Workflow uses Mistral-7b Model with specified prompt template for Toxicity moderation that identify and filter out 'Toxic' sentiment includes aggression,

Overview
1

Notes

Toxicity Moderation 

Workflow uses Mistral-7b Model with specified prompt template for Toxicity moderation that identify and filter out  'Toxic' sentiment includes aggression, hostility, or undue negativity and classifies the sentiment as 'Toxic', 'Suspicious', or 'Safe' based on the tone and content.

Text Moderation

Large Language Models (LLMs), like GPT (Generative Pretrained Transformer) variants, since they are pre-trained on diverse internet text, enabling them to understand context, nuance, and the subtleties of human language, to identify and filter out inappropriate or harmful content from digital platforms.

How to use the Toxicity Moderation workflow? 

Using Clarifai SDK

Export your PAT as an environment variable. Then, import and initialize the API Client.

Find your PAT in your security settings.

export CLARIFAI_PAT={your personal access token}

Prediction with the workflow

from clarifai.client.workflow import Workflow

workflow_url = 'https://clarifai.com/{{user_id}}/text-moderation/workflows/text-moderation-toxicity-mistral-7b'

text = 'I love this movie and i would watch it again and again!'

prediction = Workflow(workflow_url).predict_by_bytes(text.encode(), input_type="text")

# Get workflow results
print(prediction.results[0].outputs[-1].data)

Using Workflow

To utilize the Text Moderation workflow, you can input text through the Blue Plus Try your own Input button and it will filter out  'Toxic' sentiment includes aggression, hostility, or undue negativity.

  • Workflow ID
    text-moderation-toxicity-mistral-7b
  • Description
    Workflow uses Mistral-7b Model with specified prompt template for Toxicity moderation that identify and filter out 'Toxic' sentiment includes aggression, hostility, or undue negativity.
  • Last Updated
    Apr 09, 2024
  • Privacy
    PUBLIC
  • Share