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

llama : add session file format and saved sessions in main (#1169)

Browse source
This commit is contained in:
Evan Jones 2023-04-28 11:59:37 -04:00 • committed by GitHub
parent 11d902364b
commit 1481a9cf25
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
6 changed files with 156 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama.h
View file

@ -133,6 +133,10 @@ extern "C" {
// Returns the number of bytes read
LLAMA_API size_t llama_set_state_data(struct llama_context * ctx, const uint8_t * src);
// Save/load session file
LLAMA_API size_t llama_load_session_file(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out);
LLAMA_API size_t llama_save_session_file(struct llama_context * ctx, const char * path_session, const llama_token * tokens, size_t n_token_count);
// Run the llama inference to obtain the logits and probabilities for the next token.
// tokens + n_tokens is the provided batch of new tokens to process
// n_past is the number of tokens to use from previous eval calls