, \"static_weight_shapes\": true, \"test_configs.autotune_choice_desc_regex\": null, \"test_configs.autotune_choice_name_regex\": null, \"test_configs.force_extern_kernel_in_multi_template\": false, \"test_configs.graphsafe_rng_func_ignores_fallback_random\": false, \"test_configs.max_mm_configs\": null, \"test_configs.runtime_triton_dtype_assert\": false, \"trace.compile_profile\": false, \"trace.debug_dir\": null, \"trace.debug_log\": false, \"trace.dot_graph_shape\": null, \"trace.draw_orig_fx_graph\": false, \"trace.enabled\": false, \"trace.fx_graph\": true, \"trace.fx_graph_transformed\": true, \"trace.graph_diagram\": false, \"trace.info_log\": false, \"trace.ir_post_fusion\": true, \"trace.ir_pre_fusion\": true, \"trace.log_autotuning_results\": false, \"trace.log_inductor_triton_kernel_to_post_grad_node_info\": true, \"trace.log_url_for_graph_xform\": null, \"trace.output_code\": true, \"trace.save_real_tensors\": false, \"trace.upload_tar\": null, \"triton.autotune_at_compile_time\": null, \"triton.autotune_cublasLt\": true, \"triton.autotune_pointwise\": true, \"triton.codegen_upcast_to_fp32\": true, \"triton.cooperative_reductions\": false, \"triton.cudagraph_dynamic_shape_warn_limit\": 50, \"triton.cudagraph_skip_dynamic_graphs\": false, \"triton.cudagraph_support_input_mutation\": true, \"triton.cudagraph_trees\": true, \"triton.cudagraph_trees_history_recording\": false, \"triton.cudagraph_unexpected_rerecord_limit\": 128, \"triton.cudagraphs\": false, \"triton.debug_sync_graph\": false, \"triton.debug_sync_kernel\": false, \"triton.dense_indexing\": false, \"triton.descriptive_names\": \"original_aten\", \"triton.divisible_by_16\": true, \"triton.enable_persistent_tma_matmul\": false, \"triton.fast_path_cudagraph_asserts\": false, \"triton.force_cooperative_reductions\": false, \"triton.force_cudagraph_sync\": false, \"triton.force_cudagraphs_warmup\": false, \"triton.inject_relu_bug_TESTING_ONLY\": null, \"triton.max_tiles\": 2, \"triton.min_split_scan_rblock\": 256, \"triton.multi_kernel\": 0, \"triton.persistent_reductions\": true, \"triton.prefer_nd_tiling\": false, \"triton.skip_cudagraph_warmup\": false, \"triton.skip_l1_cache\": false, \"triton.slow_path_cudagraph_asserts\": true, \"triton.spill_threshold\": 16, \"triton.store_cubin\": false, \"triton.tile_reductions\": false, \"triton.tiling_prevents_pointwise_fusion\": true, \"triton.tiling_prevents_reduction_fusion\": true, \"triton.unique_kernel_names\": true, \"triton.unique_user_kernel_names\": false, \"triton.use_block_ptr\": false, \"triton_kernel_default_layout_constraint\": \"needs_fixed_stride_order\", \"unbacked_symint_fallback\": 8192, \"unroll_reductions_threshold\": 8, \"unsafe_ignore_unsupported_triton_autotune_args\": false, \"use_experimental_benchmarker\": true, \"use_fast_math\": false, \"use_mixed_mm\": true, \"verbose_progress\": false, \"warn_mix_layout\": false, \"worker_start_method\": \"subprocess\"}", "remote_cache_version": null, "inductor_fx_remote_cache_hit_count": null, "inductor_fx_remote_cache_miss_count": null, "inductor_fx_remote_cache_backend_type": null, "inductor_fx_remote_cache_hit_keys": null, "inductor_fx_remote_cache_miss_keys": null, "cuda_version": "12.6", "triton_version": "3.2.0", "feature_usage": {"aot_autograd_remote_cache": false, "fx_cache": true, "parallel_compile_post_warmup": false}, "compile_time_autotune_time_us": 72107, "is_runtime": false, "gc_time_us": 815, "tensorify_float_attempt": null, "tensorify_float_success": null, "tensorify_float_failure": null, "guard_latency_us": 0, "recompile_reason": null, "num_graph_breaks": null}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.383000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 0, "frame_compile_id": 0, "attempt": 1, "has_payload": "e69a3f19fdb54bf0b5cfb1ac101f4776"}
{
"name": "dynamo",
"ts": 1742496391383421.2,
"args": {
"compile_id": "0/0",
"distributed_ephemeral_timeout_us": 297465,
"guard_latency_us": 0,
"frame_key": "1",
"co_name": "forward",
"co_filename": "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py",
"co_firstlineno": 24,
"cache_size": 0,
"accumulated_cache_size": 0,
"guard_count": 34,
"shape_env_guard_count": 0,
"graph_op_count": 8,
"graph_node_count": 11,
"graph_input_count": 2,
"fail_type": null,
"fail_reason": null,
"fail_user_frame_filename": null,
"fail_user_frame_lineno": null,
"non_compliant_ops": [],
"compliant_custom_ops": [],
"restart_reasons": [
"Dynamic slicing with Tensor arguments\n Explanation: Creating slices with Tensor arguments is not supported. e.g. `l[:x]`, where `x` is a 1-element tensor.\n Hint: It may be possible to write Dynamo tracing rules for this code. Please report an issue to PyTorch if you encounter this graph break often and it is causing performance issues.\n\n Developer debug context: SliceVariable start: TensorVariable(), stop: TensorVariable(), step: ConstantVariable(NoneType: None)\n"
],
"dynamo_time_before_restart_s": 0.038427114486694336,
"has_guarded_code": true,
"dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": true, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.384000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "687c0255831fddccf22b4a5863500c68"}
{
"name": "cudagraphify",
"ts": 1742496391384066.2,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.384000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "a3d54f6def5f67e5899c910ea534dd2f"}
{
"name": "cudagraphify",
"ts": 1742496391384480.5,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.576000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "0144599d2fc418ea90e4f75789eb09f7"}
{
"name": "CUDAGraphTreeManager.run_eager",
"ts": 1742496391576635.2,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.578000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "bc9f190374118e42d0e3b2c9872d38bd"}
{
"name": "CUDAGraphTreeManager.run_eager",
"ts": 1742496391578298.5,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.579000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0, "has_payload": "70caaadfefcd940a70d77e0cb2c89186"}
{
"name": "dynamo",
"ts": 1742496391579291.0,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.579000 18434 site-packages/torch/_dynamo/convert_frame.py:1025] {"dynamo_start": {"stack": [{"line": 118, "name": "<module>", "filename": 1, "loc": "moe_c(input1)"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 1762, "name": "_call_impl", "filename": 2, "loc": "return forward_call(*args, **kwargs)"}, {"line": 586, "name": "_fn", "filename": 3, "loc": "return fn(*args, **kwargs)"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 1762, "name": "_call_impl", "filename": 2, "loc": "return forward_call(*args, **kwargs)"}, {"line": 24, "name": "forward", "filename": 1, "loc": "def forward(self, x: Tensor) -> Tensor:"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 43, "name": "forward", "filename": 1}]}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.580000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0, "has_payload": "a459bfff11366331916f085f9ddaada3"}
{
"name": "entire_frame_compile",
"ts": 1742496391580098.5,
"args": {
"fn_name": "_compile.compile_inner",
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.581000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 0, "describer_id": 5, "size": 48}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.582000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 1, "ndim": 3, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [1, 6, 4], "is_leaf": true, "stride": [24, 4, 1], "storage": 0, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fbce5593430>)", "describer_id": 5}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.582000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 0, "ndim": 2, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [6, 4], "is_leaf": true, "is_view": true, "stride": [4, 1], "storage": 0, "base": 1, "creation_meta": "CreationMeta.NO_GRAD_MODE", "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fba5daddbd0>)", "describer_id": 5}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.582000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 5, "id": 0, "source": "L['x']"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.584000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 1, "describer_id": 5, "size": 96}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.585000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 3, "ndim": 2, "dtype": "torch.int64", "device": "device(type='cuda', index=0)", "size": [6, 2], "is_leaf": true, "stride": [2, 1], "storage": 1, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fba5daddd10>)", "describer_id": 5}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.585000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 5, "id": 3, "source": "L['expert_indices']"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.591000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 0, "describer_id": 6, "size": 48}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.591000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 1, "ndim": 3, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [1, 6, 4], "is_leaf": true, "stride": [24, 4, 1], "storage": 0, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fbce5593430>)", "describer_id": 6}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.592000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 0, "ndim": 2, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [6, 4], "is_leaf": true, "is_view": true, "stride": [4, 1], "storage": 0, "base": 1, "creation_meta": "CreationMeta.NO_GRAD_MODE", "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fba5daddbd0>)", "describer_id": 6}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.592000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 6, "id": 0, "source": "L['x']"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.593000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 1, "describer_id": 6, "size": 96}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.593000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 2, "ndim": 2, "dtype": "torch.int64", "device": "device(type='cuda', index=0)", "size": [6, 2], "is_leaf": true, "stride": [2, 1], "storage": 1, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fba5daddd10>)", "describer_id": 6}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.594000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 6, "id": 2, "source": "L['expert_indices']"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.599000 18434 site-packages/torch/_dynamo/output_graph.py:1411] {"dynamo_output_graph": {"sizes": {"l_expert_indices_": [6, 2], "view": [12], "ordered_token_activations": [12], "div": [12], "floor": [12], "ordered_tokens": [12], "num_tokens_per_expert": [9], "cum_tokens_per_expert": [9]}}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "dfe83779fddea8b35accc22b850d6429"}
class GraphModule(torch.nn.Module):
def forward(self, L_expert_indices_: "i64[6, 2][2, 1]cuda:0"):
l_expert_indices_ = L_expert_indices_
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:66 in forward, code: ordered_token_activations = expert_indices.view(-1).argsort(stable=True)
view: "i64[12][1]cuda:0" = l_expert_indices_.view(-1)
ordered_token_activations: "i64[12][1]cuda:0" = view.argsort(stable = True); view = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:67 in forward, code: ordered_tokens =ordered_token_activations.div(num_activated_experts).floor().to(torch.int64)
div: "f32[12][1]cuda:0" = ordered_token_activations.div(2)
floor: "f32[12][1]cuda:0" = div.floor(); div = None
ordered_tokens: "i64[12][1]cuda:0" = floor.to(torch.int64); floor = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:69 in forward, code: num_tokens_per_expert = torch.histc(expert_indices, bins=self.num_experts+1, min=-1, max=self.num_experts)
num_tokens_per_expert: "i64[9][1]cuda:0" = torch.histc(l_expert_indices_, bins = 9, min = -1, max = 8); l_expert_indices_ = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:70 in forward, code: cum_tokens_per_expert = num_tokens_per_expert.cumsum(0)
cum_tokens_per_expert: "i64[9][1]cuda:0" = num_tokens_per_expert.cumsum(0); num_tokens_per_expert = None
return (ordered_token_activations, ordered_tokens, cum_tokens_per_expert)
V0320 11:46:31.600000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "913d2f8953421680bbb9e90afa28f871"}
{
"name": "backend_compile",
"ts": 1742496391599965.0,
"args": {
"fn_name": "OutputGraph.call_user_compiler",
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.600000 18434 site-packages/torch/_inductor/compile_fx.py:1965] {"inductor_pre_grad_graph": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "22f41487635a9ee7cbb3a615baccab46"}
class GraphModule(torch.nn.Module):
def forward(self, L_expert_indices_: "i64[6, 2][2, 1]cuda:0"):
l_expert_indices_ = L_expert_indices_
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:66 in forward, code: ordered_token_activations = expert_indices.view(-1).argsort(stable=True)
view: "i64[12][1]cuda:0" = l_expert_indices_.view(-1)
ordered_token_activations: "i64[12][1]cuda:0" = view.argsort(stable = True); view = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:67 in forward, code: ordered_tokens =ordered_token_activations.div(num_activated_experts).floor().to(torch.int64)
div: "f32[12][1]cuda:0" = ordered_token_activations.div(2)
floor: "f32[12][1]cuda:0" = div.floor(); div = None
ordered_tokens: "i64[12][1]cuda:0" = floor.to(torch.int64); floor = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:69 in forward, code: num_tokens_per_expert = torch.histc(expert_indices, bins=self.num_experts+1, min=-1, max=self.num_experts)
num_tokens_per_expert: "i64[9][1]cuda:0" = torch.histc(l_expert_indices_, bins = 9, min = -1, max = 8); l_expert_indices_ = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:70 in forward, code: cum_tokens_per_expert = num_tokens_per_expert.cumsum(0)
cum_tokens_per_expert: "i64[9][1]cuda:0" = num_tokens_per_expert.cumsum(0); num_tokens_per_expert = None
return (ordered_token_activations, ordered_tokens, cum_tokens_per_expert)
# graph id: 140438519379152
V0320 11:46:31.600000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "46ac57f16362e28514f9c75f2d20692a"}
{
"name": "_recursive_pre_grad_passes",
"ts": 1742496391600949.8,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.601000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "62da95718c51f5043f1661f60b9b637b"}
{
"name": "_recursive_pre_grad_passes",
"ts": 1742496391601735.0,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.606000 18434 site-packages/torch/_functorch/_aot_autograd/autograd_cache.py:491] {"artifact": {"name": "aot_forward_graph_fw_metadata", "encoding": "string"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "8f4e370d85643bdc98b42ba1a9181225"}
ViewAndMutationMeta(input_info=[InputAliasInfo(is_leaf=True,
mutates_data=False,
mutates_metadata=False,
mutations_hidden_from_autograd=True,
mutations_under_no_grad_or_inference_mode=False,
mutation_inductor_storage_resize=False,
mutates_storage_metadata=False,
requires_grad=False,
keep_input_mutations=True)],
output_info=[OutputAliasInfo(output_type=<OutputType.non_alias: 1>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=None,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=None),
OutputAliasInfo(output_type=<OutputType.non_alias: 1>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=None,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=None),
OutputAliasInfo(output_type=<OutputType.non_alias: 1>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=None,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=None)],
num_intermediate_bases=0,
keep_input_mutations=True,
traced_tangents=[],
subclass_inp_meta=[PlainTensorMeta(unwrapped_idx=0,
memory_format=None)],
subclass_fw_graph_out_meta=[PlainTensorMeta(unwrapped_idx=0,
memory_format=None),
PlainTensorMeta(unwrapped_idx=1,
memory_format=None),
PlainTensorMeta(unwrapped_idx=2,
memory_format=None)],
subclass_tangent_meta=[],
is_train=False,
traced_tangent_metas=[],
num_symints_saved_for_bw=None,
grad_enabled_mutation=None,
deterministic=None,
static_input_indices=[],
tokens={},
indices_of_inputs_that_requires_grad_with_mutations_in_bw=[],
bw_donated_idxs=None,
num_backward_tokens=0,
num_graphsafe_rng_states=0,
graphsafe_rng_state_index=None)
V0320 11:46:31.606000 18434 site-packages/torch/_functorch/_aot_autograd/autograd_cache.py:515] {"aot_inference_graph": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "cd25e35327273dec9181424bfa92977c"}
class <lambda>(torch.nn.Module):
def forward(self, arg0_1: "i64[6, 2][2, 1]cuda:0"):
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:65 in forward, code: ordered_token_activations = expert_indices.view(-1).argsort(stable=True)
view: "i64[12][1]cuda:0" = torch.ops.aten.view.default(arg0_1, [-1])
sort = torch.ops.aten.sort.stable(view, stable = True); view = None
