Model Autopsy

Parse the safetensors header of SmolLM2-360M, Qwen3-0.6B, SmolLM3-3B, and Phi-4-mini without downloading a single weight. Detect each model's GQA group size, tied-vs-untied embeddings, SwiGLU hidden ratio, and vocab size from tensor names alone. See Phi-4-mini spend 31% of its params on vocabulary where SmolLM3 spends 18%.

Time

45–60 min

Hardware

CPU · Colab

Act

III · The Current Champions

Status

Live

Artifact

A four-model architecture comparison table + parameter-distribution charts.

Clone the labs repo and run this lab as a script or open it as a notebook:

git clone https://github.com/iqbal-sk/Microscale-labs.git
cd Microscale
just setup-auto      # auto-detects CPU / CUDA / Mac
just run 04
# or:  jupyter lab labs/04-model-autopsy/lab.py

Full install options (uv, pip, or the platform-specific CUDA paths) are in the labs README.