Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Phi3 awq #1984

Merged
merged 29 commits into from
Jul 31, 2024
Merged

Phi3 awq #1984

merged 29 commits into from
Jul 31, 2024

Conversation

grimoire
Copy link
Collaborator

@grimoire grimoire commented Jul 10, 2024

@lvhan028 lvhan028 requested a review from AllentDan July 19, 2024 07:28
Copy link
Collaborator

@AllentDan AllentDan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I tested InternVL2-4B, the AWQ of PyTorch backend worked fine

'InternLMXComposer2ForCausalLM': 'InternLM2RMSNorm',
'ChatGLMForConditionalGeneration': 'RMSNorm',
}
NORM_TYPE_MAP = NORM_TYPE_MAP # legency
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

legency -> legacy
But I don't understand why we should make the NORM_TYPE_MAP=NORM_TYPE_MAP statement

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

NORM_TYPE_MAP is not used in this file, unused import will lead to lint error. And I don't know if other files will reference it or not.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Get it. May fix the typo "legency -> legacy"

@lvhan028 lvhan028 added the enhancement New feature or request label Jul 31, 2024
@lvhan028 lvhan028 merged commit 9165305 into InternLM:main Jul 31, 2024
5 checks passed
@douyh
Copy link

douyh commented Aug 7, 2024

How can I use thehttps://github.com/InternLM/lmdeploy/releases/tag/v0.5.2.post1 merged version? The latest release version 请问这个PR是还没有合入到最新的release版本中吗?

@lvhan028
Copy link
Collaborator

lvhan028 commented Aug 7, 2024

合入到 v0.5.3 中了

@douyh
Copy link

douyh commented Aug 7, 2024

合入到 v0.5.3 中了
量化的过程没有报错。
使用lmdeploy起server的时候报错了

model_name=InternVL2-4B-AWQ
CUDA_VISIBLE_DEVICES=1 lmdeploy serve api_server $model_name --backend turbomind --model-format awq  --server-port 10004 --session-len 100000 --tp 1 --server-name 0.0.0.0

错误

 undefined symbol: _ZN2at4_ops15sum_dim_IntList4callERKNS_6TensorEN3c1016OptionalArrayRefIlEEbNS5_8optionalINS5_10ScalarTypeEEE

@grimoire @lvhan028

因为提示我更新autoawq到0.1.8以上,我就pip install autoawq==0.1.8了。然后报的这个错误。

@douyh
Copy link

douyh commented Aug 7, 2024

合入到 v0.5.3 中了
量化的过程没有报错。
使用lmdeploy起server的时候报错了

model_name=InternVL2-4B-AWQ
CUDA_VISIBLE_DEVICES=1 lmdeploy serve api_server $model_name --backend turbomind --model-format awq  --server-port 10004 --session-len 100000 --tp 1 --server-name 0.0.0.0

错误

 undefined symbol: _ZN2at4_ops15sum_dim_IntList4callERKNS_6TensorEN3c1016OptionalArrayRefIlEEbNS5_8optionalINS5_10ScalarTypeEEE

@grimoire @lvhan028

因为提示我更新autoawq到0.1.8以上,我就pip install autoawq==0.1.8了。然后报的这个错误。

autoawq和transformers升级到最新版本之后解决了😊

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants