zhihui-shao

Results 7 comments of


                                            zhihui-shao

AttributeError: type object 'AttentionMaskConverter' has no attribute '_ignore_causal_mask_sdpa'

同样的问题，有人知道应该怎么解决吗

能不能通过领域数据对模型进行微调

同问如何用领域数据训练

[Bug] Training on 13B causes loss to be 0, while 7B works fine

My training loss on the V100 starts very erratically and then goes to 0

ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9) local_rank: 1

I have the same questions ,do you konw how to solve it ?

error of multi-GPU: torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9) local_rank: 0

> 我按照这个来了，但还是不行，请问还有解决办法吗

Signal 7 error while finetuning with deepspeed

me too. but ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9)

How to modify llama-Xb-hf/tokenizer_config.json from HuggingFace?

the same issue