5 Commits

Author	SHA1	Message	Date
Adam Treat	cf8a4dd868	Infinite context window through trimming.	2023-04-25 11:20:51 -04:00
Adam Treat	a79bc4233c	Implement repeat penalty for both llama and gptj in gui.	2023-04-25 08:38:29 -04:00
Adam Treat	bb78ee0025	Back out the prompt/response finding in gptj since it doesn't seem to help. ... Guard against reaching the end of the context window which we don't handle gracefully except for avoiding a crash.	2023-04-20 17:15:46 -04:00
Adam Treat	51768bfbda	Use default params unless we override them.	2023-04-20 12:07:43 -04:00
Adam Treat	55084333a9	Add llama.cpp support for loading llama based models in the gui. We now ... support loading both gptj derived models and llama derived models.	2023-04-20 06:19:09 -04:00