Jialong Wu comments

Results 20 comments of


                                            Jialong Wu

Update download_lt.py

thx a lot!!

AttributeError: module 'torch.distributed' has no attribute '_all_gather_base'

> I uninstalled the current version and installed "22.04-dev" of apex, then solved the same issue. (torch 1.9.1, cuda 11.1) Thank you very much, the solution works perfectly for me....

Would it be possible to support LoRA fine-tuned models?

mark

Support LoRA adapter

mark

Need to pin notebook version - 7.0.1 breaks install - AttributeError: module 'notebook' has no attribute 'nbextensions'

thx a lot!!

Llama inference instability in fp16 producing inf in the middle of the model

mark

[BUG] `Both max_new_tokens and max_length seem to have been set` warning trigger

And I have submited a PR(https://huggingface.co/THUDM/chatglm2-6b/discussions/105) in [HF repo](https://huggingface.co/THUDM/chatglm2-6b).

[CLI]: BrokenPipeError: [Errno 32] Broken pipe

i meet the same question, wait for fixing it.

Implementation of adapter?

mark

[Feature] <两行代码开启ChatGLM2的LoRA微调>

could you provide your dataset for tuning？