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

speculative : add tree-based sampling support

Browse source 
ggml-ci
This commit is contained in:
Georgi Gerganov 2023-10-14 17:54:02 +03:00
parent 5261aee8d8
commit 4de5a2d473
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
11 changed files with 469 additions and 192 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/llava/llava-utils.h
View file

@ -17,7 +17,7 @@ inline bool eval_image_embd(llama_context * ctx_llama, float * embd, int N, int
if (n_eval > n_batch) {
n_eval = n_batch;
}
llama_batch batch = {int32_t(n_eval), nullptr, (embd+i*n_embd), nullptr, nullptr, nullptr, *n_past, 1, 0, };
llama_batch batch = {int32_t(n_eval), nullptr, (embd+i*n_embd), nullptr, nullptr, nullptr, nullptr, *n_past, 1, 0, };
if (llama_decode(ctx_llama, batch)) {
fprintf(stderr, "%s : failed to eval\n", __func__);
return false;

2
examples/llava/llava.cpp
View file

@ -120,7 +120,7 @@ int main(int argc, char ** argv) {
const int max_tgt_len = params.n_predict < 0 ? 256 : params.n_predict;
// GG: are we sure that the should be a trailing whitespace at the end of this string?
eval_string(ctx_llama, "A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.\nUSER: ", params.n_batch, &n_past);
eval_string(ctx_llama, "A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.\nUSER:", params.n_batch, &n_past);
eval_image_embd(ctx_llama, image_embd, n_img_pos, params.n_batch, &n_past);
eval_string(ctx_llama, params.prompt.c_str(), params.n_batch, &n_past);
eval_string(ctx_llama, "\nASSISTANT:", params.n_batch, &n_past);