Chiranshu Adik

Did you get any solution for this?

I tried with small and medium models it's the same for me. I followed this tutorial to fine-tune the model. - https://towardsdatascience.com/make-your-own-rick-sanchez-bot-with-transformers-and-dialogpt-fine-tuning-f85e6d1f4e30 The author's results look really good, but surprisingly...

>  > Pls, help to remove this unwanted behavior, or suggest some ways to do it. > Thanks @alan-ai-learner You can try fine tuning it with custom dataset to...

If you have to use GPT, then best thing would be to fine-tune with super clean dataset. If you always want clean responses, try rasa, it will for sure give...

@ArEnSc , were you able to get the expected results? I also tried the same model and it's not giving quality replies. Not sure what I am missing here

@ArEnSc Thanks for your response. It's surprising to me because they mentioned that it is ready to use. I'll look up more on using a custom decoder.

@ArEnSc sure Michael, could you please tell me more on what exactly needs to be done here? I'm kinda new to this, by using a custom decoder do you mean...

@ArEnSc yes, I have been trying this and other sample decoding scripts on the readme page. None of them have worked so far, they all have issues while loading the...

@golsun @ArEnSc Thanks for your response Xiang Gao, I'll give that a try today. Will keep you both updated about how it goes.

@golsun Does a higher rank mean better response? For these simple examples with your demo notebook, the results are not as expected. Example 1- Context: Hi 0.301 gen 0.367 ranker...