ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Do not force set beta to 0 for DAPO (#3604)

Browse source
This commit is contained in:
Datta Nimmaturi 2025-11-17 12:09:36 +05:30 • committed by GitHub
commit 0f0b1fba15
1 changed files with 1 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

3
unsloth/models/rl.py
View file

@ -814,10 +814,9 @@ def _patch_trl_rl_trainers(trainer_file = "grpo_trainer"):
" if epsilon_high != 0.28:\n"
" print('Unsloth: The DAPO paper recommends `epsilon_high = 0.28` - we will set it.')\n"
" if beta != 0.0:\n"
" print('Unsloth: The DAPO paper recommends setting `beta = 0.0` to remove the KL term - we will set it.')\n"
" print(f'[WARNING] Unsloth: The DAPO paper recommends setting `beta = 0.0` to remove the KL term - You have set it to {beta}.')\n"
" mask_truncated_completions = True\n"
" epsilon_high = 0.28\n"
" beta = 0.0\n"
"\n"
)
extra_args += check_dr_grpo