Skip to content

Instantly share code, notes, and snippets.

@vukrosic
Last active October 21, 2025 21:26
Show Gist options
  • Select an option

  • Save vukrosic/c8f370a1d041f58bd28fdea9339467f8 to your computer and use it in GitHub Desktop.

Select an option

Save vukrosic/c8f370a1d041f58bd28fdea9339467f8 to your computer and use it in GitHub Desktop.
RMSNorm Tutorial.ipynb
Display the source blob
Display the rendered blob
Raw
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": [],
"gpuType": "T4",
"name": "RMSNorm Tutorial.ipynb",
"authorship_tag": "ABX9TyNGNLw7bmZ0UDXJs+0jI0HU",
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/vukrosic/c8f370a1d041f58bd28fdea9339467f8/rmsnorm.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"source": [
"Video Master RMSNorm From Scratch - Step by Step Tutorial: https://youtu.be/HgSdYtPgJnU"
],
"metadata": {
"id": "scd2-_fRqCbZ"
}
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "DT5-i_3wjKeH",
"outputId": "4458979a-0464-4077-fc62-e255eb88a3a4"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Input:\n",
"[1. 2. 3. 4. 5.]\n",
"\n",
"Input mean: 3.0000\n",
"Input std: 1.4142\n",
"\n",
"==================================================\n",
"After RMSNorm:\n",
"[0.30151133 0.60302266 0.90453399 1.20604532 1.50755665]\n",
"\n",
"Output mean: 0.9045\n",
"Output RMS: 1.0000\n",
"\n",
"==================================================\n",
"Step-by-step calculation:\n",
"1. Squared values: [ 1. 4. 9. 16. 25.]\n",
"2. Mean of squares: 11.0000\n",
"3. RMS (sqrt of mean): 3.3166\n",
"4. Divide input by RMS to get normalized output\n"
]
}
],
"source": [
"import numpy as np\n",
"\n",
"def rmsnorm(x, eps=1e-6):\n",
" \"\"\"\n",
" Root Mean Square Normalization\n",
"\n",
" Args:\n",
" x: Input array\n",
" eps: Small constant for numerical stability\n",
"\n",
" Returns:\n",
" Normalized array\n",
" \"\"\"\n",
" # Calculate RMS: sqrt of mean of squares\n",
" rms = np.sqrt(np.mean(x**2) + eps§)\n",
"\n",
" # Normalize by dividing by RMS\n",
" return x / rms\n",
"\n",
"# Example input\n",
"input_vector = np.array([1.0, 2.0, 3.0, 4.0, 5.0])\n",
"\n",
"print(\"Input:\")\n",
"print(input_vector)\n",
"print(f\"\\nInput mean: {np.mean(input_vector):.4f}\")\n",
"print(f\"Input std: {np.std(input_vector):.4f}\")\n",
"\n",
"# Apply RMSNorm\n",
"output = rmsnorm(input_vector)\n",
"\n",
"print(\"\\n\" + \"=\"*50)\n",
"print(\"After RMSNorm:\")\n",
"print(output)\n",
"print(f\"\\nOutput mean: {np.mean(output):.4f}\")\n",
"print(f\"Output RMS: {np.sqrt(np.mean(output**2)):.4f}\")\n",
"\n",
"# Show the calculation step by step\n",
"print(\"\\n\" + \"=\"*50)\n",
"print(\"Step-by-step calculation:\")\n",
"squared = input_vector**2\n",
"print(f\"1. Squared values: {squared}\")\n",
"print(f\"2. Mean of squares: {np.mean(squared):.4f}\")\n",
"print(f\"3. RMS (sqrt of mean): {np.sqrt(np.mean(squared)):.4f}\")\n",
"print(f\"4. Divide input by RMS to get normalized output\")"
]
},
{
"cell_type": "code",
"source": [
"import numpy as np\n",
"import torch\n",
"import torch.nn as nn\n",
"\n",
"# same as above\n",
"def rmsnorm_numpy(x, eps=1e-6):\n",
" \"\"\"RMSNorm using NumPy\"\"\"\n",
" rms = np.sqrt(np.mean(x**2) + eps)\n",
