U.S. flag

An official website of the United States government

Dot gov

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Https

Secure .gov websites use HTTPS
A lock () or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

  1. Home

Trojan Detection Software Challenge - llm-pretrain-apr2024-train

Data provided by  National Institute of Standards and Technology

TrojAI llm-pretrain-apr2024 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists Llama2 Large Language Models refined using fine-tuning and LoRA to perform next token prediction. A known percentage of these trained AI models have been poisoned with triggers which induces modified behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers into the model weights.

About this Dataset

Updated: 2024-09-06
Metadata Last Updated: 2024-04-16 00:00:00
Date Created: N/A
Data Provided by:
Dataset Owner: N/A

Access this data

View JSON data
Contact dataset owner Access URL
Table representation of structured data
Title Trojan Detection Software Challenge - llm-pretrain-apr2024-train
Description TrojAI llm-pretrain-apr2024 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists Llama2 Large Language Models refined using fine-tuning and LoRA to perform next token prediction. A known percentage of these trained AI models have been poisoned with triggers which induces modified behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers into the model weights.
Modified 2024-04-16 00:00:00
Publisher Name National Institute of Standards and Technology
Contact mailto:[email protected]
Keywords Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning; 
{
    "identifier": "ark:\/88434\/mds2-3235",
    "accessLevel": "public",
    "contactPoint": {
        "hasEmail": "mailto:[email protected]",
        "fn": "Michael Paul Majurski"
    },
    "programCode": [
        "006:045"
    ],
    "landingPage": "",
    "title": "Trojan Detection Software Challenge - llm-pretrain-apr2024-train",
    "description": "TrojAI llm-pretrain-apr2024 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists Llama2 Large Language Models refined using fine-tuning and LoRA to perform next token prediction. A known percentage of these trained AI models have been poisoned with triggers which induces modified behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers into the model weights.",
    "language": [
        "en"
    ],
    "distribution": [
        {
            "accessURL": "https:\/\/drive.google.com\/drive\/folders\/1eI7MsVi1qqSHvnfCUWkgNnphTk0Cth5M?usp=sharing",
            "title": "llm-pretrain-apr2024-train"
        }
    ],
    "bureauCode": [
        "006:55"
    ],
    "modified": "2024-04-16 00:00:00",
    "publisher": {
        "@type": "org:Organization",
        "name": "National Institute of Standards and Technology"
    },
    "theme": [
        "Information Technology:Software research",
        "Information Technology:Cybersecurity"
    ],
    "keyword": [
        "Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;"
    ]
}