vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Use uint32_t for seed

Browse source 
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Howard Su 2023-06-28 15:51:56 -07:00 • committed by GitHub
parent 5a16205274
commit d7d454f227
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama.h
View file

@ -198,7 +198,7 @@ extern "C" {
LLAMA_API int llama_get_kv_cache_token_count(const struct llama_context * ctx); LLAMA_API int llama_get_kv_cache_token_count(const struct llama_context * ctx);
// Sets the current rng seed. // Sets the current rng seed.
LLAMA_API void llama_set_rng_seed(struct llama_context * ctx, unsigned int seed); LLAMA_API void llama_set_rng_seed(struct llama_context * ctx, uint32_t seed);
// Returns the maximum size in bytes of the state (rng, logits, embedding // Returns the maximum size in bytes of the state (rng, logits, embedding
// and kv_cache) - will often be smaller after compacting tokens // and kv_cache) - will often be smaller after compacting tokens