From 80350daeda82edba1b6a67f3eb0d5f53745dbdf5 Mon Sep 17 00:00:00 2001
From: John <78893154+cmp-nct@users.noreply.github.com>
Date: Wed, 17 Jan 2024 00:03:58 +0100
Subject: [PATCH] Update llama.cpp

---
 llama.cpp | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/llama.cpp b/llama.cpp
index e1daf1bbb..05726c8c1 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -3438,7 +3438,8 @@ static bool llm_load_tensors(
                         if (gguf_find_tensor(ml.ctx_gguf, tn(LLM_TENSOR_OUTPUT, "weight").c_str()) >= 0) {
                             model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_OUTPUT,      "weight"), {n_embd, n_vocab});
                         } else {
-                            model.output = model.tok_embd;
+                            model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}); // needs to be on GPU
+                            ml.n_tensors++; // artificial tensor
                         }
                     }