diff --git a/wgsl/index.bs b/wgsl/index.bs
index 3a2060fab9..79a396774c 100644
--- a/wgsl/index.bs
+++ b/wgsl/index.bs
@@ -694,14 +694,14 @@ A [=diagnostic=] has the following properties:
 * A [=diagnostic/triggering location=].
 
 The <dfn dfn-for="diagnostic">severity</dfn> of a diagnostic is one of the following, ordered from greatest to least:
-: <dfn dfn-for="severity" noexport>error</a>
+: <dfn dfn-for="severity" noexport>error</dfn>
 :: The diagnostic is an error.
      This corresponds to a [=shader-creation error=] or to a [=pipeline-creation error=].
-: <dfn dfn-for="severity" noexport>warning</a>
+: <dfn dfn-for="severity" noexport>warning</dfn>
 :: The diagnostic describes an anomaly that merits the attention of the application developer, but is not an error.
-: <dfn dfn-for="severity" noexport>info</a>
+: <dfn dfn-for="severity" noexport>info</dfn>
 :: The diagnostic describes a notable condition that merits attention of the application developer, but is not an error or warning.
-: <dfn dfn-for="severity" noexport>off</a>
+: <dfn dfn-for="severity" noexport>off</dfn>
 :: The diagnostic is disabled. It will not be conveyed to the application.
 
 The name of a [=diagnostic/triggering rule=] is either:
@@ -7121,7 +7121,7 @@ first a [=read access=] gets the old value, and then a [=write access=] stores t
 A compound assignment can rewritten as different WGSL code that uses a [=simple assignment=] instead.
 The idea is to use a pointer to hold the result of evaluating the reference once.
 
-<p algorithm="translation compound assignment not vector component">For example,
+<div algorithm="translation compound assignment not vector component">For example,
 when |e1| is *not* a reference to a component inside a vector, then
 <blockquote>
 |e1|` += `|e2|;
@@ -7131,9 +7131,8 @@ can be rewritten as
 `{ let p = &(`|e1|`); *p = *p + (`|e2|`); }`
 </blockquote>
 where the identifier `p` is chosen to be different from all other identifiers in the program.
-</p>
-
-<p algorithm="translation compound assignment vector component">When
+</div>
+<div algorithm="translation compound assignment vector component">When
 |e1| is a reference to a component inside a vector, the above technique
 needs to be modified because WGSL does not allow [[#address-of-expr|taking the address]] in that case.
 For example, if <var ignore>ev</var> is a reference to a vector, the statement
@@ -9718,7 +9717,7 @@ User-defined data can be passed as input to the start of a pipeline, passed
 between stages of a pipeline or output from the end of a pipeline.
 
 Each <dfn noexport>user-defined input datum</dfn> and
-<dfn noexport>user-defined output datum</abbrev></dfn> [=shader-creation error|must=]:
+<dfn noexport>user-defined output datum</dfn> [=shader-creation error|must=]:
 * be of [=numeric scalar=] type or [=numeric vector=] type.
 * be assigned an IO location. See [[#input-output-locations]].
 
@@ -10002,7 +10001,7 @@ is determined from the size of the corresponding {{GPUBufferBinding}}:
 * Let |EBS| be the
     [=effective buffer binding size=]
     for the {{GPUBufferBinding}} bound to the pipeline binding address corresponding to the storage buffer variable.
-* Then <dfn noexport>NRuntime</sub></dfn>, i.e.
+* Then <dfn noexport>NRuntime</dfn>, i.e.
     the number of elements in the runtime-sized array,
     is the largest integer such that [=SizeOf=](|T|) &le; |EBS|.
 
@@ -10103,7 +10102,6 @@ The following table shows examples of [=NRuntime=] for the `point` member of the
   <tr><td>1025<td>31<td>[=truncate=]( ( 1025 - 16 ) &divide; 32) )
   <tr><td>1039<td>31<td>[=truncate=]( ( 1039 - 16 ) &divide; 32) )
   <tr><td>1040<td>32<td>[=truncate=]( ( 1040 - 16 ) &divide; 32) )
-  </thead>
 </table>
 </div>
 
@@ -12479,7 +12477,7 @@ An [[!IEEE-754|IEEE-754]] binary floating point type approximates the [=extended
         * A 1-bit <dfn dfn-for="ieee754" noexport>sign field</dfn>.
         * A fixed-width <dfn dfn-for="ieee754" noexport>exponent field</dfn>.
         * A fixed-width <dfn dfn-for="ieee754" noexport>trailing significand field</dfn>.
-    * An integer-valued <dfn dfn-for="ieee754" noexport>exponent bias</dfN> related to interpretation of the [=ieee754/exponent field=].
+    * An integer-valued <dfn dfn-for="ieee754" noexport>exponent bias</dfn> related to interpretation of the [=ieee754/exponent field=].
 
 The <dfn>finite range</dfn> of a floating point type is the [=interval=] [|low|, |high|],
 where |low| is the lowest finite value in the type, and |high| is the highest finite value in the type.