mirror of
https://github.com/hiyouga/LLaMA-Factory.git
synced 2025-08-04 04:32:50 +08:00
Update workflow.py
Former-commit-id: eeb5249d0b6ce0816e1fa47afc3a853c7b267cbf
This commit is contained in:
parent
ec910a87c0
commit
a419122179
@ -45,7 +45,7 @@ def run_ppo(
|
||||
mini_batch_size=training_args.per_device_train_batch_size,
|
||||
batch_size=training_args.per_device_train_batch_size * training_args.gradient_accumulation_steps,
|
||||
gradient_accumulation_steps=training_args.gradient_accumulation_steps,
|
||||
ppo_epochs=1,
|
||||
ppo_epochs=finetuning_args.ppo_epochs,
|
||||
max_grad_norm=training_args.max_grad_norm,
|
||||
seed=training_args.seed,
|
||||
optimize_device_cache=True,
|
||||
|
Loading…
x
Reference in New Issue
Block a user