getitem_1: "i64[12][1]cuda:0" = sort[1]; sort = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:66 in forward, code: ordered_tokens =ordered_token_activations.div(num_activated_experts).floor().to(torch.int64)
div: "f32[12][1]cuda:0" = torch.ops.aten.div.Tensor(getitem_1, 2)
floor: "f32[12][1]cuda:0" = torch.ops.aten.floor.default(div); div = None
convert_element_type: "i64[12][1]cuda:0" = torch.ops.prims.convert_element_type.default(floor, torch.int64); floor = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:68 in forward, code: num_tokens_per_expert = torch.histc(expert_indices, bins=self.num_experts+1, min=-1, max=self.num_experts)
histc: "i64[9][1]cuda:0" = torch.ops.aten.histc.default(arg0_1, 9, -1, 8); arg0_1 = None
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:69 in forward, code: cum_tokens_per_expert = num_tokens_per_expert.cumsum(0)
cumsum: "i64[9][1]cuda:0" = torch.ops.aten.cumsum.default(histc, 0); histc = None
return (getitem_1, convert_element_type, cumsum)
V0320 11:46:31.607000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "41f46dd7cb339000013e2bcde7b591a2"}
{
"name": "TritonBundler.read_and_emit",
"ts": 1742496391607884.5,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.608000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "5fb19ade72650692761e40fdd6f3b8b1"}
{
"name": "TritonBundler.read_and_emit",
"ts": 1742496391608276.5,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.608000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "09bf9a59c5c62026e06fa50e11c0c57d"}
{
"name": "PyCodeCache.load_by_key_path",
"ts": 1742496391608593.2,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.609000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "ae1ae322e95b506eb120b98b0cff9d7d"}
{
"name": "async_compile.precompile",
"ts": 1742496391609618.5,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.614000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "e417f2e7bc0bc11532597edb83525acc"}
{
"name": "async_compile.precompile",
"ts": 1742496391614666.2,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.615000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "34ef7c32c35f36920e13feda42d6959e"}
{
"name": "async_compile.precompile",
"ts": 1742496391615046.0,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.621000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "ee3fe2e8031b68f6798669dd93c54953"}
{
"name": "async_compile.precompile",
"ts": 1742496391621250.0,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.621000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "d498f1966d5452368b8887aa502e576d"}
{
"name": "async_compile.wait",
"ts": 1742496391621596.5,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.622000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "e762ad5f04769dbdc2d2142105d78fd6"}
{
"name": "async_compile.wait",
"ts": 1742496391622057.0,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.622000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "ce0005ee566ed5108a022597ba25e707"}
{
"name": "PyCodeCache.load_by_key_path",
"ts": 1742496391622376.8,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.622000 18434 site-packages/torch/_inductor/codecache.py:1094] {"inductor_output_code": {"filename": "/tmp/torchinductor_cdhernandez/35/c353y3obmykmgnnu6e26jxk5m3w6fm7tggdiyag3vhznrkipugwt.py"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "2ae0b606085752c67e5bd76d2d04456f"}
# AOT ID: ['1_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from cmath import nanj
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import (
grid,
split_scan_grid,
grid_combo_kernels,
start_graph,
end_graph,
cooperative_reduction_grid,
)
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
empty_strided_p2p = torch._C._distributed_c10d._SymmetricMemory.empty_strided_p2p
# kernel path: /tmp/torchinductor_cdhernandez/bj/cbj7i2qdjygnbu2cvimxmlmrnxvgdgujgz6hwl2yxnhjmu5fct5p.py
# Topologically Sorted Source Nodes: [cum_tokens_per_expert], Original ATen: [aten.cumsum]
# Source node to ATen node mapping:
# cum_tokens_per_expert => cumsum
# Graph fragment:
# %cumsum : [num_users=1] = call_function[target=torch.ops.aten.cumsum.default](args = (%histc, 0), kwargs = {})
triton_per_fused_cumsum_0 = async_compile.triton('triton_per_fused_cumsum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, DeviceProperties
triton_helpers.set_driver_to_gpu()
@triton.jit
def _triton_helper_fn_add0(arg0_0, arg1_0):
tmp0 = arg0_0 + arg1_0
return tmp0
@triton_heuristics.persistent_reduction(
size_hints={'x': 1, 'r0_': 16},
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {'in_out_ptr0': '*i64', 'xnumel': 'i32', 'r0_numel': 'i32'}, 'device': DeviceProperties(type='cuda', index=0, multi_processor_count=132, cc=90, major=9, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, warp_size=32), 'constants': {'xnumel': 1}, 'configs': [AttrsDescriptor.from_dict({'arg_properties': {'tt.divisibility': (0,), 'tt.equal_to': (1,)}, 'cls': 'AttrsDescriptor'})]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_cumsum_0', 'mutated_arg_names': ['in_out_ptr0'], 'optimize_mem': True, 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'CE326918D8FE98067798167ABD26A2E4EDFD110D9ECD4380441C64512F6D164E', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_cumsum_0(in_out_ptr0, xnumel, r0_numel, XBLOCK : tl.constexpr):
xnumel = 1
r0_numel = 9
R0_BLOCK: tl.constexpr = 16
rnumel = r0_numel
RBLOCK: tl.constexpr = R0_BLOCK
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, R0_BLOCK], True, tl.int1)
r0_index = tl.arange(0, R0_BLOCK)[None, :]
r0_offset = 0
r0_mask = r0_index < r0_numel
roffset = r0_offset
rindex = r0_index
r0_0 = r0_index
tmp0 = tl.load(in_out_ptr0 + (r0_0), r0_mask, other=0.0)
tmp1 = tmp0.to(tl.int64)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, R0_BLOCK])
tmp3, = tl.associative_scan((tmp2,), 1, _triton_helper_fn_add0)
tl.store(in_out_ptr0 + (tl.broadcast_to(r0_0, [XBLOCK, R0_BLOCK])), tmp3, r0_mask)
''', device_str='cuda')
# kernel path: /tmp/torchinductor_cdhernandez/yr/cyrbeu2cd2vywmlwwfadahmnnsjng3ji6tgqkhzkyltjpqxbxkij.py
# Topologically Sorted Source Nodes: [ordered_token_activations, div, floor, ordered_tokens], Original ATen: [aten.sort, aten.div, aten.floor, aten._to_copy]
# Source node to ATen node mapping:
# div => div
# floor => floor
# ordered_token_activations => getitem_1, sort
# ordered_tokens => convert_element_type
# Graph fragment:
# %sort : [num_users=1] = call_function[target=torch.ops.aten.sort.stable](args = (%view,), kwargs = {stable: True})
# %getitem_1 : [num_users=2] = call_function[target=operator.getitem](args = (%sort, 1), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%getitem_1, 2), kwargs = {})
# %floor : [num_users=1] = call_function[target=torch.ops.aten.floor.default](args = (%div,), kwargs = {})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%floor, torch.int64), kwargs = {})
triton_per_fused__to_copy_div_floor_sort_1 = async_compile.triton('triton_per_fused__to_copy_div_floor_sort_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, DeviceProperties
triton_helpers.set_driver_to_gpu()
@triton_heuristics.persistent_reduction(
size_hints={'x': 1, 'r0_': 16},
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {'in_ptr0': '*i64', 'out_ptr1': '*i64', 'out_ptr2': '*i64', 'xnumel': 'i32', 'r0_numel': 'i32'}, 'device': DeviceProperties(type='cuda', index=0, multi_processor_count=132, cc=90, major=9, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, warp_size=32), 'constants': {'xnumel': 1}, 'configs': [AttrsDescriptor.from_dict({'arg_properties': {'tt.divisibility': (0, 1, 2), 'tt.equal_to': (3,)}, 'cls': 'AttrsDescriptor'})]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__to_copy_div_floor_sort_1', 'mutated_arg_names': [], 'optimize_mem': True, 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'CE326918D8FE98067798167ABD26A2E4EDFD110D9ECD4380441C64512F6D164E', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__to_copy_div_floor_sort_1(in_ptr0, out_ptr1, out_ptr2, xnumel, r0_numel, XBLOCK : tl.constexpr):
xnumel = 1
r0_numel = 12
R0_BLOCK: tl.constexpr = 16
rnumel = r0_numel
RBLOCK: tl.constexpr = R0_BLOCK
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, R0_BLOCK], True, tl.int1)
r0_index = tl.arange(0, R0_BLOCK)[None, :]
r0_offset = 0
r0_mask = r0_index < r0_numel
roffset = r0_offset
rindex = r0_index
r0_0 = r0_index
tmp0 = tl.load(in_ptr0 + (r0_0), r0_mask, other=0.0)
tmp1 = r0_0
tmp2 = tmp1.to(tl.int16)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, R0_BLOCK])
tmp4 = tl.broadcast_to(tmp2, [XBLOCK, R0_BLOCK])
tmp5, tmp6, = triton_helpers.sort_with_index(tmp3, tmp4, rnumel, 1, stable=True, descending=False)
tmp7 = tmp6.to(tl.int64)
tmp8 = tmp7.to(tl.float32)
tmp9 = 0.5
tmp10 = tmp8 * tmp9
tmp11 = libdevice.floor(tmp10)
tmp12 = tmp11.to(tl.int64)
tl.store(out_ptr1 + (tl.broadcast_to(r0_0, [XBLOCK, R0_BLOCK])), tmp7, r0_mask)
tl.store(out_ptr2 + (tl.broadcast_to(r0_0, [XBLOCK, R0_BLOCK])), tmp12, r0_mask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (6, 2), (2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [num_tokens_per_expert], Original ATen: [aten.histc]
buf4 = torch.ops.aten.histc.default(arg0_1, 9, -1, 8)
buf5 = buf4
assert_size_stride(buf5, (9, ), (1, ))
del buf4
buf6 = buf5; del buf5 # reuse
# Topologically Sorted Source Nodes: [cum_tokens_per_expert], Original ATen: [aten.cumsum]
stream0 = get_raw_stream(0)
triton_per_fused_cumsum_0.run(buf6, 1, 9, grid=grid(1), stream=stream0)
buf2 = empty_strided_cuda((12, ), (1, ), torch.int64)
buf3 = empty_strided_cuda((12, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [ordered_token_activations, div, floor, ordered_tokens], Original ATen: [aten.sort, aten.div, aten.floor, aten._to_copy]
stream0 = get_raw_stream(0)
triton_per_fused__to_copy_div_floor_sort_1.run(arg0_1, buf2, buf3, 1, 12, grid=grid(1), stream=stream0)
del arg0_1
return (buf2, buf3, buf6, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((6, 2), (2, 1), device='cuda:0', dtype=torch.int64)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
V0320 11:46:31.623000 18434 site-packages/torch/_functorch/_aot_autograd/autograd_cache.py:385] {"artifact": {"name": "fx_graph_cache_hit", "encoding": "json"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "299d74d079126c7efb14fce4809b228a"}
{"triton_bundler_meta": "TritonBundlerMetadata(cached_kernel_names=[])", "key": "f4wuysroho5h6eucbqo3den7itcl2a2eudlemxu3sr44jsf7jeyk", "components": [], "cache_event_time": 1742496391623022260, "cache_state": "hit", "time_saved_ns": 2488734991}
V0320 11:46:31.623000 18434 site-packages/torch/_dynamo/utils.py:1821] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "420d5bd8239c8dce84b149f191c95855"}
{
"name": "autograd_cache_hit",
"ts": 1742496391623392.5,
"args": {
"forward_time_saved_ms": 2505.0,
"backward_time_saved_ms": 0.0,
"time_saved_ms": 2505.0,
"key": "ayoimkqin7rawz6c4rqco5iyugzgshvphhsgcbo276fv36pnwqdo",
"cache_state": "hit",
"components": [
"[pevv77ugkkqgapvijpdi3gia4ipchedxcl3fespx5lcgunzfszy] aot_config: (0, True, False, False, False, [LocalSource(local_name='expert_indices', is_input=True, is_derefed_cell_contents=False)], True, False)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False",
"[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>",
"[amesqdayivuaxpm3b2gv6qht5xmwcjywtk6bghfzcs3wq6pgu2q] gm: GraphModule()\n\n\n\ndef forward(self, L_expert_indices_ : torch.Tensor):\n l_expert_indices_ = L_expert_indices_\n view = l_expert_indices_.view(-1)\n ordered_token_activations = view.argsort(stable = True); view = None\n div = ordered_token_activations.div(2)\n floor = div.floor(); div = None\n ordered_tokens = floor.to(torch.int64); floor = None\n num_tokens_per_expert = torch.histc(l_expert_indices_, bins = 9, min = -1, max = 8); l_expert_indices_ = None\n cum_tokens_per_expert = num_tokens_per_expert.cumsum(0); num_tokens_per_expert = None\n return (ordered_token_activations, ordered_tokens, cum_tokens_per_expert)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[jn6ydv2giglse4ktis3olqidsgwoa5a4hfs3efk6xu3smtmc273] example_inputs[0]: TensorMetadata(dtype=torch.int64, shape=torch.Size([6, 2]), stride=(2, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"compile_id": "1/0"
},
"ph": "i",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0,
"s": "p"
}
V0320 11:46:31.624000 18434 site-packages/torch/_functorch/_aot_autograd/autograd_cache.py:804] {"artifact": {"name": "aotautograd_cache_hit", "encoding": "json"}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "009264c2a8d0dd72b0263ab3a8600536"}
{"forward_time_saved_ms": 2505.0, "backward_time_saved_ms": 0.0, "time_saved_ms": 2505.0, "key": "ayoimkqin7rawz6c4rqco5iyugzgshvphhsgcbo276fv36pnwqdo", "cache_state": "hit", "components": ["[pevv77ugkkqgapvijpdi3gia4ipchedxcl3fespx5lcgunzfszy] aot_config: (0, True, False, False, False, [LocalSource(local_name='expert_indices', is_input=True, is_derefed_cell_contents=False)], True, False)", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False", "[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>", "[amesqdayivuaxpm3b2gv6qht5xmwcjywtk6bghfzcs3wq6pgu2q] gm: GraphModule()\n\n\n\ndef forward(self, L_expert_indices_ : torch.Tensor):\n l_expert_indices_ = L_expert_indices_\n view = l_expert_indices_.view(-1)\n ordered_token_activations = view.argsort(stable = True); view = None\n div = ordered_token_activations.div(2)\n floor = div.floor(); div = None\n ordered_tokens = floor.to(torch.int64); floor = None\n num_tokens_per_expert = torch.histc(l_expert_indices_, bins = 9, min = -1, max = 8); l_expert_indices_ = None\n cum_tokens_per_expert = num_tokens_per_expert.cumsum(0); num_tokens_per_expert = None\n return (ordered_token_activations, ordered_tokens, cum_tokens_per_expert)\n \n# To see more debug info, please use `graph_module.print_readable()`", "[jn6ydv2giglse4ktis3olqidsgwoa5a4hfs3efk6xu3smtmc273] example_inputs[0]: TensorMetadata(dtype=torch.int64, shape=torch.Size([6, 2]), stride=(2, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ", "[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)", "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)", "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)", "[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>", "[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}", "[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}", "[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False", "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True", "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}", "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False", "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True", "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default", "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300", "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False", "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP", "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True", "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False", "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0", "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0", "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ", "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT", "[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False", "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4", "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1", "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128", "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False", "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None", "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096", "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False", "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128", "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True", "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None", "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None", "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []", "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']", "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False", "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp", "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton", "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host", "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda", "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021", "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"], "compile_id": "1/0"}
V0320 11:46:31.624000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "516205e704aeb08403592a1c9fba6a98"}
{
"name": "backend_compile",
"ts": 1742496391624498.0,
"args": {
"fn_name": "OutputGraph.call_user_compiler",
"compile_id": "1/0",
"dispatch_mode": "inference",
"requires_subclass_dispatch": false,
"cache_state": "hit",
"cache_event_time": 1742496391623392479,
"key": "ayoimkqin7rawz6c4rqco5iyugzgshvphhsgcbo276fv36pnwqdo",
"components": [
