DachengLi1/LongChat Issues
How to prepare the training data
Updated 2Update Anthropic Client
Updated 2license
Closed 1OOM issue
Closed 4flash attention rename
Closed 1train ValueError
UpdatedOutput token limit
UpdatedMaybe a bug in the preprocess?
Updated 3About the print message
Updated 2About the learning rate
Updated 1Xformers Monkey Patch Compatibility
Updated 1Longchat inference configuration
Updated 1longchat-13b-16k chat not work
Updated 9Web GUI for longchat
Updated 3The purpose of pretrain script?
Closed 2why not reuse fschat code?
Closed 8Will it support qlora?
Updated 1How to use 3090 to train 16k model?
Updated 7Multi-node training?
Closed 1Load the model for inference?
Closed 4