mirror of
https://github.com/deepseek-ai/DeepSeek-LLM.git
synced 2026-11-05 00:57:25 +00:00
[GH-ISSUE #32] AlignBench测评结果复现求助 #20
Labels
No labels
bug
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepSeek-LLM#20
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 @FoolMark on GitHub (Jan 11, 2024).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-LLM/issues/32
注意到你们的模型在alignbench上的sota表现于是尝试复现了一下
我的认知里这个应该是低于预期的(虽然没有控制变量), 我推测大概是生成过程的问题, 我这边简单参考了huggingface上提供的例子写的generate过程如下,大概就按照官方的setting改了temperature参数,其他都是default
请问如果要复现tech report中相近的精度,有没有更正确的template? 谢谢!
@DeepSeekPH commented on GitHub (Jan 15, 2024):
你好,我们只用过GPT-4 score + VLLM推理的结果。https://github.com/deepseek-ai/DeepSeek-LLM/pull/35 具体的输出可以在这里面看到。你用huggingface的方法看上去没有问题。
@FoolMark commented on GitHub (Jan 15, 2024):
感谢回复,这里也repo一下,拿你提供的结果上传测评,score是在5.90左右
download.csv
目前看来AlignBench官方提供的CritiqueLLM 模型在高分段可能和GPT4的一致性没有这么高,感觉还是以GPT4的得分为准比较妥当