From 592cc35a7a6a8ba0a9cd5d51f8a24708ce93e32b Mon Sep 17 00:00:00 2001
From: long_long_ago <149533107@qq.com>
Date: Sun, 7 Sep 2025 09:02:42 +0800
Subject: [PATCH] =?UTF-8?q?Update=205.=E5=9C=A8=E6=97=A0=E6=A0=87=E8=AE=B0?=
 =?UTF-8?q?=E6=95=B0=E6=8D=AE=E9=9B=86=E4=B8=8A=E8=BF=9B=E8=A1=8C=E9=A2=84?=
 =?UTF-8?q?=E8=AE=AD=E7=BB=83.md?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

修复代码错误
---
 cn-Book/5.在无标记数据集上进行预训练.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/cn-Book/5.在无标记数据集上进行预训练.md b/cn-Book/5.在无标记数据集上进行预训练.md
index eaa07fd..a9d5944 100644
--- a/cn-Book/5.在无标记数据集上进行预训练.md
+++ b/cn-Book/5.在无标记数据集上进行预训练.md
@@ -158,7 +158,7 @@ inputs = torch.tensor([[16833, 3626, 6100], # ["every effort moves",
                        [40, 1107, 588]])    # "I really like"]
 # Matching these inputs, the `targets` contain the token IDs we aim for the model to produce:
 targets = torch.tensor([[3626, 6100, 345 ], # [" effort moves you",
-                        [107, 588, 11311]]) # " really like chocolate"]
+                        [1107, 588, 11311]]) # " really like chocolate"]
 ```
 
 需要注意的是，目标值中展示的是输入数据向前偏移了一个位置。我们在第 2 章实现数据加载器时已介绍过这一概念。这种偏移策略对于教会模型预测序列中的下一个 token 至关重要。