add gist model generation utils to library #23
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
- Creates cache.py - Introduces global_tokens - Formats repo with ruff - Speed parity with full KV-cache
Implements window KV-Cache Compression Strategy
… Hitter methods which require attention probs.
… update() function.
Implement Scissorhands KV-cache compression & SnapKV prompt compression
commit 0528eef Author: Faisal Ladhak <[email protected]> Date: Thu Jun 13 20:30:35 2024 +0000 Reformatting with ruff commit 4c3cfc5 Author: Faisal Ladhak <[email protected]> Date: Thu Jun 13 04:35:05 2024 +0000 Added support for Qwen2 models.
Fix various bugs
Adding code for evaluation, along with Squality and refernce-based metrics.
- This is necessary for training models outside of eval.py
Update Tasks to allow for train, val, test splits to be used.
Adding four tasks from RULER
griff4692
reviewed
Jun 27, 2024
…eing modified across sub-classes.
uSaiPrashanth
force-pushed
the
gist-token-merged
branch
from
3f9be25 to
12be67c
Compare
uSaiPrashanth
force-pushed
the
gist-token-merged
branch
from
0146e6c to
9056c97
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Outdated suggestions cannot be applied.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.
Other modifications worth mentioning:
scripts/convert_hf_checkpoint.pyto support loading of finetuned Llama-3 models from safetensors state dictmodel.py(Finetuned models use a vocab size of one token higher than Llama 3)