"[pevv77ugkkqgapvijpdi3gia4ipchedxcl3fespx5lcgunzfszy] aot_config: (0, True, False, False, False, [LocalSource(local_name='expert_indices', is_input=True, is_derefed_cell_contents=False)], True, False)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False",
"[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>",
"[amesqdayivuaxpm3b2gv6qht5xmwcjywtk6bghfzcs3wq6pgu2q] gm: GraphModule()\n\n\n\ndef forward(self, L_expert_indices_ : torch.Tensor):\n l_expert_indices_ = L_expert_indices_\n view = l_expert_indices_.view(-1)\n ordered_token_activations = view.argsort(stable = True); view = None\n div = ordered_token_activations.div(2)\n floor = div.floor(); div = None\n ordered_tokens = floor.to(torch.int64); floor = None\n num_tokens_per_expert = torch.histc(l_expert_indices_, bins = 9, min = -1, max = 8); l_expert_indices_ = None\n cum_tokens_per_expert = num_tokens_per_expert.cumsum(0); num_tokens_per_expert = None\n return (ordered_token_activations, ordered_tokens, cum_tokens_per_expert)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[jn6ydv2giglse4ktis3olqidsgwoa5a4hfs3efk6xu3smtmc273] example_inputs[0]: TensorMetadata(dtype=torch.int64, shape=torch.Size([6, 2]), stride=(2, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_bypass_reason": null,
"remote_cache_enabled": false,
"local_cache_enabled": true
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.629000 18434 site-packages/torch/_dynamo/guards.py:2537] {"dynamo_cpp_guards_str": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "c7515c23d4ca93a704a65c8be7f7fc47"}
TREE_GUARD_MANAGER:
+- RootGuardManager
| +- DEFAULT_DEVICE: utils_device.CURRENT_DEVICE == None # _dynamo/output_graph.py:519 in init_ambient_guards
| +- GLOBAL_STATE: ___check_global_state()
| +- TORCH_FUNCTION_MODE_STACK: ___check_torch_function_mode_stack()
| +- GuardManager: source=L['x'], accessed_by=FrameLocalsGuardAccessor(key='x', framelocals_idx=1)
| | +- TENSOR_MATCH: check_tensor(L['x'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.bfloat16, device=0, requires_grad=False, size=[6, 4], stride=[4, 1])
| | +- NO_HASATTR: hasattr(L['x'], '_dynamo_dynamic_indices') == False
| | +- NO_TENSOR_ALIASING: check_no_aliasing(L['x'], L['expert_indices'])
| +- GuardManager: source=L['self'], accessed_by=FrameLocalsGuardAccessor(key='self', framelocals_idx=0)
| | +- TYPE_MATCH: ___check_type_id(L['self'], 11695168)
| | +- GuardManager: source=L['self'].__dict__, accessed_by=GetGenericDictGuardAccessor
| | | +- GuardManager: source=L['self'].num_experts, accessed_by=DictGetItemGuardAccessor('num_experts')
| | | | +- EQUALS_MATCH: L['self'].num_experts == 8
| +- GuardManager: source=L['expert_indices'], accessed_by=FrameLocalsGuardAccessor(key='expert_indices', framelocals_idx=2)
| | +- TENSOR_MATCH: check_tensor(L['expert_indices'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.int64, device=0, requires_grad=False, size=[6, 2], stride=[2, 1])
| | +- NO_HASATTR: hasattr(L['expert_indices'], '_dynamo_dynamic_indices') == False
| | +- NO_TENSOR_ALIASING
| | +- GuardManager: source=L['expert_indices'].view, accessed_by=GetAttrGuardAccessor(view)
| +- GuardManager: source=L['num_activated_experts'], accessed_by=FrameLocalsGuardAccessor(key='num_activated_experts', framelocals_idx=4)
| | +- EQUALS_MATCH: L['num_activated_experts'] == 2
| +- GuardManager: source=G, accessed_by=GlobalsGuardAccessor
| | +- GuardManager: source=G['torch'], accessed_by=DictGetItemGuardAccessor('torch')
| | | +- ID_MATCH: ___check_obj_id(G['torch'], 140455470383408)
| | | +- GuardManager: source=G['torch'].histc, accessed_by=GetAttrGuardAccessor(histc)
| | | | +- ID_MATCH: ___check_obj_id(G['torch'].histc, 140455466281264)
| | | +- GuardManager: source=G['torch'].int64, accessed_by=GetAttrGuardAccessor(int64)
| | | | +- EQUALS_MATCH: G['torch'].int64 == torch.int64
| | +- GuardManager: source=G['__builtins_dict___5'], accessed_by=DictGetItemGuardAccessor('__builtins_dict___5')
| | | +- GuardManager: source=G['__builtins_dict___5']['range'], accessed_by=DictGetItemGuardAccessor('range')
| | | | +- ID_MATCH: ___check_obj_id(G['__builtins_dict___5']['range'], 9674624)
Guard latency = 0.64 us
V0320 11:46:31.629000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "0c84f0da74781965ea4c3f564f5f8e70"}
{
"name": "entire_frame_compile",
"ts": 1742496391629530.8,
"args": {
"fn_name": "_compile.compile_inner",
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.629000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "a9a7ae938bb899818c30a6965f7d0bb4"}
{
"name": "gc",
"ts": 1742496391629853.2,
"args": {
"compile_id": "1/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.630000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "1dfb810cc68788e37a370489c8518e8f"}
{
"name": "gc",
"ts": 1742496391630570.2,
"args": {
"compile_id": "1/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.632000 18434 site-packages/torch/_dynamo/utils.py:1483] {"compilation_metrics": {"compile_id": "1/0", "frame_key": "2", "co_name": "forward", "co_filename": "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py", "co_firstlineno": 43, "cache_size": 0, "accumulated_cache_size": 0, "guard_count": 15, "shape_env_guard_count": 0, "graph_op_count": 7, "graph_node_count": 9, "graph_input_count": 1, "start_time": 1742496391.579609, "entire_frame_compile_time_s": 0.049432, "backend_compile_time_s": 0.024533, "inductor_compile_time_s": null, "code_gen_time_s": null, "fail_type": null, "fail_reason": null, "fail_user_frame_filename": null, "fail_user_frame_lineno": null, "non_compliant_ops": [], "compliant_custom_ops": [], "restart_reasons": ["Unknown fail_and_restart_analysis"], "dynamo_time_before_restart_s": 0.01001882553100586, "has_guarded_code": true, "remote_cache_time_saved_s": 2.488734, "structured_logging_overhead_s": 0.013371, "config_suppress_errors": false, "config_inline_inbuilt_nn_modules": true, "specialize_float": false, "dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": true, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}", "is_forward": true, "num_triton_bundles": null, "remote_fx_graph_cache_get_time_ms": null, "remote_fx_graph_cache_put_time_ms": null, "start_time_us": 1742496391579609, "duration_us": 51298, "dynamo_cumulative_compile_time_us": 49432, "aot_autograd_cumulative_compile_time_us": 24533, "inductor_cumulative_compile_time_us": null, "inductor_code_gen_cumulative_compile_time_us": null, "triton_compile_time_us": 11710, "runtime_cudagraphify_time_us": null, "runtime_triton_autotune_time_us": null, "dynamo_compile_time_before_restart_us": 10018, "cuda_synchronize_time_us": null, "distributed_ephemeral_timeout_us": 2488734, "structured_logging_overhead_us": 13371, "remote_fx_graph_cache_get_time_us": null, "remote_fx_graph_cache_put_time_us": null, "backward_cumulative_compile_time_us": null, "end_time_us": 1742496391630907, "pre_grad_pass_time_us": 785, "post_grad_pass_time_us": null, "joint_graph_pass_time_us": null, "log_format_version": 3, "inductor_config": "{\"TYPE_CHECKING\": false, \"_cache_config_ignore_prefix\": [\"trace\", \"cuda.cutlass_dir\", \"worker_start_method\", \"compile_threads\", \"post_grad_custom_post_pass\", \"post_grad_custom_pre_pass\", \"always_complex_memory_overlap_TESTING_ONLY\"], \"_collective.auto_select\": false, \"_collective.one_shot_all_reduce_threshold_bytes\": 131072, \"_fuse_ddp_bucket_size\": 25, \"_fuse_ddp_communication\": false, \"_fuse_ddp_communication_passes\": [\"fuse_ddp_with_concat_op\", \"schedule_comm_wait\"], \"_micro_pipeline_tp\": false, \"_pre_fusion_custom_pass\": null, \"_profile_var\": \"\", \"_raise_error_for_testing\": false, \"_save_config_ignore\": [\"trace.upload_tar\", \"joint_custom_pre_pass\", \"joint_custom_post_pass\", \"pre_grad_custom_pass\", \"aot_inductor.repro_level\", \"aot_inductor.dump_aoti_minifier\"], \"add_pre_grad_passes\": null, \"aggressive_fusion\": false, \"allow_buffer_reuse\": true, \"always_complex_memory_overlap_TESTING_ONLY\": false, \"always_keep_tensor_constants\": false, \"annotate_training\": false, \"aot_inductor.allow_stack_allocation\": false, \"aot_inductor.compile_wrapper_with_O0\": false, \"aot_inductor.debug_compile\": false, \"aot_inductor.debug_intermediate_value_printer\": \"0\", \"aot_inductor.dump_aoti_minifier\": false, \"aot_inductor.filtered_kernel_names\": null, \"aot_inductor.force_mmap_weights\": false, \"aot_inductor.metadata\": {}, \"aot_inductor.output_path\": \"\", \"aot_inductor.package\": false, \"aot_inductor.package_constants_in_so\": true, \"aot_inductor.package_cpp_only\": false, \"aot_inductor.presets\": {}, \"aot_inductor.raise_error_on_ignored_optimization\": true, \"aot_inductor.repro_level\": 2, \"aot_inductor.serialized_in_spec\": \"\", \"aot_inductor.serialized_out_spec\": \"\", \"aot_inductor.use_minimal_arrayref_interface\": false, \"aot_inductor.use_runtime_constant_folding\": false, \"assert_indirect_indexing\": true, \"assume_aligned_inputs\": false, \"autoheuristic_collect\": \"\", \"autoheuristic_log_path\": \"DEFAULT\", \"autoheuristic_use\": \"mixed_mm\", \"autotune_fallback_to_aten\": true, \"autotune_in_subproc\": false, \"autotune_local_cache\": true, \"autotune_multi_device\": false, \"autotune_num_choices_displayed\": 10, \"autotune_remote_cache\": null, \"b2b_gemm_pass\": false, \"batch_fusion\": true, \"benchmark_combo_kernel\": false, \"benchmark_epilogue_fusion\": true, \"benchmark_fusion\": false, \"benchmark_harness\": true, \"benchmark_kernel\": false, \"bundle_triton_into_fx_graph_cache\": true, \"bundled_autotune_remote_cache\": null, \"bw_outputs_user_visible\": true, \"can_inplace_pad_graph_input\": false, \"check_stack_no_cycles_TESTING_ONLY\": false, \"combo_kernel_allow_mixed_sizes\": 1, \"combo_kernel_foreach_dynamic_shapes\": false, \"combo_kernels\": false, \"combo_kernels_autotune\": 1, \"comment_origin\": false, \"compile_threads\": 32, \"comprehensive_padding\": true, \"compute_all_bounds\": false, \"constant_and_index_propagation\": true, \"conv_1x1_as_mm\": false, \"coordinate_descent_check_all_directions\": false, \"coordinate_descent_search_radius\": 1, \"coordinate_descent_tuning\": false, \"cpp.cxx\": [null, \"g++\"], \"cpp.descriptive_names\": \"original_aten\", \"cpp.dynamic_threads\": false, \"cpp.enable_concat_linear\": false, \"cpp.enable_floating_point_contract_flag\": \"off\", \"cpp.enable_grouped_gemm_template\": false, \"cpp.enable_kernel_profile\": false, \"cpp.enable_loop_tail_vec\": true, \"cpp.enable_tiling_heuristics\": true, \"cpp.enable_unsafe_math_opt_flag\": false, \"cpp.fallback_scatter_reduce_sum\": true, \"cpp.gemm_cache_blocking\": null, \"cpp.gemm_max_k_slices\": 1, \"cpp.gemm_thread_factors\": null, \"cpp.inject_log1p_bug_TESTING_ONLY\": null, \"cpp.inject_relu_bug_TESTING_ONLY\": null, \"cpp.max_horizontal_fusion_size\": 16, \"cpp.min_chunk_size\": 4096, \"cpp.no_redundant_loops\": true, \"cpp.simdlen\": null, \"cpp.threads\": -1, \"cpp.vec_isa_ok\": null, \"cpp.weight_prepack\": true, \"cpp_wrapper\": false, \"cpu_backend\": \"cpp\", \"cuda.arch\": null, \"cuda.compile_opt_level\": \"-O1\", \"cuda.cuda_cxx\": null, \"cuda.cutlass_backend_min_gemm_size\": 1, \"cuda.cutlass_dir\": \"/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/third_party/cutlass\", \"cuda.cutlass_instantiation_level\": \"0\", \"cuda.cutlass_max_profiling_configs\": null, \"cuda.cutlass_max_profiling_swizzle_options\": [1, 2, 4], \"cuda.cutlass_op_allowlist_regex\": null, \"cuda.cutlass_op_denylist_regex\": null, \"cuda.enable_cuda_lto\": false, \"cuda.enable_debug_info\": false, \"cuda.enable_ptxas_info\": false, \"cuda.generate_test_runner\": false, \"cuda.use_fast_math\": false, \"cuda.version\": null, \"cuda_backend\": \"triton\", \"custom_op_default_layout_constraint\": \"needs_fixed_stride_order\", \"dce\": false, \"debug\": false, \"debug_fusion\": false, \"debug_index_asserts\": false, \"debug_ir_traceback\": false, \"decompose_mem_bound_mm\": false, \"developer_warnings\": true, \"disable_cpp_codegen\": false, \"disable_padding_cpu\": true, \"disable_progress\": true, \"dynamic_scale_rblock\": true, \"efficient_conv_bn_eval_fx_passes\": false, \"emulate_precision_casts\": false, \"enable_auto_functionalized_v2\": true, \"enable_linear_binary_folding\": false, \"enabled_metric_tables\": \"\", \"epilogue_fusion\": true, \"epilogue_fusion_first\": false, \"estimate_op_runtime\": \"default\", \"external_matmul\": [], \"fallback_random\": false, \"force_disable_caches\": false, \"force_fuse_int_mm_with_mul\": false, \"force_layout_optimization\": false, \"force_pointwise_cat\": false, \"force_same_precision\": false, \"force_shape_pad\": false, \"freezing\": false, \"freezing_discard_parameters\": false, \"fx_graph_cache\": true, \"fx_graph_remote_cache\": null, \"fx_passes_numeric_check\": {\"num_iterations\": 1, \"pre_grad\": false, \"precision\": 0.0001, \"requires_optimizer\": true}, \"generate_intermediate_hooks\": false, \"global_cache_dir\": null, \"graph_partition\": false, \"group_fusion\": false, \"halide.asserts\": false, \"halide.cpu_target\": \"host\", \"halide.debug\": false, \"halide.gpu_target\": \"host-cuda\", \"halide.scan_kernels\": false, \"halide.scheduler_cpu\": \"Adams2019\", \"halide.scheduler_cuda\": \"Anderson2021\", \"implicit_fallbacks\": true, \"inplace_buffers\": true, \"inplace_padding\": true, \"inter_node_bw\": 25, \"intra_node_bw\": 300, \"is_nightly_or_source\": true, \"is_predispatch\": false, \"joint_custom_post_pass\": null, \"joint_custom_pre_pass\": null, \"joint_graph_constant_folding\": true, \"keep_output_stride\": true, \"kernel_name_max_ops\": 10, \"layout_opt_default\": \"1\", \"layout_optimization\": true, \"loop_ordering_after_fusion\": false, \"max_autotune\": false, \"max_autotune_conv_backends\": \"ATEN,TRITON\", \"max_autotune_gemm\": false, \"max_autotune_gemm_backends\": \"ATEN,TRITON,CPP\", \"max_autotune_gemm_search_space\": \"DEFAULT\", \"max_autotune_pointwise\": false, \"max_autotune_subproc_graceful_timeout_seconds\": 1.0, \"max_autotune_subproc_result_timeout_seconds\": 60.0, \"max_autotune_subproc_terminate_timeout_seconds\": 2.0, \"max_epilogue_benchmarked_choices\": 1, \"max_fusion_size\": 64, \"max_pointwise_cat_inputs\": 8, \"memory_planning\": false, \"memory_pool\": \"intermediates\", \"mixed_mm_choice\": \"heuristic\", \"nan_asserts\": false, \"optimize_scatter_upon_const_tensor\": true, \"pad_channels_last\": false, \"pad_outputs\": false, \"padding_alignment_bytes\": 128, \"padding_stride_threshold\": 1024, \"pattern_matcher\": true, \"permute_fusion\": false, \"pick_loop_orders\": true, \"post_grad_custom_post_pass\": null, \"post_grad_custom_pre_pass\": null, \"post_grad_fusion_options\": {}, \"pre_grad_custom_pass\": null, \"pre_grad_fusion_options\": {}, \"profile_bandwidth\": false, \"profile_bandwidth_output\": null, \"profile_bandwidth_regex\": \"\", \"profile_bandwidth_with_do_bench_using_profiling\": false, \"profiler_mark_wrapper_call\": false, \"prologue_fusion\": true, \"realize_acc_reads_threshold\": 8, \"realize_opcount_threshold\": 30, \"realize_reads_threshold\": 4, \"remove_pre_grad_passes\": null, \"reorder_for_compute_comm_overlap\": false, \"reorder_for_compute_comm_overlap_passes\": [\"reorder_compute_for_overlap\", \"sink_waits\", \"raise_comms\"], \"reorder_for_locality\": true, \"reorder_for_peak_memory\": true, \"rocm.arch\": [], \"rocm.ck_dir\": null, \"rocm.ck_supported_arch\": [\"gfx90a\", \"gfx942\"], \"rocm.compile_opt_level\": \"-O2\", \"rocm.flush_denormals\": true, \"rocm.generate_test_runner\": false, \"rocm.is_debug\": false, \"rocm.n_max_profiling_configs\": null, \"rocm.print_kernel_resource_usage\": false, \"rocm.rocm_home\": null, \"rocm.save_temps\": false, \"rocm.use_fast_math\": true, \"rocm.use_preselected_instances\": false, \"save_args\": false, \"scalar_asserts\": true, \"score_fusion_memory_threshold\": 10, \"search_autotune_cache\": false, \"shape_padding\": true, \"size_asserts\": true, \"sleep_sec_TESTING_ONLY\": null, \"split_cat_fx_passes\": true, \"split_reductions\": true, \"static_weight_shapes\": true, \"test_configs.autotune_choice_desc_regex\": null, \"test_configs.autotune_choice_name_regex\": null, \"test_configs.force_extern_kernel_in_multi_template\": false, \"test_configs.graphsafe_rng_func_ignores_fallback_random\": false, \"test_configs.max_mm_configs\": null, \"test_configs.runtime_triton_dtype_assert\": false, \"trace.compile_profile\": false, \"trace.debug_dir\": null, \"trace.debug_log\": false, \"trace.dot_graph_shape\": null, \"trace.draw_orig_fx_graph\": false, \"trace.enabled\": false, \"trace.fx_graph\": true, \"trace.fx_graph_transformed\": true, \"trace.graph_diagram\": false, \"trace.info_log\": false, \"trace.ir_post_fusion\": true, \"trace.ir_pre_fusion\": true, \"trace.log_autotuning_results\": false, \"trace.log_inductor_triton_kernel_to_post_grad_node_info\": true, \"trace.log_url_for_graph_xform\": null, \"trace.output_code\": true, \"trace.save_real_tensors\": false, \"trace.upload_tar\": null, \"triton.autotune_at_compile_time\": null, \"triton.autotune_cublasLt\": true, \"triton.autotune_pointwise\": true, \"triton.codegen_upcast_to_fp32\": true, \"triton.cooperative_reductions\": false, \"triton.cudagraph_dynamic_shape_warn_limit\": 