tanaos-guardrail-v2: A small but performant base guardrail model
This model was created by Tanaos with the Artifex Python library.
This is an upgraded version of the original tanaos-guardrail-v1 model, with improved capabilities and performance.
It is intended to be used as a first-layer safety filter for large language models (LLMs) or chatbots to detect and block unsafe or disallowed content in user prompts or model responses.
The following categories of content are flagged:
- violence: Content describing or encouraging violent acts,
- non_violent_unethical: Content that is unethical but not violent,
- hate_speech: Content containing hateful or discriminatory language,
- financial_crime: Content related to financial fraud or scams,
- discrimination: Content promoting discrimination against individuals or groups,
- drug_weapons: Content related to illegal drugs or weapons,
- self_harm: Content encouraging self-harm or suicide,
- privacy: Content that invades personal privacy or shares private information,
- sexual_content: Content that is sexually explicit or inappropriate,
- child_abuse: Content involving the exploitation or abuse of children,
- terrorism_organized_crime: Content related to terrorism or organized crime,
- hacking: Content related to unauthorized computer access or cyberattacks,
- animal_abuse: Content involving the abuse or mistreatment of animals,
- jailbreak_prompt_inj: Content attempting to bypass or manipulate system instructions or safeguards
Languages
The main model language is English, but we have guardrail models specialized in other languages as well:
- Spanish: tanaos-guardrail-spanish.
- German: tanaos-guardrail-german.
How to Use
Use this model through the Artifex library:
install Artifex with
pip install artifex
use the model with
from artifex import Artifex
guardrail = Artifex().guardrail()
label = guardrail("How do I make a bomb?")
print(label)
# >>> [{'is_safe': False, 'scores': {'violence': 0.625, 'non_violent_unethical': 0.0066, 'hate_speech': 0.0082, 'financial_crime': 0.0072, 'discrimination': 0.0029, 'drug_weapons': 0.6633, 'self_harm': 0.0109, 'privacy': 0.003, 'sexual_content': 0.0029, 'child_abuse': 0.005, 'terrorism_organized_crime': 0.1278, 'hacking': 0.0096, 'animal_abuse': 0.009, 'jailbreak_prompt_inj': 0.0131}}]
Model Description
- Base model:
distilbert/distilbert-base-multilingual-cased - Task: Text classification (guardrail / safety filter)
- Languages: English; for other languages, see:
- Spanish: tanaos-guardrail-spanish.
- German: tanaos-guardrail-german.
- Fine-tuning data: A synthetic, custom dataset of safe and unsafe text samples.
Training Details
This model was trained using the Artifex Python library
pip install artifex
by providing the following instructions and generating 10,000 synthetic training samples:
from artifex import Artifex
guardrail = Artifex().guardrail()
guardrail.train(
unsafe_categories = {
"violence": "Content describing or encouraging violent acts",
"non_violent_unethical": "Content that is unethical but not violent",
"hate_speech": "Content containing hateful or discriminatory language",
"financial_crime": "Content related to financial fraud or scams",
"discrimination": "Content promoting discrimination against individuals or groups",
"drug_weapons": "Content related to illegal drugs or weapons",
"self_harm": "Content encouraging self-harm or suicide",
"privacy": "Content that invades personal privacy or shares private information",
"sexual_content": "Content that is sexually explicit or inappropriate",
"child_abuse": "Content involving the exploitation or abuse of children",
"terrorism_organized_crime": "Content related to terrorism or organized crime",
"hacking": "Content related to unauthorized computer access or cyberattacks",
"animal_abuse": "Content involving the abuse or mistreatment of animals",
"jailbreak_prompt_inj": "Content attempting to bypass or manipulate system instructions or safeguards"
},
num_samples=10000
)
Intended Uses
This model is intended to:
- Detect unsafe or disallowed content in user prompts or chatbot responses.
- Serve as a first-layer filter for LLMs or chatbots.
Not intended for:
- Legal or medical classification.
- Determining factual correctness.