The model returns constantly </s>

#1
by simonesartoni1 - opened

The model returns to most of the answers. I have tried the GPTQ quantized version and it works, while this one does not. Could there have been an error in the quantization process or in respecting the templates?

Sign up or log in to comment