perplexity : faster Winogrande via batching #5024
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
ikawrakow
approved these changes
Jan 19, 2024
| batch.logits + i, | ||
| 0, 0, 0, // unused | ||
| }; | ||
|
|
There was a problem hiding this comment.
The llama_kv_cache_seq_rm call is no longer needed here?
There was a problem hiding this comment.
It's not needed because we clear the entire KV cache before each batch:
In the old implementation, it was reusing tokens from a previous batch, so the llama_kv_cache_seq_rm was used to evict the unused ones (i.e. the second sentence).
|
The score with Mistral 7B has changed in the "positive" direction as well, closer to the HF leaderboard results: |
Yes, I saw this change, but could not understand the reason. |
jordankanter
pushed a commit
to jordankanter/llama.cpp
that referenced
this pull request
Feb 3, 2024
* perplexity : faster Winogrande via batching ggml-ci * perplexity : remove unused function * perplexity : only tokenize selected tasks for Winogrande
hodlen
pushed a commit
to hodlen/llama.cpp
that referenced
this pull request
Apr 1, 2024
* perplexity : faster Winogrande via batching ggml-ci * perplexity : remove unused function * perplexity : only tokenize selected tasks for Winogrande
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Outdated suggestions cannot be applied.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.
Similar to #5017 but for the Winogrande evaluation. Here the performance gain is bigger since the tasks are smaller and we can fit more in a batch.
I'm not 100% sure about the correctness of this implementation - the final value is different from
master:It's possible that it fixes some existing issue, but it is also possible that it introduces a bug. The softmax optimization is not applied yet - it can be applied in a follow-up PR