{
  "checkpoint": "models/v004-contextual",
  "quantization": "FP32",
  "platform": "Windows-11-10.0.26200-SP0",
  "torch_version": "2.14.0+cpu",
  "threads": 2,
  "load_ms": 11.063799960538745,
  "disk_bytes": 532696,
  "artifact_sha256": "8bff26cf3b8cb4c76f971620cb281159cfe37743d953eaf68f7eace5b1534f66",
  "observed_process_rss_bytes": 281284608,
  "memory_scope": "Observed Python RSS including training-library imports and evaluation tensors, not browser or exact peak",
  "end_to_end_policy_ms": {
    "median": 1.1031999601982534,
    "p95": 1.6467999666929245
  },
  "neural_forward_ms": {
    "median": 0.6952500552870333,
    "p95": 1.0463000508025289
  },
  "feature_encoding_ms": {
    "median": 0.4278999986127019,
    "p95": 0.7296999683603644
  },
  "python_cpu_ms": {
    "median": 0.0,
    "p95": 15.625
  },
  "evaluation": {
    "validation": {
      "samples": 480,
      "action_accuracy": 1.0,
      "target_accuracy": 1.0,
      "joint_step_accuracy": 1.0,
      "action_ece": 2.777576446533203e-05,
      "target_ece": 0.0003707166906679049,
      "candidate_recall": 1.0
    },
    "test": {
      "samples": 480,
      "action_accuracy": 1.0,
      "target_accuracy": 1.0,
      "joint_step_accuracy": 1.0,
      "action_ece": 0.0,
      "target_ece": 0.0005633262626361102,
      "candidate_recall": 1.0
    },
    "novel_wording": {
      "samples": 480,
      "action_accuracy": 0.8520833253860474,
      "target_accuracy": 0.9458333253860474,
      "joint_step_accuracy": 0.8395833373069763,
      "action_ece": 0.13770944606221747,
      "target_ece": 0.046679925406351686,
      "candidate_recall": 1.0
    }
  },
  "target_vps_validated": false,
  "scope": "Synthetic single-step benchmark. Two torch threads, not two-vCPU CPU affinity or target EPYC."
}