Merge branch 'main' into fix/typos-gemma2-notebooks

bebechien · web-flow · commit 212752f8e9cb · 2025-04-21T11:06:53.000+09:00
diff --git a/CodeGemma/[CodeGemma_1]Common_use_cases.ipynb b/CodeGemma/[CodeGemma_1]Common_use_cases.ipynb
@@ -81,7 +81,7 @@
       "source": [
         "### Configure your credentials\n",
         "\n",
-        "Add your your Kaggle credentials to the Colab Secrets manager to securely store it.\n",
+        "Add your Kaggle credentials to the Colab Secrets manager to securely store it.\n",
         "\n",
         "1. Open your Google Colab notebook and click on the 🔑 Secrets tab in the left panel. <img src=\"https://storage.googleapis.com/generativeai-downloads/images/secrets.jpg\" alt=\"The Secrets tab is found on the left panel.\" width=50%>\n",
         "2. Create new secrets: `KAGGLE_USERNAME` and `KAGGLE_KEY`\n",
diff --git a/CodeGemma/[CodeGemma_1]Finetune_with_SQL.ipynb b/CodeGemma/[CodeGemma_1]Finetune_with_SQL.ipynb
@@ -401,7 +401,7 @@
         "* a python script producing a SQL query\n",
         "* two separate scripts producing respectively, python and SQL code. \n",
         "\n",
-        "CodeGemma picked the the first option. Bear it in mind!"
+        "CodeGemma picked the first option. Bear it in mind!"
       ]
     },
     {
@@ -412,7 +412,7 @@
       "source": [
         "## Fine-tuning the model with LoRA\n",
         "\n",
-        "This section of the guide focuses on training your Large Language Model (LLM) to generate SQL code fron natural language. Here, we will explore the process of fine-tuning your model to enable it to produce high quality SQL queries."
+        "This section of the guide focuses on training your Large Language Model (LLM) to generate SQL code from natural language. Here, we will explore the process of fine-tuning your model to enable it to produce high quality SQL queries."
       ]
     },
     {
@@ -943,7 +943,7 @@
       "source": [
         "This time the model picked the second option providing two separate scripts producing respectively, python and SQL code! \n",
         "\n",
-        "The model knows we 'prefer' to get a SQL query now but it didn't forget the other porgramming languages it's been trained on."
+        "The model knows we 'prefer' to get a SQL query now but it didn't forget the other programming languages it's been trained on."
       ]
     },
     {
@@ -970,8 +970,8 @@
         "id": "HIDWBva0_SX4"
       },
       "source": [
-        "# Note: The token needs to have \"write\" permisssion\n",
-        "#       You can chceck it here:\n",
+        "# Note: The token needs to have \"write\" permission\n",
+        "#       You can check it here:\n",
         "#       https://huggingface.co/settings/tokens\n",
         "model.push_to_hub(\"my-codegemma-7-finetuned-model\")"
       ]
@@ -1008,9 +1008,9 @@
         "id": "0wEjhtJawvSr"
       },
       "source": [
-        "!model=\"google/codegemma-7b-it\" # ID of the model in Hugging Face hube\n",
+        "!model=\"google/codegemma-7b-it\" # ID of the model in Hugging Face hub\n",
         "# (you can use your own fine-tuned model from\n",
-        "# the prevous step)\n",
+        "# the previous step)\n",
         "!volume=$PWD/data               # Shared directory with the Docker container\n",
         "# to avoid downloading weights every run\n",
         "\n",
diff --git a/Gemma/[Gemma_1]Advanced_Prompting_Techniques.ipynb b/Gemma/[Gemma_1]Advanced_Prompting_Techniques.ipynb
@@ -82,7 +82,7 @@
       "source": [
         "### Configure your credentials\n",
         "\n",
-        "Add your your Kaggle credentials to the Colab Secrets manager to securely store it.\n",
+        "Add your Kaggle credentials to the Colab Secrets manager to securely store it.\n",
         "\n",
         "1. Open your Google Colab notebook and click on the 🔑 Secrets tab in the left panel. <img src=\"https://storage.googleapis.com/generativeai-downloads/images/secrets.jpg\" alt=\"The Secrets tab is found on the left panel.\" width=50%>\n",
         "2. Create new secrets: `KAGGLE_USERNAME` and `KAGGLE_KEY`\n",
@@ -350,7 +350,7 @@
         }
       ],
       "source": [
-        "prompt = \"\"\"Genereate a single line of hashtags for the given topic by in the same style as the following examples:\n",
+        "prompt = \"\"\"Generate a single line of hashtags for the given topic by in the same style as the following examples:\n",
         "\n",
         "Topic: Books\n",
