From bc1fca39f33074fec4319ef46d96e09b8024c824 Mon Sep 17 00:00:00 2001
From: Andrej Karpathy <andrej.karpathy@gmail.com>
Date: Sat, 15 Nov 2025 15:43:37 +0000
Subject: [PATCH] mqa -> gqa to reduce confusion

---
 nanochat/gpt.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/nanochat/gpt.py b/nanochat/gpt.py
index b640f1e..8b220c3 100644
--- a/nanochat/gpt.py
+++ b/nanochat/gpt.py
@@ -8,7 +8,7 @@ Notable features:
 - norm after token embedding
 - no learnable params in rmsnorm
 - no bias in linear layers
-- Multi-Query Attention (MQA) support for more efficient inference
+- Group-Query Attention (GQA) support for more efficient inference
 """
 
 import math
@@ -29,7 +29,7 @@ class GPTConfig:
     vocab_size: int = 50304
     n_layer: int = 12
     n_head: int = 6 # number of query heads
-    n_kv_head: int = 6 # number of key/value heads (MQA)
+    n_kv_head: int = 6 # number of key/value heads (GQA)
     n_embd: int = 768