0.2.3

0.2.3

0.2.2

What's Changed

• fix ppo accelerate by @ell-hol in #62

Full Changelog: 0.2.1...0.2.2

0.2.1

fix a bug with the final norm in palm, thanks to @conceptofmind and @…

0.2.0

address https://github.com/lucidrains/PaLM-rlhf-pytorch/issues/41 , b…

0.1.4

old action log probs should be the true distribution in the kl div lo…

0.1.2

flash attention sdp context config only needs to be done once

0.1.1

fix assert

0.1.0

add ability to use flash attention if using pytorch 2.0, thanks to @c…

0.0.68

0.0.68

0.0.67

fix silly error in masked kl div loss, thanks to @taynoel84