-
Notifications
You must be signed in to change notification settings - Fork 3k
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
[Tokenizer] Fix tokenizer of llama3.3 #9641
[Tokenizer] Fix tokenizer of llama3.3 #9641
Conversation
Thanks for your contribution! |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #9641 +/- ##
===========================================
+ Coverage 52.76% 52.80% +0.04%
===========================================
Files 710 710
Lines 111231 111234 +3
===========================================
+ Hits 58688 58742 +54
+ Misses 52543 52492 -51 ☔ View full report in Codecov by Sentry. |
…ish19/PaddleNLP into dev_20241216_fix_llama3_tokenizer
@@ -1601,7 +1601,8 @@ def _prepare_decoder_attention_mask(attention_mask, input_shape, past_key_values | |||
expanded_attn_mask = expanded_attn_mask.astype(dtype) | |||
expanded_attn_mask = paddle.where(expanded_attn_mask, x, y).astype(dtype) | |||
else: | |||
expanded_attn_mask = paddle.where(expanded_attn_mask, 0.0, paddle.finfo(dtype).min).astype(dtype) | |||
expanded_attn_mask = paddle.where(expanded_attn_mask.cast("bool"), 0.0, paddle.finfo(dtype).min) | |||
expanded_attn_mask = expanded_attn_mask.astype(dtype) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这里PaddleNLP-CI已经验证
PR types
Bug fixes
PR changes
Others
Description
Fix tokenizer of llama3.3