]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
kv cache slot search improvements (#3493)
authorKerfuffle <redacted>
Fri, 6 Oct 2023 16:10:13 +0000 (10:10 -0600)
committerGitHub <redacted>
Fri, 6 Oct 2023 16:10:13 +0000 (10:10 -0600)
commit9ca79d5cbbc8d43f2bff951404b6a40ff1ee3788
treed53db5e2b2c7d47d38c4ae4b500d02bf8d624c6c
parent0c731ca4039ccff86ffab90eaae4ca98037c4496
kv cache slot search improvements (#3493)

* kv cache slot search improvements

* Use n_ctx in kv find slot for consistency

* Ensure kv cache head points to a valid slot in llama_decode internal

* Add some comments to prevent dumb people (like me) from getting confused.
llama.cpp