diff --git a/neural_speed/models/llama/llama_utils.cpp b/neural_speed/models/llama/llama_utils.cpp
index 6e4df8c13..73bb92c9e 100644
--- a/neural_speed/models/llama/llama_utils.cpp
+++ b/neural_speed/models/llama/llama_utils.cpp
@@ -193,7 +193,7 @@ void Llama::load(model_context* ctx, model_progress_callback progress_callback,
 
       // ffn GEMM
 
-      if (ml->verify_tensor(layers_i + ".feed_forward")) {
+      if (ml->verify_tensor(layers_i + ".feed_forward.w1.weight")) {
         layer.ffn[0] = ml->get_tensor(layers_i + ".feed_forward.w1.weight", {n_embd, n_ff}, backend);
         layer.ffn[1] = ml->get_tensor(layers_i + ".feed_forward.w2.weight", {n_ff, n_embd}, backend);
         layer.ffn[2] = ml->get_tensor(layers_i + ".feed_forward.w3.weight", {n_embd, n_ff}, backend);