From 7877e92991c242458ab31676a75846cedb985062 Mon Sep 17 00:00:00 2001
From: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
Date: Fri, 22 Sep 2023 17:55:54 +0200
Subject: [PATCH] Update sft_trainer.mdx (#808)

---
 docs/source/sft_trainer.mdx | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/source/sft_trainer.mdx b/docs/source/sft_trainer.mdx
index a22811f21c..2fc8c07e5f 100644
--- a/docs/source/sft_trainer.mdx
+++ b/docs/source/sft_trainer.mdx
@@ -356,7 +356,7 @@ pip install -U optimum
 Once you have loaded your model, wrap the `trainer.train()` call under the `with torch.backends.cuda.sdp_kernel(enable_flash=True, enable_math=False, enable_mem_efficient=False):` context manager:
 
 ```diff
-# ...
+...
 
 + with torch.backends.cuda.sdp_kernel(enable_flash=True, enable_math=False, enable_mem_efficient=False):
     trainer.train()