maocheno comments

Repositories
Issues
Comments

Results 3 comments of


                                            maocheno

'loss': 0.0 and 'grad_norm' remains the same in all steps in task lora finetuning

> I got the same issue. have you solved yet? 你好解决了吗？

'loss': 0.0 and 'grad_norm' remains the same in all steps in task lora finetuning

> probably max_length is too small, GT will be truncated? my max_length is 512. But when i use lora , i find that appears /root/lanyun-tmp/env/llava_next/lib/python3.10/site-packages/torch/utils/checkpoint.py:61: UserWarning: None of the inputs...

'loss': 0.0 and 'grad_norm' remains the same in all steps in task lora finetuning

> probably max_length is too small, GT will be truncated? Bro, you are right. When I delete the, everyting is fine. That may be will occupy too much tokens.