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

common : don't do warm-up with more than n_batch tokens (close #3058)

Browse source 
ggml-ci
This commit is contained in:
Georgi Gerganov 2023-09-07 15:32:19 +03:00
parent 783379670a
commit efac2d469f
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
common/common.cpp
View file

@ -773,7 +773,7 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
LOG("warming up the model with an empty run\n");
const std::vector<llama_token> tmp = { llama_token_bos(lctx), llama_token_eos(lctx), };
llama_eval(lctx, tmp.data(), tmp.size(), 0, params.n_threads);
llama_eval(lctx, tmp.data(), std::min(tmp.size(), (size_t) params.n_batch), 0, params.n_threads);
llama_reset_timings(lctx);
}