Huggingface demo returning the same value

[JVasselli]

I'm trying to use DialogRPT with huggingface, but I've run into the problem that the model returns the same score for every hypothesis given the same context. I tried on the demo, and get this result:

[Mikeygoldman1]

Hi @JVasselli, make sure your transformers library is up to date with the current version. This should solve your issue

[hoppityhop]

This still occurs in the demo notebook when the transformers library is up to date.