Historial de Commits

Autor SHA1 Mensaje Fecha
  Tobi Lütke d6f3688d91 Remove grpo command from default train entrypoint hace 3 meses
  Tobi Lütke 189916d6fb Move GRPO training out of default finetune pipeline hace 3 meses
  Tobi Lütke cbeeb1f89b Add wall-clock checkpoints and full eval defaults hace 3 meses
  Tobi Lütke bf1b8fc90a lots of training stuff hace 3 meses
  Tobi Lutke 38073799c0 chore: clean up finetune folder and fix training workflow hace 3 meses
  Tobi Lütke 46ff098361 Change only: format to only:lex (no space after colon) hace 3 meses
  Tobias Lütke eb1b77c8cb Deploy fine-tuned GRPO model as default query expansion (#67) hace 3 meses