From e9bd114a62921f2620c3290ab95e93fc579ea07d Mon Sep 17 00:00:00 2001
From: Liqian Chen <liqian.chen@deeplang.ai>
Date: Mon, 17 Jun 2024 14:53:02 +0800
Subject: [PATCH] Correct alignment in the seq_len diagram.

---
 vllm/attention/backends/flash_attn.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/vllm/attention/backends/flash_attn.py b/vllm/attention/backends/flash_attn.py
index 300bab72877b8..1c48e2a0bb33d 100644
--- a/vllm/attention/backends/flash_attn.py
+++ b/vllm/attention/backends/flash_attn.py
@@ -83,7 +83,7 @@ class FlashAttentionMetadata(AttentionMetadata):
     # |---------------- N iteration ---------------------|
     # |- tokenA -|......................|-- newTokens ---|
     # |---------- context_len ----------|
-    # |-------------------- seq_len ----------------------|
+    # |-------------------- seq_len ---------------------|
     #                                   |-- query_len ---|
 
     # Maximum query length in the batch. None for decoding.