AnswerDotAI / AnswerDotAI/ModernBERT

grad_norm Nan question

Open
#233 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
1.7k
Forks
145
PR merge metrics
No merged PRs in 30d

Description

{'loss': 0.2051, 'grad_norm': nan, 'learning_rate': 9.774074342405685e-06, 'epoch': 0.53}
{'loss': 0.0, 'grad_norm': nan, 'learning_rate': 9.773696745450087e-06, 'epoch': 0.53}
{'loss': 0.0, 'grad_norm': nan, 'learning_rate': 9.773318840518906e-06, 'epoch': 0.53}
How to adjust the gradient such as Nan loss and 0 when training to epoch=0.53?

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.