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

bad indent

Browse source
This commit is contained in:
Nexesenex 2024-08-24 21:02:51 +02:00
parent 5ae59714d2
commit 844d11b8f3
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/llama.cpp
View file

@ -16444,7 +16444,7 @@ static ggml_type llama_tensor_get_type(quantize_state_internal & qs, ggml_type n
else if (ftype == LLAMA_FTYPE_MOSTLY_IQ1_XS || ftype == LLAMA_FTYPE_MOSTLY_IQ1_S || ftype == LLAMA_FTYPE_MOSTLY_IQ1_M) {
if (qs.model.hparams.n_expert >= 4) new_type = GGML_TYPE_Q6_K;
else if (qs.model.hparams.n_vocab >= 127999) new_type = GGML_TYPE_IQ4_XS;
else if (qs.model.hparams.n_gqa() >= 2 || qs.model.hparams.n_expert >= 2) new_type = GGML_TYPE_Q4_K;
else if (qs.model.hparams.n_gqa() >= 2 || qs.model.hparams.n_expert >= 2) new_type = GGML_TYPE_Q4_K;
else new_type = GGML_TYPE_IQ4_XS;
}
else if (ftype == LLAMA_FTYPE_MOSTLY_IQ1_XL || ftype == LLAMA_FTYPE_MOSTLY_IQ2_XXS || ftype == LLAMA_FTYPE_MOSTLY_IQ2_XS ||