llama.cpp icon indicating copy to clipboard operation
llama.cpp copied to clipboard

llama : rename n_ctx to kv_size

Open ggerganov opened this issue 4 months ago • 3 comments

The n_ctx name is causing some confusion since it's actual meaning is the size of the KV cache, while n_ctx_train is the training context of the model

This change fixes that, but since it is a big one and touches a lot of stuff, I'm not sure if it worth merging. Maybe sometime in the future, when the time is right

Original PR: https://github.com/ggerganov/llama.cpp/pull/5546

ggerganov avatar Feb 18 '24 20:02 ggerganov