Historique des commits

Auteur SHA1 Message Date
  Tobi Lütke 189916d6fb Move GRPO training out of default finetune pipeline il y a 3 mois
  Tobi Lutke 599935754b finetune: remove orphaned files and abandoned experiments il y a 3 mois
  Tobi Lütke 102ff861d3 fix: use Qwen3 recommended sampling params to prevent repetition loops il y a 3 mois
  Tobi Lütke bf1b8fc90a lots of training stuff il y a 3 mois