fix: drop long seq even if not sample packing #2211
+41
−37
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
|
I swore someone just submitted a PR that was merged to remove this |
NanoCode012
commented
Jan 2, 2025
| @@ -171,25 +172,42 @@ def add_length(sample): | |||
| return sample | |||
|
|
|||
|
|
|||
| def drop_long_seq(sample, sequence_len=2048, min_sequence_len=2): | |||
| def drop_long_seq(sample, sequence_len=2048, min_sequence_len=None): | |||
| min_sequence_len = min_sequence_len or 2 | |||
There was a problem hiding this comment.
If we're defaulting here, should we just set the default within
to be more explicit?
NanoCode012
force-pushed
the
fix/drop_long_seq_if_not_packing
branch
from
b884f98 to
0eaa579
Compare
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Outdated suggestions cannot be applied.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.
Description
emptireported in discord https://discord.com/channels/1104757954588196865/1111279858136383509/1320021873878372436 that sequences are truncated not dropped if sample_packing is off. This PR tries to fix that. RL path seems fine, only SFT is missing.[ ] Check for pretraining case
[ ] Check handling for if
skip_prepare: TrueMotivation and Context
How has this been tested?
Untested!
Screenshots (if appropriate)
Types of changes
Social Handles (Optional)