50, \"triton.cudagraph_skip_dynamic_graphs\": false, \"triton.cudagraph_support_input_mutation\": true, \"triton.cudagraph_trees\": true, \"triton.cudagraph_trees_history_recording\": false, \"triton.cudagraph_unexpected_rerecord_limit\": 128, \"triton.cudagraphs\": false, \"triton.debug_sync_graph\": false, \"triton.debug_sync_kernel\": false, \"triton.dense_indexing\": false, \"triton.descriptive_names\": \"original_aten\", \"triton.divisible_by_16\": true, \"triton.enable_persistent_tma_matmul\": false, \"triton.fast_path_cudagraph_asserts\": false, \"triton.force_cooperative_reductions\": false, \"triton.force_cudagraph_sync\": false, \"triton.force_cudagraphs_warmup\": false, \"triton.inject_relu_bug_TESTING_ONLY\": null, \"triton.max_tiles\": 2, \"triton.min_split_scan_rblock\": 256, \"triton.multi_kernel\": 0, \"triton.persistent_reductions\": true, \"triton.prefer_nd_tiling\": false, \"triton.skip_cudagraph_warmup\": false, \"triton.skip_l1_cache\": false, \"triton.slow_path_cudagraph_asserts\": true, \"triton.spill_threshold\": 16, \"triton.store_cubin\": false, \"triton.tile_reductions\": false, \"triton.tiling_prevents_pointwise_fusion\": true, \"triton.tiling_prevents_reduction_fusion\": true, \"triton.unique_kernel_names\": true, \"triton.unique_user_kernel_names\": false, \"triton.use_block_ptr\": false, \"triton_kernel_default_layout_constraint\": \"needs_fixed_stride_order\", \"unbacked_symint_fallback\": 8192, \"unroll_reductions_threshold\": 8, \"unsafe_ignore_unsupported_triton_autotune_args\": false, \"use_experimental_benchmarker\": true, \"use_fast_math\": false, \"use_mixed_mm\": true, \"verbose_progress\": false, \"warn_mix_layout\": false, \"worker_start_method\": \"subprocess\"}", "remote_cache_version": null, "inductor_fx_remote_cache_hit_count": null, "inductor_fx_remote_cache_miss_count": null, "inductor_fx_remote_cache_backend_type": null, "inductor_fx_remote_cache_hit_keys": null, "inductor_fx_remote_cache_miss_keys": null, "cuda_version": "12.6", "triton_version": "3.2.0", "feature_usage": {"aot_autograd_remote_cache": false, "parallel_compile_post_warmup": false}, "compile_time_autotune_time_us": null, "is_runtime": false, "gc_time_us": 717, "tensorify_float_attempt": null, "tensorify_float_success": null, "tensorify_float_failure": null, "guard_latency_us": 0, "recompile_reason": null, "num_graph_breaks": null}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1}
V0320 11:46:31.632000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 1, "frame_compile_id": 0, "attempt": 1, "has_payload": "fac24d6cc3b51c00a703758ff500aee5"}
{
"name": "dynamo",
"ts": 1742496391632364.5,
"args": {
"compile_id": "1/0",
"distributed_ephemeral_timeout_us": 2488734,
"guard_latency_us": 0,
"frame_key": "2",
"co_name": "forward",
"co_filename": "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py",
"co_firstlineno": 43,
"cache_size": 0,
"accumulated_cache_size": 0,
"guard_count": 15,
"shape_env_guard_count": 0,
"graph_op_count": 7,
"graph_node_count": 9,
"graph_input_count": 1,
"fail_type": null,
"fail_reason": null,
"fail_user_frame_filename": null,
"fail_user_frame_lineno": null,
"non_compliant_ops": [],
"compliant_custom_ops": [],
"restart_reasons": [
"Unknown fail_and_restart_analysis"
],
"dynamo_time_before_restart_s": 0.01001882553100586,
"has_guarded_code": true,
"dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": true, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.632000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "caf166dcf1f7a19f66ee28fac406784a"}
{
"name": "cudagraphify",
"ts": 1742496391632788.5,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.633000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "d1c72b76b35cc2f9c878435018da1883"}
{
"name": "cudagraphify",
"ts": 1742496391633192.5,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.633000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "977e7e369dbd7c55a6f0323ac41a033e"}
{
"name": "CUDAGraphTreeManager.run_eager",
"ts": 1742496391633594.0,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.634000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "d2ae96f4f75c2b91a8f1431f8365b7c3"}
{
"name": "CUDAGraphTreeManager.run_eager",
"ts": 1742496391634243.2,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.636000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "78bdbf5ce489e5cdec7e2aeb6b65159f"}
{
"name": "dynamo",
"ts": 1742496391636104.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.636000 18434 site-packages/torch/_dynamo/convert_frame.py:1025] {"dynamo_start": {"stack": [{"line": 118, "name": "<module>", "filename": 1, "loc": "moe_c(input1)"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 1762, "name": "_call_impl", "filename": 2, "loc": "return forward_call(*args, **kwargs)"}, {"line": 586, "name": "_fn", "filename": 3, "loc": "return fn(*args, **kwargs)"}, {"line": 1751, "name": "_wrapped_call_impl", "filename": 2, "loc": "return self._call_impl(*args, **kwargs)"}, {"line": 1762, "name": "_call_impl", "filename": 2, "loc": "return forward_call(*args, **kwargs)"}, {"line": 32, "name": "torch_dynamo_resume_in_forward_at_32", "filename": 1}]}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.636000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "ae03d545bc98ec846ed7e025ac94e55c"}
{
"name": "entire_frame_compile",
"ts": 1742496391636824.0,
"args": {
"fn_name": "_compile.compile_inner",
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.638000 18434 site-packages/torch/_subclasses/meta_utils.py:248] {"describe_storage": {"id": 0, "describer_id": 8, "size": 48}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.638000 18434 site-packages/torch/_subclasses/meta_utils.py:462] {"describe_tensor": {"id": 0, "ndim": 2, "dtype": "torch.bfloat16", "device": "device(type='cuda', index=0)", "size": [6, 4], "is_leaf": true, "stride": [4, 1], "storage": 0, "view_func": "_CustomViewFunc(func=<built-in method _view_func_unsafe of Tensor object at 0x7fba64095950>)", "describer_id": 8}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.639000 18434 site-packages/torch/_subclasses/meta_utils.py:1869] {"describe_source": {"describer_id": 8, "id": 0, "source": "L['___stack0']"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.641000 18434 site-packages/torch/_dynamo/output_graph.py:1411] {"dynamo_output_graph": {"sizes": {"l_stack0_": [6, 4], "reshape": [1, 6, 4]}}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "16da9cd2d91f4294bcc9a62d14e80e1f"}
class GraphModule(torch.nn.Module):
def forward(self, L_stack0_: "bf16[6, 4][4, 1]cuda:0"):
l_stack0_ = L_stack0_
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:33 in torch_dynamo_resume_in_forward_at_32, code: return out.reshape(batch_size, num_tokens, self.dim)
reshape: "bf16[1, 6, 4][24, 4, 1]cuda:0" = l_stack0_.reshape(1, 6, 4); l_stack0_ = None
return (reshape,)
V0320 11:46:31.642000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "561b5da9fe6af11df2fa05a4dfd5ca26"}
{
"name": "backend_compile",
"ts": 1742496391642213.5,
"args": {
"fn_name": "OutputGraph.call_user_compiler",
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.642000 18434 site-packages/torch/_inductor/compile_fx.py:1965] {"inductor_pre_grad_graph": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "f154de758256f3bc5579e6dc3478378b"}
class GraphModule(torch.nn.Module):
def forward(self, L_stack0_: "bf16[6, 4][4, 1]cuda:0"):
l_stack0_ = L_stack0_
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:33 in torch_dynamo_resume_in_forward_at_32, code: return out.reshape(batch_size, num_tokens, self.dim)
reshape: "bf16[1, 6, 4][24, 4, 1]cuda:0" = l_stack0_.reshape(1, 6, 4); l_stack0_ = None
return (reshape,)
# graph id: 140438521768240
V0320 11:46:31.643000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "aa39414f4a053cdab48855911504d428"}
{
"name": "_recursive_pre_grad_passes",
"ts": 1742496391643028.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.643000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "e4d19a96627ad0a96143073580b1f2f7"}
{
"name": "_recursive_pre_grad_passes",
"ts": 1742496391643504.2,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.647000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "cbb96694a5c310d2097ff6256cf48458"}
{
"name": "create_aot_dispatcher_function",
"ts": 1742496391647584.2,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.654000 18434 site-packages/torch/_functorch/_aot_autograd/dispatch_and_compile_graph.py:214] {"artifact": {"name": "aot_forward_graph_fw_metadata", "encoding": "string"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "7619f92ea17c3deda482a8b9d5cd136a"}
ViewAndMutationMeta(input_info=[InputAliasInfo(is_leaf=True,
mutates_data=False,
mutates_metadata=False,
mutations_hidden_from_autograd=True,
mutations_under_no_grad_or_inference_mode=False,
mutation_inductor_storage_resize=False,
mutates_storage_metadata=False,
requires_grad=False,
keep_input_mutations=True)],
output_info=[OutputAliasInfo(output_type=<OutputType.alias_of_input: 2>,
raw_type=<class 'torch._subclasses.functional_tensor.FunctionalTensor'>,
base_idx=0,
dynamic_dims=set(),
requires_grad=False,
functional_tensor=<torch._functorch._aot_autograd.functional_utils.FunctionalTensorMetadataEq object at 0x7fba5da6d760>)],
num_intermediate_bases=0,
keep_input_mutations=True,
traced_tangents=[],
subclass_inp_meta=[PlainTensorMeta(unwrapped_idx=0,
memory_format=None)],
subclass_fw_graph_out_meta=[PlainTensorMeta(unwrapped_idx=0,
memory_format=None)],
subclass_tangent_meta=[],
is_train=False,
traced_tangent_metas=None,
num_symints_saved_for_bw=None,
grad_enabled_mutation=None,
deterministic=None,
static_input_indices=[],
tokens={},
indices_of_inputs_that_requires_grad_with_mutations_in_bw=[],
bw_donated_idxs=None,
num_backward_tokens=0,
num_graphsafe_rng_states=0,
graphsafe_rng_state_index=None)
V0320 11:46:31.655000 18434 site-packages/torch/_functorch/_aot_autograd/dispatch_and_compile_graph.py:232] {"aot_inference_graph": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "7242a297a068fa2b48230f18b5cb0ecd"}
class <lambda>(torch.nn.Module):
def forward(self, arg0_1: "bf16[6, 4][4, 1]cuda:0"):
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:33 in torch_dynamo_resume_in_forward_at_32, code: return out.reshape(batch_size, num_tokens, self.dim)
view: "bf16[1, 6, 4][24, 4, 1]cuda:0" = torch.ops.aten.view.default(arg0_1, [1, 6, 4]); arg0_1 = None
return (view,)
V0320 11:46:31.655000 18434 site-packages/torch/_functorch/_aot_autograd/jit_compile_runtime_wrappers.py:185] {"artifact": {"name": "torch._functorch.config", "encoding": "string"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "64d79258b64ad8a5b3d76727b5c240ae"}
{
"TYPE_CHECKING": false,
"functionalize_rng_ops": false,
"fake_tensor_allow_meta": true,
"debug_assert": false,
"debug_partitioner": false,
"decompose_custom_triton_ops": true,
"static_weight_shapes": true,
"cse": true,
"enable_autograd_cache": true,
"enable_remote_autograd_cache": null,
"view_replay_for_aliased_outputs": true,
"max_dist_from_bw": 1000,
"ban_recompute_used_far_apart": true,
"ban_recompute_long_fusible_chains": true,
"ban_recompute_materialized_backward": true,
"ban_recompute_not_in_allowlist": true,
"ban_recompute_reductions": true,
"recompute_views": false,
"activation_memory_budget": 1.0,
"activation_memory_budget_runtime_estimator": "flops",
"activation_memory_budget_solver": "dp",
"visualize_memory_budget_pareto": false,
"aggressive_recomputation": false,
"fake_tensor_allow_unsafe_data_ptr_access": true,
"unlift_effect_tokens": true,
"fake_tensor_crossref": false,
"fake_tensor_propagate_real_tensors": false,
"donated_buffer": true,
"torch_compile_graph_format": "svg",
"generate_fake_kernels_from_real_mismatches": false,
"graphsafe_rng_functionalization": true,
"strict_autograd_cache": false,
"disable_guess_zero_tangent_for_mutated_input_subclass": false
}
V0320 11:46:31.655000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "b813d21c5639d80f2afc11323564bc2e"}
{
"name": "compile_fx.<locals>.fw_compiler_base",
"ts": 1742496391655870.0,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.656000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "7cb827eeaeeae91158e06650a6c5d6db"}
{
"name": "_recursive_joint_graph_passes",
"ts": 1742496391656278.2,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.657000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "608d3f098f1fc75531d5c042a469f7db"}
{
"name": "_recursive_joint_graph_passes",
"ts": 1742496391657946.0,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.658000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "7980c80723811cdfc7faf6ffda871c44"}
{
"name": "inductor_compile",
"ts": 1742496391658323.8,
"args": {
"fn_name": "compile_fx_inner",
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.665000 18434 site-packages/torch/_inductor/compile_fx.py:934] {"artifact": {"name": "fx_graph_runnable", "encoding": "string"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "f352d0e17438b9a57ce362b4102ad557"}
import os
os.environ['TORCH_CUDA_ARCH_LIST'] = '8.0'
os.environ['TORCH_SHOW_CPP_STACKTRACES'] = '0'
os.environ['TORCHDYNAMO_CAPTURE_SCALAR_OUTPUTS'] = '1'
os.environ['TORCH_TRACE'] = '/home/cdhernandez/tl_parse_dir'
os.environ['TORCHINDUCTOR_CACHE_DIR'] = '/tmp/torchinductor_cdhernandez'
os.environ['TRITON_CACHE_DIR'] = '/tmp/torchinductor_cdhernandez/triton/0'
import torch
from torch import tensor, device
import torch.fx as fx
from torch._dynamo.testing import rand_strided
from math import inf
import torch._inductor.inductor_prims
import torch._dynamo.config
import torch._inductor.config
import torch._functorch.config
import torch.fx.experimental._config
torch._inductor.config.triton.cudagraphs = True
torch._functorch.config.functionalize_rng_ops = False
torch._functorch.config.fake_tensor_allow_unsafe_data_ptr_access = True
torch._functorch.config.unlift_effect_tokens = True
isolate_fails_code_str = None
# torch version: 2.7.0.dev20250304+cu126
# torch cuda version: 12.6
# torch git version: d13587dfbffb169f02bb1e660256fdb186bd9e72
# CUDA Info:
# nvcc: NVIDIA (R) Cuda compiler driver
# Copyright (c) 2005-2024 NVIDIA Corporation
# Built on Tue_Oct_29_23:50:19_PDT_2024
# Cuda compilation tools, release 12.6, V12.6.85
# Build cuda_12.6.r12.6/compiler.35059454_0
# GPU Hardware Info:
# NVIDIA H100 : 8
from torch.nn import *
class Repro(torch.nn.Module):
def __init__(self) -> None:
super().__init__()
def forward(self, arg0_1):
view = torch.ops.aten.view.default(arg0_1, [1, 6, 4]); arg0_1 = None
return (view,)
def load_args(reader):
buf0 = reader.storage(None, 48, device=device(type='cuda', index=0), dtype_hint=torch.bfloat16)
reader.tensor(buf0, (6, 4), dtype=torch.bfloat16, is_leaf=True) # arg0_1
load_args._version = 0
mod = Repro()
if __name__ == '__main__':
from torch._dynamo.repro.after_aot import run_repro
with torch.no_grad():
run_repro(mod, load_args, accuracy=False, command='run', save_dir=None, tracing_mode='real', check_str=None)
# To run it separately, do
# mod, args = run_repro(mod, load_args, accuracy=False, command='get_args', save_dir=None, tracing_mode='real', check_str=None)
# mod(*args)
V0320 11:46:31.667000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "15ee6953b0648dda8f19a62ea7ec18dd"}
{
"name": "_recursive_post_grad_passes",
"ts": 1742496391667138.8,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.674000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "51059ff62a45def9d2c902b668afb355"}
{
"name": "_recursive_post_grad_passes",
"ts": 1742496391674815.8,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.675000 18434 site-packages/torch/_inductor/compile_fx.py:997] {"inductor_post_grad_graph": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "a4d3740ffce8edd3fee0aa767b3791a4"}
class <lambda>(torch.nn.Module):
def forward(self, arg0_1: "bf16[6, 4][4, 1]cuda:0"):
# File: /data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py:33 in torch_dynamo_resume_in_forward_at_32, code: return out.reshape(batch_size, num_tokens, self.dim)
view: "bf16[1, 6, 4][24, 4, 1]cuda:0" = torch.ops.aten.reshape.default(arg0_1, [1, 6, 4]); arg0_1 = None
return (view,)
V0320 11:46:31.680000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "4f755e27ceec06c28f89d84c7ea1c3ec"}
{
"name": "GraphLowering.run",
"ts": 1742496391680461.2,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.681000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "ab7ca24f35b70a19b23fbc884bb382d5"}
{
"name": "GraphLowering.run",
"ts": 1742496391681875.5,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.682000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "30f5c352916b5ee6d0e6268089652100"}
{
"name": "GraphLowering.compile_to_fn",
