Malaysian SFT
Collection
SFT using LoRA and DoRA including reasoning. • 9 items • Updated
LoRA SFT openai/gpt-oss-120b on initial mesolitica/Malaysian-Reasoning
We follow the same rank as https://huggingface.co/Scicom-intl/gpt-oss-20b-Malaysian-Reasoning-SFT-v0.1#we-only-upload-the-best-model, all linear layers with experts, 256 rank 512 alpha.
Source code at https://github.com/Scicom-AI-Enterprise-Organization/small-ablation/blob/main/malaysian-reasoning
Special thanks to https://www.scitix.ai/ for H100 Node!
Base model
openai/gpt-oss-120b