Closed keyboardAnt closed 1 year ago
I believe this can be fixed by running pip install --upgrade transformers
Yes like arjun said this isn't an evaluation harness issue you need to upgrade your transformers version to >=4.28.1
where the gpt_bigcode
architecture was introduced to be able to load the model.
I couldn't load
HuggingFaceH4/starchat-beta
using our existing codebase on a single machine with multiple 40GB GPUs. See loading instructions on the model card.