general-eval-card / schema /output-schema.json
Avijit Ghosh
fixed a lot of bugs, centralized schema
49d5ba7
{
"id": "string",
"systemName": "string",
"url": "string",
"provider": "string",
"version": "string",
"modelTag": "string",
"knowledgeCutoff": "YYYY-MM-DD",
"modelType": "foundational | fine-tuned | na",
"inputModalities": ["string"],
"outputModalities": ["string"],
"deploymentContexts": ["string"],
"evaluationDate": "YYYY-MM-DD",
"evaluator": "string",
"selectedCategories": [
"language-communication",
"problem-solving"
],
"categoryEvaluations": {
"language-communication": {
"benchmarkAnswers": {
"A1": "yes",
"A2": "no",
"A3": "na",
"A4": "yes",
"A5": "no",
"A6": "yes"
},
"processAnswers": {
"B1": "yes",
"B2": "no",
"B3": "yes",
"B4": "no",
"B5": "yes",
"B6": "na"
},
"benchmarkSources": {
"A1": [
{
"id": "1",
"benchmarkName": "MMLU",
"version": "2023-05",
"taskVariants": "multiple choice",
"metrics": "accuracy",
"url": "https://example.org",
"description": "string",
"sourceType": "external",
"score": "86.4%",
"confidenceInterval": "95% CI [85,88]",
"customFields": {}
}
]
},
"processSources": {
"B1": [
{
"id": "7",
"url": "https://example.org/doc",
"description": "string",
"sourceType": "internal",
"documentType": "Research Paper",
"title": "string",
"author": "string",
"organization": "string",
"date": "YYYY-MM-DD",
"customFields": {}
}
]
},
"additionalAspects": "string",
"score": {
"benchmarkScore": 0,
"processScore": 0,
"totalScore": 0,
"status": "strong"
}
}
}
}