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

parallel : process system prompt once + configurable paramters + llama API

Browse source
This commit is contained in:
Georgi Gerganov 2023-09-19 17:00:42 +03:00
parent 82e20e9ba0
commit 4b5f3cd6bf
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
9 changed files with 187 additions and 93 deletions
Download patch file Download diff file Expand all files Collapse all files

20
common/common.cpp
View file

@ -317,6 +317,18 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
break;
}
params.n_chunks = std::stoi(argv[i]);
} else if (arg == "-np" || arg == "--parallel") {
if (++i >= argc) {
invalid_param = true;
break;
}
params.n_parallel = std::stoi(argv[i]);
} else if (arg == "-ns" || arg == "--sequences") {
if (++i >= argc) {
invalid_param = true;
break;
}
params.n_sequences = std::stoi(argv[i]);
} else if (arg == "-m" || arg == "--model") {
if (++i >= argc) {
invalid_param = true;
@ -360,6 +372,8 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
params.multiline_input = true;
} else if (arg == "--simple-io") {
params.simple_io = true;
} else if (arg == "--hot-plug") {
params.hot_plug = true;
} else if (arg == "--color") {
params.use_color = true;
} else if (arg == "--mlock") {
@ -659,6 +673,9 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
printf(" --keep N number of tokens to keep from the initial prompt (default: %d, -1 = all)\n", params.n_keep);
printf(" --draft N number of tokens to draft for speculative decoding (default: %d)\n", params.n_draft);
printf(" --chunks N max number of chunks to process (default: %d, -1 = all)\n", params.n_chunks);
printf(" -np N, --parallel N number of parallel sequences to decode (default: %d)\n", params.n_parallel);
printf(" -ns N, --sequences N number of sequences to decode (default: %d)\n", params.n_sequences);
printf(" --hot-plug enable hot-plugging of new sequences for decoding (default: disabled)\n");
if (llama_mlock_supported()) {
printf(" --mlock force system to keep model in RAM rather than swapping or compressing\n");
}
@ -781,7 +798,7 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
std::vector<llama_token> tmp = { llama_token_bos(lctx), llama_token_eos(lctx), };
llama_decode(lctx, llama_batch_get_one(tmp.data(), std::min(tmp.size(), (size_t) params.n_batch), 0, 0), params.n_threads);
llama_kv_cache_rm_tokens(lctx, -1, -1);
llama_kv_cache_tokens_rm(lctx, -1, -1);
llama_reset_timings(lctx);
}
@ -1253,6 +1270,7 @@ void dump_non_result_info_yaml(FILE * stream, const gpt_params & params, const l
fprintf(stream, "rope_freq_scale: %f # default: 1.0\n", params.rope_freq_scale);
fprintf(stream, "seed: %d # default: -1 (random seed)\n", params.seed);
fprintf(stream, "simple_io: %s # default: false\n", params.simple_io ? "true" : "false");
fprintf(stream, "hot_plug: %s # default: false\n", params.hot_plug ? "true" : "false");
fprintf(stream, "temp: %f # default: 0.8\n", params.temp);
const std::vector<float> tensor_split_vector(params.tensor_split, params.tensor_split + LLAMA_MAX_DEVICES);

3
common/common.h
View file

@ -43,6 +43,8 @@ struct gpt_params {
int32_t n_keep = 0; // number of tokens to keep from initial prompt
int32_t n_draft = 16; // number of tokens to draft during speculative decoding
int32_t n_chunks = -1; // max number of chunks to process (-1 = unlimited)
int32_t n_parallel = 1; // number of parallel sequences to decode
int32_t n_sequences = 1; // number of sequences to decode
int32_t n_gpu_layers = -1; // number of layers to store in VRAM (-1 - use default)
int32_t n_gpu_layers_draft = -1; // number of layers to store in VRAM for the draft model (-1 - use default)
int32_t main_gpu = 0; // the GPU that is used for scratch and small tensors
@ -108,6 +110,7 @@ struct gpt_params {
bool interactive_first = false; // wait for user input immediately
bool multiline_input = false; // reverse the usage of `\`
bool simple_io = false; // improves compatibility with subprocesses and limited consoles
bool hot_plug = false; // hot-plug new sequences for decoding
bool input_prefix_bos = false; // prefix BOS to user inputs, preceding input_prefix
bool ignore_eos = false; // ignore generated EOS tokens