diff --git a/src/llama-arch.cpp b/src/llama-arch.cpp
index a7260f495..101d08717 100644
--- a/src/llama-arch.cpp
+++ b/src/llama-arch.cpp
@@ -1024,6 +1024,9 @@ static const std::map<llm_arch, std::map<llm_tensor, const char *>> LLM_TENSOR_N
             { LLM_TENSOR_OUTPUT,          "output" },
             { LLM_TENSOR_ATTN_NORM,       "blk.%d.attn_norm" },
             { LLM_TENSOR_ATTN_QKV,        "blk.%d.attn_qkv" },
+            { LLM_TENSOR_ATTN_Q,             "blk.%d.attn_q" },
+            { LLM_TENSOR_ATTN_K,             "blk.%d.attn_k" },
+            { LLM_TENSOR_ATTN_V,             "blk.%d.attn_v" },
             { LLM_TENSOR_ATTN_OUT,        "blk.%d.attn_output" },
             { LLM_TENSOR_FFN_NORM,        "blk.%d.ffn_norm" },
             { LLM_TENSOR_FFN_UP,          "blk.%d.ffn_up" },
diff --git a/src/llama-model.cpp b/src/llama-model.cpp
index 7026c009f..e27169ca1 100644
--- a/src/llama-model.cpp
+++ b/src/llama-model.cpp
@@ -3068,6 +3068,7 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
                 } break;
             case LLM_ARCH_CHATGLM:
                 {
+                    printf("Loading ChatGLM model...\n");
                     tok_embd   = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD,      "weight"), {n_embd, n_vocab}, 0);
 
                     // output
@@ -3076,12 +3077,14 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
 
                     for (int i = 0; i < n_layer; ++i) {
                         auto & layer = layers[i];
+                        printf("Loading layer %d...\n",i);
 
                         layer.attn_norm = create_tensor(tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd}, 0);
                         layer.wqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, llama_model_loader::TENSOR_NOT_REQUIRED);
                         layer.bqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "bias", i),   {n_embd + 2*n_embd_gqa}, llama_model_loader::TENSOR_NOT_REQUIRED);
 
                         if(layer.wqkv == nullptr){
+                            printf("Loading W q k v %d...\n",i);
                             layer.wq = create_tensor(tn(LLM_TENSOR_ATTN_Q,   "weight", i), {n_embd, n_embd_head_k * n_head}, 0);
                             layer.wk = create_tensor(tn(LLM_TENSOR_ATTN_K,   "weight", i), {n_embd, n_embd_k_gqa}, 0);
                             layer.wv = create_tensor(tn(LLM_TENSOR_ATTN_V,   "weight", i), {n_embd, n_embd_v_gqa}, 0);