"ts": 1742496391682262.2,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.682000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "fa180604fe20e62f9a25ea1c4253bac8"}
{
"name": "code_gen",
"ts": 1742496391682567.2,
"args": {
"fn_name": "GraphLowering.compile_to_module",
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.682000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "2acf1a5284c3c8d9042ddb29b1077f45"}
{
"name": "GraphLowering.codegen",
"ts": 1742496391682859.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.684000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "d0aabdc06f25a611234c52cbeff05366"}
{
"name": "Scheduler.__init__",
"ts": 1742496391684083.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.685000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "608e8a368989c372314b725742602b53"}
{
"name": "Scheduler.fused_nodes",
"ts": 1742496391684991.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.685000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "b55e93a93f125d1ac1bd69924f829596"}
{
"name": "Scheduler.fused_nodes",
"ts": 1742496391685634.2,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.688000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "67d913216fdc5d957aea8eda5c8f254e"}
{
"name": "Scheduler.__init__",
"ts": 1742496391688181.2,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.689000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "824753faa1f3bacda51f7c1928640035"}
{
"name": "Scheduler.codegen",
"ts": 1742496391688979.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.689000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "3c365648a1b318941d39b91108edcaaa"}
{
"name": "Scheduler.codegen",
"ts": 1742496391689541.2,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.690000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "69efa39edcad11543eb6b337122f5335"}
{
"name": "PythonWrapperCodegen.generate",
"ts": 1742496391690036.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.690000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "979df08a7c5dcb6cc3e30c731c896508"}
{
"name": "PythonWrapperCodegen.generate",
"ts": 1742496391690862.8,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.691000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "f521a15fff641b47b48ac51823a7201b"}
{
"name": "GraphLowering.codegen",
"ts": 1742496391691399.8,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.691000 18434 site-packages/torch/_inductor/graph.py:2102] {"inductor_output_code": {"filename": "/tmp/torchinductor_cdhernandez/5p/c5psi7ayigmrebf76mn3eebs23tyqdo2uox7jispav4owoxk2h4n.py"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "ad3b8af7916ca276607f1fb8ee4d7a07"}
# AOT ID: ['2_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from cmath import nanj
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
empty_strided_p2p = torch._C._distributed_c10d._SymmetricMemory.empty_strided_p2p
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (6, 4), (4, 1))
return (reinterpret_tensor(arg0_1, (1, 6, 4), (24, 4, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((6, 4), (4, 1), device='cuda:0', dtype=torch.bfloat16)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
V0320 11:46:31.692000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "f5a8d9ff25fb85866781ccb6812acdb2"}
{
"name": "PyCodeCache.load_by_key_path",
"ts": 1742496391692205.2,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.693000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "f91940b7dd34c8d70b241739fb1c250f"}
{
"name": "async_compile.wait",
"ts": 1742496391692993.8,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.693000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "b0a94d9ba25ec94e301bc238869c3f29"}
{
"name": "async_compile.wait",
"ts": 1742496391693867.2,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.694000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "fa358a71656e542dc80f64cef60a9fc9"}
{
"name": "PyCodeCache.load_by_key_path",
"ts": 1742496391694480.5,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.694000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "9fb1c405d30d772addd72b3569d696e2"}
{
"name": "code_gen",
"ts": 1742496391694882.8,
"args": {
"fn_name": "GraphLowering.compile_to_module",
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.695000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "bd966890d171c5a25b118702a5efd28b"}
{
"name": "GraphLowering.compile_to_fn",
"ts": 1742496391695151.5,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.697000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "8e8f2af591d44bd27c079c905316ef58"}
{
"name": "TritonBundler.collect",
"ts": 1742496391697042.0,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.697000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "3252606eea86979da4987dd08558a76d"}
{
"name": "TritonBundler.collect",
"ts": 1742496391697404.0,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.698000 18434 site-packages/torch/_dynamo/utils.py:1821] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "5a6953286c70d06e09cdb2900e3ea1d0"}
{
"name": "fx_graph_cache_miss",
"ts": 1742496391658824.8,
"args": {
"key": "fcfvuxfnee47bbhr5teznyrrxrmc6qnd4fz6wi7a4enco7y2evak",
"components": [
"[xdmm3lnwfgoh27blh34weneseyvwq6xjm4ktxoshjnexqzyazls] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1):\n view = torch.ops.aten.view.default(arg0_1, [1, 6, 4]); arg0_1 = None\n return (view,)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[mnxeqpr2t7dsiq37scjai7m4zrevtkj5q3e4g4atbx3e3sraox6] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([6, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] fx_kwargs[static_input_idxs]: []",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_event_time": 1742496391661868687,
"cache_state": "miss",
"triton_bundler_meta": "TritonBundlerMetadata(cached_kernel_names=[])",
"time_taken_ns": 38196366,
"compile_id": "2/0"
},
"ph": "i",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0,
"s": "p"
}
V0320 11:46:31.698000 18434 site-packages/torch/_inductor/compile_fx.py:827] {"artifact": {"name": "fx_graph_cache_miss", "encoding": "json"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "3226c6e8b88ea5f1e9bfd2c50e2b1999"}
{"key": "fcfvuxfnee47bbhr5teznyrrxrmc6qnd4fz6wi7a4enco7y2evak", "components": ["[xdmm3lnwfgoh27blh34weneseyvwq6xjm4ktxoshjnexqzyazls] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1):\n view = torch.ops.aten.view.default(arg0_1, [1, 6, 4]); arg0_1 = None\n return (view,)\n \n# To see more debug info, please use `graph_module.print_readable()`", "[mnxeqpr2t7dsiq37scjai7m4zrevtkj5q3e4g4atbx3e3sraox6] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([6, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ", "[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False", "[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] fx_kwargs[static_input_idxs]: []", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0", "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)", "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)", "[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>", "[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}", "[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}", "[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False", "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True", "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}", "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False", "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True", "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default", "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300", "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False", "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP", "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True", "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False", "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0", "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0", "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ", "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT", "[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False", "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4", "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1", "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128", "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False", "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None", "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096", "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False", "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128", "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True", "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None", "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None", "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []", "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']", "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False", "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp", "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton", "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host", "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda", "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021", "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"], "cache_event_time": 1742496391661868687, "cache_state": "miss", "triton_bundler_meta": "TritonBundlerMetadata(cached_kernel_names=[])", "time_taken_ns": 38196366, "compile_id": "2/0"}
V0320 11:46:31.699000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "c7ca3838c76d967ccdb24f262916ec59"}
{
"name": "inductor_compile",
"ts": 1742496391699238.5,
"args": {
"fn_name": "compile_fx_inner",
"compile_id": "2/0",
"is_backward": false,
"cache_state": "miss",
"cache_event_time": 1742496391658824674,
"key": "fcfvuxfnee47bbhr5teznyrrxrmc6qnd4fz6wi7a4enco7y2evak",
"components": [
"[xdmm3lnwfgoh27blh34weneseyvwq6xjm4ktxoshjnexqzyazls] gm: <lambda>()\n\n\n\ndef forward(self, arg0_1):\n view = torch.ops.aten.view.default(arg0_1, [1, 6, 4]); arg0_1 = None\n return (view,)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[mnxeqpr2t7dsiq37scjai7m4zrevtkj5q3e4g4atbx3e3sraox6] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([6, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[lmglpn4zi7vob56n34r2j2rk7flv5xfgrcvmo7xcpirqsitygqx] fx_kwargs[boxed_forward_device_index]: BoxedDeviceIndex(value=None)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[cpp_wrapper]: False",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[extern_node_serializer]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] fx_kwargs[is_backward]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] fx_kwargs[is_inference]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] fx_kwargs[layout_opt]: None",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] fx_kwargs[static_input_idxs]: []",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inputs_to_check[0]: 0",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_bypass_reason": null,
"remote_cache_enabled": false,
"local_cache_enabled": true
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.699000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "1665d37ab6ce040811fe97c5a3734859"}
{
"name": "compile_fx.<locals>.fw_compiler_base",
"ts": 1742496391699943.5,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.701000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "b06601e13d8aaddfe6181310e62b98ba"}
{
"name": "create_aot_dispatcher_function",
"ts": 1742496391701816.5,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.702000 18434 site-packages/torch/_dynamo/utils.py:1821] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "37a4b34b2f9bdb1674e519c7eda5f49c"}
{
"name": "autograd_cache_miss",
"ts": 1742496391647566.8,
"args": {
"key": "as4rt2cudo6yxkjgccxoo645fncgkonhj6mrrbztfkct5lj2d4yk",
"cache_state": "miss",
"components": [
"[awhgiqomghe2o6qosp7uujukp43cqtz4cjrjrtiz7eeuj5xmxwz] aot_config: (0, True, False, False, False, [LocalSource(local_name='___stack0', is_input=True, is_derefed_cell_contents=False)], True, False)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False",
"[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>",
"[whawwbj2hebcdz7zb27x3t2mexvzmiinc36efyaebz56oduurrp] gm: GraphModule()\n\n\n\ndef forward(self, L_stack0_ : torch.Tensor):\n l_stack0_ = L_stack0_\n reshape = l_stack0_.reshape(1, 6, 4); l_stack0_ = None\n return (reshape,)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[mnxeqpr2t7dsiq37scjai7m4zrevtkj5q3e4g4atbx3e3sraox6] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([6, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"compile_id": "2/0"
},
"ph": "i",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0,
"s": "p"
}
V0320 11:46:31.702000 18434 site-packages/torch/_functorch/_aot_autograd/autograd_cache.py:804] {"artifact": {"name": "aotautograd_cache_miss", "encoding": "json"}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "c82392832d0683709e2a6d4dd648012c"}
{"key": "as4rt2cudo6yxkjgccxoo645fncgkonhj6mrrbztfkct5lj2d4yk", "cache_state": "miss", "components": ["[awhgiqomghe2o6qosp7uujukp43cqtz4cjrjrtiz7eeuj5xmxwz] aot_config: (0, True, False, False, False, [LocalSource(local_name='___stack0', is_input=True, is_derefed_cell_contents=False)], True, False)", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False", "[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>", "[whawwbj2hebcdz7zb27x3t2mexvzmiinc36efyaebz56oduurrp] gm: GraphModule()\n\n\n\ndef forward(self, L_stack0_ : torch.Tensor):\n l_stack0_ = L_stack0_\n reshape = l_stack0_.reshape(1, 6, 4); l_stack0_ = None\n return (reshape,)\n \n# To see more debug info, please use `graph_module.print_readable()`", "[mnxeqpr2t7dsiq37scjai7m4zrevtkj5q3e4g4atbx3e3sraox6] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([6, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ", "[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)", "[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)", "[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)", "[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>", "[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}", "[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}", "[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order", "[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False", "[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True", "[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}", "[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False", "[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True", "[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default", "[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300", "[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False", "[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP", "[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True", "[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False", "[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0", "[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0", "[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ", "[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm", "[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT", "[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False", "[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4", "[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1", "[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False", "[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None", "[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128", "[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False", "[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None", "[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096", "[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False", "[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True", "[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128", "[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False", "[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False", "[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True", "[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False", "[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ", "[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False", "[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2", "[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None", "[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None", "[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None", "[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None", "[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []", "[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']", "[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True", "[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False", "[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp", "[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton", "[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host", "[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda", "[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021", "[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False", "[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None", "[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None", "[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"], "compile_id": "2/0"}
V0320 11:46:31.703000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "25a0710a8a6d13b8daee0f55e0c8a166"}
{
"name": "backend_compile",
"ts": 1742496391703244.2,
"args": {
"fn_name": "OutputGraph.call_user_compiler",
"compile_id": "2/0",
"requires_subclass_dispatch": false,
"dispatch_mode": "inference",
"cache_state": "miss",
"cache_event_time": 1742496391647566726,
"key": "as4rt2cudo6yxkjgccxoo645fncgkonhj6mrrbztfkct5lj2d4yk",
"components": [
"[awhgiqomghe2o6qosp7uujukp43cqtz4cjrjrtiz7eeuj5xmxwz] aot_config: (0, True, False, False, False, [LocalSource(local_name='___stack0', is_input=True, is_derefed_cell_contents=False)], True, False)",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] grad_enabled: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] disable_amp: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] deterministic_algorithms: False",
"[4vsokyzruw6gtx7f4unng6zysynpvihwdqjimute22pv5vahzue] autograd_config: <bytes>",
"[whawwbj2hebcdz7zb27x3t2mexvzmiinc36efyaebz56oduurrp] gm: GraphModule()\n\n\n\ndef forward(self, L_stack0_ : torch.Tensor):\n l_stack0_ = L_stack0_\n reshape = l_stack0_.reshape(1, 6, 4); l_stack0_ = None\n return (reshape,)\n \n# To see more debug info, please use `graph_module.print_readable()`",
"[mnxeqpr2t7dsiq37scjai7m4zrevtkj5q3e4g4atbx3e3sraox6] example_inputs[0]: TensorMetadata(dtype=torch.bfloat16, shape=torch.Size([6, 4]), stride=(4, 1), device=device(type='cuda', index=0), layout=torch.strided, memory_format=torch.contiguous_format, storage_offset=0, storage_bytes=None, requires_grad=False, is_quantized=False, is_conj=False, is_neg=False, is_inference=False, is_sparse=False, is_coalesced=None, dense_dim=None, sparse_dim=None)",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] cache_key_tag: ",
"[moqyx73pd52qgfefpqfwjkxjzf4hj25dyzowq4prptxxsmfw2vg] fx_kwargs[cudagraphs]: BoxedBool(value=True)",
"[du4vyrfyozrfxcf6kk6ma7oqwatapifazeelfsawmsiu6gjdtxp] deterministic_algorithms_settings: (False, False, True)",
"[qiptf2633zubseuei4bkisoq3not35l6lud6p23p4qmcsxiw2uq] cuda_matmul_settings: (False, True, True)",
"[tksgklcqguoxgz5l6pbyor5fyut6nbzoxa5y2d4xzybucylsgep] torch_version: <bytes>",
"[poglqjwowp4gnkmehjby2lvdjrwuo5tbxa2gayd6smgasl2hgsd] system_info[device]: {'name': 'NVIDIA H100'}",
"[eskkzcpfmeu7knqfrfq5omiy5ohm6gipq6pazw3ot6wfd3nyjo7] system_info[version]: {'triton': '3.2.039ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-83cd0eef733233d69c83b5cd7dc1e0aeae88e1b36011de7f2d8ee0629def3ec1-39ace39686dc8ae6b537d14ae140575242b5833746b519e6e329b355f5d9e364-23d635e690d670bf61798e1259674b78c0ed5ba222ab6a455f329f27a758fc2d-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-2598a211b07d5b0b37b6353a297af274ab6539023b4ae93e99360f95a7dad109-417ffa3f51a8f5a8dd947662e07bf79c2b5d1f20c0ea8f3f94cdcd327947d669-e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855-ea8d8a4744679bb406b914485b1743669eb685f24949cd210f8fb72c103d2e46-747b5e566049ac66b1bbb28a97a3dbce27460dd1183d43b663b3ea5fedfdf436-467b09d6ea706a4813bfca3aed14f0090fbfcbf7e30ab542cb86be4137972f6c-6e4a7df0c1f6cb799f488ee1d6efd3df754fc18aac6e7209923bb022c36c7c4e-6bd038079b94067d628eea9264530fec03d4d7027ff834e89b9d6fbc31ae8bd9-5d15c5bebef8d7aa51b21fd187e5faa95eba4a213254355bc69e0648013599f7-26a8aeaf21e759191f173b578316e243ad970387442a1b238481c6a1b2eecdc4-170b6c22d072485ef44734b7a8c2ab95c771ce78c5a26e890f8d103acfd17cd5-72bc1771d8b160fbafcd5390d1ee6cb72440c10ad4834451465a9e93b42f5d1c-7b506c955ee5646862dae548968e6523d75c37ad4750c214d9ab0f6918ecc88a-89e4844782de5fdff4868ac2846514570a6e280a6b92e91af2e180904043c833-115ada51f797cd098ddc5e4b95e8369a885571b15639694b6801f59e46eab55e-0e48b5e1e95136642ccfe62dc3d0a739a2c20a7b5ee13e9c23c6cecd68cdeb70-2e4af7e3ca937a7dd4c7207eb6ee3f8f7ebc2f30912b52a7204eeac040bd2a0f-b3aeb8ae6cb1e94085254a3c33612134752c72444ec219965dd7fa52e010a762-c89868706a4edff5f861293d19345031a95b843b7b7a1f523b30ead360e3837e-34ca3a350389b7cd59c72f6cf94f68e319a0e4394a858d381fb59344a7cc052e', 'cuda': '12.6'}",
"[ux7o4nonjsze4jm6seusgn5qk65ezpeuahecvd32ppgwbwkfvix] system_info[hash]: 7317a65ff551a72a4745171d67351cb23ae4e49ff28466d2bdbb30d72ea6f849",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[TYPE_CHECKING]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[can_inplace_pad_graph_input]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[enable_auto_functionalized_v2]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_progress]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[verbose_progress]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[fx_graph_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[fx_graph_remote_cache]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bundle_triton_into_fx_graph_cache]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_local_cache]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[autotune_remote_cache]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[bundled_autotune_remote_cache]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_disable_caches]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[sleep_sec_TESTING_ONLY]: None",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[custom_op_default_layout_constraint]: needs_fixed_stride_order",
"[pikr7bbcoixfzftsazp5ggufhdklj24babfry77bl4nuvyrrcp4] inductor_config[triton_kernel_default_layout_constraint]: needs_fixed_stride_order",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp_wrapper]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[dce]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[static_weight_shapes]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[size_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[nan_asserts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[scalar_asserts]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pick_loop_orders]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[inplace_buffers]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[allow_buffer_reuse]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[memory_planning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[use_fast_math]: False",
"[x75won4jmsgeb63pcvwr2y4eteyzzdhmf5rv6xhjppie4hx2yu5] inductor_config[memory_pool]: intermediates",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_harness]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[epilogue_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[prologue_fusion]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[epilogue_fusion_first]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[pattern_matcher]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[b2b_gemm_pass]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_pre_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[joint_custom_post_pass]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[pre_grad_custom_pass]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_cat_fx_passes]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[efficient_conv_bn_eval_fx_passes]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[is_predispatch]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[group_fusion]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[batch_fusion]: True",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[pre_grad_fusion_options]: {}",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[post_grad_fusion_options]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_locality]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[dynamic_scale_rblock]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_fuse_int_mm_with_mul]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_mixed_mm]: True",
"[zwmmbkdkarexuhbigurz5lfnhx64tht7fznecjkrvznh6rzivbv] inductor_config[fx_passes_numeric_check]: {'pre_grad': False, 'precision': 0.0001, 'num_iterations': 1, 'requires_optimizer': True}",
"[v2td5s4lnsvyxvaevy4chx6kc5h3mm2axazbgwimqule5zrzao7] inductor_config[mixed_mm_choice]: heuristic",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[reorder_for_compute_comm_overlap]: False",
"[ssupi7bu3rrhdpg2jyegzncu3kg3nnhklyliqvutaxgs7y7k3dx] inductor_config[reorder_for_compute_comm_overlap_passes]: ['reorder_compute_for_overlap', 'sink_waits', 'raise_comms']",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[reorder_for_peak_memory]: True",
"[lxxtoqhcoepwfokeiibd575gnxo3uzwiv4hmpomlwkpzqz3qzsh] inductor_config[estimate_op_runtime]: default",
"[yezuzjtg4h3jjur4jwtwiehbyixa7eonq4tqsqmwqve2lvvmrem] inductor_config[intra_node_bw]: 300",
"[5fxczt3ciyxitdhizb7sfsgn7fhpczcqsngttnt5ot2wyctk7co] inductor_config[inter_node_bw]: 25",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[use_experimental_benchmarker]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_pointwise]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[max_autotune_gemm]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[autotune_num_choices_displayed]: 10",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[graph_partition]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_same_precision]: False",
"[2y7luesktjrque3nr7qtxnum2mkbeegzdrsvkm3rvdlhqboajhx] inductor_config[max_autotune_gemm_backends]: ATEN,TRITON,CPP",
"[uqlsbif4zxd75vt522p52txyuguieipi2lwz5g5awt56lccqk7s] inductor_config[max_autotune_conv_backends]: ATEN,TRITON",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[max_autotune_gemm_search_space]: DEFAULT",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[autotune_fallback_to_aten]: True",
"[wft6ljqsfr3x4m7fa5zuyb7cwknky4irrxz4bjr6uzr2yiopxqj] inductor_config[unbacked_symint_fallback]: 8192",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[search_autotune_cache]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[save_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_in_subproc]: False",
"[iglov24t7x5ruci344aer2tm6nqshi4veuw4wxlssxtu46cx76m] inductor_config[max_autotune_subproc_result_timeout_seconds]: 60.0",
"[bh33ranllcgilhgmgr3qvygzxjm6isq5iexnfm3zx6fnr2zwlp2] inductor_config[max_autotune_subproc_graceful_timeout_seconds]: 1.0",
"[pwoh5aypf4fxbntdvwt67rppxorqos6xr3w7qzeun6kblbfg2ga] inductor_config[max_autotune_subproc_terminate_timeout_seconds]: 2.0",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[autotune_multi_device]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_tuning]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[coordinate_descent_check_all_directions]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[coordinate_descent_search_radius]: 1",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[autoheuristic_collect]: ",
"[jwbrgxes7vjqumngs5hyj6gn5nytv2whnppnzngvaagfmawhkkd] inductor_config[autoheuristic_use]: mixed_mm",
"[jvchmi66fvqzlemhr5fcqorz5trfdtdalzfagtj2aolmimwqhdq] inductor_config[autoheuristic_log_path]: DEFAULT",
"[4p2fdjlvxrcw7c7fvzm5huhtqxnro4kvkx56f7p5zyrxqkwooov] inductor_config[layout_opt_default]: 1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[layout_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_layout_optimization]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[keep_output_stride]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[warn_mix_layout]: False",
"[lkkae3meylaixfif4thncru4hjqeaislawjoghffrbwuscaagei] inductor_config[realize_reads_threshold]: 4",
"[rr5m5hsocoyodldz7vcvaizdwvm2rt34evmqdxvng7wz3tufvo6] inductor_config[realize_opcount_threshold]: 30",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[realize_acc_reads_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[fallback_random]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[implicit_fallbacks]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aggressive_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_fusion]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[enabled_metric_tables]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[loop_ordering_after_fusion]: False",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[score_fusion_memory_threshold]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[benchmark_epilogue_fusion]: True",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[max_epilogue_benchmarked_choices]: 1",
"[jykiys6ynafs3zdylwa5ggq6j655mxeh42d6mtdi22gffkrmiac] inductor_config[max_fusion_size]: 64",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[max_pointwise_cat_inputs]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_pointwise_cat]: False",
"[yttmfmxblgcbsvbokguzowcorrcxz5uunxtcvsbe6nijgcx45he] inductor_config[unroll_reductions_threshold]: 8",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[comment_origin]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[conv_1x1_as_mm]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[split_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_kernel]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[constant_and_index_propagation]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[always_keep_tensor_constants]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[assert_indirect_indexing]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[compute_all_bounds]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernels]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[benchmark_combo_kernel]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernels_autotune]: 1",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[combo_kernel_allow_mixed_sizes]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[combo_kernel_foreach_dynamic_shapes]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[joint_graph_constant_folding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_index_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[emulate_precision_casts]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[is_nightly_or_source]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[developer_warnings]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[optimize_scatter_upon_const_tensor]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[add_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[remove_pre_grad_passes]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[global_cache_dir]: None",
"[j6c55jha5r2sdys2rwq7uqhtleea5dgjcye7nicfgft36v7xfvp] inductor_config[kernel_name_max_ops]: 10",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[shape_padding]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[comprehensive_padding]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_channels_last]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[disable_padding_cpu]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[padding_alignment_bytes]: 128",
"[dnnw5ks3yxrp7mwvihb2hh4tqx35ye637xt33x64kw4fvz2nyzg] inductor_config[padding_stride_threshold]: 1024",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[pad_outputs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[bw_outputs_user_visible]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[force_shape_pad]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[permute_fusion]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profiler_mark_wrapper_call]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[generate_intermediate_hooks]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[debug_ir_traceback]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[profile_bandwidth_regex]: ",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[profile_bandwidth_output]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[profile_bandwidth_with_do_bench_using_profiling]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[disable_cpp_codegen]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[freezing_discard_parameters]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[decompose_mem_bound_mm]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[assume_aligned_inputs]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[unsafe_ignore_unsupported_triton_autotune_args]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[check_stack_no_cycles_TESTING_ONLY]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[enable_linear_binary_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[annotate_training]: False",
"[sz3im5ogc6asp7g4uqocnovype63tkdexzfrniv6hn2oank3biu] inductor_config[cpp.threads]: -1",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.no_redundant_loops]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.dynamic_threads]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.simdlen]: None",
"[g7rrnbg5yonzux3cfj5ovre5lob3ayda7qcfpxjvtwmiz4uicii] inductor_config[cpp.min_chunk_size]: 4096",
"[c7zj4qytmety6keurs3hsh5wn7foxp3dqx4kym2ucszzcb2ngrf] inductor_config[cpp.cxx]: (None, 'g++')",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_kernel_profile]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.weight_prepack]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_relu_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.inject_log1p_bug_TESTING_ONLY]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.vec_isa_ok]: None",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[cpp.descriptive_names]: original_aten",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[cpp.max_horizontal_fusion_size]: 16",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.fallback_scatter_reduce_sum]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_unsafe_math_opt_flag]: False",
"[ijs44lspkinjvhcs7uff7n3noc53jvsp4yfljjh22mafhb7khxe] inductor_config[cpp.enable_floating_point_contract_flag]: off",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_tiling_heuristics]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_grouped_gemm_template]: False",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cpp.gemm_max_k_slices]: 1",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_cache_blocking]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cpp.gemm_thread_factors]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[cpp.enable_loop_tail_vec]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cpp.enable_concat_linear]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraphs]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_trees]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_skip_dynamic_graphs]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.slow_path_cudagraph_asserts]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cudagraph_trees_history_recording]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.cudagraph_support_input_mutation]: True",
"[ljdqgtysl3vdf7j6attlz5gmjg2ncihnveojfyubosplmkrjgra] inductor_config[triton.cudagraph_unexpected_rerecord_limit]: 128",
"[tuax46wac7rfv2trf5gcps6vleo3cq44lbnrdxtprvo3ljjaddj] inductor_config[triton.cudagraph_dynamic_shape_warn_limit]: 50",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraph_sync]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cudagraphs_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.fast_path_cudagraph_asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_cudagraph_warmup]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_graph]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.debug_sync_kernel]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.dense_indexing]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[triton.max_tiles]: 2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.prefer_nd_tiling]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_pointwise]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.autotune_cublasLt]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.autotune_at_compile_time]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.tile_reductions]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_pointwise_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.tiling_prevents_reduction_fusion]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.unique_kernel_names]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.unique_user_kernel_names]: False",
"[yrty22bseefglnysuoec4ji7j2rnaggdj3g33zzj7avogwfmgdw] inductor_config[triton.descriptive_names]: original_aten",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.persistent_reductions]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.cooperative_reductions]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.force_cooperative_reductions]: False",
"[vrl5ktomgtzox5xucd3np6vug3vyj6hwwzahqijuwpmamlv7ohi] inductor_config[triton.multi_kernel]: 0",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.divisible_by_16]: True",
"[fv6slhtedtydps5s5u2etitscliblzcidyitqf7krsv4e23fzk6] inductor_config[triton.min_split_scan_rblock]: 256",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.store_cubin]: False",
"[ebt2ncs4f5y7dn7btzi76mnouepvzad474tmp5iju4wiuumjl4s] inductor_config[triton.spill_threshold]: 16",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.use_block_ptr]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[triton.inject_relu_bug_TESTING_ONLY]: None",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[triton.codegen_upcast_to_fp32]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.enable_persistent_tma_matmul]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[triton.skip_l1_cache]: False",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.output_path]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.debug_compile]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.compile_wrapper_with_O0]: False",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[aot_inductor.debug_intermediate_value_printer]: 0",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[aot_inductor.filtered_kernel_names]: None",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_in_spec]: ",
"[v3hzzlv4tjgvp3pyhmzagjd25orl6n7nynoa7svlhhwk73b7u3c] inductor_config[aot_inductor.serialized_out_spec]: ",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_runtime_constant_folding]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.force_mmap_weights]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.package_cpp_only]: False",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.metadata]: {}",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.raise_error_on_ignored_optimization]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.dump_aoti_minifier]: False",
"[pr5nr4a7dthirgd2ljo3d2xakc63ywxugusu6mkmr6gmpeliyib] inductor_config[aot_inductor.repro_level]: 2",
"[4bryyl4ahh5whyg3zwqebpwmjnx6w77nqgqbdjlowju6lkqtn7w] inductor_config[aot_inductor.presets]: {}",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.allow_stack_allocation]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[aot_inductor.use_minimal_arrayref_interface]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[aot_inductor.package_constants_in_so]: True",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.arch]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.version]: None",
"[tvyftmtdmezlejo2xllu7awzv4pzc4vm4fub4b3gpl5jptjkosi] inductor_config[cuda.compile_opt_level]: -O1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_cuda_lto]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_ptxas_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.enable_debug_info]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.use_fast_math]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_max_profiling_configs]: None",
"[xaicuyqjstadzph6cgvxowlzizkts6kzmfupsnbyaorxh37cppz] inductor_config[cuda.cutlass_max_profiling_swizzle_options]: [1, 2, 4]",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cuda_cxx]: None",
"[aghvyrrgwvxijco2pk5wzc3cgmmthrbmgxitiibxuuscxdwrjd3] inductor_config[cuda.cutlass_backend_min_gemm_size]: 1",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[cuda.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_allowlist_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[cuda.cutlass_op_denylist_regex]: None",
"[ngkkx5e6z7erl6da23zb2cmsctz4yvaqyameyg5hbqln4wrhh7x] inductor_config[cuda.cutlass_instantiation_level]: 0",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[rocm.arch]: []",
"[3w3j4h2iiu3addwyb5alaeecz3so7teb23hp4d5n3b46w5n73ur] inductor_config[rocm.ck_supported_arch]: ['gfx90a', 'gfx942']",
"[klfqjprnpfhcdurgvuikvc4rpd5ynkpk77toousr5h3u5roty6p] inductor_config[rocm.compile_opt_level]: -O2",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.is_debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.save_temps]: False",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.use_fast_math]: True",
"[cev5uo2jlwdhw2uyzcm7vr6cl23azjfw437f5r5lskm7spucos6] inductor_config[rocm.flush_denormals]: True",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.print_kernel_resource_usage]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.rocm_home]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.ck_dir]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.generate_test_runner]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[rocm.n_max_profiling_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[rocm.use_preselected_instances]: False",
"[bsvfcwwoczx2rlkdz2eta6doujsymyihmi46hhwk6clrrvwcb6m] inductor_config[cpu_backend]: cpp",
"[caw4ly2z672k6kjfahoxwpajp5idhhtrpgf3ma2clylcp7c7aid] inductor_config[cuda_backend]: triton",
"[ljhgflgihidopsfsdcbqynv27nceykby3nutyd5jlcpq7n6e7l4] inductor_config[halide.cpu_target]: host",
"[wx7vmsmrdpk5ue2txlywp3lj3faqmdjphs5fgg2ehzsyno7uovg] inductor_config[halide.gpu_target]: host-cuda",
"[svgytlua5wcyeia7wq7e6zgh5tsueikrnzchmdmouvmkpfsc2zq] inductor_config[halide.scheduler_cuda]: Anderson2021",
"[k5ogk6345jvklsnu7g2njqstiz2g6pm5wmqpgg3kasrmuqwjvl6] inductor_config[halide.scheduler_cpu]: Adams2019",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.asserts]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.debug]: False",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[halide.scan_kernels]: False",
"[h25wqx6vliw4j5rtzzbv6latydxyei3deyg6v7wzvnzryfktuki] inductor_config[external_matmul]: []",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.force_extern_kernel_in_multi_template]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.max_mm_configs]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.runtime_triton_dtype_assert]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_name_regex]: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] inductor_config[test_configs.autotune_choice_desc_regex]: None",
"[esstihe2nyydk4mhzpvox3qkajyu5y5t23hk3fi2me7jn75xi3o] inductor_config[test_configs.graphsafe_rng_func_ignores_fallback_random]: False",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_pre_pass: None",
"[tquy2we2efmowuj4wuqzcfcfdcrkzkzmwdae6hprj7fa64jpusq] post_grad_custom_post_pass: None"
],
"cache_bypass_reason": null,
"remote_cache_enabled": false,
"local_cache_enabled": true
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.705000 18434 site-packages/torch/_dynamo/guards.py:2537] {"dynamo_cpp_guards_str": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "55eb9a90b3539a56b3cd3941e1ab4951"}
TREE_GUARD_MANAGER:
+- RootGuardManager
| +- DEFAULT_DEVICE: utils_device.CURRENT_DEVICE == None # _dynamo/output_graph.py:519 in init_ambient_guards
| +- GLOBAL_STATE: ___check_global_state()
| +- TORCH_FUNCTION_MODE_STACK: ___check_torch_function_mode_stack()
| +- GuardManager: source=L['self'], accessed_by=FrameLocalsGuardAccessor(key='self', framelocals_idx=1)
| | +- TYPE_MATCH: ___check_type_id(L['self'], 97578784)
| | +- GuardManager: source=L['self'].__dict__, accessed_by=GetGenericDictGuardAccessor
| | | +- GuardManager: source=L['self'].dim, accessed_by=DictGetItemGuardAccessor('dim')
| | | | +- EQUALS_MATCH: L['self'].dim == 4
| +- GuardManager: source=L['___stack0'], accessed_by=FrameLocalsGuardAccessor(key='___stack0', framelocals_idx=0)
| | +- TENSOR_MATCH: check_tensor(L['___stack0'], Tensor, DispatchKeySet(CUDA, BackendSelect, ADInplaceOrView, AutogradCUDA), torch.bfloat16, device=0, requires_grad=False, size=[6, 4], stride=[4, 1])
| | +- NO_HASATTR: hasattr(L['___stack0'], '_dynamo_dynamic_indices') == False
| | +- GuardManager: source=L['___stack0'].reshape, accessed_by=GetAttrGuardAccessor(reshape)
| +- GuardManager: source=L['batch_size'], accessed_by=FrameLocalsGuardAccessor(key='batch_size', framelocals_idx=2)
| | +- EQUALS_MATCH: L['batch_size'] == 1
| +- GuardManager: source=L['num_tokens'], accessed_by=FrameLocalsGuardAccessor(key='num_tokens', framelocals_idx=3)
| | +- EQUALS_MATCH: L['num_tokens'] == 6
Guard latency = 0.41 us
V0320 11:46:31.705000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "46ce030a0a884ab42a6f0c3d92017562"}
{
"name": "entire_frame_compile",
"ts": 1742496391705913.5,
"args": {
"fn_name": "_compile.compile_inner",
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.706000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "12931e1d8517ed6fc84189510bd67537"}
{
"name": "gc",
"ts": 1742496391706223.5,
"args": {
"compile_id": "2/0"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.706000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "a45cc79912c62b7a2506b4a788e31bf2"}
{
"name": "gc",
"ts": 1742496391706885.2,
"args": {
"compile_id": "2/0"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.708000 18434 site-packages/torch/_dynamo/utils.py:1483] {"compilation_metrics": {"compile_id": "2/0", "frame_key": "3", "co_name": "torch_dynamo_resume_in_forward_at_32", "co_filename": "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py", "co_firstlineno": 32, "cache_size": 0, "accumulated_cache_size": 0, "guard_count": 11, "shape_env_guard_count": 0, "graph_op_count": 1, "graph_node_count": 3, "graph_input_count": 1, "start_time": 1742496391.636421, "entire_frame_compile_time_s": 0.069089, "backend_compile_time_s": 0.06103, "inductor_compile_time_s": 0.040914, "code_gen_time_s": 0.012315, "fail_type": null, "fail_reason": null, "fail_user_frame_filename": null, "fail_user_frame_lineno": null, "non_compliant_ops": [], "compliant_custom_ops": [], "restart_reasons": [], "dynamo_time_before_restart_s": 0.0, "has_guarded_code": true, "remote_cache_time_saved_s": null, "structured_logging_overhead_s": 0.024896, "config_suppress_errors": false, "config_inline_inbuilt_nn_modules": true, "specialize_float": false, "dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": true, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}", "is_forward": true, "num_triton_bundles": null, "remote_fx_graph_cache_get_time_ms": null, "remote_fx_graph_cache_put_time_ms": null, "start_time_us": 1742496391636421, "duration_us": 70789, "dynamo_cumulative_compile_time_us": 69089, "aot_autograd_cumulative_compile_time_us": 61030, "inductor_cumulative_compile_time_us": 40914, "inductor_code_gen_cumulative_compile_time_us": 12315, "triton_compile_time_us": 873, "runtime_cudagraphify_time_us": null, "runtime_triton_autotune_time_us": null, "dynamo_compile_time_before_restart_us": 0, "cuda_synchronize_time_us": null, "distributed_ephemeral_timeout_us": null, "structured_logging_overhead_us": 24896, "remote_fx_graph_cache_get_time_us": null, "remote_fx_graph_cache_put_time_us": null, "backward_cumulative_compile_time_us": null, "end_time_us": 1742496391707210, "pre_grad_pass_time_us": 475, "post_grad_pass_time_us": 7677, "joint_graph_pass_time_us": 1667, "log_format_version": 3, "inductor_config": "{\"TYPE_CHECKING\": false, \"_cache_config_ignore_prefix\": [\"trace\", \"cuda.cutlass_dir\", \"worker_start_method\", \"compile_threads\", \"post_grad_custom_post_pass\", \"post_grad_custom_pre_pass\", \"always_complex_memory_overlap_TESTING_ONLY\"], \"_collective.auto_select\": false, \"_collective.one_shot_all_reduce_threshold_bytes\": 131072, \"_fuse_ddp_bucket_size\": 25, \"_fuse_ddp_communication\": false, \"_fuse_ddp_communication_passes\": [\"fuse_ddp_with_concat_op\", \"schedule_comm_wait\"], \"_micro_pipeline_tp\": false, \"_pre_fusion_custom_pass\": null, \"_profile_var\": \"\", \"_raise_error_for_testing\": false, \"_save_config_ignore\": [\"trace.upload_tar\", \"joint_custom_pre_pass\", \"joint_custom_post_pass\", \"pre_grad_custom_pass\", \"aot_inductor.repro_level\", \"aot_inductor.dump_aoti_minifier\"], \"add_pre_grad_passes\": null, \"aggressive_fusion\": false, \"allow_buffer_reuse\": true, \"always_complex_memory_overlap_TESTING_ONLY\": false, \"always_keep_tensor_constants\": false, \"annotate_training\": false, \"aot_inductor.allow_stack_allocation\": false, \"aot_inductor.compile_wrapper_with_O0\": false, \"aot_inductor.debug_compile\": false, \"aot_inductor.debug_intermediate_value_printer\": \"0\", \"aot_inductor.dump_aoti_minifier\": false, \"aot_inductor.filtered_kernel_names\": null, \"aot_inductor.force_mmap_weights\": false, \"aot_inductor.metadata\": {}, \"aot_inductor.output_path\": \"\", \"aot_inductor.package\": false, \"aot_inductor.package_constants_in_so\": true, \"aot_inductor.package_cpp_only\": false, \"aot_inductor.presets\": {}, \"aot_inductor.raise_error_on_ignored_optimization\": true, \"aot_inductor.repro_level\": 2, \"aot_inductor.serialized_in_spec\": \"\", \"aot_inductor.serialized_out_spec\": \"\", \"aot_inductor.use_minimal_arrayref_interface\": false, \"aot_inductor.use_runtime_constant_folding\": false, \"assert_indirect_indexing\": true, \"assume_aligned_inputs\": false, \"autoheuristic_collect\": \"\", \"autoheuristic_log_path\": \"DEFAULT\", \"autoheuristic_use\": \"mixed_mm\", \"autotune_fallback_to_aten\": true, \"autotune_in_subproc\": false, \"autotune_local_cache\": true, \"autotune_multi_device\": false, \"autotune_num_choices_displayed\": 