From dbfeb596597c7d0e27b688dd1b65f05aab29e973 Mon Sep 17 00:00:00 2001
From: sihan-bzwj <25531470@bjtu.edu.cn>
Date: Sat, 8 Aug 2026 16:11:12 +0800
Subject: [PATCH 1/2] Fix max_length/max_new_tokens conflict in Chapter 1.3

## Summary

Fix the text-generation example in Chapter 1.3 by replacing `max_length` with `max_new_tokens`.

In recent versions of `transformers`, the text-generation pipeline may already define `max_new_tokens`, which causes a conflict when the course example explicitly passes `max_length`. Since `max_new_tokens` takes precedence, the example may not control the generated output length as intended.

## Changes

* Replace `max_length=30` with `max_new_tokens=30`
* Update the surrounding explanation to use `max_new_tokens`
* Clarify that generation length is measured in tokens, not words

This PR intentionally updates only the English source. Localized versions can be synchronized separately through the translation workflow.

Fixes #1285
---
 chapters/en/chapter1/3.mdx | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/chapters/en/chapter1/3.mdx b/chapters/en/chapter1/3.mdx
index 7efc4ef8e..6141d75ab 100644
--- a/chapters/en/chapter1/3.mdx
+++ b/chapters/en/chapter1/3.mdx
@@ -146,10 +146,10 @@ generator("In this course, we will teach you how to")
                     'HTTP'}]
 ```
 
-You can control how many different sequences are generated with the argument `num_return_sequences` and the total length of the output text with the argument `max_length`.
+You can control how many different sequences are generated with the argument `num_return_sequences` and the total length of the output text with the argument `max_new_tokens`.
 
 > [!TIP]
-> ✏️ **Try it out!** Use the `num_return_sequences` and `max_length` arguments to generate two sentences of 15 words each.
+> ✏️ **Try it out!** Use the `num_return_sequences` and `max_new_tokens` arguments to generate two sentences of 15 words each.
 
 ## Using any model from the Hub in a pipeline[[using-any-model-from-the-hub-in-a-pipeline]]
 
@@ -163,7 +163,7 @@ from transformers import pipeline
 generator = pipeline("text-generation", model="HuggingFaceTB/SmolLM2-360M")
 generator(
     "In this course, we will teach you how to",
-    max_length=30,
+    max_new_tokens=30,
     num_return_sequences=2,
 )
 ```

From bec9bad9c12846e8073165925194b1fab51f0764 Mon Sep 17 00:00:00 2001
From: sihan-bzwj <25531470@bjtu.edu.cn>
Date: Sat, 8 Aug 2026 16:28:17 +0800
Subject: [PATCH 2/2] Clarify max_new_tokens argument in documentation

Updated text to clarify the maximum number of newly generated tokens.
---
 chapters/en/chapter1/3.mdx | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/chapters/en/chapter1/3.mdx b/chapters/en/chapter1/3.mdx
index 6141d75ab..c0e1ad5b1 100644
--- a/chapters/en/chapter1/3.mdx
+++ b/chapters/en/chapter1/3.mdx
@@ -146,10 +146,10 @@ generator("In this course, we will teach you how to")
                     'HTTP'}]
 ```
 
-You can control how many different sequences are generated with the argument `num_return_sequences` and the total length of the output text with the argument `max_new_tokens`.
+You can control how many different sequences are generated with the argument `num_return_sequences` and the maximum number of newly generated tokens with the argument `max_new_tokens`.
 
 > [!TIP]
-> ✏️ **Try it out!** Use the `num_return_sequences` and `max_new_tokens` arguments to generate two sentences of 15 words each.
+> ✏️ **Try it out!** Use the `num_return_sequences` and `max_new_tokens` arguments to generate two sequences with at most 15 new tokens each.
 
 ## Using any model from the Hub in a pipeline[[using-any-model-from-the-hub-in-a-pipeline]]