Published finedtuned model prompts for token?
#72
by
gnumanth
- opened
I train a simple dataset and published it here but seeing the error in the model interface:
You are trying to access a gated repo. Make sure to request access at https://huggingface.co/google/gemma-7b and pass a token having permission to this repo either by logging in with `huggingface-cli login` or by passing `token=<your_token>`.
Also, I am not seeing the "use with transformers" button (needs an additional step to add the model to transformers?)
gnumanth
changed discussion title from
Published model prompts for token?
to Published finedtuned model prompts for token?
Looks like it was gated, now I am seeing:
The API does not support running gated models for community model with framework: peft
Hi there! You trained a model with PEFT. The error is about Hugging Face free serverless API services, which support only some models. You now do have a Use with PEFT button!
osanseviero
changed discussion status to
closed
Thank you @osanseviero
Finally noticed, but PEFT is heavy. SFTTrainer from trl has an option to use transformers ? This was via transformers and I see the Interface API supporting it.