add support to extend context with xpos rope #181
Merged
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
NanoCode012
reviewed
Jun 10, 2023
| @@ -127,6 +127,14 @@ def load_model( | |||
| # TODO: Check if this would overwrite previous additional_special_tokens | |||
| tokenizer.add_special_tokens({"additional_special_tokens": [MEM_TOKEN]}) | |||
|
|
|||
| if cfg.is_llama_derived_model and cfg.xpos_rope: | |||
There was a problem hiding this comment.
Perhaps it should be elif from above. We should maybe not allow multiple patches?
There was a problem hiding this comment.
it works with xformers_attention,it's specific to the rotary embedding layer which most of the above doesn't touch
There was a problem hiding this comment.
Could you also add this to Readme please?
mkeoliya
pushed a commit
to mkeoliya/axolotl
that referenced
this pull request
Dec 15, 2023
…/xpos-rope add support to extend context with xpos rope
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Outdated suggestions cannot be applied.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.
I'm able to generate a PoC with openllama extended to 4k with this (will release soon)