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

Update rl_replacements.py (#1754)

Browse source 
Fix typo in comment: know -> now.

This was printed when running the Llama3.1_(8B)-GRPO.ipynb example notebook, so I'd expect others to run into it as well.
This commit is contained in:
Seth Weidman 2025-02-19 02:12:07 -08:00 • committed by GitHub
commit 9f354c1447
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
unsloth/models/rl_replacements.py
View file

@ -268,7 +268,7 @@ def grpo_trainer_fix_batch_size(RLTrainer_source, RLConfig_source):
check_batch_size = \
"div = per_device_train_batch_size // num_generations\n"\
"if div * num_generations != per_device_train_batch_size:\n"\
" print('Unsloth: We know expect `per_device_train_batch_size` to be a multiple of `num_generations`.\\n"\
" print('Unsloth: We now expect `per_device_train_batch_size` to be a multiple of `num_generations`.\\n"\
"We will change the batch size of ' + str(per_device_train_batch_size) + ' to the `num_generations` of ' + str(num_generations))\n"\
" per_device_train_batch_size = num_generations\n"
return check_batch_size