I am trying to run LLaMA 2 on google colab but I get bitsandbytes installation error. I have confirmed that it is in fact installed (version 0.43.0). I have restarted the kernel and done everything I could think of. Is there a compatibility issue? How can I figure it out?
I saw the answer here and tried it. I have the exact same versions of tokenizers, torchaudio, torchvision, and transformers except that I get +cu121 and not 118 but the cuda on colab is 12.2. could it be the issue?
I have tried the following lines of code:
!pip install -U bitsandbytes
!pip install -i https://pypi.org/simple/ bitsandbytes
!pip install bitsandbytes