From 1c8b3e4c4464baab8507a7279bc1367985b3488f Mon Sep 17 00:00:00 2001
From: Gabe Goodhart <ghart@us.ibm.com>
Date: Mon, 23 Sep 2024 13:59:54 -0600
Subject: [PATCH] fix: Allow "output" layer in granite moe architecture
 (convert and cpp)

Branch: GraniteMoE

Co-Authored-By: git@compilade.net
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
---
 gguf-py/gguf/constants.py | 1 +
 src/llama.cpp             | 1 +
 2 files changed, 2 insertions(+)

diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py
index fed7418be..bd83be144 100644
--- a/gguf-py/gguf/constants.py
+++ b/gguf-py/gguf/constants.py
@@ -1247,6 +1247,7 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
     MODEL_ARCH.GRANITE_MOE: [
         MODEL_TENSOR.TOKEN_EMBD,
         MODEL_TENSOR.OUTPUT_NORM,
+        MODEL_TENSOR.OUTPUT,
         MODEL_TENSOR.ATTN_NORM,
         MODEL_TENSOR.ATTN_Q,
         MODEL_TENSOR.ATTN_K,
diff --git a/src/llama.cpp b/src/llama.cpp
index 65aa6cbeb..f268dd91e 100644
--- a/src/llama.cpp
+++ b/src/llama.cpp
@@ -1485,6 +1485,7 @@ static const std::map<llm_arch, std::map<llm_tensor, std::string>> LLM_TENSOR_NA
         {
             { LLM_TENSOR_TOKEN_EMBD,      "token_embd" },
             { LLM_TENSOR_OUTPUT_NORM,     "output_norm" },
+            { LLM_TENSOR_OUTPUT,          "output" },
             { LLM_TENSOR_ATTN_NORM,       "blk.%d.attn_norm" },
             { LLM_TENSOR_ATTN_Q,          "blk.%d.attn_q" },
             { LLM_TENSOR_ATTN_K,          "blk.%d.attn_k" },