llama.cpp
llama : KV cache view API + better KV cache management
#4170
Merged

llama : KV cache view API + better KV cache management #4170

ggerganov merged 4 commits into master from kv-cache-opts
ggerganov
ggerganov llama : keep track of used KV cells + better KV cache management
79cb8f00
ggerganov ggerganov added need feedback
ggerganov ggerganov requested a review from KerfuffleV2 KerfuffleV2 1 year ago
WeirdConstructor
ggerganov llama : zero KV cache used upon clear
671f639c
KerfuffleV2
KerfuffleV2
KerfuffleV2
KerfuffleV2 approved these changes on 2023-11-23
KerfuffleV2
KerfuffleV2
KerfuffleV2 llama : allow exporting a view of the KV cache (#4180)
5df7d06c
ggerganov ggerganov changed the title llama : keep track of used KV cells + better KV cache management llama : KV cache view API + better KV cache management 1 year ago
ggerganov
KerfuffleV2
KerfuffleV2 commented on 2023-11-23
ggerganov common : add -dkvc arg for enabling kv cache dumps
f8e9f114
ggerganov ggerganov merged 6b0a7420 into master 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone