0
followers ·
1 following Architecte de souveraineté numérique. Mechanistic interpretability, refusal surgery, hybrid SSM–attention. La conscience ne se code pas — elle se conquiert
replied to Felladrin 'spost about 1 hour ago I've open-sourced the trainer I've been using to build tiny language models from scratch, together with the 95M base model I trained with it.
The trainer runs on Deno (https://deno.com, cross-platform), trains on WebGPU, and it writes GGUF directly. No Python/PyTorch. The weights live in a GGUF file from the first step to the last, so every checkpoint is already something llama.cpp can load.
The model is https://huggingface.co/Felladrin/Minueza-3-95M-Base: 94.7M parameters, 1.95B tokens seen, 8192 context.
And here’s the repository on GitHub: https://github.com/felladrin/gguf-trainer
Here on Hugging Face, I published the optimizer state next to the weights, so you can continue the pretraining instead of starting over.
Or start your own from nothing: `deno run -A cli.ts demo` trains a tiny one end to end in under a minute.
And the docs are written for coding agents, so you can point your agent of choice at the GitHub repo and have it drive the whole pipeline. View all activity None yet
