Harness Text Generation with nateraw/wizard-mega-13b-awq Cognitive Actions

22 Apr 2025
Harness Text Generation with nateraw/wizard-mega-13b-awq Cognitive Actions

In the ever-evolving landscape of natural language processing, the nateraw/wizard-mega-13b-awq API offers a powerful toolset for developers seeking to integrate advanced text generation capabilities into their applications. The Cognitive Actions provided in this spec allow you to utilize the wizard-mega-13b model, which is optimized for speed, quality, and accuracy in generating human-like text. With the ability to fine-tune the output using various parameters, developers can create engaging and contextually relevant text effortlessly.

Prerequisites

Before diving into the integration of the Cognitive Actions, ensure you have the following:

  • An API key for accessing the Cognitive Actions platform.
  • Basic knowledge of making HTTP requests and handling JSON data.
  • Familiarity with Python programming for implementing the conceptual usage examples.

The authentication process typically involves passing your API key as a bearer token in the headers of your requests, enabling secure and authorized access to the API's functionalities.

Cognitive Actions Overview

Generate Text with Wizard-Mega-13B

The Generate Text with Wizard-Mega-13B action leverages the wizard-mega-13b model quantized with AWQ specifically for text generation. This action enables developers to produce coherent and contextually appropriate text based on a given message. By adjusting parameters like top-k, top-p, temperature, and presence penalty, developers can fine-tune the output to meet their requirements.

Input

The input for this action is structured as a JSON object with the following schema:

{
  "topK": 50,
  "topP": 0.95,
  "message": "write me a short itinerary for my dog's birthday party.",
  "temperature": 0.8,
  "maxNewTokens": 512,
  "presencePenalty": 1
}
  • topK (integer, optional): Specifies the number of highest probability tokens to consider. Default is 50.
  • topP (number, optional): Sets a cumulative probability threshold for token selection. Default is 1.0.
  • message (string, required): The prompt or message for which you want to generate text. Example: "write me a short itinerary for my dog's birthday party."
  • temperature (number, optional): Controls the randomness of the output. Default is 1.0.
  • maxNewTokens (integer, optional): Maximum number of tokens to generate in the output. Default is 512.
  • presencePenalty (number, optional): Adjusts the likelihood of repeating tokens. Default is 1.15.

Output

The output of this action is a string containing the generated text. For example:

Great idea! Here's a short itinerary for your dog's birthday party:

1. Welcome: Start the party by welcoming all the guests and their dogs. Hand out goody bags with treats, toys, and other surprises for the dogs.
2. Games: Set up some fun games for the dogs to play, such as "hot dog race" or "doggy limbo." Make sure to have plenty of water on hand for the dogs to cool off after playing.
3. Photo booth: Create a photo booth with props and costumes for the dogs and their owners to take pictures. You can even hire a professional photographer to capture the memories of the day.
4. Cake time: Cut the birthday cake and let the dogs (and humans) enjoy it together. Make sure the cake is safe for dogs to eat and does not contain any ingredients that could harm them.
5. Goodbyes: As the party comes to an end, thank everyone for coming and send them home with more treats and goodies to continue the celebration.

Conceptual Usage Example (Python)

Here’s a conceptual Python code snippet demonstrating how to call the Generate Text action using a hypothetical Cognitive Actions execution endpoint. Replace the placeholders with actual values as necessary.

import requests
import json

# Replace with your Cognitive Actions API key and endpoint
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute" # Hypothetical endpoint

action_id = "6acb0e19-b99a-4d11-9f99-7d5e800662c1"  # Action ID for Generate Text with Wizard-Mega-13B

# Construct the input payload based on the action's requirements
payload = {
    "topK": 50,
    "topP": 0.95,
    "message": "write me a short itinerary for my dog's birthday party.",
    "temperature": 0.8,
    "maxNewTokens": 512,
    "presencePenalty": 1
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json"
}

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json={"action_id": action_id, "inputs": payload}  # Hypothetical structure
    )
    response.raise_for_status()  # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully:")
    print(json.dumps(result, indent=2))

except requests.exceptions.RequestException as e:
    print(f"Error executing action {action_id}: {e}")
    if e.response is not None:
        print(f"Response status: {e.response.status_code}")
        try:
            print(f"Response body: {e.response.json()}")
        except json.JSONDecodeError:
            print(f"Response body: {e.response.text}")

In this code snippet, we first define the API key and endpoint, then construct the input payload according to the action's schema. The response from the API is handled to ensure the output is printed in a readable format.

Conclusion

Integrating the nateraw/wizard-mega-13b-awq Cognitive Actions into your applications can significantly enhance user engagement by providing dynamic text generation capabilities. By utilizing the adjustable parameters, you can tailor the output to suit different contexts and requirements. Explore various use cases, such as content creation, chatbots, or personalized messaging, to leverage the full potential of these powerful Cognitive Actions. Happy coding!

← Back to All Articles