eagle0504/multireward-grpo-fintech-an-qwen2.5-1.5b
04
Add multi-seed training curves + held-out eval comparison
Add multi-seed training curves + held-out eval comparison
Upload metrics.json with huggingface_hub
GRPO (AN advantage) — final checkpoint
Upload metrics.json with huggingface_hub
GRPO (AN advantage) — final checkpoint
initial commit