" return x / rms\n",
"\n",
"class RMSNorm(nn.Module):\n",
" \"\"\"RMSNorm using PyTorch\"\"\"\n",
" def __init__(self, dim, eps=1e-6):\n",
" super().__init__()\n",
" self.eps = eps\n",
" # Learnable scale parameter (like gamma in LayerNorm)\n",
" self.weight = nn.Parameter(torch.ones(dim))\n",
"\n",
" def forward(self, x):\n",
" # Calculate RMS\n",
" rms = torch.sqrt(torch.mean(x**2, dim=-1, keepdim=True) + self.eps)\n",
" # Normalize and scale\n",
" return (x / rms) * self.weight\n",
"\n",
"# Example input\n",
"input_vector = np.array([1.0, 2.0, 3.0, 4.0, 5.0])\n",
"\n",
"print(\"=\"*60)\n",
"print(\"NUMPY VERSION\")\n",
"print(\"=\"*60)\n",
"print(f\"Input: {input_vector}\")\n",
"print(f\"Input RMS: {np.sqrt(np.mean(input_vector**2)):.4f}\")\n",
"\n",
"output_np = rmsnorm_numpy(input_vector)\n",
"print(f\"\\nAfter RMSNorm: {output_np}\")\n",
"print(f\"Output RMS: {np.sqrt(np.mean(output_np**2)):.4f}\")\n",
"print(f\"Values are now scaled to have RMS ≈ 1.0\")\n",
"\n",
"print(\"\\n\" + \"=\"*60)\n",
"print(\"PYTORCH VERSION\")\n",
"print(\"=\"*60)\n",
"\n",
"# Convert to PyTorch tensor\n",
"input_tensor = torch.tensor([1.0, 2.0, 3.0, 4.0, 5.0])\n",
"print(f\"Input: {input_tensor}\")\n",
"\n",
"# Create RMSNorm layer\n",
"rmsnorm_layer = RMSNorm(dim=5)\n",
"print(f\"Learnable weights: {rmsnorm_layer.weight.data}\")\n",
"\n",
"output_torch = rmsnorm_layer(input_tensor)\n",
"print(f\"\\nAfter RMSNorm: {output_torch}\")\n",
"print(f\"Output RMS: {torch.sqrt(torch.mean(output_torch**2)):.4f}\")\n",
"\n",
"# Show with different scale inputs\n",
"print(\"\\n\" + \"=\"*60)\n",
"print(\"SAME PATTERN, DIFFERENT SCALES:\")\n",
"print(\"=\"*60)\n",
"small = np.array([0.1, 0.2, 0.3, 0.4, 0.5])\n",
"large = np.array([10.0, 20.0, 30.0, 40.0, 50.0])\n",
"\n",
"print(f\"Small input RMS: {np.sqrt(np.mean(small**2)):.4f}\")\n",
"print(f\"After RMSNorm RMS: {np.sqrt(np.mean(rmsnorm_numpy(small)**2)):.4f}\")\n",
"print(f\"\\nLarge input RMS: {np.sqrt(np.mean(large**2)):.4f}\")\n",
"print(f\"After RMSNorm RMS: {np.sqrt(np.mean(rmsnorm_numpy(large)**2)):.4f}\")\n",
"print(\"\\n→ Both normalized to similar scale!\")"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "vfLWQTBFnN1E",
"outputId": "fb31bf07-bc75-491f-892f-4c6097b3a472"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"============================================================\n",
"NUMPY VERSION\n",
"============================================================\n",
"Input: [1. 2. 3. 4. 5.]\n",
"Input RMS: 3.3166\n",
"\n",
"After RMSNorm: [0.30151133 0.60302266 0.90453399 1.20604532 1.50755665]\n",
"Output RMS: 1.0000\n",
"Values are now scaled to have RMS ≈ 1.0\n",
"\n",
"============================================================\n",
"PYTORCH VERSION\n",
"============================================================\n",
"Input: tensor([1., 2., 3., 4., 5.])\n",
"Learnable weights: tensor([1., 1., 1., 1., 1.])\n",
"\n",
"After RMSNorm: tensor([0.3015, 0.6030, 0.9045, 1.2060, 1.5076], grad_fn=<MulBackward0>)\n",
"Output RMS: 1.0000\n",
"\n",
"============================================================\n",
"SAME PATTERN, DIFFERENT SCALES:\n",
"============================================================\n",
