kyy/llm_trainer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

support SimPO #3900

Browse Source
This commit is contained in:
hiyouga
2024-05-26 23:46:33 +08:00
parent 063f91cc80
commit cb63b32986
19 changed files with 145 additions and 339 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/llamafactory/train/utils.py
View File

@@ -90,7 +90,7 @@ def create_ref_model(
)
)
ref_model_args = ModelArguments(**ref_model_args_dict)
ref_finetuning_args = FinetuningArguments(finetuning_type="lora")
ref_finetuning_args = FinetuningArguments()
tokenizer = load_tokenizer(ref_model_args)["tokenizer"]
ref_model = load_model(
tokenizer, ref_model_args, ref_finetuning_args, is_trainable=False, add_valuehead=add_valuehead
@@ -146,7 +146,7 @@ def create_reward_model(
)
)
reward_model_args = ModelArguments(**reward_model_args_dict)
reward_finetuning_args = FinetuningArguments(finetuning_type="lora")
reward_finetuning_args = FinetuningArguments()
tokenizer = load_tokenizer(reward_model_args)["tokenizer"]
reward_model = load_model(
tokenizer, reward_model_args, reward_finetuning_args, is_trainable=False, add_valuehead=True