zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-07-12 20:51:13 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

quick prompting fix

Browse Source
This commit is contained in:
Goekdeniz-Guelmez 2025-02-12 19:24:35 +01:00
parent a7273f6a56
commit 8179b99436
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
llms/mlx_lm/tuner/datasets.py
View File

@ -30,7 +30,7 @@ class GRPODataset:
prompt_tokens = tokenizer.apply_chat_template(
[
{'role': 'system', 'content': """A conversation between User and Assistant. The user asks a question, and the Assistant solves it.
The assistantfirst thinks about the reasoning process in the mind and then provides the user with the answer.
The assistant first thinks about the reasoning process in the mind and then provides the user with the answer.
The reasoning process and answer are enclosed within <think> </think> and <answer> </answer> tags, respectively, i.e., <think> reasoning process here </think><answer> answer here </answer>."""},
{'role': 'user', 'content': prompt_str}
],
@ -39,7 +39,7 @@ class GRPODataset:
else:
if use_prompt:
prompt_tokens = tokenizer.encode(f"""A conversation between User and Assistant. The user asks a question, and the Assistant solves it.
The assistantfirst thinks about the reasoning process in the mind and then provides the user with the answer.
The assistant first thinks about the reasoning process in the mind and then provides the user with the answer.
The reasoning process and answer are enclosed within <think> </think> and <answer> </answer> tags, respectively, i.e., <think> reasoning process here </think><answer> answer here </answer>.
User: {prompt_str} Assistant: """)
else: