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

Support special tokens and not adding BOS to prompt in speculative

Browse source
This commit is contained in:
Branden Butler 2023-11-10 09:59:22 -06:00
parent 2923f17f6f
commit c2d0041704
1 changed files with 6 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

8
examples/speculative/speculative.cpp
View file

@ -94,9 +94,13 @@ int main(int argc, char ** argv) {
}
}
// tokenize the prompt
// Tokenize the prompt
const bool add_bos = llama_vocab_type(llama_get_model(ctx_tgt)) == LLAMA_VOCAB_TYPE_SPM;
LOG("add_bos: %d\n", add_bos);
std::vector<llama_token> inp;
inp = ::llama_tokenize(ctx_tgt, params.prompt, true);
inp = ::llama_tokenize(ctx_tgt, params.prompt, add_bos, true);
const int max_context_size = llama_n_ctx(ctx_tgt);
const int max_tokens_list_size = max_context_size - 4;