diff --git a/website/docs/api/doc.jade b/website/docs/api/doc.jade
index 86c4dd65a..fab1dd86b 100644
--- a/website/docs/api/doc.jade
+++ b/website/docs/api/doc.jade
@@ -164,8 +164,8 @@ p
         +cell #[code other]
         +cell -
         +cell
-            | The object to compare with. By default, accepts #[code Doc],
-            | #[code Span], #[code Token] and #[code Lexeme] objects.
+            |  The object to compare with. By default, accepts #[code Doc],
+            |  #[code Span], #[code Token] and #[code Lexeme] objects.
 
     +footrow
         +cell return
diff --git a/website/docs/api/span.jade b/website/docs/api/span.jade
index f071f5abc..a07ee25d9 100644
--- a/website/docs/api/span.jade
+++ b/website/docs/api/span.jade
@@ -156,8 +156,8 @@ p
         +cell #[code other]
         +cell -
         +cell
-            | The object to compare with. By default, accepts #[code Doc],
-            | #[code Span], #[code Token] and #[code Lexeme] objects.
+            |  The object to compare with. By default, accepts #[code Doc],
+            |  #[code Span], #[code Token] and #[code Lexeme] objects.
 
     +footrow
         +cell return
diff --git a/website/docs/api/vocab.jade b/website/docs/api/vocab.jade
index 96356cb41..7490bccf4 100644
--- a/website/docs/api/vocab.jade
+++ b/website/docs/api/vocab.jade
@@ -70,8 +70,8 @@ p Create the vocabulary.
         +cell #[code lex_attr_getters]
         +cell dict
         +cell
-            | A dictionary mapping attribute IDs to functions to compute them.
-            | Defaults to #[code None].
+            |  A dictionary mapping attribute IDs to functions to compute them.
+            |  Defaults to #[code None].
 
     +row
         +cell #[code lemmatizer]
diff --git a/website/docs/usage/processing-text.jade b/website/docs/usage/processing-text.jade
index b26538dcb..205986e8a 100644
--- a/website/docs/usage/processing-text.jade
+++ b/website/docs/usage/processing-text.jade
@@ -73,7 +73,7 @@ p
     |  one-by-one. After a long and bitter struggle, the global interpreter
     |  lock was freed around spaCy's main parsing loop in v0.100.3. This means
     |  that the #[code .pipe()] method will be significantly faster in most
-    | practical situations, because it allows shared memory parallelism.
+    |  practical situations, because it allows shared memory parallelism.
 
 +code.
     for doc in nlp.pipe(texts, batch_size=10000, n_threads=3):