| { |
| "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 |
| ] |
| } |
| } |
|
|