laqwenta_sft_model / README.md
jeanprbt's picture
doc: update model card
d6c3e4b verified
metadata
library_name: transformers
datasets:
  - meta-math/MetaMathQA
base_model:
  - Qwen/Qwen3-0.6B-Base
pipeline_tag: text-generation

This model is a fine-tuned version of Qwen3-0.6B-Base, fine-tuned on a sub-sample of 6k pairs from MetaMathQA dataset using SFT and the trl library. It is the first step of LaQwenTa, a light-weight STEM QA answering model for educational purposes.