{
  "_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/mds2-3235"
    }
  ],
  "@type": [
    "nrdp:DataPublication",
    "nrdp:PublicDataResource",
    "dcat:Dataset"
  ],
  "_extensionSchemas": [
    "https://data.nist.gov/od/dm/nerdm-schema/pub/v0.7#/definitions/PublicDataResource"
  ],
  "@id": "ark:/88434/mds2-3235",
  "ediid": "ark:/88434/mds2-3235",
  "version": "1.0.1",
  "doi": "doi:10.18434/mds2-3235",
  "title": "Trojan Detection Software Challenge - llm-pretrain-apr2024-train",
  "contactPoint": {
    "fn": "Michael Paul Majurski",
    "hasEmail": "mailto:michael.majurski@nist.gov"
  },
  "modified": "2024-04-16 00:00:00",
  "status": "available",
  "landingPage": "https://data.nist.gov/od/id/mds2-3235",
  "description": [
    "TrojAI llm-pretrain-apr2024 Train 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: Software research",
    "Information Technology: Cybersecurity"
  ],
  "topic": [
    {
      "@type": "Concept",
      "scheme": "https://data.nist.gov/od/dm/nist-themes/v1.1",
      "tag": "Information Technology: Software research"
    },
    {
      "@type": "Concept",
      "scheme": "https://data.nist.gov/od/dm/nist-themes/v1.1",
      "tag": "Information Technology: Cybersecurity"
    }
  ],
  "accessLevel": "public",
  "license": "https://www.nist.gov/open/license",
  "components": [
    {
      "accessURL": "https://drive.google.com/drive/folders/1eI7MsVi1qqSHvnfCUWkgNnphTk0Cth5M?usp=sharing",
      "title": "llm-pretrain-apr2024-train",
      "@type": [
        "nrdp:AccessPage",
        "dcat:Distribution"
      ],
      "@id": "#drive/folders/1eI7MsVi1qqSHvnfCUWkgNnphTk0Cth5M?usp=sharing",
      "_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": "2025-02-07T19:02:12.414035",
  "revised": "2024-04-19T05:50:18.197417",
  "issued": null,
  "firstIssued": "2024-04-19T05:50:18.197417",
  "_editStatus": "done",
  "releaseHistory": {
    "@id": "ark:/88434/mds2-3235.rel",
    "@type": [
      "nrdr:ReleaseHistory"
    ],
    "hasRelease": [
      {
        "version": "1.0.0",
        "issued": "2024-04-16",
        "@id": "ark:/88434/mds2-3235/pdr:v/1.0.0",
        "location": "https://data.nist.gov/od/id/ark:/88434/mds2-3235/pdr:v/1.0.0",
        "description": "initial release"
      },
      {
        "version": "1.0.1",
        "issued": "2024-04-16 00:00:00",
        "@id": "ark:/88434/mds2-3235/pdr:v/1.0.1",
        "location": "https://data.nist.gov/od/id/ark:/88434/mds2-3235/pdr:v/1.0.1",
        "description": "metadata update"
      }
    ]
  },
  "authors": [
    {
      "familyName": "Majurski",
      "fn": "Michael  Majurski",
      "givenName": "Michael",
      "middleName": "",
      "affiliation": [
        {
          "title": "National Institute of Standards and Technology",
          "@type": "org:Organization",
          "@id": "ror:05xpvk416"
        }
      ],
      "orcid": "0000-0001-9663-3803",
      "@type": "foaf:Person"
    },
    {
      "familyName": "Blattner",
      "fn": "Timothy  Blattner",
      "givenName": "Timothy",
      "middleName": "",
      "affiliation": [
        {
          "title": "National Institute of Standards and Technology",
          "@type": "org:Organization",
          "@id": "ror:05xpvk416"
        }
      ],
      "orcid": "0009-0005-5628-2777",
      "@type": "foaf:Person"
    },
    {
      "familyName": "Juba",
      "fn": "Derek  Juba",
      "givenName": "Derek",
      "middleName": "",
      "affiliation": [
        {
          "title": "National Institute of Standards and Technology",
          "@type": "org:Organization",
          "@id": "ror:05xpvk416"
        }
      ],
      "orcid": "0000-0003-3055-7128",
      "@type": "foaf:Person"
    }
  ]
}