{
  "default": {
    "builder_name": "parquet",
    "config_description": "MedARC processed outputs grouped by model and environment.",
    "config_name": "default",
    "data_files": {
      "train": [
        "baichuan-m2/agentclinic.parquet",
        "baichuan-m2/careqa_open.parquet",
        "baichuan-m2/healthbench-consensus.parquet",
        "baichuan-m2/healthbench-hard.parquet",
        "baichuan-m2/healthbench.parquet",
        "baichuan-m2/med_dialog.parquet",
        "baichuan-m2/medagentbenchv2.parquet",
        "baichuan-m2/medcasereasoning.parquet",
        "baichuan-m2/medec.parquet",
        "baichuan-m2/medexqa.parquet",
        "baichuan-m2/medicationqa.parquet",
        "baichuan-m2/medrbench-1turn.parquet",
        "baichuan-m2/medrbench-free_turn.parquet",
        "baichuan-m2/medrbench-oracle.parquet",
        "baichuan-m2/mtsamples_procedures.parquet",
        "baichuan-m2/mtsamples_replicate.parquet",
        "baichuan-m2/pubhealthbench_freeform.parquet",
        "baichuan-m3/agentclinic.parquet",
        "baichuan-m3/careqa_open.parquet",
        "baichuan-m3/healthbench-consensus.parquet",
        "baichuan-m3/healthbench-hard.parquet",
        "baichuan-m3/healthbench.parquet",
        "baichuan-m3/med_dialog.parquet",
        "baichuan-m3/medcasereasoning.parquet",
        "baichuan-m3/medec.parquet",
        "baichuan-m3/medexqa.parquet",
        "baichuan-m3/medicationqa.parquet",
        "baichuan-m3/medrbench-1turn.parquet",
        "baichuan-m3/medrbench-free_turn.parquet",
        "baichuan-m3/medrbench-oracle.parquet",
        "baichuan-m3/mtsamples_procedures.parquet",
        "baichuan-m3/mtsamples_replicate.parquet",
        "baichuan-m3/pubhealthbench_freeform.parquet",
        "gemini-3-pro-preview/agentclinic.parquet",
        "gemini-3-pro-preview/careqa_open.parquet",
        "gemini-3-pro-preview/healthbench-consensus.parquet",
        "gemini-3-pro-preview/healthbench-hard.parquet",
        "gemini-3-pro-preview/healthbench.parquet",
        "gemini-3-pro-preview/med_dialog.parquet",
        "gemini-3-pro-preview/medcasereasoning.parquet",
        "gemini-3-pro-preview/medec.parquet",
        "gemini-3-pro-preview/medexqa.parquet",
        "gemini-3-pro-preview/medicationqa.parquet",
        "gemini-3-pro-preview/medrbench-1turn.parquet",
        "gemini-3-pro-preview/medrbench-free_turn.parquet",
        "gemini-3-pro-preview/medrbench-oracle.parquet",
        "gemini-3-pro-preview/mtsamples_procedures.parquet",
        "gemini-3-pro-preview/mtsamples_replicate.parquet",
        "gemini-3-pro-preview/pubhealthbench_freeform.parquet",
        "glm-4.7-fp8/agentclinic.parquet",
        "glm-4.7-fp8/careqa_open.parquet",
        "glm-4.7-fp8/healthbench-consensus.parquet",
        "glm-4.7-fp8/healthbench-hard.parquet",
        "glm-4.7-fp8/healthbench.parquet",
        "glm-4.7-fp8/med_dialog.parquet",
        "glm-4.7-fp8/medagentbenchv2.parquet",
        "glm-4.7-fp8/medcasereasoning.parquet",
        "glm-4.7-fp8/medec.parquet",
        "glm-4.7-fp8/medexqa.parquet",
        "glm-4.7-fp8/medicationqa.parquet",
        "glm-4.7-fp8/medrbench-1turn.parquet",
        "glm-4.7-fp8/medrbench-free_turn.parquet",
        "glm-4.7-fp8/medrbench-oracle.parquet",
        "glm-4.7-fp8/mtsamples_procedures.parquet",
        "glm-4.7-fp8/mtsamples_replicate.parquet",
        "glm-4.7-fp8/pubhealthbench_freeform.parquet",
        "gpt-5_2/agentclinic.parquet",
        "gpt-5_2/careqa_open.parquet",
        "gpt-5_2/healthbench-consensus.parquet",
        "gpt-5_2/healthbench-hard.parquet",
        "gpt-5_2/healthbench.parquet",
        "gpt-5_2/med_dialog.parquet",
        "gpt-5_2/medcasereasoning.parquet",
        "gpt-5_2/medec.parquet",
        "gpt-5_2/medexqa.parquet",
        "gpt-5_2/medicationqa.parquet",
        "gpt-5_2/medrbench-1turn.parquet",
        "gpt-5_2/medrbench-free_turn.parquet",
        "gpt-5_2/medrbench-oracle.parquet",
        "gpt-5_2/mtsamples_procedures.parquet",
        "gpt-5_2/mtsamples_replicate.parquet",
        "gpt-5_2/pubhealthbench_freeform.parquet",
        "gpt-oss-120b-high/agentclinic.parquet",
        "gpt-oss-120b-high/careqa_open.parquet",
        "gpt-oss-120b-high/healthbench-consensus.parquet",
        "gpt-oss-120b-high/healthbench-hard.parquet",
        "gpt-oss-120b-high/healthbench.parquet",
        "gpt-oss-120b-high/med_dialog.parquet",
        "gpt-oss-120b-high/medagentbenchv2.parquet",
        "gpt-oss-120b-high/medcasereasoning.parquet",
        "gpt-oss-120b-high/medec.parquet",
        "gpt-oss-120b-high/medexqa.parquet",
        "gpt-oss-120b-high/medicationqa.parquet",
        "gpt-oss-120b-high/medrbench-1turn.parquet",
        "gpt-oss-120b-high/medrbench-free_turn.parquet",
        "gpt-oss-120b-high/medrbench-oracle.parquet",
        "gpt-oss-120b-high/mtsamples_procedures.parquet",
        "gpt-oss-120b-high/mtsamples_replicate.parquet",
        "gpt-oss-120b-high/pubhealthbench_freeform.parquet",
        "gpt-oss-20b-high/agentclinic.parquet",
        "gpt-oss-20b-high/careqa_open.parquet",
        "gpt-oss-20b-high/healthbench-consensus.parquet",
        "gpt-oss-20b-high/healthbench-hard.parquet",
        "gpt-oss-20b-high/healthbench.parquet",
        "gpt-oss-20b-high/med_dialog.parquet",
        "gpt-oss-20b-high/medagentbenchv2.parquet",
        "gpt-oss-20b-high/medcasereasoning.parquet",
        "gpt-oss-20b-high/medec.parquet",
        "gpt-oss-20b-high/medexqa.parquet",
        "gpt-oss-20b-high/medicationqa.parquet",
        "gpt-oss-20b-high/medrbench-1turn.parquet",
        "gpt-oss-20b-high/medrbench-free_turn.parquet",
        "gpt-oss-20b-high/medrbench-oracle.parquet",
        "gpt-oss-20b-high/mtsamples_procedures.parquet",
        "gpt-oss-20b-high/mtsamples_replicate.parquet",
        "gpt-oss-20b-high/pubhealthbench_freeform.parquet",
        "gpt_5_1/agentclinic.parquet",
        "gpt_5_1/careqa_open.parquet",
        "gpt_5_1/healthbench-consensus.parquet",
        "gpt_5_1/healthbench-hard.parquet",
        "gpt_5_1/healthbench.parquet",
        "gpt_5_1/med_dialog.parquet",
        "gpt_5_1/medagentbenchv2.parquet",
        "gpt_5_1/medcasereasoning.parquet",
        "gpt_5_1/medec.parquet",
        "gpt_5_1/medexqa.parquet",
        "gpt_5_1/medicationqa.parquet",
        "gpt_5_1/medrbench-1turn.parquet",
        "gpt_5_1/medrbench-free_turn.parquet",
        "gpt_5_1/medrbench-oracle.parquet",
        "gpt_5_1/mtsamples_procedures.parquet",
        "gpt_5_1/mtsamples_replicate.parquet",
        "gpt_5_1/pubhealthbench_freeform.parquet",
        "qwen-235b-a22b/agentclinic.parquet",
        "qwen-235b-a22b/careqa_open.parquet",
        "qwen-235b-a22b/healthbench-consensus.parquet",
        "qwen-235b-a22b/healthbench-hard.parquet",
        "qwen-235b-a22b/healthbench.parquet",
        "qwen-235b-a22b/med_dialog.parquet",
        "qwen-235b-a22b/medagentbenchv2.parquet",
        "qwen-235b-a22b/medcasereasoning.parquet",
        "qwen-235b-a22b/medec.parquet",
        "qwen-235b-a22b/medexqa.parquet",
        "qwen-235b-a22b/medicationqa.parquet",
        "qwen-235b-a22b/medrbench-1turn.parquet",
        "qwen-235b-a22b/medrbench-free_turn.parquet",
        "qwen-235b-a22b/medrbench-oracle.parquet",
        "qwen-235b-a22b/mtsamples_procedures.parquet",
        "qwen-235b-a22b/mtsamples_replicate.parquet",
        "qwen-235b-a22b/pubhealthbench_freeform.parquet",
        "qwen-3-8b/agentclinic.parquet",
        "qwen-3-8b/careqa_open.parquet",
        "qwen-3-8b/healthbench-consensus.parquet",
        "qwen-3-8b/healthbench-hard.parquet",
        "qwen-3-8b/healthbench.parquet",
        "qwen-3-8b/med_dialog.parquet",
        "qwen-3-8b/medagentbenchv2.parquet",
        "qwen-3-8b/medcasereasoning.parquet",
        "qwen-3-8b/medec.parquet",
        "qwen-3-8b/medexqa.parquet",
        "qwen-3-8b/medicationqa.parquet",
        "qwen-3-8b/medrbench-1turn.parquet",
        "qwen-3-8b/medrbench-free_turn.parquet",
        "qwen-3-8b/medrbench-oracle.parquet",
        "qwen-3-8b/mtsamples_procedures.parquet",
        "qwen-3-8b/mtsamples_replicate.parquet",
        "qwen-3-8b/pubhealthbench_freeform.parquet",
        "qwen-30b-a3b/agentclinic.parquet",
        "qwen-30b-a3b/careqa_open.parquet",
        "qwen-30b-a3b/healthbench-consensus.parquet",
        "qwen-30b-a3b/healthbench-hard.parquet",
        "qwen-30b-a3b/healthbench.parquet",
        "qwen-30b-a3b/med_dialog.parquet",
        "qwen-30b-a3b/medagentbenchv2.parquet",
        "qwen-30b-a3b/medcasereasoning.parquet",
        "qwen-30b-a3b/medec.parquet",
        "qwen-30b-a3b/medexqa.parquet",
        "qwen-30b-a3b/medicationqa.parquet",
        "qwen-30b-a3b/medrbench-1turn.parquet",
        "qwen-30b-a3b/medrbench-free_turn.parquet",
        "qwen-30b-a3b/medrbench-oracle.parquet",
        "qwen-30b-a3b/mtsamples_procedures.parquet",
        "qwen-30b-a3b/mtsamples_replicate.parquet",
        "qwen-30b-a3b/pubhealthbench_freeform.parquet",
        "sonnet-4_5/agentclinic.parquet",
        "sonnet-4_5/careqa_open.parquet",
        "sonnet-4_5/healthbench-consensus.parquet",
        "sonnet-4_5/healthbench-hard.parquet",
        "sonnet-4_5/healthbench.parquet",
        "sonnet-4_5/med_dialog.parquet",
        "sonnet-4_5/medagentbenchv2.parquet",
        "sonnet-4_5/medcasereasoning.parquet",
        "sonnet-4_5/medec.parquet",
        "sonnet-4_5/medexqa.parquet",
        "sonnet-4_5/medicationqa.parquet",
        "sonnet-4_5/medrbench-1turn.parquet",
        "sonnet-4_5/medrbench-free_turn.parquet",
        "sonnet-4_5/medrbench-oracle.parquet",
        "sonnet-4_5/mtsamples_procedures.parquet",
        "sonnet-4_5/mtsamples_replicate.parquet",
        "sonnet-4_5/pubhealthbench_freeform.parquet"
      ]
    },
    "dataset_size": null,
    "download_checksums": null,
    "download_size": null,
    "extras": {
      "processed_at": "2026-03-02T15:32:58Z",
      "processed_with_args": {
        "clean": false,
        "dry_run": false,
        "exclude_datasets": [],
        "exclude_models": [],
        "hf_max_files_per_commit": 20,
        "hf_pull_policy": null,
        "hf_repo": "bwarner/inference-scratch-llm",
        "hf_request_timeout": 300.0,
        "hf_retries": 5,
        "max_results_missing_pct": 2.5,
        "max_workers": 8,
        "replace_envs": [],
        "replace_models": [],
        "status": [
          "completed"
        ]
      }
    },
    "features": null,
    "homepage": null,
    "license": null,
    "splits": {
      "train": {
        "dataset_name": "default",
        "name": "train",
        "num_bytes": null,
        "num_examples": null
      }
    },
    "version": "0.0.0"
  }
}