fix llama alibi pretrain

[lugimzzz]

PR types

Bug fixes

PR changes

Models

Description

修复alibi 预训练 问题

[paddle-bot]

Thanks for your contribution!

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.

---

zhangjunjun04 seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[codecov]

Codecov Report

Attention: Patch coverage is 8.33333% with 11 lines in your changes missing coverage. Please review.

Project coverage is 55.79%. Comparing base (4e971f8) to head (c1f21bf).
Report is 1 commits behind head on develop.

Files	Patch %	Lines
paddlenlp/transformers/llama/modeling_pp.py	0.00%	11 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #8668      +/-   ##
===========================================
- Coverage    55.80%   55.79%   -0.01%     
===========================================
  Files          620      620              
  Lines        96642    96651       +9     
===========================================
  Hits         53928    53928              
- Misses       42714    42723       +9     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.