{
  "commands": {
    "convert": [
      "PYTHON",
      "LLAMA_CPP/convert_hf_to_gguf.py",
      "STAGING",
      "--outfile",
      "INTERMEDIATE",
      "--outtype",
      "f16"
    ],
    "quantize": [
      "LLAMA_QUANTIZE",
      "--output-tensor-type",
      "f16",
      "--token-embedding-type",
      "f16",
      "INTERMEDIATE",
      "OUTPUT",
      "Q4_K_M"
    ]
  },
  "comparison_to_f32_reference": {
    "decode": {
      "max_abs_error": 0.00035304948687553406,
      "mean_abs_error": 0.00010476766328793019,
      "top1_agreement": true,
      "top5_overlap": 5
    },
    "prefill": {
      "max_abs_error": 0.00022038817405700684,
      "mean_abs_error": 6.761292752344161e-05,
      "top1_agreement": true,
      "top5_overlap": 5
    }
  },
  "converter": "convert_hf_to_gguf.py",
  "converter_log_sha256": "5e83ed6cc964b76599a9aabff0b1677c6c12b1405dc8d20035f8e7ac3a9f4195",
  "direct_token_validation": {
    "all_logits_finite": true,
    "decode_token_id": 13,
    "exactly_reproducible": true,
    "prefill_token_ids": [
      3,
      5,
      7,
      11
    ]
  },
  "effective_eog_token_ids": [
    2
  ],
  "eog_validation_range": [
    0,
    127
  ],
  "format": "GGUF",
  "intermediate_f16_sha256": "75dc83a81398e18d3291b4c6505f2d6cefb396cfbed20edfc1a8aeec1aa6e3c5",
  "intermediate_retained": false,
  "llama_cpp_revision": "40b740ad05c531b9d57aca6698c3ed553a9e784c",
  "model_sha256": "8224570d8c01310dc10083e28da8b077503469fe06d701f44b047011065fda43",
  "model_size_bytes": 87936,
  "notes": "Q4_K_M is lossy; numerical differences are recorded as observations, not pass thresholds.",
  "quantization_requested": "Q4_K_M",
  "quantizer": "llama-quantize",
  "quantizer_fallback_observed": true,
  "quantizer_log_sha256": "1fe9befa42eff0d3f2696731a9838c9f2ce313206c9fcf2450e5c7a4da4c7fc9",
  "schema_version": 1,
  "source_auxiliary_tokenizer_sha256": "517673e2ef9f9b9a667d03914009d1cec25514c1d8a1f8e0eded4d9f44eab94d",
  "source_package_model_sha256": "b5fd132288d1a8e8a4c3dd57ff7bc863ba4e3aa7d0c58b8fd3a178f4b4c70f7f",
  "staging_special_token_normalization": {
    "applied": false,
    "gguf_primary_eos_token_id": 2,
    "reason": "GGUF exposes one primary EOS metadata field; all source EOS IDs remain documented by tokenizer metadata.",
    "source_eos_token_id": 2
  },
  "tensor_type_histogram": {
    "F16": 13,
    "F32": 5,
    "Q5_0": 1,
    "Q8_0": 1
  },
  "tiny_dimension_tensor_overrides": {
    "output.weight": "F16",
    "reason": "These tensors can be too narrow for the Q8_0 types selected by the Q4_K_M mixture.",
    "token_embd.weight": "F16"
  },
  "tokenizer_smoke": {
    "hf_native_ids_equal": true,
    "text": "a",
    "token_ids": [
      126,
      124
    ]
  }
}