Make rotary freqs buffer non-persistent #1168
Merged
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Outdated suggestions cannot be applied.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.
closes #1161 .
Makes the
inv_freq
rotary embedding buffer not saved in checkpoints, in keeping with NeoX, Mistral, and Llama in Huggingface post-4.36 .Since this is a breaking change to saved checkpoints in GPT-NeoX which do store it, added a flag so people can turn it back on and load from the checkpoint.