| --- |
| title: Weight-Space Programming Lab |
| emoji: 🧮 |
| colorFrom: gray |
| colorTo: indigo |
| sdk: static |
| app_file: index.html |
| pinned: false |
| short_description: "Hands-on labs: fix a small LLM by editing its weights." |
| --- |
| |
| # Weight-space programming lab |
|
|
| This Space holds nine hands-on labs that reproduce and probe Experiment 8 of "Sparks of Weight-Space Programming". In that experiment, an AI agent fixed the looping of Qwen3.5-0.8B. It edited 131,072 weights with arithmetic and did no training. |
|
|
| - Read the labs on the [live site](https://thegovind-weight-space-programming.static.hf.space/index.html). |
| - Start with the [Quick start](https://thegovind-weight-space-programming.static.hf.space/quickstart.html) page. |
| - Browse the code in the [code/ folder](https://huggingface.co/spaces/thegovind/weight-space-programming/tree/main/code/). |
|
|
| ## Get the code |
|
|
| The first line below installs `uv`, which runs the lab's Python scripts with the packages they need and can fetch Python itself. The second line puts `uv` on your PATH (the folders the shell searches for commands) in this terminal. The `uvx` command then runs `hf`, the download tool from `huggingface_hub`, without installing it, and fetches the `code/` folder: |
|
|
| ```bash |
| curl -LsSf https://astral.sh/uv/install.sh | sh |
| source $HOME/.local/bin/env |
| uvx --from huggingface_hub hf download thegovind/weight-space-programming \ |
| --repo-type space --include "code/*" --local-dir ~/weight-space-programming |
| cd ~/weight-space-programming/code |
| ``` |
|
|
| If you already have `uv`, skip the first two lines. |
|
|
| ## Credits and licenses |
|
|
| - The project reproduces Experiment 8 of the blog post ["Sparks of Weight-Space Programming"](https://rikiyat.github.io/blog/weight-space-programming/) by the authors of "LLMs can understand and write neural network weights". It uses their [public logs](https://github.com/RikiyaT/lm_understand-logs). |
| - The agent's 8 texts in `code/labs/data/astra_pairs.json` and the quoted log excerpts come from [RikiyaT/lm_understand-logs](https://github.com/RikiyaT/lm_understand-logs). They are copyright the authors of "LLMs can understand and write neural network weights" and licensed [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). We extracted them from the logs. The selection and the JSON layout are ours. The labs, measurements and figures come from our own runs. |
| - `code/wsp/grading.py` includes answer-extraction and grading code copied from the paper's harness (`exp8_math/code/mathqa.py`). It is used under the MIT License. See [code/NOTICE.md](https://huggingface.co/spaces/thegovind/weight-space-programming/blob/main/code/NOTICE.md). |
| - Benchmarks: MATH-500 (from MATH, MIT License), GSM8K (MIT License) and IFEval (Apache License 2.0). The benchmark files are not included. `code/scripts/prepare_data.py` downloads them again from their public copies on Hugging Face. See [code/data/README.md](https://huggingface.co/spaces/thegovind/weight-space-programming/blob/main/code/data/README.md). |
| - No license is granted yet for this project's own code and text. |
|
|
