LLM inference in C/C++
C++ ggml
model : load relevant arrays with n_layer_all (#28173)
Sigbjørn Skjæretd11b3cc7ed2b
diff --git a/src/llama-model.cpp b/src/llama-model.cppindex ec91b7c6d0..6c79971ced 100644--- a/src/llama-model.cpp+++ b/src/llama-model.cpp@@ -1284,8 +1284,8 @@ void llama_model_base::load_hparams(llama_model_loader & ml) {std::fill(hparams.swiglu_clamp_exp.begin(), hparams.swiglu_clamp_exp.end(), 0.0f);std::fill(hparams.swiglu_clamp_shexp.begin(), hparams.swiglu_clamp_shexp.end(), 0.0f);- ml.get_key_or_arr(LLM_KV_FEED_FORWARD_LENGTH, hparams.n_ff_arr, hparams.n_layer(), false);- ml.get_key_or_arr(LLM_KV_ATTENTION_HEAD_COUNT, hparams.n_head_arr, hparams.n_layer(), false);+ ml.get_key_or_arr(LLM_KV_FEED_FORWARD_LENGTH, hparams.n_ff_arr, hparams.n_layer_all, false);+ ml.get_key_or_arr(LLM_KV_ATTENTION_HEAD_COUNT, hparams.n_head_arr, hparams.n_layer_all, false);// Populate deepstack_mapping_arr - initialized to -1 (no deepstack)std::fill(hparams.deepstack_mapping_arr.begin(), hparams.deepstack_mapping_arr.end(), -1);@@ -1293,7 +1293,7 @@ void llama_model_base::load_hparams(llama_model_loader & ml) {// n_head_kv is optional, default to n_headhparams.n_head_kv_arr = hparams.n_head_arr;- ml.get_key_or_arr(LLM_KV_ATTENTION_HEAD_COUNT_KV, hparams.n_head_kv_arr, hparams.n_layer(), false);+ ml.get_key_or_arr(LLM_KV_ATTENTION_HEAD_COUNT_KV, hparams.n_head_kv_arr, hparams.n_layer_all, false);bool rope_finetuned = false;ml.get_key(LLM_KV_ROPE_SCALING_FINETUNED, rope_finetuned, false);