docs: update Minimal_RAG.ipynb

relevent -> relevant

Gemma/Minimal_RAG.ipynb

@@ -191,7 +191,7 @@
   {
    "cell_type": "markdown",
    "source": [
-    "## Retrieve the relevent chunks"
+    "## Retrieve the relevant chunks"
    ],
    "metadata": {
     "id": "iLBZLeZpQL5t"
@@ -200,7 +200,7 @@
   {
    "cell_type": "markdown",
    "source": [
-    "Given a user question 'where can I get PaliGemma?', you will use Unisim to retrieve the relevent chunks.\n",
+    "Given a user question 'where can I get PaliGemma?', you will use Unisim to retrieve the relevant chunks.\n",
     "\n",
     "First, compute the similarities between the user question and all the text chunks (passages)."
    ],
@@ -722,7 +722,7 @@
   {
    "cell_type": "markdown",
    "source": [
-    "Identify the top 3 most relevent passages."
+    "Identify the top 3 most relevant passages."
    ],
    "metadata": {
     "id": "J5Oe5A0cm16y"
@@ -773,7 +773,7 @@
    "source": [
     "prompt_template = \"\"\"You are an expert in answering user questions. You always understand user questions well, and then provide high-quality answers based on the information provided in the context.\n",
     "\n",
-    "If the provided context does not contain relevent information, just respond \"I could not find the answer based on the context you provided.\"\n",
+    "If the provided context does not contain relevant information, just respond \"I could not find the answer based on the context you provided.\"\n",
     "\n",
     "User question: {}\n",
     "\n",
@@ -821,7 +821,7 @@
      "text": [
       "You are an expert in answering user questions. You always understand user questions well, and then provide high-quality answers based on the information provided in the context.\n",
       "\n",
-      "If the provided context does not contain relevent information, just respond \"I could not find the answer based on the context you provided.\"\n",
+      "If the provided context does not contain relevant information, just respond \"I could not find the answer based on the context you provided.\"\n",
       "\n",
       "User question: where can I find PaliGemma?\n",
       "\n",
@@ -1023,7 +1023,7 @@
    "source": [
     "Gemma is able to provide the correct answer based on the retrieved context.\n",
     "\n",
-    "In this cookbook the sample document [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) is pretty short, so after chunking there aren't many passages to search through. To make the cookbook minimal, we did exhaustive search to find the relevent search.\n",
+    "In this cookbook the sample document [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) is pretty short, so after chunking there aren't many passages to search through. To make the cookbook minimal, we did exhaustive search to find the relevant search.\n",
     "\n",
     "In real world use cases, there may be a lot of chunks to go through for a single query, in which case you will need to use Approximate Nearest Neighbor (ANN) for efficiency. This is usually directly supported by vector databases. UniSim also supports ANN, please consult UniSim documentation and its [Colab](https://github.com/google/unisim/blob/main/notebooks/unisim_text_demo.ipynb) on indexing and searching.\n",
     "\n",