Fix it's/its mistakes

src/llm/walkthrough/Walkthrough00_Intro.tsx

@@ -60,7 +60,7 @@ export function walkthroughIntro(args: IWalkthroughArgs) {
 
     let c0 = commentary(wt, null, 0)`Welcome to the walkthrough of the GPT large language model! Here we'll explore the model _nano-gpt_, with a mere 85,000 parameters.
 
-It's goal is a simple one: take a sequence of six letters: ${embed(ExampleInputOutput)}
+Its goal is a simple one: take a sequence of six letters: ${embed(ExampleInputOutput)}
 and sort them in alphabetical order, i.e. to "ABBBCC".`;
 
     if (c0.t > 0) {
@@ -96,9 +96,9 @@ and sort them in alphabetical order, i.e. to "ABBBCC".`;
     let tokenStr = c_str('_token_', 0, DimStyle.Token);
     let tokenIdxStr = c_str('_token index_', 0, DimStyle.TokenIdx);
 
-    commentary(wt, t6)`We call each of these letters a ${tokenStr}, and the set of the model's different tokens make up it's _vocabulary_:${embed(TokenVocab)}
+    commentary(wt, t6)`We call each of these letters a ${tokenStr}, and the set of the model's different tokens make up its _vocabulary_:${embed(TokenVocab)}
 
-    From this table, each token is assigned a number, it's ${tokenIdxStr}. And now we can enter this sequence of numbers into the model:${embed(ExampleTokenValues)}\n`;
+    From this table, each token is assigned a number, its ${tokenIdxStr}. And now we can enter this sequence of numbers into the model:${embed(ExampleTokenValues)}\n`;
     breakAfter();
 
     let t7 = afterTime(null, 1.5, 0.5);

src/llm/walkthrough/Walkthrough07_Mlp.tsx

@@ -78,7 +78,7 @@ We then project the vector back down to length ${c_dimRef('C', DimStyle.C)} with
     breakAfter();
 
 commentary(wt)`
-Like in the self-attention + projection section, we add the result of the MLP to it's input, element-wise.
+Like in the self-attention + projection section, we add the result of the MLP to its input, element-wise.
 `;
     breakAfter();