ec893798b7
* tests : verify that RoPE is "additive" * llama : replace ggml_diag_mask_inf with ggml_add (custom -inf mask) * ggml : ggml_rope now takes a vector with positions instead of n_past * metal : add rope_f16 kernel + optimize cpy kernels * llama : unified KV cache + batch inference API * llama : add new llama_decode() API that works with llama_batch * llama : add cell_max heuristic for more efficient kv_cache * llama : extend llama_kv_cache API * llama : more robust cell_max heuristic + wip shift * metal : disable concurrency optimization * llama : add llama_kv_cache_shift_seq + no more context swaps * llama : apply K-cache roping for Falcon and Baichuan * speculative : fix KV cache management * parallel : example for serving multiple users in parallel * parallel : disable hot-plug to avoid cache fragmentation * fixes : speculative KV cache + llama worst-case graph * llama : extend batch API to select which logits to output * llama : fix worst case graph build * ggml-cuda : update rope implementation for parallel decoding (#3254) * ggml-cuda : update rope implementation for parallel decoding * better solution for p0 computation * fix rope * simpler rope implementation --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * make : add parallel to build + fix static functions in llama.cpp * simple : fix token counting * parallel : various improvements * llama : fix cell_max logic + rename functions * parallel : try smaller batches when the KV cache is fragmented * parallel : fix sequence termination criteria * llama : silence errors KV cache errors * parallel : remove new line from prompt * parallel : process system prompt once + configurable paramters + llama API * parallel : remove question with short answers * parallel : count cache misses * parallel : print misses on each request * parallel : minor * llama : fix n_kv to never become 0 * parallel : rename hot-plug to continuous-batching * llama : improve llama_batch API + simplify parallel example * simple : add parallel decoding support * simple : improve comments + free batch * ggml-cuda : add rope f16, restore performance with parallel decoding (#3272) * ggml-cuda : add rope f16, restore performance * offload KQ_mask with all models * fix rope shift --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * llama : disable MPI for now ggml-ci * train : make KQ_pos memory buffer permanent via dummy scale op * ggml : revert change to ggml_cpy, add ggml_cont_Nd instead (#3275) ggml-ci * parallel : fix bug (extra BOS) + smaller token_prev array * parallel : fix cases where the input prompts can overflow the batch * parallel : add disabled experimental batch chunking in powers of two * llama : llama.h formatting + comments * simple : add README.md * llama : fix kv cache heuristic when context is less than 32 * parallel : fix crash when `-n -1` * llama : simplify returns if/else branches * metal : use mm kernels for batch size > 2 * examples : utilize new llama_get_logits_ith() * examples : add example for batched decoding * examples : do not eval prompt 2 times (close #3348) * server : clear the KV cache beyond n_past before llama_decode * server : avoid context swaps by shifting the KV cache --------- Co-authored-by: slaren <slarengh@gmail.com>
105 lines
2.9 KiB
C++
105 lines
2.9 KiB
C++
#include "build-info.h"
|
|
#include "common.h"
|
|
#include "llama.h"
|
|
|
|
#include <ctime>
|
|
|
|
#if defined(_MSC_VER)
|
|
#pragma warning(disable: 4244 4267) // possible loss of data
|
|
#endif
|
|
|
|
int main(int argc, char ** argv) {
|
|
gpt_params params;
|
|
|
|
if (!gpt_params_parse(argc, argv, params)) {
|
|
return 1;
|
|
}
|
|
|
|
params.embedding = true;
|
|
|
|
print_build_info();
|
|
|
|
if (params.seed == LLAMA_DEFAULT_SEED) {
|
|
params.seed = time(NULL);
|
|
}
|
|
|
|
fprintf(stderr, "%s: seed = %u\n", __func__, params.seed);
|
|
|
|
std::mt19937 rng(params.seed);
|
|
if (params.random_prompt) {
|
|
params.prompt = gpt_random_prompt(rng);
|
|
}
|
|
|
|
llama_backend_init(params.numa);
|
|
|
|
llama_model * model;
|
|
llama_context * ctx;
|
|
|
|
// load the model
|
|
std::tie(model, ctx) = llama_init_from_gpt_params(params);
|
|
if (model == NULL) {
|
|
fprintf(stderr, "%s: error: unable to load model\n", __func__);
|
|
return 1;
|
|
}
|
|
|
|
const int n_ctx_train = llama_n_ctx_train(ctx);
|
|
if (params.n_ctx > n_ctx_train) {
|
|
fprintf(stderr, "%s: warning: model was trained on only %d context tokens (%d specified)\n",
|
|
__func__, n_ctx_train, params.n_ctx);
|
|
}
|
|
|
|
// print system information
|
|
{
|
|
fprintf(stderr, "\n");
|
|
fprintf(stderr, "system_info: n_threads = %d / %d | %s\n",
|
|
params.n_threads, std::thread::hardware_concurrency(), llama_print_system_info());
|
|
}
|
|
|
|
int n_past = 0;
|
|
|
|
// tokenize the prompt
|
|
auto embd_inp = ::llama_tokenize(ctx, params.prompt, true);
|
|
|
|
if (params.verbose_prompt) {
|
|
fprintf(stderr, "\n");
|
|
fprintf(stderr, "%s: prompt: '%s'\n", __func__, params.prompt.c_str());
|
|
fprintf(stderr, "%s: number of tokens in prompt = %zu\n", __func__, embd_inp.size());
|
|
for (int i = 0; i < (int) embd_inp.size(); i++) {
|
|
fprintf(stderr, "%6d -> '%s'\n", embd_inp[i], llama_token_to_piece(ctx, embd_inp[i]).c_str());
|
|
}
|
|
fprintf(stderr, "\n");
|
|
}
|
|
|
|
if (embd_inp.size() > (size_t)params.n_ctx) {
|
|
fprintf(stderr, "%s: error: prompt is longer than the context window (%zu tokens, n_ctx = %d)\n",
|
|
__func__, embd_inp.size(), params.n_ctx);
|
|
return 1;
|
|
}
|
|
|
|
while (!embd_inp.empty()) {
|
|
int n_tokens = std::min(params.n_batch, (int) embd_inp.size());
|
|
if (llama_decode(ctx, llama_batch_get_one(embd_inp.data(), n_tokens, n_past, 0), params.n_threads)) {
|
|
fprintf(stderr, "%s : failed to eval\n", __func__);
|
|
return 1;
|
|
}
|
|
n_past += n_tokens;
|
|
embd_inp.erase(embd_inp.begin(), embd_inp.begin() + n_tokens);
|
|
}
|
|
|
|
const int n_embd = llama_n_embd(ctx);
|
|
const auto embeddings = llama_get_embeddings(ctx);
|
|
|
|
for (int i = 0; i < n_embd; i++) {
|
|
printf("%f ", embeddings[i]);
|
|
}
|
|
printf("\n");
|
|
|
|
llama_print_timings(ctx);
|
|
llama_free(ctx);
|
|
llama_free_model(model);
|
|
|
|
llama_backend_free();
|
|
|
|
return 0;
|
|
}
|