{
  "_schema": "https://data.nist.gov/od/dm/nerdm-schema/v0.7#",
  "@context": [
    "https://data.nist.gov/od/dm/nerdm-pub-context.jsonld",
    {
      "@base": "ark:/88434/mds2-3905"
    }
  ],
  "@type": [
    "nrdp:PublicDataResource",
    "dcat:Dataset"
  ],
  "_extensionSchemas": [
    "https://data.nist.gov/od/dm/nerdm-schema/pub/v0.7#/definitions/PublicDataResource"
  ],
  "@id": "ark:/88434/mds2-3905",
  "ediid": "ark:/88434/mds2-3905",
  "version": "1.0.0",
  "doi": "doi:10.18434/mds2-3905",
  "title": "Trojan Detection Software Challenge - llm-pretrain-apr2024-test",
  "contactPoint": {
    "fn": "Michael Majurski",
    "hasEmail": "mailto:michael.majurski@nist.gov"
  },
  "modified": "2024-04-16",
  "status": "available",
  "landingPage": "https://data.nist.gov/od/id/mds2-3905",
  "description": [
    "TrojAI llm-pretrain-apr2024 Test Dataset",
    "This 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."
  ],
  "keyword": [
    "Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;"
  ],
  "theme": [
    "Information Technology: Cybersecurity",
    "Information Technology: Software research"
  ],
  "topic": [
    {
      "@type": "Concept",
      "scheme": "https://data.nist.gov/od/dm/nist-themes/v1.1",
      "tag": "Information Technology: Cybersecurity"
    },
    {
      "@type": "Concept",
      "scheme": "https://data.nist.gov/od/dm/nist-themes/v1.1",
      "tag": "Information Technology: Software research"
    }
  ],
  "accessLevel": "public",
  "license": "https://www.nist.gov/open/license",
  "components": [
    {
      "accessURL": "https://drive.google.com/drive/folders/1klFznLk0BkQSnW0Am4QulVViEBQPB5WY?usp=drive_link",
      "title": "llm-pretrain-apr2024-test",
      "@type": [
        "nrdp:AccessPage",
        "dcat:Distribution"
      ],
      "@id": "#drive/folders/1klFznLk0BkQSnW0Am4QulVViEBQPB5WY?usp=drive_link",
      "_extensionSchemas": [
        "https://data.nist.gov/od/dm/nerdm-schema/pub/v0.7#/definitions/AccessPage"
      ]
    }
  ],
  "publisher": {
    "name": "National Institute of Standards and Technology",
    "@type": "org:Organization"
  },
  "language": [
    "en"
  ],
  "bureauCode": [
    "006:55"
  ],
  "programCode": [
    "006:052"
  ],
  "annotated": "2025-08-26T19:48:47.224601",
  "revised": "2025-08-26T19:48:47.224601",
  "issued": null,
  "firstIssued": "2025-08-26T19:48:47.224601"
}