tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-27 12:05:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama-model : fix the reported size class for nomic-embed-text-v2-moe (#13223)

Browse Source
This commit is contained in:
Jared Van Bortel
2025-05-01 03:09:41 -04:00
committed by GitHub
parent 8d33d740c3
commit a70183eb00
2 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/llama-model.cpp
View File

@ -40,6 +40,7 @@ const char * llm_type_name(llm_type type) {
case LLM_TYPE_335M: return "335M";
case LLM_TYPE_410M: return "410M";
case LLM_TYPE_450M: return "450M";
case LLM_TYPE_475M: return "475M";
case LLM_TYPE_770M: return "770M";
case LLM_TYPE_780M: return "780M";
case LLM_TYPE_0_5B: return "0.5B";
@ -707,7 +708,11 @@ void llama_model::load_hparams(llama_model_loader & ml) {
ml.get_key(LLM_KV_MOE_EVERY_N_LAYERS, hparams.moe_every_n_layers, 0);
if (hparams.n_layer == 12 && hparams.n_embd == 768) {
type = LLM_TYPE_137M;
if (arch == LLM_ARCH_NOMIC_BERT) {
type = LLM_TYPE_137M;
} else if (arch == LLM_ARCH_NOMIC_BERT_MOE && hparams.moe_every_n_layers == 2) {
type = LLM_TYPE_475M;
}
}
} break;
case LLM_ARCH_BLOOM: