edits to hyperparameters, prompt engineering

Author: Alexis Cook

notebooks/llmu/Parameters_for_Controlling_Outputs.ipynb

@@ -31,10 +31,9 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "The notebook has 3 sections:\n",
+    "The notebook has 2 sections:\n",
     "- **Model Type** - Select a variation of the Command model.\n",
-    "- **Randomness** - Use the `temperature` parameter to control the level of randomness of the model.\n",
-    "- **Conciseness** - Set the `preamble` parameter to an empty string to make model responses more concise."
+    "- **Randomness** - Use the `temperature` parameter to control the level of randomness of the model."
    ]
   },
   {
@@ -131,9 +130,7 @@
    "source": [
     "## Model Type\n",
     "\n",
-    "When calling the Chat endpoint, use the `model` parameter to choose from several variations of the Command model. \n",
-    "\n",
-    "[See the documentation](https://docs.cohere.com/docs/models#command) for the most updated list of available Cohere models."
+    "When calling the Chat endpoint, use the `model` parameter to choose from several variations of the Command model. In the example, we select [Command R+](https://docs.cohere.com/docs/command-r-plus)."
    ]
   },
   {
@@ -179,6 +176,8 @@
     "id": "Df4lLD-P3aj_"
    },
    "source": [
+    "[See the documentation](https://docs.cohere.com/docs/models#command) for the most updated list of available Cohere models.\n",
+    "\n",
     "## Randomness\n",
     "\n",
     "You can use the `temperature` parameter to control the level of randomness of the model. It is a value between 0 and 1. As you increase the temperature, the model gets more creative and random. Temperature can be tuned for different problems, and most people will find that the default temperature of 0.3 is a good starting point.\n",
@@ -284,131 +283,6 @@
     "                       model=\"command-r-plus\")\n",
     "    print(response.text)"
    ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Conciseness\n",
-    "\n",
-    "Sometimes, the model provides more context than you need to address a query. For instance, consider what happens when we ask the model to generate code."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "\n",
-       "  <style>\n",
-       "    pre {\n",
-       "        white-space: pre-wrap;\n",
-       "    }\n",
-       "  </style>\n",
-       "  "
-      ],
-      "text/plain": [
-       "<IPython.core.display.HTML object>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Here is an example of how to load a CSV file as a Pandas DataFrame:\n",
-      "```python\n",
-      "import pandas as pd\n",
-      "\n",
-      "# Load the CSV file into a Pandas DataFrame\n",
-      "df = pd.read_csv('filename.csv')\n",
-      "\n",
-      "# Display the first few rows of the DataFrame\n",
-      "print(df.head())\n",
-      "```\n",
-      "\n",
-      "In this example, we use the `pd.read_csv()` function to load the CSV file into a Pandas DataFrame. The `read_csv()` function takes the file path or URL of the CSV file as an argument and returns a Pandas DataFrame object.\n",
-      "\n",
-      "We then use the `head()` function to display the first few rows of the DataFrame, which can be useful for quickly inspecting the data.\n",
-      "\n",
-      "Note that you need to replace `'filename.csv'` with the actual file path or URL of your CSV file.\n"
-     ]
-    }
-   ],
-   "source": [
-    "response = co.chat(message=\"Generate code for loading a csv file as a pandas dataframe\", model=\"command-r-plus\")\n",
-    "print(response.text)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The model answers the question, and the response is relatively detailed.\n",
-    "\n",
-    "We can get the model to shorten its response by setting the `preamble` parameter to an empty string."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "\n",
-       "  <style>\n",
-       "    pre {\n",
-       "        white-space: pre-wrap;\n",
-       "    }\n",
-       "  </style>\n",
-       "  "
-      ],
-      "text/plain": [
-       "<IPython.core.display.HTML object>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Here is an example of how to load a CSV file as a Pandas DataFrame:\n",
-      "```python\n",
-      "import pandas as pd\n",
-      "\n",
-      "# Load the CSV file into a Pandas DataFrame\n",
-      "df = pd.read_csv('filename.csv')\n",
-      "\n",
-      "# Display the first few rows of the DataFrame\n",
-      "print(df.head())\n",
-      "```\n",
-      "\n",
-      "This code uses the `pd.read_csv()` function to load the CSV file into a Pandas DataFrame. The `print(df.head())` function then displays the first few rows of the DataFrame.\n"
-     ]
-    }
-   ],
-   "source": [
-    "response = co.chat(message=\"Generate code for loading a csv file as a pandas dataframe\",\n",
-    "                   model=\"command-r-plus\",\n",
-    "                   preamble=\"\")\n",
-    "print(response.text)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "As you have explored in this notebook, the Chat endpoint is a versatile tool that empowers developers with a useful array of options and parameters."
-   ]
   }
  ],
  "metadata": {

notebooks/llmu/Prompt_Engineering_Basics.ipynb

@@ -145,7 +145,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 4,
    "id": "0d4130e9",
    "metadata": {},
    "outputs": [
@@ -170,7 +170,7 @@
    ],
    "source": [
     "def generate_text(message):\n",
-    "    stream = co.chat_stream(message=message, model=\"command-r-plus\", preamble=\"\")\n",
+    "    stream = co.chat_stream(message=message, model=\"command-r-plus\")\n",
     "    for event in stream:\n",
     "        if event.event_type == \"text-generation\":\n",
     "            print(event.text, end='')"
@@ -196,7 +196,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 5,
    "id": "3802a5b5",
    "metadata": {},
    "outputs": [
@@ -222,9 +222,9 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Wireless Earbuds:\n",
+      "Sure! Here is a concise product description for wireless earbuds: \n",
       "\n",
-      "Experience true wireless freedom with our latest earbuds. Immerse yourself in your favorite tunes or take calls with crystal-clear clarity, all without the hassle of tangled cords. Our earbuds offer a secure and comfortable fit, ensuring they stay put during any activity. With a sleek and compact design, these earbuds are the perfect companion for your on-the-go lifestyle."
+      "\"Experience unparalleled sound with our wireless earbuds. Immerse yourself in rich, high-quality audio and enjoy the freedom of movement without tangles or hassles. Our earbuds offer a secure and comfortable fit, ensuring they stay put during any activity. With a sleek and portable design, these earbuds are the perfect companion for your active lifestyle.\""
      ]
     }
    ],
@@ -250,7 +250,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 6,
    "id": "a542f651",
    "metadata": {},
    "outputs": [
@@ -276,17 +276,20 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Hook: Experience True Wireless Freedom with Superior Sound\n",
+      "Here is a concise product description for wireless earbuds: \n",
       "\n",
-      "Solution: Introducing our cutting-edge wireless earbuds, designed to revolutionize your audio experience.\n",
+      "**Hook:** Tired of tangled cords and restricted movement? \n",
       "\n",
-      "Features and Benefits:\n",
-      "- Immersive Audio: Enjoy superior sound quality with deep bass and crystal-clear highs, ensuring an immersive listening experience.\n",
-      "- True Wireless Technology: Free yourself from tangles and restrictions with a stable Bluetooth connection, providing seamless connectivity and true wireless freedom.\n",
-      "- Comfortable Design: The ergonomic design ensures a secure and comfortable fit, allowing you to enjoy your music without distraction.\n",
-      "- Extended Battery Life: With a long-lasting battery, you can listen for hours on end without worrying about recharging.\n",
+      "**Solution:** Go wireless with our latest earbuds! \n",
       "\n",
-      "Call to Action: Upgrade your audio experience with our wireless earbuds. Immerse yourself in superior sound, enjoy the freedom of true wireless technology, and benefit from the comfort and convenience they offer."
+      "**Features and Benefits:** \n",
+      "- Sleek and ergonomic design ensures a secure and comfortable fit. \n",
+      "- Advanced Bluetooth technology for seamless connectivity and uninterrupted audio. \n",
+      "- Immersive sound quality with powerful bass and crystal-clear highs. \n",
+      "- Built-in microphone for hands-free calls and voice assistant access. \n",
+      "- Convenient charging case provides on-the-go power. \n",
+      "\n",
+      "**Call to Action:** Experience true wireless freedom and elevate your audio experience with our cutting-edge earbuds. Order now and enjoy uninterrupted music, calls, and more!"
      ]
     }
    ],
@@ -315,7 +318,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 7,
    "id": "1b5278a7",
    "metadata": {},
    "outputs": [
@@ -341,7 +344,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "The email expresses gratitude and appreciation to the team members for their hard work, dedication, and contributions that led to the success of the recent community meetup event."
+      "The email expresses gratitude to the team members for their hard work, dedication, and guidance in organizing a successful community meetup, and looks forward to future collaborations."
      ]
     }
    ],
@@ -381,7 +384,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 8,
    "id": "7a671108",
    "metadata": {},
    "outputs": [
@@ -438,7 +441,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 9,
    "id": "83f9631e",
    "metadata": {},
    "outputs": [
@@ -466,22 +469,25 @@
      "text": [
       "Here is a list of potential FAQs based on the provided text: \n",
       "\n",
-      "**Q: What does the Cohere Platform offer to developers and organizations?**\n",
-      "**A:** The Cohere Platform provides an API that gives developers and organizations access to cutting-edge Large Language Models (LLMs) without requiring expertise in machine learning. \n",
+      "- **Q: What does the Cohere Platform offer to developers and organizations?**\n",
+      "   A: The Cohere Platform offers an API that provides access to advanced Large Language Models (LLMs) without requiring machine learning expertise. It simplifies the process by handling data curation, model development, training, and serving. \n",
+      "\n",
+      "- **Q: What are the key language processing capabilities of the Cohere Platform?**\n",
+      "   A: The platform offers two main capabilities: text generation and text embedding. Text generation involves completing a prompt with a stream of generated text, like writing a haiku. Text embedding returns a numerical representation of the semantic meaning of a text input, useful for sentiment analysis and other measurements. \n",
       "\n",
-      "**Q: What are the key advantages of using the Cohere Platform?**\n",
-      "**A:** The platform handles all the complex tasks associated with LLMs, including data curation, model development, distributed training, and model serving. This allows users to focus on creating value in applied areas. \n",
+      "- **Q: How does the Cohere Platform benefit developers?**\n",
+      "   A: Developers can focus on creating valuable applications without getting bogged down by the complexities of building language processing capabilities from scratch. The platform handles the heavy lifting, allowing developers to save time and effort. \n",
       "\n",
-      "**Q: What are the two main language processing capabilities provided by the Cohere Platform?**\n",
-      "**A:** The platform offers text generation and text embedding capabilities. Text generation involves providing a prompt and receiving a completion, such as asking for a haiku. Text embedding returns a list of numbers representing the semantic meaning of the input text, useful for sentiment analysis and other applications. \n",
+      "- **Q: What is the difference between text generation and text embedding?**\n",
+      "   A: Text generation is about creating new text based on a prompt, like generating a haiku or continuing a story. Text embedding, on the other hand, translates text into a series of numbers that represent its semantic meaning, enabling quantitative analysis and understanding of the text's context. \n",
       "\n",
-      "**Q: What is an example of text generation using the Cohere Platform?**\n",
-      "**A:** One example is prompting the model to write a haiku, and the model will generate an original haiku as a response. \n",
+      "- **Q: Can I use the Cohere Platform for sentiment analysis?**\n",
+      "   A: Yes, the text embedding capability of the Cohere Platform is particularly useful for sentiment analysis. By converting text into numerical representations, you can quantitatively analyze and understand the sentiment or emotional tone expressed in a given piece of text. \n",
       "\n",
-      "**Q: How does text embedding work, and what is its utility?**\n",
-      "**A:** Text embedding translates a piece of text into a list of numbers that captures its semantic meaning. This is valuable for tasks that require understanding the sentiment or underlying meaning of a text passage, enabling quantitative analysis. \n",
+      "- **Q: Does the Cohere Platform require machine learning expertise to use?**\n",
+      "   A: No, the platform is designed to abstract away the complexities of machine learning. Developers can utilize the API without needing to understand the intricate details of machine learning or natural language processing. \n",
       "\n",
-      "These FAQs address some of the key points a user might inquire about regarding the Cohere Platform and its capabilities."
+      "These FAQs aim to address common inquiries that users might have about the Cohere Platform and its language processing capabilities."
      ]
     }
    ],