cernenv-trainer / training

Commit History

sft+reward-fix: training/training_script.py
2b97998
verified

anugrahhu commited on

sft+reward-fix: training/sft_warmstart.py
a8d4d87
verified

anugrahhu commited on

vanilla GRPO: backport EvidenceCallback for live evidence/*.csv + plots
11307a1
verified

anugrahhu commited on

fix: disable fast_inference (vLLM not installed) in training/evaluate.py
8f805e2
verified

anugrahhu commited on

fix: disable fast_inference (vLLM not installed) in training/training_unsloth.py
f82f913
verified

anugrahhu commited on

Update CERNenv Space
0a6c641
verified

anugrahhu commited on