llama : fix comment for "output.weight" tensor

diff --git a/llama.cpp b/llama.cpp
index 4a646eb911621ea65f292c2213ccba207c2935ca..33ee4fbb5947461a0f2685a6fb7ce73354c22bf1 100644 (file)
--- a/llama.cpp
+++ b/llama.cpp
@@ -1618,8 +1618,8 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s
         // quantize only 2D tensors
         quantize &= (tensor.ne.size() == 2);
 
-        // GG: uncomment this to keep the output layer in FP16
-        //if (tensor.name.rfind("output")) {
+        // uncomment this to keep the output layer in FP16
+        //if (tensor.name == "output.weight") {
         //    quantize = false;
         //}