early exit #34244
Merged
early exit #34244
+130
−29
Conversation
This file contains 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
| inputs_tensor: Optional[torch.Tensor] = None, | ||
| logits_processor: "LogitsProcessorList" = None, | ||
| ): | ||
| # TODO(joao): somehow check whether the model supports early exit |
There was a problem hiding this comment.
should we add _supports_early_exist ? hasattr(model, "active_layers")
There was a problem hiding this comment.
IMO it depends on how the model is structured/trained 👀
- if the model is expected to have early exit at ANY layer because the lm head is compatible with all layers -> there is no way to detect unless we manually add an argument in the config, which is... brittle. Probably I would suggest to not do any check for now?
- If the model is expected to have early exit on specific layers, store those layers in the config and check that attribute here.
WDYT?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
What does this PR do?