"Small input RMS: 0.3317\n",
"After RMSNorm RMS: 1.0000\n",
"\n",
"Large input RMS: 33.1662\n",
"After RMSNorm RMS: 1.0000\n",
"\n",
"→ Both normalized to similar scale!\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"import torch\n",
"import torch.nn as nn\n",
"import torch.optim as optim\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"\n",
"class RMSNorm(nn.Module):\n",
" \"\"\"RMSNorm with learnable scale\"\"\"\n",
" def __init__(self, dim, eps=1e-6):\n",
" super().__init__()\n",
" self.eps = eps\n",
" self.weight = nn.Parameter(torch.ones(dim))\n",
"\n",
" def forward(self, x):\n",
" rms = torch.sqrt(torch.mean(x**2, dim=-1, keepdim=True) + self.eps)\n",
" return (x / rms) * self.weight\n",
"\n",
"class SimpleNet(nn.Module):\n",
" \"\"\"Simple neural network with RMSNorm\"\"\"\n",
" def __init__(self, input_dim, hidden_dim, output_dim):\n",
" super().__init__()\n",
" self.fc1 = nn.Linear(input_dim, hidden_dim)\n",
" self.rmsnorm = RMSNorm(hidden_dim)\n",
" self.fc2 = nn.Linear(hidden_dim, output_dim)\n",
" self.relu = nn.ReLU()\n",
"\n",
" def forward(self, x):\n",
" x = self.fc1(x)\n",
" x = self.rmsnorm(x) # Normalize after first layer\n",
" x = self.relu(x)\n",
" x = self.fc2(x)\n",
" return x\n",
"\n",
"# Create a simple dataset: learn XOR-like pattern\n",
"torch.manual_seed(42)\n",
"X = torch.tensor([[0., 0.], [0., 1.], [1., 0.], [1., 1.]])\n",
"y = torch.tensor([[0.], [1.], [1.], [0.]]) # XOR\n",
"\n",
"print(\"=\"*60)\n",
"print(\"TRAINING A SIMPLE NEURAL NETWORK WITH RMSNorm\")\n",
"print(\"=\"*60)\n",
"print(f\"Task: Learn XOR function\")\n",
"print(f\"Input:\\n{X}\")\n",
"print(f\"Target:\\n{y.squeeze()}\")\n",
"\n",
"# Create model\n",
"model = SimpleNet(input_dim=2, hidden_dim=4, output_dim=1)\n",
"criterion = nn.MSELoss()\n",
"optimizer = optim.Adam(model.parameters(), lr=0.1)\n",
"\n",
"# Track RMSNorm weights during training\n",
"weight_history = []\n",
"loss_history = []\n",
"\n",
"print(f\"\\nInitial RMSNorm weights: {model.rmsnorm.weight.data.numpy()}\")\n",
"\n",
"# Train\n",
"epochs = 500\n",
"for epoch in range(epochs):\n",
" optimizer.zero_grad()\n",
" output = model(X)\n",
" loss = criterion(output, y)\n",
" loss.backward()\n",
" optimizer.step()\n",
"\n",
" # Record weights and loss\n",
" weight_history.append(model.rmsnorm.weight.data.clone().numpy())\n",
" loss_history.append(loss.item())\n",
"\n",
" if (epoch + 1) % 100 == 0:\n",
" print(f\"Epoch {epoch+1}/{epochs}, Loss: {loss.item():.6f}\")\n",
"\n",
"print(f\"\\nFinal RMSNorm weights: {model.rmsnorm.weight.data.numpy()}\")\n",
"\n",
"# Test the model\n",
"print(\"\\n\" + \"=\"*60)\n",
"print(\"FINAL PREDICTIONS:\")\n",
"print(\"=\"*60)\n",
"with torch.no_grad():\n",
" predictions = model(X)\n",
" for i, (inp, pred, target) in enumerate(zip(X, predictions, y)):\n",
" print(f\"Input: {inp.numpy()} → Predicted: {pred.item():.4f}, Target: {target.item():.4f}\")\n",
"\n",
"# Plot the weight evolution\n",
"weight_history = np.array(weight_history)\n",
"fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))\n",
"\n",
"# Plot 1: RMSNorm weights over time\n",
"for i in range(4):\n",
" ax1.plot(weight_history[:, i], label=f'Weight {i}', linewidth=2)\n",
