fix: Compute the full sum in llama-eval-callback #15637
Merged
+32
−18
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
… printed values This makes it much easier to compare between llama.cpp and transformers! https://github.com/ggml-org/llama.cpp/issues/nemotron-nano-15409 Branch: gabe-l-hart/nvidia-nemotron-nano-15409 Signed-off-by: Gabe Goodhart <[email protected]>
|
@ggerganov I'm assigning you on this since you gave a 👍 to my comment on the other PR, but feel free to forward |
ggerganov
approved these changes
Aug 28, 2025
Minh141120
pushed a commit
to menloresearch/llama.cpp
that referenced
this pull request
Aug 29, 2025
… printed values (ggml-org#15637) This makes it much easier to compare between llama.cpp and transformers! https://github.com/ggml-org/llama.cpp/issues/nemotron-nano-15409 Branch: gabe-l-hart/nvidia-nemotron-nano-15409 Signed-off-by: Gabe Goodhart <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
Previously,
llama-eval-callback's printedsumwas only the sum of the values that got printed since the printing function compacted its view by jumping indices. This PR makes a double pass over each tensor, once iterating all values to compute thesumand once as before to create the printed view. This makes it much easier to compare between llama.cpp and transformers.