ChatGLM-Finetuning icon indicating copy to clipboard operation
ChatGLM-Finetuning copied to clipboard

基于ChatGLM-6B、ChatGLM2-6B、ChatGLM3-6B模型,进行下游具体任务微调,涉及Freeze、Lora、P-tuning、全参微调等

Results 67 ChatGLM-Finetuning issues
Sort by recently updated
recently updated
newest added

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 144.00 MiB (GPU 0; 31.75 GiB total capacity; 30.26 GiB already allocated; 68.69 MiB free; 30.33 GiB reserved in total by PyTorch)...

RT,现在的模型会把超出最长限度的文本截断,但是这在做文本摘要的时候可能出现标签(输出)的部分内容事实上对应的是被截断的文本。请问如何解决这一问题,即保留完整文本进行训练?

[2023-04-13 00:21:53,594] [INFO] [comm.py:639:init_distributed] Not using the DeepSpeed or dist launchers, attempting to detect MPI environment... 在这卡住好久, 请问您的MPI是哪个版本呀。mpi4py

[INFO] [comm.py:570:init_distributed] Not using the DeepSpeed or dist launchers, attempting to detect MPI environment...有人遇到这个问题吗