CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-03 09:22:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : do not allocate KV cache for "vocab_only == true" (#682)

Browse Source 
Fixes sanitizer CI
This commit is contained in:
Stephan Walter
2023-04-02 07:18:53 +00:00
committed by GitHub
parent c4f89d8d73
commit 81040f10aa
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
llama.cpp
View File

@@ -1608,7 +1608,7 @@ struct llama_context * llama_init_from_file(
} }
// reserve memory for context buffers // reserve memory for context buffers
{ if (!params.vocab_only) {
if (!kv_cache_init(ctx->model.hparams, ctx->model.kv_self, memory_type, ctx->model.hparams.n_ctx)) { if (!kv_cache_init(ctx->model.hparams, ctx->model.kv_self, memory_type, ctx->model.hparams.n_ctx)) {
fprintf(stderr, "%s: kv_cache_init() failed for self-attention cache\n", __func__); fprintf(stderr, "%s: kv_cache_init() failed for self-attention cache\n", __func__);
llama_free(ctx); llama_free(ctx);