What exactly is the SFT model post-trained on?

#23
by Phaserblast - opened

The SFT model is fine-tuned from the base model, but I can't find any documentation on ๐Ÿ‘‰what.๐Ÿ‘ˆ Like, what genre(s) of music is this more likely to generate in the "high" quality as explained in the docs? Please explain because if you're targeting a particular style of music, it would be helpful to know what this is tuned on.

Sign up or log in to comment