bert : various improvements #3
Merged
Conversation
This file contains hidden or 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
ImportError: Using `low_cpu_mem_usage=True` or a `device_map` requires Accelerate: `pip install accelerate`
ggerganov
force-pushed
the
gg/mac-improvements
branch
from
8b9845a to
8fbd461
Compare
|
Awesome, thanks for the fixes! |
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Outdated suggestions cannot be applied.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 ran the instructions following the README on a MacOS device and found some minor issues
The
Vtensor does not need to be transposed explicitly - it can be done earlier during theggml_permute(). This should improve the performance a little bit since it will save an extra copyRegarding the
ggml_soft_max_ext()change that I mentioned in thellama.cppissue - it's not going to work because it assumes the mask can be broadcasted across the batches, but it is not the case here. So the way it is implemented is good