wassname/SimPO
1
0
Fork 0
mirror of https://github.com/wassname/SimPO.git synced 2026-06-27 16:43:59 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix on-policy data order

Browse Source
This commit is contained in:
Yu Meng
2024-08-22 16:06:04 -04:00
parent 54545e803b
commit 2dcc4350f8
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
on_policy_data_gen/decode.py
+1 -1
View File
@@ -30,7 +30,7 @@ tokenizer = llm.get_tokenizer()
train_dataset= load_dataset(data_dir, split='train_prefs') train_dataset= load_dataset(data_dir, split='train_prefs')
prompts = list(set(train_dataset['prompt'])) prompts = sorted(list(set(train_dataset['prompt'])))
conversations = [tokenizer.apply_chat_template([{'role': 'user', 'content': prompt}], tokenize=False, add_generation_prompt=True) for prompt in prompts] conversations = [tokenizer.apply_chat_template([{'role': 'user', 'content': prompt}], tokenize=False, add_generation_prompt=True) for prompt in prompts]