How is this model different from the one listed on Hugging Face by Meta?

#13

by mhamzaathar - opened Jul 11

Discussion

mhamzaathar

Jul 11

Is this a fine-tuned version? To perform better on the leaderboard?

MaziyarPanahi

Owner Jul 11

Hi,

This model is a merge of my other fine-tuned models (v0.4 and v0.2 if I am not mistaken), then it was DPO further. I use a group of datasets I like and I think they are more diverse and open to conversations, the fact it does better on the leaderboard is just a nice coincidence :)

MaziyarPanahi changed discussion status to closed Jul 11

AndyLuo1029

Jul 23

Hi Dear Author,
Could you please list the dataset you used to tune this model? I'm curious about what kind of dataset can make the model performs so good on the leaderboard.

Thanks!

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment