deepseek-ai/DeepSeek-R1
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepSeek-R1.git synced 2026-11-05 00:47:42 +00:00
Code Issues 45 Projects Packages Wiki Activity

[GH-ISSUE #402] 请问huggingface上的DeepSeek-R1-Distill-Qwen为什么要更新tokenizer_config.json，导致返回的content中<think>不匹配 #316

New issue
Closed
opened 2026-10-26 21:19:18 +00:00 by bsakaguchi · 4 comments
bsakaguchi commented 2026-10-26 21:19:18 +00:00
Owner
Copy link

Originally created by @Flynn-Zh on GitHub (Feb 14, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-R1/issues/402

如题

Image

Originally created by @Flynn-Zh on GitHub (Feb 14, 2025). Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-R1/issues/402 如题 ![Image](https://github.com/user-attachments/assets/0d941879-6e92-4ffc-bcdf-01b7da9082a7)
bsakaguchi 2026-10-26 21:19:18 +00:00
bsakaguchi commented 2026-10-26 21:19:20 +00:00
Author
Owner
Copy link

@liwei-rk commented on GitHub (Feb 14, 2025):

同问，所有的R1系列模型都加了；可能是有时候模型不输出标签，官方强行在Assitant后面加了一个标签 模型去补全。搞不清楚。

<!-- gh-comment-id:2659446994 --> @liwei-rk commented on GitHub (Feb 14, 2025): 同问，所有的R1系列模型都加了；可能是有时候模型不输出<think>标签，官方强行在Assitant后面加了一个<think>标签 模型去补全。搞不清楚。
bsakaguchi commented 2026-10-26 21:19:20 +00:00
Author
Owner
Copy link

@echozyr2001 commented on GitHub (Feb 28, 2025):

在某些情况下，比如询问一个简单的问题：“今天天气怎么样”，会输出 <think>\n\n</think>，也就是模型跳过了思考阶段，思维链没有起作用，这时的回答质量也会非常糟糕。

修改 tokenizer_config.json 文件之后，简单来说就是可以强制模型进行思考，提高回答质量，代价是输出结果只有 </think>。

<!-- gh-comment-id:2690014336 --> @echozyr2001 commented on GitHub (Feb 28, 2025): 在某些情况下，比如询问一个简单的问题：“今天天气怎么样”，会输出 `<think>\n\n</think>`，也就是模型跳过了思考阶段，思维链没有起作用，这时的回答质量也会非常糟糕。 修改 tokenizer_config.json 文件之后，简单来说就是可以强制模型进行思考，提高回答质量，代价是输出结果只有 `</think>`。
bsakaguchi commented 2026-10-26 21:19:20 +00:00
Author
Owner
Copy link

@github-actions[bot] commented on GitHub (Apr 26, 2025):

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!

<!-- gh-comment-id:2831642443 --> @github-actions[bot] commented on GitHub (Apr 26, 2025): This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!
bsakaguchi commented 2026-10-26 21:19:20 +00:00
Author
Owner
Copy link

@github-actions[bot] commented on GitHub (May 29, 2025):

false

<!-- gh-comment-id:2917883410 --> @github-actions[bot] commented on GitHub (May 29, 2025): false
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
v1.0.0
Labels
Clear labels   
bug

   
closed-as-stale

   
closed-as-stale

   
enhancement

   
pull-request
Mirrored from GitHub Pull Request

  
stale

   
stale
No labels
bug
closed-as-stale
closed-as-stale
enhancement
pull-request
stale
stale
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepSeek-R1#316
No description provided.