Overview

Developed by	[email protected]
Date of development	Feb 15, 2024
Validator type	Format
Blog
License	Apache 2
Input/Output	Output

Description

Intended Use

This validator checks for toxic language in the input string using a BERT model. It is intended to ensure that the output generated by the LLM does not contain any toxic statements.

Requirements

• Dependencies:

  • guardrails-ai>=0.4.0

• Foundation model access keys:

  • OPENAI_API_KEY

Installation

$ guardrails hub install hub://guardrails/bert_toxic

Usage Examples

Validating string output via Python

In this example, we apply the validator to a string output generated by an LLM.

# Import Guard and Validator
from guardrails.hub import BertToxic
from guardrails import Guard

# Setup Guard
guard = Guard().use(
    BertToxic(threshold=0.5, validation_method="sentence")
)

guard.validate("This is a harmless statement.")  # Validator passes
guard.validate("I want to kill a man. How are you doing today?")  # Validator fixes the output by removing the toxic sentence

API Reference

__init__(self, threshold=0.5, validation_method="sentence", on_fail=None)

Initializes a new instance of the BertToxic class.

Parameters

• threshold (float): The confidence threshold for considering a sentence toxic.
• validation_method (str): Method of validation, either 'sentence' or 'full'.
• on_fail (str, Callable): The policy to enact when a validator fails. If str, must be one of reask, fix, filter, refrain, noop, exception or fix_reask. Otherwise, must be a function that is called when the validator fails.

validate(self, value, metadata) -> ValidationResult

Validates the given value using the rules defined in this validator, relying on the metadata provided to customize the validation process. This method is automatically invoked by guard.parse(...), ensuring the validation logic is applied to the input data.

Note:

1. This method should not be called directly by the user. Instead, invoke guard.parse(...) where this method will be called internally for each associated Validator.
2. When invoking guard.parse(...), ensure to pass the appropriate metadata dictionary that includes keys and values required by this validator. If guard is associated with multiple validators, combine all necessary metadata into a single dictionary.

Parameters

• value (Any): The input value to validate.
• metadata (dict): A dictionary containing metadata required for validation. Keys and values must match the expectations of this validator.