mirror of
https://github.com/deepseek-ai/DeepSeek-R1.git
synced 2026-11-05 00:47:42 +00:00
[GH-ISSUE #606] 有人发现对Distill模型SFT之后幻觉变的非常严重吗 #487
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepSeek-R1#487
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @dayL-W on GitHub (Apr 1, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-R1/issues/606
如题：
我使用DeepSeek-R1-Distill-Qwen-7B在自己的数据集上做SFT，发现训练之后的模型幻觉变的非常严重。
有人发现这个现象吗？为什么导致的，怎样解决呢？
@github-actions[bot] commented on GitHub (May 12, 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!
@Kenwwww commented on GitHub (May 19, 2025):
用全参sft，加大数据量和轮次，混入更多任务数据集
@Evi233 commented on GitHub (Jul 1, 2025):
reinforcement learning after SFT
@github-actions[bot] commented on GitHub (Aug 2, 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!
@github-actions[bot] commented on GitHub (Aug 18, 2025):
false