"ax1.set_xlabel('Epoch', fontsize=12)\n",
"ax1.set_ylabel('Weight Value', fontsize=12)\n",
"ax1.set_title('RMSNorm Learnable Weights During Training', fontsize=14, fontweight='bold')\n",
"ax1.legend()\n",
"ax1.grid(True, alpha=0.3)\n",
"ax1.axhline(y=1.0, color='red', linestyle='--', alpha=0.5, label='Initial value')\n",
"\n",
"# Plot 2: Training loss\n",
"ax2.plot(loss_history, color='orange', linewidth=2)\n",
"ax2.set_xlabel('Epoch', fontsize=12)\n",
"ax2.set_ylabel('Loss (MSE)', fontsize=12)\n",
"ax2.set_title('Training Loss Over Time', fontsize=14, fontweight='bold')\n",
"ax2.grid(True, alpha=0.3)\n",
"ax2.set_yscale('log')\n",
"\n",
"plt.tight_layout()\n",
"plt.savefig('rmsnorm_training.png', dpi=150, bbox_inches='tight')\n",
"print(\"\\n✓ Plot saved as 'rmsnorm_training.png'\")\n",
"plt.show()\n",
"\n",
"print(\"\\n\" + \"=\"*60)\n",
"print(\"KEY OBSERVATIONS:\")\n",
"print(\"=\"*60)\n",
"print(\"• RMSNorm weights started at 1.0 (neutral scaling)\")\n",
"print(\"• During training, they learned to scale different dimensions\")\n",
"print(\"• This helps the model emphasize important features\")\n",
"print(\"• Each weight adjusts the importance of one hidden dimension\")"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 980
},
"id": "i_FUTNqpgmuq",
"outputId": "0d306d57-6130-4389-d7a8-421e04c4bf18"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"============================================================\n",
"TRAINING A SIMPLE NEURAL NETWORK WITH RMSNorm\n",
"============================================================\n",
"Task: Learn XOR function\n",
"Input:\n",
"tensor([[0., 0.],\n",
" [0., 1.],\n",
" [1., 0.],\n",
" [1., 1.]])\n",
"Target:\n",
"tensor([0., 1., 1., 0.])\n",
"\n",
"Initial RMSNorm weights: [1. 1. 1. 1.]\n",
"Epoch 100/500, Loss: 0.000022\n",
"Epoch 200/500, Loss: 0.000000\n",
"Epoch 300/500, Loss: 0.000235\n",
"Epoch 400/500, Loss: 0.000000\n",
"Epoch 500/500, Loss: 0.000000\n",
"\n",
"Final RMSNorm weights: [0.78946674 1.6595931 0.78331435 0.54245394]\n",
"\n",
"============================================================\n",
"FINAL PREDICTIONS:\n",
"============================================================\n",
"Input: [0. 0.] → Predicted: -0.0000, Target: 0.0000\n",
"Input: [0. 1.] → Predicted: 1.0000, Target: 1.0000\n",
"Input: [1. 0.] → Predicted: 1.0000, Target: 1.0000\n",
"Input: [1. 1.] → Predicted: -0.0000, Target: 0.0000\n",
"\n",
"✓ Plot saved as 'rmsnorm_training.png'\n"
]
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"<Figure size 1400x500 with 2 Axes>"
],
"image/png": "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\n"
},
"metadata": {}
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"\n",
"============================================================\n",
"KEY OBSERVATIONS:\n",
"============================================================\n",
"• RMSNorm weights started at 1.0 (neutral scaling)\n",
"• During training, they learned to scale different dimensions\n",
"• This helps the model emphasize important features\n",
"• Each weight adjusts the importance of one hidden dimension\n"
]
}
]
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment