{
  "project_id": "orielith9",
  "release": "3.0",
  "configs": {
    "manuscript": {
      "rows": 43,
      "split": "train",
      "path": "data/manuscript.jsonl",
      "fields": {
        "id": "str",
        "section_index": "int",
        "title": "str",
        "text": "str",
        "source_path": "str",
        "record_type": "str",
        "evidence_status": "str"
      }
    },
    "design_parameters": {
      "rows": 25,
      "split": "train",
      "path": "data/design_parameters.jsonl",
      "fields": {
        "id": "str",
        "parameter": "str",
        "value_json": "str",
        "source_path": "str",
        "evidence_status": "str"
      }
    },
    "layers": {
      "rows": 8,
      "split": "train",
      "path": "data/layers.jsonl",
      "fields": {
        "id": "str",
        "layer": "str",
        "z0_um": "int",
        "z1_um": "int",
        "thickness_um": "int",
        "role": "str",
        "source_path": "str",
        "evidence_status": "str"
      }
    },
    "formulations": {
      "rows": 16,
      "split": "train",
      "path": "data/formulations.jsonl",
      "fields": {
        "id": "str",
        "formulation": "str",
        "ingredient": "str",
        "amount_mmol": "str",
        "MW_g_mol": "str",
        "mass_g": "str",
        "group_amount_mmol": "str",
        "qualification": "str",
        "source_path": "str",
        "evidence_status": "str"
      }
    },
    "technology_compatibility": {
      "rows": 4,
      "split": "train",
      "path": "data/technology_compatibility.jsonl",
      "fields": {
        "id": "str",
        "profile": "str",
        "present_method": "str",
        "material_interface": "str",
        "status": "str",
        "not_claimed": "str",
        "qualification": "str",
        "source_path": "str",
        "evidence_status": "str"
      }
    },
    "frontier_comparisons": {
      "rows": 8,
      "split": "train",
      "path": "data/frontier_comparisons.jsonl",
      "fields": {
        "id": "str",
        "system": "str",
        "evidence_type": "str",
        "specific_metric": "str",
        "form_or_scope": "str",
        "recovery_or_memory_scope": "str",
        "compatibility_limit": "str",
        "primary_source": "str",
        "source_path": "str",
        "evidence_status": "str"
      }
    },
    "reference_model": {
      "rows": 72,
      "split": "train",
      "path": "data/reference_model.jsonl",
      "fields": {
        "id": "str",
        "model_state_z": "float",
        "model_host_kb": "float",
        "model_total_force": "float",
        "active_force": "float",
        "reference_force": "float",
        "true_translation": "float",
        "reconstructed_tether_extension": "float",
        "true_tether_extension": "float",
        "source_path": "str",
        "evidence_status": "str"
      }
    },
    "capture_control": {
      "rows": 121,
      "split": "train",
      "path": "data/capture_control.jsonl",
      "fields": {
        "id": "str",
        "model_time_s": "float",
        "opened_fraction_m": "float",
        "captured_fraction_r": "float",
        "g_N_m": "float",
        "receptor_force_fraction": "float",
        "Ktotal_N_m": "float",
        "source_path": "str",
        "evidence_status": "str"
      }
    }
  },
  "split_note": "train is the Hub storage convention only; no training/evaluation partition is implied",
  "experimental_records": 0,
  "transformations": [
    "manuscript split at explicit page markers; text preserved after outer whitespace trimming",
    "design JSON top-level values encoded as JSON strings to preserve heterogeneous types",
    "CSV rows preserved; layer coordinates cast to integers and model coordinates to floats",
    "deterministic IDs and explicit provenance/evidence labels added"
  ]
}