We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 7e50c4a commit f10dfd3Copy full SHA for f10dfd3
src/llama-model.cpp
@@ -672,7 +672,7 @@ void llama_model::load_hparams(llama_model_loader & ml) {
672
ml.get_key(LLM_KV_ATTENTION_LAYERNORM_EPS, hparams.f_norm_eps);
673
ml.get_key(LLM_KV_ATTENTION_CAUSAL, hparams.causal_attn);
674
ml.get_key(LLM_KV_POOLING_TYPE, hparams.pooling_type);
675
- ml.get_key(LLM_KV_MOE_EVERY_N_LAYERS, hparams.moe_every_n_layers);
+ ml.get_key(LLM_KV_MOE_EVERY_N_LAYERS, hparams.moe_every_n_layers, 0);
676
677
if (hparams.n_layer == 12 && hparams.n_embd == 768) {
678
type = LLM_TYPE_137M;
0 commit comments