]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)
authorKerfuffle <redacted>
Sun, 29 Oct 2023 17:31:40 +0000 (11:31 -0600)
committerGitHub <redacted>
Sun, 29 Oct 2023 17:31:40 +0000 (11:31 -0600)
commit6e08281e588bbba1a5d180290a94a43f167f3a1a
tree46add394417eb2b5929793ca879c793a478fd3f8
parent2046eb4345e62c4575b3cdc0115a51db89f3fb70
Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)

* Extend llama_kv_cache_seq_rm to allow matichng any sequence

* Replace llama_kv_cache_tokens_rm with llama_kv_cache_clear

Use llama_kv_cache_clear for cache clearing

Change calls to llama_kv_cache_tokens_rm that want to delete by position to use llama_kv_cache_seq_rm functionality
common/common.cpp
examples/batched-bench/batched-bench.cpp
examples/llama-bench/llama-bench.cpp
examples/main/main.cpp
examples/perplexity/perplexity.cpp
examples/server/server.cpp
llama.cpp
llama.h