{
  "id": "open-platypus",
  "name": "Open-Platypus",
  "identifiers": {
    "huggingface": [
      "garage-bAInd/Open-Platypus"
    ]
  },
  "builder": "Garage-bAInd (Platypus authors)",
  "release_date": {
    "value": "2023-08-03",
    "source": "https://huggingface.co/api/datasets/garage-bAInd/Open-Platypus",
    "status": "partial",
    "note": "HF dataset repo creation date, not the builder's own release announcement."
  },
  "availability": {
    "value": "available",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/datasets/garage-bAInd/Open-Platypus"
  },
  "content": {
    "value": "Card: 'focused on improving LLM logical reasoning skills and ... used to train the Platypus2 models'. A filtered mix (keyword search, then Sentence Transformers to drop questions above 80% similarity) of PRM800K, MATH, ScienceQA, SciBench, ReClor, TheoremQA, leetcode-solutions-python-testgen-gpt4, airoboros-gpt4-1.4.1, tigerbot-kaggle-leetcodesolutions, ARB and openassistant-guanaco.",
    "source": "https://huggingface.co/datasets/garage-bAInd/Open-Platypus",
    "status": "recorded",
    "note": "Two components have 'gpt4' in their names; the card does not say what share of the mix is model-generated, so no model-parent edge is recorded."
  },
  "primary_sources": [
    "https://huggingface.co/datasets/garage-bAInd/Open-Platypus"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "created from primary sources (tranche 2 dataset records)",
      "by": "wilson-pruitt + claude"
    }
  ]
}