lucidrains/PaLM-rlhf-pytorch Issues
Flash Attention 2
ClosedHow to use lora?
UpdatedModel Name
Closed 3A few questions on training
Updated 3norm.gamma not used during backprop
Closed 2KL divergence loss
Closed 1train your reward model issue
Updated 1mask raised error
Closed 2Value function
UpdatedDo you need cuda for this?
Closed 1value function input
Closed 1The loss function of reward model.
Updated 2KL_div/ratio on policy
ClosedEncoder-Decoder
Closed 39Training the reward model
Closed 8PaLM-rlhf-pytorch Roadmap
Closed 4Help with computational power
Closed 4Simple Web Interface
Closed 2Palm
ClosedI'm dumb
Closed 1Can I train a model on my own data?
Closed 1GPU requirements
Closed 3