         "#BooksLover #Books #MyBooks #BestBook #BookOfTheYear\n",
diff --git a/Gemma/[Gemma_1]Basics_with_HF.ipynb b/Gemma/[Gemma_1]Basics_with_HF.ipynb
@@ -881,8 +881,8 @@
         }
       ],
       "source": [
-        "# Note: The token needs to have \"write\" permisssion\n",
-        "#       You can chceck it here:\n",
+        "# Note: The token needs to have \"write\" permission\n",
+        "#       You can check it here:\n",
         "#       https://huggingface.co/settings/tokens\n",
         "model.push_to_hub(\"my-gemma-2-finetuned-model\")"
       ]
@@ -921,9 +921,9 @@
       },
       "outputs": [],
       "source": [
-        "!model=\"google/gemma-1.1-2b-it\" # ID of the model in Hugging Face hube\n",
+        "!model=\"google/gemma-1.1-2b-it\" # ID of the model in Hugging Face hub\n",
         "# (you can use your own fine-tuned model from\n",
-        "# the prevous step)\n",
+        "# the previous step)\n",
         "!volume=$PWD/data               # Shared directory with the Docker container\n",
         "# to avoid downloading weights every run\n",
         "\n",
diff --git a/Gemma/[Gemma_1]Common_use_cases.ipynb b/Gemma/[Gemma_1]Common_use_cases.ipynb
@@ -81,7 +81,7 @@
       "source": [
         "### Configure your credentials\n",
         "\n",
-        "Add your your Kaggle credentials to the Colab Secrets manager to securely store it.\n",
+        "Add your Kaggle credentials to the Colab Secrets manager to securely store it.\n",
         "\n",
         "1. Open your Google Colab notebook and click on the 🔑 Secrets tab in the left panel. <img src=\"https://storage.googleapis.com/generativeai-downloads/images/secrets.jpg\" alt=\"The Secrets tab is found on the left panel.\" width=50%>\n",
         "2. Create new secrets: `KAGGLE_USERNAME` and `KAGGLE_KEY`\n",
diff --git a/Gemma/[Gemma_1]Finetune_distributed.ipynb b/Gemma/[Gemma_1]Finetune_distributed.ipynb
@@ -1260,7 +1260,7 @@
       "source": [
         "# What's next\n",
         "\n",
-        "In this tutorial, you learned how to chat with the Gemma 7B model and fine-tune it to speak like a pirate, using Keras on JAX. You also learned how to load and train the large model in a distributed manner, on powerful TPUs, uising model parallelism.\n",
+        "In this tutorial, you learned how to chat with the Gemma 7B model and fine-tune it to speak like a pirate, using Keras on JAX. You also learned how to load and train the large model in a distributed manner, on powerful TPUs, using model parallelism.\n",
         "\n",
         "Here are a few suggestions for what else to learn, about Keras and JAX:\n",
         "* [Distributed training with Keras 3](https://keras.io/guides/distribution/).\n",
diff --git a/Gemma/[Gemma_1]Minimal_RAG.ipynb b/Gemma/[Gemma_1]Minimal_RAG.ipynb
@@ -85,9 +85,9 @@
         "\n",
         "### Chunking the data\n",
         "\n",
-        "To improve the relevance of content returned by the vector database during  retrieval, break down large documents into smaller pieces or chunks while ingesting the document.\n",
+        "To improve the relevance of content returned by the vector database during retrieval, break down large documents into smaller pieces or chunks while ingesting the document.\n",
         "\n",
-        "In this cookcook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."
+        "In this cookbook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."
       ]
     },
     {
@@ -828,7 +828,7 @@
         "id": "uXLpmtoeU0gx"
       },
       "source": [
-        "Now load the Gemma model in quanzied 4-bit mode using Hugging Face."
+        "Now load the Gemma model in quantized 4-bit mode using Hugging Face."
       ]
     },
     {
diff --git a/Gemma/[Gemma_1]RAG_with_ChromaDB.ipynb b/Gemma/[Gemma_1]RAG_with_ChromaDB.ipynb
@@ -87,7 +87,7 @@
         "\n",
         "To improve the relevance of content returned by the vector database during  retrieval, break down large documents into smaller pieces or chunks while ingesting the document.\n",
         "\n",
-        "In this cookcook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."
+        "In this cookbook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."
       ]
     },
     {
@@ -400,7 +400,7 @@
       "source": [
         "### Generate the answer\n",
         "\n",
-        "Now load the Gemma model in quanzied 4-bit mode using Hugging Face."
+        "Now load the Gemma model in quantized 4-bit mode using Hugging Face."
       ]
     },
     {
diff --git a/Gemma/[Gemma_1]data_parallel_inference_in_jax_tpu.ipynb b/Gemma/[Gemma_1]data_parallel_inference_in_jax_tpu.ipynb
@@ -217,7 +217,7 @@
         "## Load the Model\n",
         "You will use the latest [Gemma-2B](https://huggingface.co/google/gemma-1.1-2b-it), this model offers 2 billion parameters, ensuring a lightweight footprint.\n",
         "\n",
-        "The Gemma model can be loaded using the familiar [`from_pretrained`](https://huggingface.co/docs/transformers/v4.38.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method in Transformers. This method downloads the model weights from the Hugging Face Hub the first time it is called, and subsequently intialises the Gemma model using these weights.\n"
+        "The Gemma model can be loaded using the familiar [`from_pretrained`](https://huggingface.co/docs/transformers/v4.38.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method in Transformers. This method downloads the model weights from the Hugging Face Hub the first time it is called, and subsequently initializes the Gemma model using these weights.\n"
       ]
     },
     {

Original file line number	Diff line number	Diff line change
`@@ -85,9 +85,9 @@`
`85`	`85`	`"\n",`
`86`	`86`	`"### Chunking the data\n",`
`87`	`87`	`"\n",`
`88`		`- "To improve the relevance of content returned by the vector database during retrieval, break down large documents into smaller pieces or chunks while ingesting the document.\n",`
	`88`	`+ "To improve the relevance of content returned by the vector database during retrieval, break down large documents into smaller pieces or chunks while ingesting the document.\n",`
`89`	`89`	`"\n",`
`90`		`- "In this cookcook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."`
	`90`	`+ "In this cookbook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."`
`91`	`91`	`]`
`92`	`92`	`},`
`93`	`93`	`{`
`@@ -828,7 +828,7 @@`
`828`	`828`	`"id": "uXLpmtoeU0gx"`
`829`	`829`	`},`
`830`	`830`	`"source": [`
`831`		`- "Now load the Gemma model in quanzied 4-bit mode using Hugging Face."`
	`831`	`+ "Now load the Gemma model in quantized 4-bit mode using Hugging Face."`
`832`	`832`	`]`
`833`	`833`	`},`
`834`	`834`	`{`
Original file line number	Diff line number	Diff line change
`@@ -87,7 +87,7 @@`
`87`	`87`	`"\n",`
`88`	`88`	`"To improve the relevance of content returned by the vector database during retrieval, break down large documents into smaller pieces or chunks while ingesting the document.\n",`
`89`	`89`	`"\n",`
`90`		`- "In this cookcook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."`
	`90`	`+ "In this cookbook, you will use the [Google I/O 2024 Gemma family expansion launch blog](https://developers.googleblog.com/en/gemma-family-and-toolkit-expansion-io-2024/) as the sample document and Google's [Open Source HtmlChunker](https://github.com/google/labs-prototypes/tree/main/seeds/chunker-python) to chunk it up into passages."`
`91`	`91`	`]`
`92`	`92`	`},`
`93`	`93`	`{`
`@@ -400,7 +400,7 @@`
`400`	`400`	`"source": [`
`401`	`401`	`"### Generate the answer\n",`
`402`	`402`	`"\n",`
`403`		`- "Now load the Gemma model in quanzied 4-bit mode using Hugging Face."`
	`403`	`+ "Now load the Gemma model in quantized 4-bit mode using Hugging Face."`
`404`	`404`	`]`
`405`	`405`	`},`
`406`	`406`	`{`
Original file line number	Diff line number	Diff line change
`@@ -217,7 +217,7 @@`
`217`	`217`	`"## Load the Model\n",`
`218`	`218`	`"You will use the latest [Gemma-2B](https://huggingface.co/google/gemma-1.1-2b-it), this model offers 2 billion parameters, ensuring a lightweight footprint.\n",`
`219`	`219`	`"\n",`
`220`		- "The Gemma model can be loaded using the familiar [`from_pretrained`](https://huggingface.co/docs/transformers/v4.38.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method in Transformers. This method downloads the model weights from the Hugging Face Hub the first time it is called, and subsequently intialises the Gemma model using these weights.\n"
	`220`	+ "The Gemma model can be loaded using the familiar [`from_pretrained`](https://huggingface.co/docs/transformers/v4.38.1/en/main_classes/model#transformers.FlaxPreTrainedModel.from_pretrained) method in Transformers. This method downloads the model weights from the Hugging Face Hub the first time it is called, and subsequently initializes the Gemma model using these weights.\n"
`221`	`221`	`]`
`222`	`222`	`},`
`223`	`223`	`{`