10, \"autotune_remote_cache\": null, \"b2b_gemm_pass\": false, \"batch_fusion\": true, \"benchmark_combo_kernel\": false, \"benchmark_epilogue_fusion\": true, \"benchmark_fusion\": false, \"benchmark_harness\": true, \"benchmark_kernel\": false, \"bundle_triton_into_fx_graph_cache\": true, \"bundled_autotune_remote_cache\": null, \"bw_outputs_user_visible\": true, \"can_inplace_pad_graph_input\": false, \"check_stack_no_cycles_TESTING_ONLY\": false, \"combo_kernel_allow_mixed_sizes\": 1, \"combo_kernel_foreach_dynamic_shapes\": false, \"combo_kernels\": false, \"combo_kernels_autotune\": 1, \"comment_origin\": false, \"compile_threads\": 32, \"comprehensive_padding\": true, \"compute_all_bounds\": false, \"constant_and_index_propagation\": true, \"conv_1x1_as_mm\": false, \"coordinate_descent_check_all_directions\": false, \"coordinate_descent_search_radius\": 1, \"coordinate_descent_tuning\": false, \"cpp.cxx\": [null, \"g++\"], \"cpp.descriptive_names\": \"original_aten\", \"cpp.dynamic_threads\": false, \"cpp.enable_concat_linear\": false, \"cpp.enable_floating_point_contract_flag\": \"off\", \"cpp.enable_grouped_gemm_template\": false, \"cpp.enable_kernel_profile\": false, \"cpp.enable_loop_tail_vec\": true, \"cpp.enable_tiling_heuristics\": true, \"cpp.enable_unsafe_math_opt_flag\": false, \"cpp.fallback_scatter_reduce_sum\": true, \"cpp.gemm_cache_blocking\": null, \"cpp.gemm_max_k_slices\": 1, \"cpp.gemm_thread_factors\": null, \"cpp.inject_log1p_bug_TESTING_ONLY\": null, \"cpp.inject_relu_bug_TESTING_ONLY\": null, \"cpp.max_horizontal_fusion_size\": 16, \"cpp.min_chunk_size\": 4096, \"cpp.no_redundant_loops\": true, \"cpp.simdlen\": null, \"cpp.threads\": -1, \"cpp.vec_isa_ok\": null, \"cpp.weight_prepack\": true, \"cpp_wrapper\": false, \"cpu_backend\": \"cpp\", \"cuda.arch\": null, \"cuda.compile_opt_level\": \"-O1\", \"cuda.cuda_cxx\": null, \"cuda.cutlass_backend_min_gemm_size\": 1, \"cuda.cutlass_dir\": \"/home/cdhernandez/.conda/envs/pytorch-3.12/lib/python3.12/site-packages/third_party/cutlass\", \"cuda.cutlass_instantiation_level\": \"0\", \"cuda.cutlass_max_profiling_configs\": null, \"cuda.cutlass_max_profiling_swizzle_options\": [1, 2, 4], \"cuda.cutlass_op_allowlist_regex\": null, \"cuda.cutlass_op_denylist_regex\": null, \"cuda.enable_cuda_lto\": false, \"cuda.enable_debug_info\": false, \"cuda.enable_ptxas_info\": false, \"cuda.generate_test_runner\": false, \"cuda.use_fast_math\": false, \"cuda.version\": null, \"cuda_backend\": \"triton\", \"custom_op_default_layout_constraint\": \"needs_fixed_stride_order\", \"dce\": false, \"debug\": false, \"debug_fusion\": false, \"debug_index_asserts\": false, \"debug_ir_traceback\": false, \"decompose_mem_bound_mm\": false, \"developer_warnings\": true, \"disable_cpp_codegen\": false, \"disable_padding_cpu\": true, \"disable_progress\": true, \"dynamic_scale_rblock\": true, \"efficient_conv_bn_eval_fx_passes\": false, \"emulate_precision_casts\": false, \"enable_auto_functionalized_v2\": true, \"enable_linear_binary_folding\": false, \"enabled_metric_tables\": \"\", \"epilogue_fusion\": true, \"epilogue_fusion_first\": false, \"estimate_op_runtime\": \"default\", \"external_matmul\": [], \"fallback_random\": false, \"force_disable_caches\": false, \"force_fuse_int_mm_with_mul\": false, \"force_layout_optimization\": false, \"force_pointwise_cat\": false, \"force_same_precision\": false, \"force_shape_pad\": false, \"freezing\": false, \"freezing_discard_parameters\": false, \"fx_graph_cache\": true, \"fx_graph_remote_cache\": null, \"fx_passes_numeric_check\": {\"num_iterations\": 1, \"pre_grad\": false, \"precision\": 0.0001, \"requires_optimizer\": true}, \"generate_intermediate_hooks\": false, \"global_cache_dir\": null, \"graph_partition\": false, \"group_fusion\": false, \"halide.asserts\": false, \"halide.cpu_target\": \"host\", \"halide.debug\": false, \"halide.gpu_target\": \"host-cuda\", \"halide.scan_kernels\": false, \"halide.scheduler_cpu\": \"Adams2019\", \"halide.scheduler_cuda\": \"Anderson2021\", \"implicit_fallbacks\": true, \"inplace_buffers\": true, \"inplace_padding\": true, \"inter_node_bw\": 25, \"intra_node_bw\": 300, \"is_nightly_or_source\": true, \"is_predispatch\": false, \"joint_custom_post_pass\": null, \"joint_custom_pre_pass\": null, \"joint_graph_constant_folding\": true, \"keep_output_stride\": true, \"kernel_name_max_ops\": 10, \"layout_opt_default\": \"1\", \"layout_optimization\": true, \"loop_ordering_after_fusion\": false, \"max_autotune\": false, \"max_autotune_conv_backends\": \"ATEN,TRITON\", \"max_autotune_gemm\": false, \"max_autotune_gemm_backends\": \"ATEN,TRITON,CPP\", \"max_autotune_gemm_search_space\": \"DEFAULT\", \"max_autotune_pointwise\": false, \"max_autotune_subproc_graceful_timeout_seconds\": 1.0, \"max_autotune_subproc_result_timeout_seconds\": 60.0, \"max_autotune_subproc_terminate_timeout_seconds\": 2.0, \"max_epilogue_benchmarked_choices\": 1, \"max_fusion_size\": 64, \"max_pointwise_cat_inputs\": 8, \"memory_planning\": false, \"memory_pool\": \"intermediates\", \"mixed_mm_choice\": \"heuristic\", \"nan_asserts\": false, \"optimize_scatter_upon_const_tensor\": true, \"pad_channels_last\": false, \"pad_outputs\": false, \"padding_alignment_bytes\": 128, \"padding_stride_threshold\": 1024, \"pattern_matcher\": true, \"permute_fusion\": false, \"pick_loop_orders\": true, \"post_grad_custom_post_pass\": null, \"post_grad_custom_pre_pass\": null, \"post_grad_fusion_options\": {}, \"pre_grad_custom_pass\": null, \"pre_grad_fusion_options\": {}, \"profile_bandwidth\": false, \"profile_bandwidth_output\": null, \"profile_bandwidth_regex\": \"\", \"profile_bandwidth_with_do_bench_using_profiling\": false, \"profiler_mark_wrapper_call\": false, \"prologue_fusion\": true, \"realize_acc_reads_threshold\": 8, \"realize_opcount_threshold\": 30, \"realize_reads_threshold\": 4, \"remove_pre_grad_passes\": null, \"reorder_for_compute_comm_overlap\": false, \"reorder_for_compute_comm_overlap_passes\": [\"reorder_compute_for_overlap\", \"sink_waits\", \"raise_comms\"], \"reorder_for_locality\": true, \"reorder_for_peak_memory\": true, \"rocm.arch\": [], \"rocm.ck_dir\": null, \"rocm.ck_supported_arch\": [\"gfx90a\", \"gfx942\"], \"rocm.compile_opt_level\": \"-O2\", \"rocm.flush_denormals\": true, \"rocm.generate_test_runner\": false, \"rocm.is_debug\": false, \"rocm.n_max_profiling_configs\": null, \"rocm.print_kernel_resource_usage\": false, \"rocm.rocm_home\": null, \"rocm.save_temps\": false, \"rocm.use_fast_math\": true, \"rocm.use_preselected_instances\": false, \"save_args\": false, \"scalar_asserts\": true, \"score_fusion_memory_threshold\": 10, \"search_autotune_cache\": false, \"shape_padding\": true, \"size_asserts\": true, \"sleep_sec_TESTING_ONLY\": null, \"split_cat_fx_passes\": true, \"split_reductions\": true, \"static_weight_shapes\": true, \"test_configs.autotune_choice_desc_regex\": null, \"test_configs.autotune_choice_name_regex\": null, \"test_configs.force_extern_kernel_in_multi_template\": false, \"test_configs.graphsafe_rng_func_ignores_fallback_random\": false, \"test_configs.max_mm_configs\": null, \"test_configs.runtime_triton_dtype_assert\": false, \"trace.compile_profile\": false, \"trace.debug_dir\": null, \"trace.debug_log\": false, \"trace.dot_graph_shape\": null, \"trace.draw_orig_fx_graph\": false, \"trace.enabled\": false, \"trace.fx_graph\": true, \"trace.fx_graph_transformed\": true, \"trace.graph_diagram\": false, \"trace.info_log\": false, \"trace.ir_post_fusion\": true, \"trace.ir_pre_fusion\": true, \"trace.log_autotuning_results\": false, \"trace.log_inductor_triton_kernel_to_post_grad_node_info\": true, \"trace.log_url_for_graph_xform\": null, \"trace.output_code\": true, \"trace.save_real_tensors\": false, \"trace.upload_tar\": null, \"triton.autotune_at_compile_time\": null, \"triton.autotune_cublasLt\": true, \"triton.autotune_pointwise\": true, \"triton.codegen_upcast_to_fp32\": true, \"triton.cooperative_reductions\": false, \"triton.cudagraph_dynamic_shape_warn_limit\": 50, \"triton.cudagraph_skip_dynamic_graphs\": false, \"triton.cudagraph_support_input_mutation\": true, \"triton.cudagraph_trees\": true, \"triton.cudagraph_trees_history_recording\": false, \"triton.cudagraph_unexpected_rerecord_limit\": 128, \"triton.cudagraphs\": false, \"triton.debug_sync_graph\": false, \"triton.debug_sync_kernel\": false, \"triton.dense_indexing\": false, \"triton.descriptive_names\": \"original_aten\", \"triton.divisible_by_16\": true, \"triton.enable_persistent_tma_matmul\": false, \"triton.fast_path_cudagraph_asserts\": false, \"triton.force_cooperative_reductions\": false, \"triton.force_cudagraph_sync\": false, \"triton.force_cudagraphs_warmup\": false, \"triton.inject_relu_bug_TESTING_ONLY\": null, \"triton.max_tiles\": 2, \"triton.min_split_scan_rblock\": 256, \"triton.multi_kernel\": 0, \"triton.persistent_reductions\": true, \"triton.prefer_nd_tiling\": false, \"triton.skip_cudagraph_warmup\": false, \"triton.skip_l1_cache\": false, \"triton.slow_path_cudagraph_asserts\": true, \"triton.spill_threshold\": 16, \"triton.store_cubin\": false, \"triton.tile_reductions\": false, \"triton.tiling_prevents_pointwise_fusion\": true, \"triton.tiling_prevents_reduction_fusion\": true, \"triton.unique_kernel_names\": true, \"triton.unique_user_kernel_names\": false, \"triton.use_block_ptr\": false, \"triton_kernel_default_layout_constraint\": \"needs_fixed_stride_order\", \"unbacked_symint_fallback\": 8192, \"unroll_reductions_threshold\": 8, \"unsafe_ignore_unsupported_triton_autotune_args\": false, \"use_experimental_benchmarker\": true, \"use_fast_math\": false, \"use_mixed_mm\": true, \"verbose_progress\": false, \"warn_mix_layout\": false, \"worker_start_method\": \"subprocess\"}", "remote_cache_version": null, "inductor_fx_remote_cache_hit_count": null, "inductor_fx_remote_cache_miss_count": null, "inductor_fx_remote_cache_backend_type": null, "inductor_fx_remote_cache_hit_keys": null, "inductor_fx_remote_cache_miss_keys": null, "cuda_version": "12.6", "triton_version": "3.2.0", "feature_usage": {"aot_autograd_remote_cache": false, "fx_cache": true, "triton_bundling": true}, "compile_time_autotune_time_us": null, "is_runtime": false, "gc_time_us": 661, "tensorify_float_attempt": null, "tensorify_float_success": null, "tensorify_float_failure": null, "guard_latency_us": 0, "recompile_reason": null, "num_graph_breaks": 1}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0}
V0320 11:46:31.708000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "frame_id": 2, "frame_compile_id": 0, "attempt": 0, "has_payload": "acca088009f68c5889844a744e962ca5"}
{
"name": "dynamo",
"ts": 1742496391708733.5,
"args": {
"compile_id": "2/0",
"num_graph_breaks": 1,
"guard_latency_us": 0,
"frame_key": "3",
"co_name": "torch_dynamo_resume_in_forward_at_32",
"co_filename": "/data/users/cdhernandez/gpt-fast/mixtral-moe/even_smaller_moe_repro.py",
"co_firstlineno": 32,
"cache_size": 0,
"accumulated_cache_size": 0,
"guard_count": 11,
"shape_env_guard_count": 0,
"graph_op_count": 1,
"graph_node_count": 3,
"graph_input_count": 1,
"fail_type": null,
"fail_reason": null,
"fail_user_frame_filename": null,
"fail_user_frame_lineno": null,
"non_compliant_ops": [],
"compliant_custom_ops": [],
"restart_reasons": [],
"dynamo_time_before_restart_s": 0.0,
"has_guarded_code": true,
"dynamo_config": "{\"_autograd_backward_strict_mode_conditional_banned_ops\": [\"stride\", \"storage_offset\", \"is_contiguous\"], \"_unsafe_skip_fsdp_module_guards\": false, \"accumulated_recompile_limit\": 256, \"allow_complex_guards_as_runtime_asserts\": false, \"allow_empty_graphs\": false, \"allow_ignore_mark_dynamic\": false, \"allow_rnn\": false, \"allow_unspec_int_on_nn_module\": false, \"allowed_functions_module_string_ignorelist\": [\"torch._decomp\", \"torch._prims\", \"torch._refs\", \"torch.distributions\", \"torch.testing\"], \"assume_static_by_default\": true, \"automatic_dynamic_local_pgo\": true, \"automatic_dynamic_remote_pgo\": null, \"automatic_dynamic_shapes\": true, \"automatic_dynamic_shapes_mark_as\": \"dynamic\", \"capture_autograd_function\": true, \"capture_dynamic_output_shape_ops\": false, \"capture_func_transforms\": true, \"capture_scalar_outputs\": true, \"capture_sparse_compute\": true, \"compiled_autograd\": false, \"compiled_autograd_kwargs_override\": {}, \"cprofile\": false, \"cudagraph_backend_keep_input_mutation\": false, \"cudagraph_backend_support_input_mutation\": false, \"dead_code_elimination\": true, \"disable\": false, \"do_not_emit_runtime_asserts\": false, \"dynamic_shapes\": true, \"enable_compiler_collectives\": false, \"enable_cpp_framelocals_guard_eval\": true, \"enable_cpp_guard_manager\": true, \"enable_cpp_symbolic_shape_guards\": false, \"enable_faithful_generator_behavior\": true, \"enable_trace_contextlib\": true, \"error_on_nested_fx_trace\": true, \"error_on_nested_jit_trace\": true, \"error_on_recompile\": false, \"fail_on_recompile_limit_hit\": false, \"fake_tensor_cache_crosscheck_enabled\": false, \"fake_tensor_cache_enabled\": true, \"force_nn_module_property_static_shapes\": true, \"force_parameter_static_shapes\": true, \"force_unspec_int_unbacked_size_like_on_torchrec_kjt\": false, \"guard_nn_modules\": true, \"guard_nn_modules_using_dict_tags\": true, \"inline_inbuilt_nn_modules\": true, \"issue_3_13_0_warning\": true, \"minimum_call_count\": 1, \"numpy_default_complex\": \"complex128\", \"numpy_default_float\": \"float64\", \"numpy_default_int\": \"int64\", \"only_allow_pt2_compliant_ops\": false, \"optimize_ddp\": true, \"optimize_ddp_lazy_compile\": false, \"prefer_deferred_runtime_asserts_over_guards\": false, \"prepare_freezing\": false, \"raise_on_ctx_manager_usage\": true, \"raise_on_unsafe_aot_autograd\": false, \"recompile_limit\": 8, \"record_compile_time_instruction_count\": false, \"replay_record_enabled\": false, \"report_guard_failures\": true, \"rewrite_assert_with_torch_assert\": true, \"run_gc_after_compile\": true, \"skip_code_recursive_on_recompile_limit_hit\": true, \"skip_fsdp_guards\": true, \"skip_fsdp_hooks\": true, \"skip_nnmodule_hook_guards\": true, \"skip_no_tensor_aliasing_guards_on_parameters\": true, \"skip_tensor_guards_with_matching_dict_tags\": true, \"skip_torchrec\": true, \"skipfiles_inline_module_allowlist\": {}, \"specialize_float\": false, \"specialize_int\": false, \"suppress_errors\": false, \"trace_numpy\": true, \"track_nodes_for_deduplication\": false, \"use_graph_deduplication\": false, \"use_lazy_graph_module\": true, \"use_numpy_random_stream\": false, \"verify_correctness\": false}"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.709000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "c72a1fd5b26fcac1f48deb9c5bfcdb21"}
{
"name": "cudagraphify",
"ts": 1742496391709145.0,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.709000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "5bb443cc77150af0845a959b8499b8b3"}
{
"name": "cudagraphify",
"ts": 1742496391709544.5,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.709000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "8d5430a35aace520ad433724c7b5cacf"}
{
"name": "CUDAGraphTreeManager.run_eager",
"ts": 1742496391709926.2,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.710000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "8cf0c44caf260535e6788d1e150006b6"}
{
"name": "CUDAGraphTreeManager.run_eager",
"ts": 1742496391710433.5,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.710000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "4767d60ef1d72138a9c9a374ed48fc83"}
{
"name": "CUDAGraphTreeManager.record_function",
"ts": 1742496391710892.2,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.916000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "90d59a5fc20a95d4263a45a5c2a2661c"}
{
"name": "CUDAGraphTreeManager.record_function",
"ts": 1742496391916429.2,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:31.917000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "8077f181f5d711580c33a5268b68c4cb"}
{
"name": "CUDAGraphTreeManager.record_function",
"ts": 1742496391917196.8,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:32.107000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "66d26326481b6712d27c30b368686216"}
{
"name": "CUDAGraphTreeManager.record_function",
"ts": 1742496392107434.0,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:32.109000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "e58c5cd8ef665f530f2d1d47322a6974"}
{
"name": "CUDAGraphTreeManager.record_function",
"ts": 1742496392109425.8,
"args": {
"compile_id": "None"
},
"ph": "B",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
V0320 11:46:32.294000 18434 site-packages/torch/_dynamo/utils.py:1782] {"chromium_event": {}, "has_payload": "d33a93a47ee6d62123d8fa8383459af4"}
{
"name": "CUDAGraphTreeManager.record_function",
"ts": 1742496392294734.2,
"args": {
"compile_id": "None"
},
"ph": "E",
"cat": "dynamo_timed",
"tid": 0,
"pid": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment