financial-task-env / train_sft.py

Commit History

train_sft: drop fp16, prefer bf16 (MPS-compatible without grad scaler)
35fa944

bpHigh Claude Opus 4.7 (1M context) commited on

Phase 10: SFT training script (Qwen2.5-Coder-3B + LoRA via TRL)
c803cd5

bpHigh Claude Opus 4.7 (1M context) commited on