{
  "harness": "lm-evaluation-harness 0.4.13",
  "fewshot": 0,
  "limit_per_task": 500,
  "device": "cpu",
  "nova_model": "plasmova/nova-v3",
  "baseline_model": "HuggingFaceTB/SmolLM2-135M",
  "comparison": {
    "hellaswag": {
      "metric": "acc_norm",
      "nova": 0.436,
      "smollm2": 0.442,
      "delta": -0.006000000000000005,
      "nova_stderr": 0.0221989546414768,
      "smollm2_stderr": 0.02223197069632112
    },
    "winogrande": {
      "acc": {
        "nova": 0.526,
        "smollm2": 0.52,
        "delta": 0.006000000000000005,
        "nova_stderr": 0.02235279165091416,
        "smollm2_stderr": 0.02236516042423134
      },
      "acc_norm": {
        "nova": 0.522,
        "smollm2": 0.506,
        "delta": 0.016000000000000014,
        "nova_stderr": 0.02236139673920787,
        "smollm2_stderr": 0.022381462412439324
      }
    }
  },
  "nova_evaluation_seconds": "452.16297420003684",
  "baseline_evaluation_seconds": "185.393080800015"
}