Jeff Tang 7edf3d8df0 llm as a judge running now пре 1 месец
..
deepspeed_zero3.yaml c88e10fab6 grpo llama 3.2 3b with 3 reward functions пре 1 месец
grpo-llama323b-text2sql.yaml 7edf3d8df0 llm as a judge running now пре 1 месец
grpo_text2sql.py 7edf3d8df0 llm as a judge running now пре 1 месец
requirements.txt 57c05170eb added llm as a judge reward func пре 1 месец