add conv1d layer type (used in gpt2)#50
Merged
zywilliamli merged 7 commits intoEleutherAI:mainfrom Oct 16, 2025
Merged
Conversation
luciaquirke
reviewed
Oct 16, 2025
luciaquirke
reviewed
Oct 16, 2025
bergson/gradients.py
Outdated
| """Process the incoming gradient wrt the output of the module.""" | ||
| # Sanity checks | ||
| assert isinstance(module, nn.Linear), "Expected a Linear module" | ||
| assert isinstance(module, LayerAdapter.supported_modules), "Expected a supported module" |
Collaborator
There was a problem hiding this comment.
Could we print the supported modules here?
|
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
gpt2 type models use transformers.pytorch_utils.Conv1D instead of nn.Linear which is not currently tracked by gradient collector, this pr adds that layer type.
tested manually using python -m bergson runs/test --model openai-community/gpt2 --dataset NeelNanda/pile-10k --truncation, also added unit tests for it.