Pretrain multipack v2 #1470
Merged
Pretrain multipack v2 #1470
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
|
Thank you for the attention to pretraining |
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Outdated suggestions cannot be applied.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.
This PR enables more traditional pretraining, concatenating datasets, but not handling cross attention between the packed sequences. This helps with filling the context length when a sample doesn't come close to using the configured context length.
the multipack buffer size is how far ahead to look to try to pack samples together for pretraining/streaming datasets. What happens is that we pull in N samples and attempt to optimally pack those since we can't pack the entire dataset b/c they are usually too large to pre-process.