tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-30 20:58:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : add llama_sampling API + move grammar in libllama

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2024-08-05 10:08:25 +03:00
parent b69a480af4
commit f648ca2cee
48 changed files with 2481 additions and 2590 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
common/common.h
View File

@ -77,8 +77,6 @@ struct cpu_params {
};
struct gpt_params {
uint32_t seed = LLAMA_DEFAULT_SEED; // RNG seed
int32_t n_predict = -1; // new tokens to predict
int32_t n_ctx = 0; // context size
int32_t n_batch = 2048; // logical batch size for prompt processing (must be >=32 to use BLAS)
@ -120,8 +118,7 @@ struct gpt_params {
enum llama_pooling_type pooling_type = LLAMA_POOLING_TYPE_UNSPECIFIED; // pooling type for embeddings
enum llama_attention_type attention_type = LLAMA_ATTENTION_TYPE_UNSPECIFIED; // attention type for embeddings
// // sampling parameters
struct llama_sampling_params sparams;
struct gpt_sampling_params sparams;
std::string model = ""; // model path
std::string model_draft = ""; // draft model for speculative decoding
@ -185,7 +182,6 @@ struct gpt_params {
bool flash_attn = false; // flash attention
bool input_prefix_bos = false; // prefix BOS to user inputs, preceding input_prefix
bool ignore_eos = false; // ignore generated EOS tokens
bool logits_all = false; // return logits for all tokens in the batch
bool use_mmap = true; // use mmap for faster loads
bool use_mlock = false; // use mlock to keep model in memory