edumunozsala / llama-2-7b-4bit-python-coder Goto Github PK

hi

thz for so comprehensive guide which it is really helpful for me to understanding the state-of-art in the field. However, I could not run the code which you shared on Colab.

There are the Erorrs :
#1 on # Define the training arguments box: ValueError: Your setup doesn't support bf16/gpu. You need torch>=1.10, using Ampere GPU with cuda>=11.0
so I charged it from "fp16 = False, bf16 = True" to "fp16 = True, bf16 = False" on the Setting Global Parameters cell. is it ok to charge it?

#2 after # train cell called OutOfMemoryError since "CUDA out of memory. Tried to allocate 2.00 GiB (GPU 0; 14.75 GiB total capacity; 10.23 GiB already allocated; 790.81 MiB free; 12.92 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF"
this time I have no ideas to fix :(