-
Notifications
You must be signed in to change notification settings - Fork 25.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
RoPE loses precision for Llama / Gemma + Gemma logits.float() #29285
Merged
+23
−8
Commits on Feb 26, 2024
-
Llama - Force float32 since bfloat16 loses precision on long contexts
Configuration menu - View commit details
-
Copy full SHA for 7a25720 - Browse repository at this point
Copy the full SHA 7a25720View commit details -
Configuration menu - View commit details
-
Copy full SHA for db8237f - Browse repository at this point
Copy the full SHA db8237fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 3de95c4 - Browse repository at this point
Copy the full SHA 3de95c4View commit details
Commits on Feb 27, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 9e5cbb0 - Browse repository at this point
Copy the full SHA 9e5cbb0View commit details -
Configuration menu - View commit details
-
Copy full SHA for 99d564e - Browse repository at this point
Copy the full SHA 99d564eView commit details -
Configuration menu - View commit details
-
Copy full SHA for d0c08bf - Browse repository at this point
Copy the full SHA d0c08bfView commit details
Commits on Feb 28, 2024
-
Configuration menu - View commit details
-
Copy full SHA for bd3a214 - Browse repository at this point
Copy the full SHA bd3a214View commit details -
Configuration menu - View commit details
-
Copy full SHA for abffebb - Browse repository at this point
Copy the full SHA abffebbView commit details -
Configuration menu - View commit details
-
Copy full SHA for c2e31bf - Browse repository at this point
Copy the full SHA c2e31bfView commit details -
Update src/transformers/models/gemma/modeling_gemma.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for f487800 - Browse repository at this point
Copy the full SHA f487800View commit details -
Update src/transformers/models/llama/modeling_llama.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for c852675 - Browse repository at this point
Copy the full SHA c852675View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1a50a4b - Browse repository at this point
Copy the full SHA 1a50a4bView commit details -
Configuration menu - View commit details
-
Copy full SHA for b860a22 - Browse repository at this point
Copy the full SHA b860a22View commit details -
Configuration menu - View commit details
-
Copy full SHA for 790e4a3 - Browse repository at this point
Copy the full SHA 790e4a3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 06c7634 - Browse repository at this point
Copy the full SHA 06c7634View commit details -
Configuration menu - View commit details
-
Copy full SHA for aa03a43 - Browse repository at this point
Copy the full SHA aa03a43View commit details -
Configuration menu - View commit details
-
Copy full SHA for 5730a50 - Browse repository at this point
Copy the full SHA 5730a50View commit details -
Configuration menu - View commit details
-
Copy full SHA for 31cea3b - Browse repository at this point
Copy the full SHA 31cea3bView commit details -
Configuration menu - View commit details
-
Copy full SHA for ae9957f - Browse repository at this point
Copy the full SHA ae9957fView commit details -
Configuration menu - View commit details
-
Copy full SHA for ec9ef17 - Browse repository at this point
Copy the full SHA ec9ef17View commit details
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.