diff options
| author | saood06 <saood05@gmail.com> | 2025-02-10 09:40:38 -0600 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-02-10 17:40:38 +0200 |
| commit | a366a3d17d8f2de0eb8c3d9eddc7b5840fb5761a (patch) | |
| tree | ae0cb943fb4b83cb9e24d1a51d15550d5d7f0903 /common/common.cpp | |
| parent | c12f73ba6153d162f36434cb48e36dd3649b7701 (diff) | |
Load all MoE experts during warmup and make warmup 1 token (#198)
* Load all MoE experts during warmup
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
* Unify warmup to one token
---------
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
Diffstat (limited to 'common/common.cpp')
| -rw-r--r-- | common/common.cpp | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/common/common.cpp b/common/common.cpp index 6219f0ce..44678d7a 100644 --- a/common/common.cpp +++ b/common/common.cpp @@ -2169,8 +2169,10 @@ struct llama_init_result llama_init_from_gpt_params(gpt_params & params) { if (bos != -1) { tmp.push_back(bos); } - tmp.push_back(eos); - + else + { + tmp.push_back(eos); + } if (llama_model_has_encoder(model)) { llama_encode(lctx, llama_batch_get_one(tmp.data(), tmp.size(), 0, 0)); llama_token decoder_start_token_id = llama_model_decoder_start_token(model); |