diff --git a/docs/source-doc-builder/pipeline.mdx b/docs/source-doc-builder/pipeline.mdx
index d40029a0..30f92bf1 100644
--- a/docs/source-doc-builder/pipeline.mdx
+++ b/docs/source-doc-builder/pipeline.mdx
@@ -300,7 +300,7 @@ customize this part. Currently, the 🤗 Tokenizers library supports:
 -   `models.WordPiece`
 
 For more details about each model and its behavior, you can check
-[here](components.html#models)
+[here](components#models)
 
 ## Post-Processing
 
diff --git a/docs/source-doc-builder/quicktour.mdx b/docs/source-doc-builder/quicktour.mdx
index 85deb193..b7cd1400 100644
--- a/docs/source-doc-builder/quicktour.mdx
+++ b/docs/source-doc-builder/quicktour.mdx
@@ -287,7 +287,7 @@ with the `Tokenizer.encode` method:
 
 This applied the full pipeline of the tokenizer on the text, returning
 an `Encoding` object. To learn more
-about this pipeline, and how to apply (or customize) parts of it, check out `this page <pipeline>`.
+about this pipeline, and how to apply (or customize) parts of it, check out [this page](pipeline).
 
 This `Encoding` object then has all the
 attributes you need for your deep learning model (or other). The
diff --git a/docs/source-doc-builder/training_from_memory.mdx b/docs/source-doc-builder/training_from_memory.mdx
index aea2a6b9..121af3d9 100644
--- a/docs/source-doc-builder/training_from_memory.mdx
+++ b/docs/source-doc-builder/training_from_memory.mdx
@@ -1,6 +1,6 @@
 # Training from memory
 
-In the [Quicktour](quicktour.html), we saw how to build and train a
+In the [Quicktour](quicktour), we saw how to build and train a
 tokenizer using text files, but we can actually use any Python Iterator.
 In this section we'll see a few different ways of training our
 tokenizer.
@@ -22,7 +22,7 @@ takes care of normalizing the input using the NFKC Unicode normalization
 method, and uses a [`~tokenizers.pre_tokenizers.ByteLevel`] pre-tokenizer with the corresponding decoder.
 
 For more information on the components used here, you can check
-[here](components.html)
+[here](components).
 
 ## The most basic way
 
diff --git a/docs/source/pipeline.rst b/docs/source/pipeline.rst
index d843ee57..e5ab8fe9 100644
--- a/docs/source/pipeline.rst
+++ b/docs/source/pipeline.rst
@@ -261,7 +261,7 @@ how to customize this part. Currently, the 🤗 Tokenizers library supports:
 - :entity:`models.WordLevel`
 - :entity:`models.WordPiece`
 
-For more details about each model and its behavior, you can check `here <components.html#models>`__
+For more details about each model and its behavior, you can check `here <components#models>`__
 
 
 .. _post-processing:
diff --git a/docs/source/tutorials/python/training_from_memory.rst b/docs/source/tutorials/python/training_from_memory.rst
index ef15813e..3fe92495 100644
--- a/docs/source/tutorials/python/training_from_memory.rst
+++ b/docs/source/tutorials/python/training_from_memory.rst
@@ -1,7 +1,7 @@
 Training from memory
 ----------------------------------------------------------------------------------------------------
 
-In the `Quicktour <quicktour.html>`__, we saw how to build and train a tokenizer using text files,
+In the `Quicktour <quicktour>`__, we saw how to build and train a tokenizer using text files,
 but we can actually use any Python Iterator. In this section we'll see a few different ways of
 training our tokenizer.
 
@@ -18,7 +18,7 @@ This tokenizer is based on the :class:`~tokenizers.models.Unigram` model. It tak
 normalizing the input using the NFKC Unicode normalization method, and uses a
 :class:`~tokenizers.pre_tokenizers.ByteLevel` pre-tokenizer with the corresponding decoder.
 
-For more information on the components used here, you can check `here <components.html>`__
+For more information on the components used here, you can check `here <components>`__
 
 The most basic way
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~