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

[InternLM] Add support for InternLM #26302

Merged
merged 3 commits into from
Sep 26, 2023
Merged

Conversation

Rocketknight1
Copy link
Member

InternLM is based on the LLaMA code but adds a config.bias parameter. We can support those models by adding config.bias to LLaMA, and preserve backward compatibility by defaulting it to False

@HuggingFaceDocBuilderDev
Copy link

HuggingFaceDocBuilderDev commented Sep 20, 2023

The documentation is not available anymore as the PR was closed or merged.

Copy link
Collaborator

@ArthurZucker ArthurZucker left a comment

Choose a reason for hiding this comment

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

Thanks! Quick question, do they use dropout after attention?
Let's also add a InternLM.md, add it to the doc etc 😉

src/transformers/models/llama/configuration_llama.py Outdated Show resolved Hide resolved
@Rocketknight1 Rocketknight1 changed the title Add config.bias to LLaMA for InternLM [InternLM] Add support for InternLM Sep 21, 2023
@Rocketknight1
Copy link
Member Author

@ArthurZucker Also, to answer the question, there is no Dropout anywhere in the LLaMA code or the InternLM code.

@Rocketknight1 Rocketknight1 merged commit 6ba63ac into main Sep 26, 2023
@Rocketknight1 Rocketknight1 deleted the port_internlm_as_llama branch September 26, 2023 15:52
blbadger pushed a commit to blbadger/transformers that referenced this pull request Nov 8, 2023
* Add config.bias to LLaMA to allow InternLM models to be ported as LLaMA checkpoints

* Rename bias -> attention_bias and add docstring
EduardoPach pushed a commit to EduardoPach/transformers that referenced this pull request Nov 18, 2023
* Add config.bias to LLaMA to allow InternLM models to be ported as LLaMA checkpoints

* Rename bias -> attention_bias and add docstring
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants