KapiCoder-Mongo-Small / evaluation /homologation_v1_summary.json
esilva's picture
Add v3 weights (v0.1.0-preview)
4ba57bc verified
Raw History Blame Contribute Delete
141 kB
{
"schema": "kapi-hf-eval-summary/1",
"kind": "homologation_v1",
"label": "Internal evaluation, synthetic out-of-distribution homologation; LLM- and grammar-written, not written by people; not fully independent of the v3 training data. Aggregates only; no item text.",
"model": "esilva/KapiCoder-Mongo-Small",
"format": "homolog-v1-eval-1",
"set_sha256": "aca0aae51533a95d830c1f296d621f9203a7d76475658fb1e2dcc8f389495332",
"n": 6030,
"models": {
"base_1.5b": {
"overall": {
"n": 6030,
"exact_match": 0.1856,
"normalized_match": 0.1874,
"edit_similarity": 0.3732,
"accepted_prefix_tokens": 2.2488,
"syntax_valid": 0.6774,
"mongo_valid": 0.6048,
"exec_valid": 0.4915,
"ceiling_exec_valid": 0.9307,
"exec_over_ceiling": 0.5281,
"exec_given_ref_ok": 0.5276,
"n_ref_ok": 5612,
"nonexistent_fields_rate": 0.0818,
"empty_rate": 0.0141,
"normalized_match_ci95": [
0.1778,
0.1974
],
"exec_over_ceiling_ci95": [
0.5151,
0.5409
]
},
"by_context_kind": {
"full": {
"n": 3720,
"exact_match": 0.1392,
"normalized_match": 0.1398,
"edit_similarity": 0.3221,
"accepted_prefix_tokens": 2.1505,
"syntax_valid": 0.6231,
"mongo_valid": 0.5567,
"exec_valid": 0.45,
"ceiling_exec_valid": 0.9237,
"exec_over_ceiling": 0.4872,
"exec_given_ref_ok": 0.4863,
"n_ref_ok": 3436,
"nonexistent_fields_rate": 0.0715,
"empty_rate": 0.0108
},
"irrelevant": {
"n": 1155,
"exact_match": 0.2442,
"normalized_match": 0.2485,
"edit_similarity": 0.43,
"accepted_prefix_tokens": 2.3446,
"syntax_valid": 0.7238,
"mongo_valid": 0.6442,
"exec_valid": 0.5342,
"ceiling_exec_valid": 0.9455,
"exec_over_ceiling": 0.565,
"exec_given_ref_ok": 0.565,
"n_ref_ok": 1092,
"nonexistent_fields_rate": 0.0961,
"empty_rate": 0.0208
},
"none": {
"n": 1155,
"exact_match": 0.2762,
"normalized_match": 0.2797,
"edit_similarity": 0.4809,
"accepted_prefix_tokens": 2.4693,
"syntax_valid": 0.8061,
"mongo_valid": 0.7203,
"exec_valid": 0.5827,
"ceiling_exec_valid": 0.9385,
"exec_over_ceiling": 0.6209,
"exec_given_ref_ok": 0.6208,
"n_ref_ok": 1084,
"nonexistent_fields_rate": 0.1004,
"empty_rate": 0.0182
}
},
"by_cut_kind": {
"key": {
"n": 959,
"exact_match": 0.1846,
"normalized_match": 0.1856,
"edit_similarity": 0.353,
"accepted_prefix_tokens": 1.4943,
"syntax_valid": 0.6486,
"mongo_valid": 0.6173,
"exec_valid": 0.5036,
"ceiling_exec_valid": 0.9166,
"exec_over_ceiling": 0.5494,
"exec_given_ref_ok": 0.5495,
"n_ref_ok": 879,
"nonexistent_fields_rate": 0.0563,
"empty_rate": 0.0
},
"member": {
"n": 600,
"exact_match": 0.4583,
"normalized_match": 0.4583,
"edit_similarity": 0.5483,
"accepted_prefix_tokens": 1.2783,
"syntax_valid": 0.85,
"mongo_valid": 0.6467,
"exec_valid": 0.575,
"ceiling_exec_valid": 0.89,
"exec_over_ceiling": 0.6461,
"exec_given_ref_ok": 0.6461,
"n_ref_ok": 534,
"nonexistent_fields_rate": 0.0767,
"empty_rate": 0.0
},
"multiline": {
"n": 480,
"exact_match": 0.0187,
"normalized_match": 0.025,
"edit_similarity": 0.3463,
"accepted_prefix_tokens": 3.7896,
"syntax_valid": 0.6167,
"mongo_valid": 0.5542,
"exec_valid": 0.4188,
"ceiling_exec_valid": 0.9583,
"exec_over_ceiling": 0.437,
"exec_given_ref_ok": 0.437,
"n_ref_ok": 460,
"nonexistent_fields_rate": 0.1437,
"empty_rate": 0.0
},
"operator": {
"n": 841,
"exact_match": 0.2331,
"normalized_match": 0.2342,
"edit_similarity": 0.3963,
"accepted_prefix_tokens": 2.4388,
"syntax_valid": 0.6338,
"mongo_valid": 0.5945,
"exec_valid": 0.5125,
"ceiling_exec_valid": 0.9489,
"exec_over_ceiling": 0.5401,
"exec_given_ref_ok": 0.5401,
"n_ref_ok": 798,
"nonexistent_fields_rate": 0.0559,
"empty_rate": 0.0
},
"partial": {
"n": 600,
"exact_match": 0.3333,
"normalized_match": 0.3333,
"edit_similarity": 0.4636,
"accepted_prefix_tokens": 2.0317,
"syntax_valid": 0.72,
"mongo_valid": 0.6367,
"exec_valid": 0.4717,
"ceiling_exec_valid": 0.91,
"exec_over_ceiling": 0.5184,
"exec_given_ref_ok": 0.5183,
"n_ref_ok": 546,
"nonexistent_fields_rate": 0.1217,
"empty_rate": 0.01
},
"rest": {
"n": 720,
"exact_match": 0.0778,
"normalized_match": 0.0778,
"edit_similarity": 0.2764,
"accepted_prefix_tokens": 2.2569,
"syntax_valid": 0.7056,
"mongo_valid": 0.6028,
"exec_valid": 0.475,
"ceiling_exec_valid": 0.9319,
"exec_over_ceiling": 0.5097,
"exec_given_ref_ok": 0.5097,
"n_ref_ok": 671,
"nonexistent_fields_rate": 0.0528,
"empty_rate": 0.1069
},
"stage": {
"n": 720,
"exact_match": 0.0222,
"normalized_match": 0.0306,
"edit_similarity": 0.3191,
"accepted_prefix_tokens": 3.5264,
"syntax_valid": 0.6056,
"mongo_valid": 0.5903,
"exec_valid": 0.4139,
"ceiling_exec_valid": 0.9486,
"exec_over_ceiling": 0.4363,
"exec_given_ref_ok": 0.4363,
"n_ref_ok": 683,
"nonexistent_fields_rate": 0.1361,
"empty_rate": 0.0028
},
"stmt_start": {
"n": 269,
"exact_match": 0.1227,
"normalized_match": 0.1227,
"edit_similarity": 0.4425,
"accepted_prefix_tokens": 2.6729,
"syntax_valid": 0.8439,
"mongo_valid": 0.6208,
"exec_valid": 0.6283,
"ceiling_exec_valid": 0.9777,
"exec_over_ceiling": 0.6426,
"exec_given_ref_ok": 0.6312,
"n_ref_ok": 263,
"nonexistent_fields_rate": 0.1004,
"empty_rate": 0.0
},
"value": {
"n": 841,
"exact_match": 0.1867,
"normalized_match": 0.1867,
"edit_similarity": 0.306,
"accepted_prefix_tokens": 1.6504,
"syntax_valid": 0.6195,
"mongo_valid": 0.5862,
"exec_valid": 0.4899,
"ceiling_exec_valid": 0.9251,
"exec_over_ceiling": 0.5296,
"exec_given_ref_ok": 0.5296,
"n_ref_ok": 778,
"nonexistent_fields_rate": 0.0488,
"empty_rate": 0.0
}
},
"by_domain": {
"adtech": {
"n": 3057,
"exact_match": 0.1835,
"normalized_match": 0.1861,
"edit_similarity": 0.3734,
"accepted_prefix_tokens": 2.2512,
"syntax_valid": 0.6817,
"mongo_valid": 0.6042,
"exec_valid": 0.4953,
"ceiling_exec_valid": 0.9303,
"exec_over_ceiling": 0.5324,
"exec_given_ref_ok": 0.5313,
"n_ref_ok": 2844,
"nonexistent_fields_rate": 0.0818,
"empty_rate": 0.0144
},
"telecom": {
"n": 2973,
"exact_match": 0.1877,
"normalized_match": 0.1887,
"edit_similarity": 0.373,
"accepted_prefix_tokens": 2.2462,
"syntax_valid": 0.6731,
"mongo_valid": 0.6054,
"exec_valid": 0.4877,
"ceiling_exec_valid": 0.931,
"exec_over_ceiling": 0.5238,
"exec_given_ref_ok": 0.5238,
"n_ref_ok": 2768,
"nonexistent_fields_rate": 0.0817,
"empty_rate": 0.0138
}
},
"by_author": {
"grammar": {
"n": 3220,
"exact_match": 0.1261,
"normalized_match": 0.1261,
"edit_similarity": 0.2908,
"accepted_prefix_tokens": 1.8161,
"syntax_valid": 0.5894,
"mongo_valid": 0.5509,
"exec_valid": 0.3829,
"ceiling_exec_valid": 0.8702,
"exec_over_ceiling": 0.44,
"exec_given_ref_ok": 0.439,
"n_ref_ok": 2802,
"nonexistent_fields_rate": 0.0761,
"empty_rate": 0.0081
},
"llm": {
"n": 2810,
"exact_match": 0.2537,
"normalized_match": 0.2577,
"edit_similarity": 0.4676,
"accepted_prefix_tokens": 2.7445,
"syntax_valid": 0.7783,
"mongo_valid": 0.6665,
"exec_valid": 0.616,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.616,
"exec_given_ref_ok": 0.616,
"n_ref_ok": 2810,
"nonexistent_fields_rate": 0.0883,
"empty_rate": 0.021
}
},
"by_author_context": {
"grammar/full": {
"n": 2121,
"exact_match": 0.099,
"normalized_match": 0.099,
"edit_similarity": 0.2651,
"accepted_prefix_tokens": 1.9043,
"syntax_valid": 0.5413,
"mongo_valid": 0.5101,
"exec_valid": 0.3446,
"ceiling_exec_valid": 0.8661,
"exec_over_ceiling": 0.3979,
"exec_given_ref_ok": 0.3963,
"n_ref_ok": 1837,
"nonexistent_fields_rate": 0.0707,
"empty_rate": 0.0057
},
"grammar/irrelevant": {
"n": 553,
"exact_match": 0.1374,
"normalized_match": 0.1374,
"edit_similarity": 0.2987,
"accepted_prefix_tokens": 1.5443,
"syntax_valid": 0.6239,
"mongo_valid": 0.5714,
"exec_valid": 0.4014,
"ceiling_exec_valid": 0.8861,
"exec_over_ceiling": 0.453,
"exec_given_ref_ok": 0.4531,
"n_ref_ok": 490,
"nonexistent_fields_rate": 0.0922,
"empty_rate": 0.0163
},
"grammar/none": {
"n": 546,
"exact_match": 0.2198,
"normalized_match": 0.2198,
"edit_similarity": 0.3827,
"accepted_prefix_tokens": 1.7491,
"syntax_valid": 0.7418,
"mongo_valid": 0.6886,
"exec_valid": 0.5128,
"ceiling_exec_valid": 0.87,
"exec_over_ceiling": 0.5894,
"exec_given_ref_ok": 0.5895,
"n_ref_ok": 475,
"nonexistent_fields_rate": 0.0806,
"empty_rate": 0.0092
},
"llm/full": {
"n": 1599,
"exact_match": 0.1926,
"normalized_match": 0.1939,
"edit_similarity": 0.3977,
"accepted_prefix_tokens": 2.4772,
"syntax_valid": 0.7317,
"mongo_valid": 0.6185,
"exec_valid": 0.5897,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.5897,
"exec_given_ref_ok": 0.5897,
"n_ref_ok": 1599,
"nonexistent_fields_rate": 0.0725,
"empty_rate": 0.0175
},
"llm/irrelevant": {
"n": 602,
"exact_match": 0.3422,
"normalized_match": 0.3505,
"edit_similarity": 0.5507,
"accepted_prefix_tokens": 3.0797,
"syntax_valid": 0.8156,
"mongo_valid": 0.711,
"exec_valid": 0.6561,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.6561,
"exec_given_ref_ok": 0.6561,
"n_ref_ok": 602,
"nonexistent_fields_rate": 0.0997,
"empty_rate": 0.0249
},
"llm/none": {
"n": 609,
"exact_match": 0.3268,
"normalized_match": 0.3333,
"edit_similarity": 0.5689,
"accepted_prefix_tokens": 3.1149,
"syntax_valid": 0.8637,
"mongo_valid": 0.7488,
"exec_valid": 0.6453,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.6453,
"exec_given_ref_ok": 0.6453,
"n_ref_ok": 609,
"nonexistent_fields_rate": 0.1182,
"empty_rate": 0.0263
}
}
},
"t1_distill": {
"overall": {
"n": 6030,
"exact_match": 0.3687,
"normalized_match": 0.3697,
"edit_similarity": 0.6188,
"accepted_prefix_tokens": 2.7891,
"syntax_valid": 0.9902,
"mongo_valid": 0.962,
"exec_valid": 0.7859,
"ceiling_exec_valid": 0.9307,
"exec_over_ceiling": 0.8444,
"exec_given_ref_ok": 0.8437,
"n_ref_ok": 5612,
"nonexistent_fields_rate": 0.1007,
"empty_rate": 0.0017,
"normalized_match_ci95": [
0.3576,
0.382
],
"exec_over_ceiling_ci95": [
0.8342,
0.854
]
},
"by_context_kind": {
"full": {
"n": 3720,
"exact_match": 0.3151,
"normalized_match": 0.3153,
"edit_similarity": 0.5746,
"accepted_prefix_tokens": 2.7946,
"syntax_valid": 0.9892,
"mongo_valid": 0.9613,
"exec_valid": 0.7879,
"ceiling_exec_valid": 0.9237,
"exec_over_ceiling": 0.853,
"exec_given_ref_ok": 0.8519,
"n_ref_ok": 3436,
"nonexistent_fields_rate": 0.0919,
"empty_rate": 0.0022
},
"irrelevant": {
"n": 1155,
"exact_match": 0.4511,
"normalized_match": 0.4545,
"edit_similarity": 0.689,
"accepted_prefix_tokens": 2.7307,
"syntax_valid": 0.9913,
"mongo_valid": 0.9645,
"exec_valid": 0.7931,
"ceiling_exec_valid": 0.9455,
"exec_over_ceiling": 0.8388,
"exec_given_ref_ok": 0.8388,
"n_ref_ok": 1092,
"nonexistent_fields_rate": 0.1134,
"empty_rate": 0.0
},
"none": {
"n": 1155,
"exact_match": 0.4589,
"normalized_match": 0.4597,
"edit_similarity": 0.6908,
"accepted_prefix_tokens": 2.8294,
"syntax_valid": 0.9922,
"mongo_valid": 0.9619,
"exec_valid": 0.7723,
"ceiling_exec_valid": 0.9385,
"exec_over_ceiling": 0.8229,
"exec_given_ref_ok": 0.8229,
"n_ref_ok": 1084,
"nonexistent_fields_rate": 0.116,
"empty_rate": 0.0017
}
},
"by_cut_kind": {
"key": {
"n": 959,
"exact_match": 0.3566,
"normalized_match": 0.3566,
"edit_similarity": 0.5737,
"accepted_prefix_tokens": 1.8749,
"syntax_valid": 0.9958,
"mongo_valid": 0.9698,
"exec_valid": 0.7737,
"ceiling_exec_valid": 0.9166,
"exec_over_ceiling": 0.8441,
"exec_given_ref_ok": 0.8441,
"n_ref_ok": 879,
"nonexistent_fields_rate": 0.1157,
"empty_rate": 0.0
},
"member": {
"n": 600,
"exact_match": 0.785,
"normalized_match": 0.785,
"edit_similarity": 0.8609,
"accepted_prefix_tokens": 1.95,
"syntax_valid": 0.9917,
"mongo_valid": 0.9567,
"exec_valid": 0.8233,
"ceiling_exec_valid": 0.89,
"exec_over_ceiling": 0.9251,
"exec_given_ref_ok": 0.9251,
"n_ref_ok": 534,
"nonexistent_fields_rate": 0.0417,
"empty_rate": 0.0
},
"multiline": {
"n": 480,
"exact_match": 0.0229,
"normalized_match": 0.0271,
"edit_similarity": 0.4367,
"accepted_prefix_tokens": 4.6229,
"syntax_valid": 0.9917,
"mongo_valid": 0.9521,
"exec_valid": 0.65,
"ceiling_exec_valid": 0.9583,
"exec_over_ceiling": 0.6783,
"exec_given_ref_ok": 0.6783,
"n_ref_ok": 460,
"nonexistent_fields_rate": 0.25,
"empty_rate": 0.0
},
"operator": {
"n": 841,
"exact_match": 0.4554,
"normalized_match": 0.4554,
"edit_similarity": 0.6871,
"accepted_prefix_tokens": 2.9382,
"syntax_valid": 0.9881,
"mongo_valid": 0.9584,
"exec_valid": 0.8502,
"ceiling_exec_valid": 0.9489,
"exec_over_ceiling": 0.896,
"exec_given_ref_ok": 0.896,
"n_ref_ok": 798,
"nonexistent_fields_rate": 0.0606,
"empty_rate": 0.0
},
"partial": {
"n": 600,
"exact_match": 0.6083,
"normalized_match": 0.6083,
"edit_similarity": 0.7702,
"accepted_prefix_tokens": 2.38,
"syntax_valid": 0.9933,
"mongo_valid": 0.9633,
"exec_valid": 0.735,
"ceiling_exec_valid": 0.91,
"exec_over_ceiling": 0.8077,
"exec_given_ref_ok": 0.8077,
"n_ref_ok": 546,
"nonexistent_fields_rate": 0.1333,
"empty_rate": 0.0017
},
"rest": {
"n": 720,
"exact_match": 0.2875,
"normalized_match": 0.2875,
"edit_similarity": 0.5632,
"accepted_prefix_tokens": 2.6264,
"syntax_valid": 0.9792,
"mongo_valid": 0.9458,
"exec_valid": 0.7556,
"ceiling_exec_valid": 0.9319,
"exec_over_ceiling": 0.8108,
"exec_given_ref_ok": 0.8107,
"n_ref_ok": 671,
"nonexistent_fields_rate": 0.0764,
"empty_rate": 0.0097
},
"stage": {
"n": 720,
"exact_match": 0.0347,
"normalized_match": 0.0403,
"edit_similarity": 0.4278,
"accepted_prefix_tokens": 4.6514,
"syntax_valid": 0.9875,
"mongo_valid": 0.9778,
"exec_valid": 0.7264,
"ceiling_exec_valid": 0.9486,
"exec_over_ceiling": 0.7658,
"exec_given_ref_ok": 0.7657,
"n_ref_ok": 683,
"nonexistent_fields_rate": 0.1583,
"empty_rate": 0.0028
},
"stmt_start": {
"n": 269,
"exact_match": 0.2379,
"normalized_match": 0.2379,
"edit_similarity": 0.5815,
"accepted_prefix_tokens": 3.7212,
"syntax_valid": 0.9777,
"mongo_valid": 0.9368,
"exec_valid": 0.7732,
"ceiling_exec_valid": 0.9777,
"exec_over_ceiling": 0.7908,
"exec_given_ref_ok": 0.7757,
"n_ref_ok": 263,
"nonexistent_fields_rate": 0.1561,
"empty_rate": 0.0
},
"value": {
"n": 841,
"exact_match": 0.4221,
"normalized_match": 0.4221,
"edit_similarity": 0.6483,
"accepted_prefix_tokens": 1.7729,
"syntax_valid": 0.9976,
"mongo_valid": 0.9738,
"exec_valid": 0.9037,
"ceiling_exec_valid": 0.9251,
"exec_over_ceiling": 0.9769,
"exec_given_ref_ok": 0.9769,
"n_ref_ok": 778,
"nonexistent_fields_rate": 0.0107,
"empty_rate": 0.0
}
},
"by_domain": {
"adtech": {
"n": 3057,
"exact_match": 0.3749,
"normalized_match": 0.3759,
"edit_similarity": 0.6244,
"accepted_prefix_tokens": 2.8911,
"syntax_valid": 0.9908,
"mongo_valid": 0.9627,
"exec_valid": 0.7975,
"ceiling_exec_valid": 0.9303,
"exec_over_ceiling": 0.8573,
"exec_given_ref_ok": 0.8558,
"n_ref_ok": 2844,
"nonexistent_fields_rate": 0.0932,
"empty_rate": 0.002
},
"telecom": {
"n": 2973,
"exact_match": 0.3623,
"normalized_match": 0.3633,
"edit_similarity": 0.613,
"accepted_prefix_tokens": 2.6842,
"syntax_valid": 0.9896,
"mongo_valid": 0.9613,
"exec_valid": 0.774,
"ceiling_exec_valid": 0.931,
"exec_over_ceiling": 0.8314,
"exec_given_ref_ok": 0.8313,
"n_ref_ok": 2768,
"nonexistent_fields_rate": 0.1083,
"empty_rate": 0.0013
}
},
"by_author": {
"grammar": {
"n": 3220,
"exact_match": 0.3758,
"normalized_match": 0.3761,
"edit_similarity": 0.633,
"accepted_prefix_tokens": 3.027,
"syntax_valid": 0.9975,
"mongo_valid": 0.9969,
"exec_valid": 0.7898,
"ceiling_exec_valid": 0.8702,
"exec_over_ceiling": 0.9076,
"exec_given_ref_ok": 0.9061,
"n_ref_ok": 2802,
"nonexistent_fields_rate": 0.0724,
"empty_rate": 0.0
},
"llm": {
"n": 2810,
"exact_match": 0.3605,
"normalized_match": 0.3623,
"edit_similarity": 0.6025,
"accepted_prefix_tokens": 2.5164,
"syntax_valid": 0.9819,
"mongo_valid": 0.9221,
"exec_valid": 0.7815,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.7815,
"exec_given_ref_ok": 0.7815,
"n_ref_ok": 2810,
"nonexistent_fields_rate": 0.1331,
"empty_rate": 0.0036
}
},
"by_author_context": {
"grammar/full": {
"n": 2121,
"exact_match": 0.3215,
"normalized_match": 0.322,
"edit_similarity": 0.5933,
"accepted_prefix_tokens": 3.1061,
"syntax_valid": 0.9972,
"mongo_valid": 0.9972,
"exec_valid": 0.7845,
"ceiling_exec_valid": 0.8661,
"exec_over_ceiling": 0.9058,
"exec_given_ref_ok": 0.9036,
"n_ref_ok": 1837,
"nonexistent_fields_rate": 0.0726,
"empty_rate": 0.0
},
"grammar/irrelevant": {
"n": 553,
"exact_match": 0.4629,
"normalized_match": 0.4629,
"edit_similarity": 0.6965,
"accepted_prefix_tokens": 2.8101,
"syntax_valid": 0.9982,
"mongo_valid": 0.9964,
"exec_valid": 0.8065,
"ceiling_exec_valid": 0.8861,
"exec_over_ceiling": 0.9102,
"exec_given_ref_ok": 0.9102,
"n_ref_ok": 490,
"nonexistent_fields_rate": 0.0687,
"empty_rate": 0.0
},
"grammar/none": {
"n": 546,
"exact_match": 0.4982,
"normalized_match": 0.4982,
"edit_similarity": 0.7231,
"accepted_prefix_tokens": 2.9396,
"syntax_valid": 0.9982,
"mongo_valid": 0.9963,
"exec_valid": 0.793,
"ceiling_exec_valid": 0.87,
"exec_over_ceiling": 0.9115,
"exec_given_ref_ok": 0.9116,
"n_ref_ok": 475,
"nonexistent_fields_rate": 0.0751,
"empty_rate": 0.0
},
"llm/full": {
"n": 1599,
"exact_match": 0.3064,
"normalized_match": 0.3064,
"edit_similarity": 0.5499,
"accepted_prefix_tokens": 2.3815,
"syntax_valid": 0.9787,
"mongo_valid": 0.9137,
"exec_valid": 0.7924,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.7924,
"exec_given_ref_ok": 0.7924,
"n_ref_ok": 1599,
"nonexistent_fields_rate": 0.1176,
"empty_rate": 0.005
},
"llm/irrelevant": {
"n": 602,
"exact_match": 0.4402,
"normalized_match": 0.4468,
"edit_similarity": 0.6822,
"accepted_prefix_tokens": 2.6578,
"syntax_valid": 0.985,
"mongo_valid": 0.9352,
"exec_valid": 0.7807,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.7807,
"exec_given_ref_ok": 0.7807,
"n_ref_ok": 602,
"nonexistent_fields_rate": 0.1545,
"empty_rate": 0.0
},
"llm/none": {
"n": 609,
"exact_match": 0.4236,
"normalized_match": 0.4253,
"edit_similarity": 0.6618,
"accepted_prefix_tokens": 2.7307,
"syntax_valid": 0.9869,
"mongo_valid": 0.931,
"exec_valid": 0.7537,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.7537,
"exec_given_ref_ok": 0.7537,
"n_ref_ok": 609,
"nonexistent_fields_rate": 0.1527,
"empty_rate": 0.0033
}
}
},
"t1_v3": {
"overall": {
"n": 6030,
"exact_match": 0.4516,
"normalized_match": 0.4537,
"edit_similarity": 0.7011,
"accepted_prefix_tokens": 3.8992,
"syntax_valid": 0.9972,
"mongo_valid": 0.9718,
"exec_valid": 0.8307,
"ceiling_exec_valid": 0.9307,
"exec_over_ceiling": 0.8926,
"exec_given_ref_ok": 0.892,
"n_ref_ok": 5612,
"nonexistent_fields_rate": 0.0788,
"empty_rate": 0.0,
"normalized_match_ci95": [
0.4412,
0.4663
],
"exec_over_ceiling_ci95": [
0.8839,
0.9007
]
},
"by_context_kind": {
"full": {
"n": 3720,
"exact_match": 0.3812,
"normalized_match": 0.3828,
"edit_similarity": 0.6471,
"accepted_prefix_tokens": 3.7911,
"syntax_valid": 0.9965,
"mongo_valid": 0.9704,
"exec_valid": 0.8204,
"ceiling_exec_valid": 0.9237,
"exec_over_ceiling": 0.8882,
"exec_given_ref_ok": 0.8874,
"n_ref_ok": 3436,
"nonexistent_fields_rate": 0.075,
"empty_rate": 0.0
},
"irrelevant": {
"n": 1155,
"exact_match": 0.5524,
"normalized_match": 0.555,
"edit_similarity": 0.782,
"accepted_prefix_tokens": 3.8952,
"syntax_valid": 0.9965,
"mongo_valid": 0.974,
"exec_valid": 0.8511,
"ceiling_exec_valid": 0.9455,
"exec_over_ceiling": 0.9002,
"exec_given_ref_ok": 0.9002,
"n_ref_ok": 1092,
"nonexistent_fields_rate": 0.0823,
"empty_rate": 0.0
},
"none": {
"n": 1155,
"exact_match": 0.5775,
"normalized_match": 0.581,
"edit_similarity": 0.7941,
"accepted_prefix_tokens": 4.2511,
"syntax_valid": 1.0,
"mongo_valid": 0.974,
"exec_valid": 0.8433,
"ceiling_exec_valid": 0.9385,
"exec_over_ceiling": 0.8986,
"exec_given_ref_ok": 0.8985,
"n_ref_ok": 1084,
"nonexistent_fields_rate": 0.0874,
"empty_rate": 0.0
}
},
"by_cut_kind": {
"key": {
"n": 959,
"exact_match": 0.439,
"normalized_match": 0.44,
"edit_similarity": 0.6624,
"accepted_prefix_tokens": 2.4703,
"syntax_valid": 0.999,
"mongo_valid": 0.975,
"exec_valid": 0.8373,
"ceiling_exec_valid": 0.9166,
"exec_over_ceiling": 0.9135,
"exec_given_ref_ok": 0.9135,
"n_ref_ok": 879,
"nonexistent_fields_rate": 0.0688,
"empty_rate": 0.0
},
"member": {
"n": 600,
"exact_match": 0.835,
"normalized_match": 0.835,
"edit_similarity": 0.895,
"accepted_prefix_tokens": 2.17,
"syntax_valid": 0.9983,
"mongo_valid": 0.9617,
"exec_valid": 0.8267,
"ceiling_exec_valid": 0.89,
"exec_over_ceiling": 0.9289,
"exec_given_ref_ok": 0.9288,
"n_ref_ok": 534,
"nonexistent_fields_rate": 0.04,
"empty_rate": 0.0
},
"multiline": {
"n": 480,
"exact_match": 0.1062,
"normalized_match": 0.1125,
"edit_similarity": 0.5706,
"accepted_prefix_tokens": 8.2354,
"syntax_valid": 0.9812,
"mongo_valid": 0.9646,
"exec_valid": 0.7021,
"ceiling_exec_valid": 0.9583,
"exec_over_ceiling": 0.7327,
"exec_given_ref_ok": 0.7326,
"n_ref_ok": 460,
"nonexistent_fields_rate": 0.1896,
"empty_rate": 0.0
},
"operator": {
"n": 841,
"exact_match": 0.5743,
"normalized_match": 0.5755,
"edit_similarity": 0.7736,
"accepted_prefix_tokens": 3.8359,
"syntax_valid": 1.0,
"mongo_valid": 0.9762,
"exec_valid": 0.8859,
"ceiling_exec_valid": 0.9489,
"exec_over_ceiling": 0.9336,
"exec_given_ref_ok": 0.9336,
"n_ref_ok": 798,
"nonexistent_fields_rate": 0.0523,
"empty_rate": 0.0
},
"partial": {
"n": 600,
"exact_match": 0.6567,
"normalized_match": 0.6567,
"edit_similarity": 0.8213,
"accepted_prefix_tokens": 2.8333,
"syntax_valid": 0.9983,
"mongo_valid": 0.9717,
"exec_valid": 0.7817,
"ceiling_exec_valid": 0.91,
"exec_over_ceiling": 0.859,
"exec_given_ref_ok": 0.859,
"n_ref_ok": 546,
"nonexistent_fields_rate": 0.1133,
"empty_rate": 0.0
},
"rest": {
"n": 720,
"exact_match": 0.4,
"normalized_match": 0.4,
"edit_similarity": 0.6704,
"accepted_prefix_tokens": 4.0847,
"syntax_valid": 1.0,
"mongo_valid": 0.9667,
"exec_valid": 0.8375,
"ceiling_exec_valid": 0.9319,
"exec_over_ceiling": 0.8987,
"exec_given_ref_ok": 0.8987,
"n_ref_ok": 671,
"nonexistent_fields_rate": 0.0681,
"empty_rate": 0.0
},
"stage": {
"n": 720,
"exact_match": 0.0958,
"normalized_match": 0.1056,
"edit_similarity": 0.5092,
"accepted_prefix_tokens": 6.6472,
"syntax_valid": 0.9958,
"mongo_valid": 0.9847,
"exec_valid": 0.7778,
"ceiling_exec_valid": 0.9486,
"exec_over_ceiling": 0.8199,
"exec_given_ref_ok": 0.8199,
"n_ref_ok": 683,
"nonexistent_fields_rate": 0.1389,
"empty_rate": 0.0
},
"stmt_start": {
"n": 269,
"exact_match": 0.368,
"normalized_match": 0.3717,
"edit_similarity": 0.7071,
"accepted_prefix_tokens": 4.8662,
"syntax_valid": 0.9963,
"mongo_valid": 0.9517,
"exec_valid": 0.8699,
"ceiling_exec_valid": 0.9777,
"exec_over_ceiling": 0.8897,
"exec_given_ref_ok": 0.8783,
"n_ref_ok": 263,
"nonexistent_fields_rate": 0.0892,
"empty_rate": 0.0
},
"value": {
"n": 841,
"exact_match": 0.4958,
"normalized_match": 0.4958,
"edit_similarity": 0.7118,
"accepted_prefix_tokens": 2.2901,
"syntax_valid": 0.9988,
"mongo_valid": 0.975,
"exec_valid": 0.9061,
"ceiling_exec_valid": 0.9251,
"exec_over_ceiling": 0.9795,
"exec_given_ref_ok": 0.9794,
"n_ref_ok": 778,
"nonexistent_fields_rate": 0.0107,
"empty_rate": 0.0
}
},
"by_domain": {
"adtech": {
"n": 3057,
"exact_match": 0.4544,
"normalized_match": 0.4576,
"edit_similarity": 0.7054,
"accepted_prefix_tokens": 4.0203,
"syntax_valid": 0.9974,
"mongo_valid": 0.9702,
"exec_valid": 0.8328,
"ceiling_exec_valid": 0.9303,
"exec_over_ceiling": 0.8952,
"exec_given_ref_ok": 0.8942,
"n_ref_ok": 2844,
"nonexistent_fields_rate": 0.0772,
"empty_rate": 0.0
},
"telecom": {
"n": 2973,
"exact_match": 0.4487,
"normalized_match": 0.4497,
"edit_similarity": 0.6967,
"accepted_prefix_tokens": 3.7746,
"syntax_valid": 0.997,
"mongo_valid": 0.9734,
"exec_valid": 0.8285,
"ceiling_exec_valid": 0.931,
"exec_over_ceiling": 0.8899,
"exec_given_ref_ok": 0.8898,
"n_ref_ok": 2768,
"nonexistent_fields_rate": 0.0804,
"empty_rate": 0.0
}
},
"by_author": {
"grammar": {
"n": 3220,
"exact_match": 0.3835,
"normalized_match": 0.3835,
"edit_similarity": 0.641,
"accepted_prefix_tokens": 3.0596,
"syntax_valid": 0.9988,
"mongo_valid": 0.9981,
"exec_valid": 0.8,
"ceiling_exec_valid": 0.8702,
"exec_over_ceiling": 0.9193,
"exec_given_ref_ok": 0.9183,
"n_ref_ok": 2802,
"nonexistent_fields_rate": 0.0528,
"empty_rate": 0.0
},
"llm": {
"n": 2810,
"exact_match": 0.5295,
"normalized_match": 0.5342,
"edit_similarity": 0.77,
"accepted_prefix_tokens": 4.8612,
"syntax_valid": 0.9954,
"mongo_valid": 0.9416,
"exec_valid": 0.8658,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.8658,
"exec_given_ref_ok": 0.8658,
"n_ref_ok": 2810,
"nonexistent_fields_rate": 0.1085,
"empty_rate": 0.0
}
},
"by_author_context": {
"grammar/full": {
"n": 2121,
"exact_match": 0.3234,
"normalized_match": 0.3234,
"edit_similarity": 0.5932,
"accepted_prefix_tokens": 3.075,
"syntax_valid": 0.9986,
"mongo_valid": 0.9976,
"exec_valid": 0.7888,
"ceiling_exec_valid": 0.8661,
"exec_over_ceiling": 0.9107,
"exec_given_ref_ok": 0.9091,
"n_ref_ok": 1837,
"nonexistent_fields_rate": 0.0528,
"empty_rate": 0.0
},
"grammar/irrelevant": {
"n": 553,
"exact_match": 0.4792,
"normalized_match": 0.4792,
"edit_similarity": 0.7223,
"accepted_prefix_tokens": 3.0362,
"syntax_valid": 0.9982,
"mongo_valid": 0.9982,
"exec_valid": 0.83,
"ceiling_exec_valid": 0.8861,
"exec_over_ceiling": 0.9367,
"exec_given_ref_ok": 0.9367,
"n_ref_ok": 490,
"nonexistent_fields_rate": 0.0488,
"empty_rate": 0.0
},
"grammar/none": {
"n": 546,
"exact_match": 0.5201,
"normalized_match": 0.5201,
"edit_similarity": 0.7445,
"accepted_prefix_tokens": 3.0238,
"syntax_valid": 1.0,
"mongo_valid": 1.0,
"exec_valid": 0.8132,
"ceiling_exec_valid": 0.87,
"exec_over_ceiling": 0.9347,
"exec_given_ref_ok": 0.9347,
"n_ref_ok": 475,
"nonexistent_fields_rate": 0.0568,
"empty_rate": 0.0
},
"llm/full": {
"n": 1599,
"exact_match": 0.4578,
"normalized_match": 0.4615,
"edit_similarity": 0.7186,
"accepted_prefix_tokens": 4.7411,
"syntax_valid": 0.9937,
"mongo_valid": 0.9343,
"exec_valid": 0.8624,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.8624,
"exec_given_ref_ok": 0.8624,
"n_ref_ok": 1599,
"nonexistent_fields_rate": 0.1044,
"empty_rate": 0.0
},
"llm/irrelevant": {
"n": 602,
"exact_match": 0.6196,
"normalized_match": 0.6246,
"edit_similarity": 0.8369,
"accepted_prefix_tokens": 4.6844,
"syntax_valid": 0.995,
"mongo_valid": 0.9518,
"exec_valid": 0.8704,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.8704,
"exec_given_ref_ok": 0.8704,
"n_ref_ok": 602,
"nonexistent_fields_rate": 0.113,
"empty_rate": 0.0
},
"llm/none": {
"n": 609,
"exact_match": 0.6289,
"normalized_match": 0.6355,
"edit_similarity": 0.8386,
"accepted_prefix_tokens": 5.3514,
"syntax_valid": 1.0,
"mongo_valid": 0.9507,
"exec_valid": 0.8703,
"ceiling_exec_valid": 1.0,
"exec_over_ceiling": 0.8703,
"exec_given_ref_ok": 0.8703,
"n_ref_ok": 609,
"nonexistent_fields_rate": 0.1149,
"empty_rate": 0.0
}
}
}
},
"val_v2c": {
"base_1.5b": {
"n": 2000,
"normalized_match": 0.142,
"exact_match": 0.142,
"edit_similarity": 0.3057,
"syntax_valid": 0.667,
"mongo_valid": 0.595,
"exec_valid": 0.316,
"ceiling_exec_valid": 0.6255,
"exec_over_ceiling": 0.5052,
"nonexistent_fields_rate": 0.051,
"nonexistent_fields_full": 0.0376,
"by_context_kind": {
"full": {
"normalized_match": 0.1475,
"exec_valid": 0.3364
},
"irrelevant": {
"normalized_match": 0.1061,
"exec_valid": 0.2172
},
"none": {
"normalized_match": 0.129,
"exec_valid": 0.2258
}
}
},
"t1_distill": {
"n": 2000,
"normalized_match": 0.4105,
"exact_match": 0.4105,
"edit_similarity": 0.6568,
"syntax_valid": 0.9955,
"mongo_valid": 0.9945,
"exec_valid": 0.595,
"ceiling_exec_valid": 0.6255,
"exec_over_ceiling": 0.9512,
"nonexistent_fields_rate": 0.029,
"nonexistent_fields_full": 0.0158,
"by_context_kind": {
"full": {
"normalized_match": 0.4159,
"exec_valid": 0.6199
},
"irrelevant": {
"normalized_match": 0.4091,
"exec_valid": 0.5051
},
"none": {
"normalized_match": 0.3548,
"exec_valid": 0.4452
}
}
},
"t1_v3": {
"n": 2000,
"normalized_match": 0.41,
"exact_match": 0.41,
"edit_similarity": 0.6536,
"syntax_valid": 0.998,
"mongo_valid": 0.9975,
"exec_valid": 0.6015,
"ceiling_exec_valid": 0.6255,
"exec_over_ceiling": 0.9616,
"nonexistent_fields_rate": 0.021,
"nonexistent_fields_full": 0.0121,
"by_context_kind": {
"full": {
"normalized_match": 0.4183,
"exec_valid": 0.6211
},
"irrelevant": {
"normalized_match": 0.3838,
"exec_valid": 0.5303
},
"none": {
"normalized_match": 0.3548,
"exec_valid": 0.4839
}
}
}
},
"pairs": {
"t1_distill_vs_base_1.5b": {
"context_kind": {
"n_aligned": 6030,
"overall": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 2.2487562189054726,
"mean_b": 2.789054726368159,
"delta": 0.5402985074626866,
"ci_low": 0.4485820895522389,
"ci_high": 0.6321766169154228,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 0.37318669983416247,
"mean_b": 0.6188011608623549,
"delta": 0.24561446102819237,
"ci_low": 0.23526142454394694,
"ci_high": 0.2561746745439469,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.1855721393034826,
"rate_b": 0.3686567164179104,
"delta": 0.18308457711442785,
"better": 1307,
"worse": 203,
"equal": 4520,
"p": 1.4405600645418062e-197,
"ci_low": 0.17147180762852407,
"ci_high": 0.19502487562189055,
"superior": true,
"p_holm": 2.5930081161752513e-196
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.49154228855721394,
"rate_b": 0.7859038142620233,
"delta": 0.2943615257048093,
"better": 2104,
"worse": 329,
"equal": 3597,
"p": 9.56375726e-316,
"ci_low": 0.2800995024875622,
"ci_high": 0.3086235489220564,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.9127514525e-314
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6048092868988392,
"rate_b": 0.9620232172470978,
"delta": 0.3572139303482587,
"better": 2194,
"worse": 40,
"equal": 3796,
"p": 0.0,
"ci_low": 0.3449419568822554,
"ci_high": 0.369983416252073,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9182421227197347,
"rate_b": 0.8993366500829187,
"delta": -0.01890547263681592,
"better": 305,
"worse": 419,
"equal": 5306,
"p": 2.5795404069987098e-05,
"ci_low": -0.02769485903814262,
"ci_high": -0.010281923714759536,
"superior": false,
"p_holm": 0.00010318161627994839
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.18739635157545606,
"rate_b": 0.36965174129353234,
"delta": 0.18225538971807628,
"better": 1308,
"worse": 209,
"equal": 4513,
"p": 2.0345053924323034e-194,
"ci_low": 0.17064676616915422,
"ci_high": 0.19419568822553898,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 3.4586591671349157e-193
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6774461028192371,
"rate_b": 0.9902155887230514,
"delta": 0.31276948590381426,
"better": 1919,
"worse": 33,
"equal": 4078,
"p": 0.0,
"ci_low": 0.30082918739635156,
"ci_high": 0.3250414593698176,
"superior": true,
"p_holm": 0.0
}
],
"groups": {
"full": {
"n": 3720,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 3720,
"lower_is_better": false,
"mean_a": 2.150537634408602,
"mean_b": 2.7946236559139783,
"delta": 0.6440860215053763,
"ci_low": 0.5311760752688172,
"ci_high": 0.7588709677419355,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 3720,
"lower_is_better": false,
"mean_a": 0.3220927956989247,
"mean_b": 0.5746390860215054,
"delta": 0.25254629032258064,
"ci_low": 0.23996857459677418,
"ci_high": 0.26556272849462365,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.139247311827957,
"rate_b": 0.3150537634408602,
"delta": 0.17580645161290323,
"better": 735,
"worse": 81,
"equal": 2904,
"p": 1.0249568878429672e-132,
"ci_low": 0.1620967741935484,
"ci_high": 0.18978494623655914,
"superior": true,
"p_holm": 1.6399310205487475e-131
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.45,
"rate_b": 0.7879032258064517,
"delta": 0.3379032258064516,
"better": 1451,
"worse": 194,
"equal": 2075,
"p": 6.806754608883945e-238,
"ci_low": 0.3193548387096774,
"ci_high": 0.35618279569892475,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.2932833756879496e-236
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.5567204301075269,
"rate_b": 0.9612903225806452,
"delta": 0.40456989247311825,
"better": 1528,
"worse": 23,
"equal": 2169,
"p": 0.0,
"ci_low": 0.3881720430107527,
"ci_high": 0.4206989247311828,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.928494623655914,
"rate_b": 0.9080645161290323,
"delta": -0.02043010752688172,
"better": 174,
"worse": 250,
"equal": 3296,
"p": 0.00026101422945458636,
"ci_low": -0.03118279569892473,
"ci_high": -0.00967741935483871,
"superior": false,
"p_holm": 0.0007830426883637591
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.13978494623655913,
"rate_b": 0.3153225806451613,
"delta": 0.17553763440860215,
"better": 736,
"worse": 83,
"equal": 2901,
"p": 1.4039628267736683e-131,
"ci_low": 0.16182795698924732,
"ci_high": 0.18951612903225806,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 2.1059442401605025e-130
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.6231182795698925,
"rate_b": 0.989247311827957,
"delta": 0.36612903225806454,
"better": 1385,
"worse": 23,
"equal": 2312,
"p": 0.0,
"ci_low": 0.3497311827956989,
"ci_high": 0.381989247311828,
"superior": true,
"p_holm": 0.0
}
]
},
"irrelevant": {
"n": 1155,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 2.3445887445887448,
"mean_b": 2.7307359307359307,
"delta": 0.38614718614718613,
"ci_low": 0.1774891774891775,
"ci_high": 0.6043290043290044,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 0.4300482251082251,
"mean_b": 0.6890493506493506,
"delta": 0.25900112554112553,
"ci_low": 0.2331739329004329,
"ci_high": 0.2849483398268398,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.24415584415584415,
"rate_b": 0.4510822510822511,
"delta": 0.20692640692640693,
"better": 290,
"worse": 51,
"equal": 814,
"p": 1.0054457699912198e-41,
"ci_low": 0.17835497835497835,
"ci_high": 0.2354978354978355,
"superior": true,
"p_holm": 9.049011929920978e-41
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.5341991341991342,
"rate_b": 0.793073593073593,
"delta": 0.25887445887445887,
"better": 357,
"worse": 58,
"equal": 740,
"p": 1.2878161280010718e-53,
"ci_low": 0.22857142857142856,
"ci_high": 0.29004329004329005,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.2878161280010719e-52
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.6441558441558441,
"rate_b": 0.9645021645021645,
"delta": 0.3203463203463203,
"better": 378,
"worse": 8,
"equal": 769,
"p": 1.4728552360527823e-100,
"ci_low": 0.29264069264069265,
"ci_high": 0.34805194805194806,
"superior": true,
"p_holm": 2.0619973304738953e-99
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.9038961038961039,
"rate_b": 0.8865800865800866,
"delta": -0.017316017316017316,
"better": 64,
"worse": 84,
"equal": 1007,
"p": 0.11803924678172636,
"ci_low": -0.0380952380952381,
"ci_high": 0.0025974025974025974,
"superior": false,
"p_holm": 0.23607849356345273
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.24848484848484848,
"rate_b": 0.45454545454545453,
"delta": 0.20606060606060606,
"better": 290,
"worse": 52,
"equal": 813,
"p": 3.3200986933768434e-41,
"ci_low": 0.1774891774891775,
"ci_high": 0.23463203463203464,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 2.6560789547014747e-40
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.7238095238095238,
"rate_b": 0.9913419913419913,
"delta": 0.2675324675324675,
"better": 313,
"worse": 4,
"equal": 838,
"p": 3.1321784783752422e-87,
"ci_low": 0.24155844155844156,
"ci_high": 0.2935064935064935,
"superior": true,
"p_holm": 4.071832021887815e-86
}
]
},
"none": {
"n": 1155,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 2.469264069264069,
"mean_b": 2.8294372294372296,
"delta": 0.3601731601731602,
"ci_low": 0.16277056277056276,
"ci_high": 0.5627705627705628,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 0.48088735930735926,
"mean_b": 0.690789264069264,
"delta": 0.20990190476190476,
"ci_low": 0.18358663203463205,
"ci_high": 0.2359251753246753,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.2761904761904762,
"rate_b": 0.4588744588744589,
"delta": 0.18268398268398267,
"better": 282,
"worse": 71,
"equal": 802,
"p": 6.925330255223634e-31,
"ci_low": 0.1523809523809524,
"ci_high": 0.21298701298701297,
"superior": true,
"p_holm": 4.15519815313418e-30
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.5826839826839827,
"rate_b": 0.7722943722943723,
"delta": 0.18961038961038962,
"better": 296,
"worse": 77,
"equal": 782,
"p": 2.1825485902965193e-31,
"ci_low": 0.15844155844155844,
"ci_high": 0.22077922077922077,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.5277840132075635e-30
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.7203463203463204,
"rate_b": 0.9619047619047619,
"delta": 0.24155844155844156,
"better": 288,
"worse": 9,
"equal": 858,
"p": 3.5538904469071796e-73,
"ci_low": 0.2164285714285716,
"ci_high": 0.2675324675324675,
"superior": true,
"p_holm": 4.2646685362886157e-72
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.8995670995670996,
"rate_b": 0.883982683982684,
"delta": -0.015584415584415584,
"better": 67,
"worse": 85,
"equal": 1003,
"p": 0.16772199317919748,
"ci_low": -0.03636363636363636,
"ci_high": 0.004329004329004329,
"superior": false,
"p_holm": 0.23607849356345273
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.27965367965367965,
"rate_b": 0.4597402597402597,
"delta": 0.1800865800865801,
"better": 282,
"worse": 74,
"equal": 799,
"p": 1.0098733511899805e-29,
"ci_low": 0.1497835497835498,
"ci_high": 0.21038961038961038,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 5.049366755949903e-29
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.806060606060606,
"rate_b": 0.9922077922077922,
"delta": 0.18614718614718614,
"better": 221,
"worse": 6,
"equal": 928,
"p": 1.6941577103211408e-57,
"ci_low": 0.16363636363636364,
"ci_high": 0.20952380952380953,
"superior": true,
"p_holm": 1.863573481353255e-56
}
]
}
},
"groups_skipped": {}
},
"author": {
"n_aligned": 6030,
"overall": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 2.2487562189054726,
"mean_b": 2.789054726368159,
"delta": 0.5402985074626866,
"ci_low": 0.4485820895522389,
"ci_high": 0.6321766169154228,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 0.37318669983416247,
"mean_b": 0.6188011608623549,
"delta": 0.24561446102819237,
"ci_low": 0.23526142454394694,
"ci_high": 0.2561746745439469,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.1855721393034826,
"rate_b": 0.3686567164179104,
"delta": 0.18308457711442785,
"better": 1307,
"worse": 203,
"equal": 4520,
"p": 1.4405600645418062e-197,
"ci_low": 0.17147180762852407,
"ci_high": 0.19502487562189055,
"superior": true,
"p_holm": 1.4405600645418063e-196
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.49154228855721394,
"rate_b": 0.7859038142620233,
"delta": 0.2943615257048093,
"better": 2104,
"worse": 329,
"equal": 3597,
"p": 9.56375726e-316,
"ci_low": 0.2800995024875622,
"ci_high": 0.3086235489220564,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.3389260167e-314
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6048092868988392,
"rate_b": 0.9620232172470978,
"delta": 0.3572139303482587,
"better": 2194,
"worse": 40,
"equal": 3796,
"p": 0.0,
"ci_low": 0.3449419568822554,
"ci_high": 0.369983416252073,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9182421227197347,
"rate_b": 0.8993366500829187,
"delta": -0.01890547263681592,
"better": 305,
"worse": 419,
"equal": 5306,
"p": 2.5795404069987098e-05,
"ci_low": -0.02769485903814262,
"ci_high": -0.010281923714759536,
"superior": false,
"p_holm": 5.1590808139974196e-05
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.18739635157545606,
"rate_b": 0.36965174129353234,
"delta": 0.18225538971807628,
"better": 1308,
"worse": 209,
"equal": 4513,
"p": 2.0345053924323034e-194,
"ci_low": 0.17064676616915422,
"ci_high": 0.19419568822553898,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.831054853189073e-193
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6774461028192371,
"rate_b": 0.9902155887230514,
"delta": 0.31276948590381426,
"better": 1919,
"worse": 33,
"equal": 4078,
"p": 0.0,
"ci_low": 0.30082918739635156,
"ci_high": 0.3250414593698176,
"superior": true,
"p_holm": 0.0
}
],
"groups": {
"grammar": {
"n": 3220,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 3220,
"lower_is_better": false,
"mean_a": 1.8161490683229813,
"mean_b": 3.0270186335403726,
"delta": 1.2108695652173913,
"ci_low": 1.0956521739130434,
"ci_high": 1.3301242236024844,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 3220,
"lower_is_better": false,
"mean_a": 0.2908079192546584,
"mean_b": 0.6330357142857143,
"delta": 0.34222779503105594,
"ci_low": 0.3284316257763975,
"ci_high": 0.3557838066770186,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.12608695652173912,
"rate_b": 0.37577639751552794,
"delta": 0.2496894409937888,
"better": 839,
"worse": 35,
"equal": 2346,
"p": 7.216279331318949e-201,
"ci_low": 0.23385093167701862,
"ci_high": 0.265527950310559,
"superior": true,
"p_holm": 7.937907264450844e-200
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.3829192546583851,
"rate_b": 0.7897515527950311,
"delta": 0.40683229813664595,
"better": 1397,
"worse": 87,
"equal": 1736,
"p": 1.1874189546903685e-304,
"ci_low": 0.38819875776397517,
"ci_high": 0.4254658385093168,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.543644641097479e-303
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.5509316770186335,
"rate_b": 0.9968944099378882,
"delta": 0.44596273291925465,
"better": 1438,
"worse": 2,
"equal": 1780,
"p": 0.0,
"ci_low": 0.4282608695652174,
"ci_high": 0.46366459627329193,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.9239130434782609,
"rate_b": 0.927639751552795,
"delta": 0.0037267080745341614,
"better": 183,
"worse": 171,
"equal": 2866,
"p": 0.5588488229617014,
"ci_low": -0.007763975155279503,
"ci_high": 0.015217391304347827,
"superior": false,
"p_holm": 0.5588488229617014
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.12608695652173912,
"rate_b": 0.3760869565217391,
"delta": 0.25,
"better": 840,
"worse": 35,
"equal": 2345,
"p": 3.758285052644206e-201,
"ci_low": 0.2341614906832298,
"ci_high": 0.26583850931677017,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 4.509942063173047e-200
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.5894409937888199,
"rate_b": 0.9975155279503105,
"delta": 0.4080745341614907,
"better": 1315,
"worse": 1,
"equal": 1904,
"p": 0.0,
"ci_low": 0.3906832298136646,
"ci_high": 0.4254658385093168,
"superior": true,
"p_holm": 0.0
}
]
},
"llm": {
"n": 2810,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 2810,
"lower_is_better": false,
"mean_a": 2.7444839857651244,
"mean_b": 2.516370106761566,
"delta": -0.22811387900355873,
"ci_low": -0.3608540925266904,
"ci_high": -0.09217081850533808,
"superior": false
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 2810,
"lower_is_better": false,
"mean_a": 0.46758516014234874,
"mean_b": 0.6024896797153025,
"delta": 0.13490451957295374,
"ci_low": 0.11944684519572954,
"ci_high": 0.15020412010676157,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.2537366548042705,
"rate_b": 0.3604982206405694,
"delta": 0.10676156583629894,
"better": 468,
"worse": 168,
"equal": 2174,
"p": 1.1582932129215375e-33,
"ci_low": 0.08967971530249111,
"ci_high": 0.12384341637010676,
"superior": true,
"p_holm": 5.7914660646076875e-33
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.6160142348754448,
"rate_b": 0.7814946619217081,
"delta": 0.16548042704626334,
"better": 707,
"worse": 242,
"equal": 1861,
"p": 1.9060073309706465e-53,
"ci_low": 0.14483985765124555,
"ci_high": 0.18612099644128113,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.1436043985823879e-52
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.6665480427046263,
"rate_b": 0.9220640569395018,
"delta": 0.25551601423487547,
"better": 756,
"worse": 38,
"equal": 2016,
"p": 2.4517989932927237e-174,
"ci_low": 0.23843416370106763,
"ci_high": 0.27259786476868325,
"superior": true,
"p_holm": 1.961439194634179e-173
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.9117437722419929,
"rate_b": 0.8669039145907473,
"delta": -0.044839857651245554,
"better": 122,
"worse": 248,
"equal": 2440,
"p": 5.329604207306828e-11,
"ci_low": -0.058362989323843414,
"ci_high": -0.03131672597864769,
"superior": false,
"p_holm": 1.5988812621920485e-10
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.2576512455516014,
"rate_b": 0.3622775800711744,
"delta": 0.10462633451957296,
"better": 468,
"worse": 174,
"equal": 2168,
"p": 4.964410510210872e-32,
"ci_low": 0.08754448398576513,
"ci_high": 0.12206405693950179,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.985764204084349e-31
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.7782918149466193,
"rate_b": 0.9818505338078292,
"delta": 0.20355871886120996,
"better": 604,
"worse": 32,
"equal": 2174,
"p": 6.539367210840716e-138,
"ci_low": 0.18753558718861219,
"ci_high": 0.2195729537366548,
"superior": true,
"p_holm": 4.577557047588501e-137
}
]
}
},
"groups_skipped": {}
}
},
"t1_v3_vs_t1_distill": {
"context_kind": {
"n_aligned": 6030,
"overall": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 2.789054726368159,
"mean_b": 3.8991708126036486,
"delta": 1.1101160862354893,
"ci_low": 0.9971807628524046,
"ci_high": 1.2273673300165837,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 0.6188011608623549,
"mean_b": 0.7011165505804311,
"delta": 0.08231538971807628,
"ci_low": 0.07611636194029851,
"ci_high": 0.08865810199004974,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.3686567164179104,
"rate_b": 0.451575456053068,
"delta": 0.08291873963515754,
"better": 600,
"worse": 100,
"equal": 5330,
"p": 9.348895347113809e-88,
"ci_low": 0.07462686567164178,
"ci_high": 0.09137645107794362,
"superior": true,
"p_holm": 2.150245929836176e-86
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.7859038142620233,
"rate_b": 0.8306799336650083,
"delta": 0.04477611940298507,
"better": 442,
"worse": 172,
"equal": 5416,
"p": 2.414329124019748e-28,
"ci_low": 0.03681592039800995,
"ci_high": 0.052570480928689885,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 4.587225335637521e-27
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9620232172470978,
"rate_b": 0.9718076285240465,
"delta": 0.00978441127694859,
"better": 82,
"worse": 23,
"equal": 5925,
"p": 5.956502026439826e-09,
"ci_low": 0.006467661691542288,
"ci_high": 0.013101160862354893,
"superior": true,
"p_holm": 6.552152229083809e-08
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.8993366500829187,
"rate_b": 0.9212271973466003,
"delta": 0.021890547263681594,
"better": 296,
"worse": 164,
"equal": 5570,
"p": 7.782239199341524e-10,
"ci_low": 0.015091210613598673,
"ci_high": 0.028855721393034824,
"superior": true,
"p_holm": 9.338687039209828e-09
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.36965174129353234,
"rate_b": 0.4537313432835821,
"delta": 0.08407960199004975,
"better": 608,
"worse": 101,
"equal": 5321,
"p": 4.35933007640598e-89,
"ci_low": 0.075787728026534,
"ci_high": 0.09253731343283582,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.046239218337435e-87
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9902155887230514,
"rate_b": 0.9971807628524046,
"delta": 0.006965174129353234,
"better": 59,
"worse": 17,
"equal": 5954,
"p": 1.3968718113140024e-06,
"ci_low": 0.0041459369817578775,
"ci_high": 0.00978441127694859,
"superior": true,
"p_holm": 1.3968718113140024e-05
}
],
"groups": {
"full": {
"n": 3720,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 3720,
"lower_is_better": false,
"mean_a": 2.7946236559139783,
"mean_b": 3.7911290322580644,
"delta": 0.996505376344086,
"ci_low": 0.850268817204301,
"ci_high": 1.1508131720430108,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 3720,
"lower_is_better": false,
"mean_a": 0.5746390860215054,
"mean_b": 0.6471357795698924,
"delta": 0.0724966935483871,
"ci_low": 0.06445746639784947,
"ci_high": 0.08039365658602149,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.3150537634408602,
"rate_b": 0.3811827956989247,
"delta": 0.06612903225806452,
"better": 303,
"worse": 57,
"equal": 3360,
"p": 1.2206754770942516e-41,
"ci_low": 0.05618279569892473,
"ci_high": 0.0760752688172043,
"superior": true,
"p_holm": 2.5634185018979285e-40
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.7879032258064517,
"rate_b": 0.8204301075268817,
"delta": 0.032526881720430105,
"better": 252,
"worse": 131,
"equal": 3337,
"p": 6.314946543298559e-10,
"ci_low": 0.022311827956989248,
"ci_high": 0.04247311827956989,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 8.209430506288127e-09
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.9612903225806452,
"rate_b": 0.9704301075268817,
"delta": 0.00913978494623656,
"better": 52,
"worse": 18,
"equal": 3650,
"p": 5.849547318695842e-05,
"ci_low": 0.004838709677419355,
"ci_high": 0.013709677419354839,
"superior": true,
"p_holm": 0.0005264592586826258
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.9080645161290323,
"rate_b": 0.925,
"delta": 0.016935483870967744,
"better": 168,
"worse": 105,
"equal": 3447,
"p": 0.00016477217809691969,
"ci_low": 0.008064516129032258,
"ci_high": 0.025537634408602152,
"superior": true,
"p_holm": 0.0011534052466784378
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.3153225806451613,
"rate_b": 0.3827956989247312,
"delta": 0.0674731182795699,
"better": 309,
"worse": 58,
"equal": 3353,
"p": 1.678335502796201e-42,
"ci_low": 0.057526881720430106,
"ci_high": 0.07741935483870968,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 3.692338106151642e-41
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.989247311827957,
"rate_b": 0.996505376344086,
"delta": 0.007258064516129033,
"better": 40,
"worse": 13,
"equal": 3667,
"p": 0.0002685401188191605,
"ci_low": 0.0034946236559139786,
"ci_high": 0.01129032258064516,
"superior": true,
"p_holm": 0.0013427005940958026
}
]
},
"irrelevant": {
"n": 1155,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 2.7307359307359307,
"mean_b": 3.895238095238095,
"delta": 1.1645021645021645,
"ci_low": 0.9246753246753247,
"ci_high": 1.4130086580086578,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 0.6890493506493506,
"mean_b": 0.7820035497835498,
"delta": 0.09295419913419914,
"ci_low": 0.07836375541125543,
"ci_high": 0.10772298051948052,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.4510822510822511,
"rate_b": 0.5523809523809524,
"delta": 0.1012987012987013,
"better": 140,
"worse": 23,
"equal": 992,
"p": 1.1750354186024887e-21,
"ci_low": 0.08051948051948052,
"ci_high": 0.12294372294372294,
"superior": true,
"p_holm": 1.997560211624231e-20
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.793073593073593,
"rate_b": 0.8510822510822511,
"delta": 0.05800865800865801,
"better": 88,
"worse": 21,
"equal": 1046,
"p": 6.112691638091427e-11,
"ci_low": 0.04069264069264069,
"ci_high": 0.07532467532467532,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 8.557768293327998e-10
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.9645021645021645,
"rate_b": 0.974025974025974,
"delta": 0.009523809523809525,
"better": 16,
"worse": 5,
"equal": 1134,
"p": 0.02660369873046875,
"ci_low": 0.0017316017316017316,
"ci_high": 0.017316017316017316,
"superior": true,
"p_holm": 0.0532073974609375
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.8865800865800866,
"rate_b": 0.9177489177489178,
"delta": 0.03116883116883117,
"better": 64,
"worse": 28,
"equal": 1063,
"p": 0.00022250112957580387,
"ci_low": 0.015584415584415584,
"ci_high": 0.047619047619047616,
"superior": true,
"p_holm": 0.0013350067774548233
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.45454545454545453,
"rate_b": 0.554978354978355,
"delta": 0.10043290043290043,
"better": 140,
"worse": 24,
"equal": 991,
"p": 4.04851304031129e-21,
"ci_low": 0.07965367965367966,
"ci_high": 0.12207792207792208,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 6.477620864498064e-20
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.9913419913419913,
"rate_b": 0.9965367965367965,
"delta": 0.005194805194805195,
"better": 10,
"worse": 4,
"equal": 1141,
"p": 0.1795654296875,
"ci_low": -0.0008658008658008658,
"ci_high": 0.011255411255411256,
"superior": false,
"p_holm": 0.1795654296875
}
]
},
"none": {
"n": 1155,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 2.8294372294372296,
"mean_b": 4.251082251082251,
"delta": 1.4216450216450216,
"ci_low": 1.1523809523809523,
"ci_high": 1.7073593073593074,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 0.690789264069264,
"mean_b": 0.7940896969696969,
"delta": 0.10330043290043289,
"ci_low": 0.08851722943722945,
"ci_high": 0.11877723160173158,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.4588744588744589,
"rate_b": 0.5774891774891775,
"delta": 0.11861471861471862,
"better": 157,
"worse": 20,
"equal": 978,
"p": 1.4656075428029792e-27,
"ci_low": 0.09783549783549783,
"ci_high": 0.14025974025974025,
"superior": true,
"p_holm": 2.6380935770453625e-26
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.7722943722943723,
"rate_b": 0.8432900432900433,
"delta": 0.070995670995671,
"better": 102,
"worse": 20,
"equal": 1033,
"p": 1.9641987858977157e-14,
"ci_low": 0.05281385281385281,
"ci_high": 0.08917748917748917,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 2.946298178846574e-13
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.9619047619047619,
"rate_b": 0.974025974025974,
"delta": 0.012121212121212121,
"better": 14,
"worse": 0,
"equal": 1141,
"p": 0.0001220703125,
"ci_low": 0.006060606060606061,
"ci_high": 0.01904761904761905,
"superior": true,
"p_holm": 0.0009765625
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.883982683982684,
"rate_b": 0.9125541125541126,
"delta": 0.02857142857142857,
"better": 64,
"worse": 31,
"equal": 1060,
"p": 0.0009246444117719318,
"ci_low": 0.012121212121212121,
"ci_high": 0.045021645021645025,
"superior": true,
"p_holm": 0.0036985776470877274
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.4597402597402597,
"rate_b": 0.580952380952381,
"delta": 0.12121212121212122,
"better": 159,
"worse": 19,
"equal": 977,
"p": 1.0293288050049541e-28,
"ci_low": 0.10043290043290043,
"ci_high": 0.14285714285714285,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 2.0586576100099084e-27
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.9922077922077922,
"rate_b": 1.0,
"delta": 0.007792207792207792,
"better": 9,
"worse": 0,
"equal": 1146,
"p": 0.00390625,
"ci_low": 0.003463203463203463,
"ci_high": 0.012987012987012988,
"superior": true,
"p_holm": 0.01171875
}
]
}
},
"groups_skipped": {}
},
"author": {
"n_aligned": 6030,
"overall": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 2.789054726368159,
"mean_b": 3.8991708126036486,
"delta": 1.1101160862354893,
"ci_low": 0.9971807628524046,
"ci_high": 1.2273673300165837,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 0.6188011608623549,
"mean_b": 0.7011165505804311,
"delta": 0.08231538971807628,
"ci_low": 0.07611636194029851,
"ci_high": 0.08865810199004974,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.3686567164179104,
"rate_b": 0.451575456053068,
"delta": 0.08291873963515754,
"better": 600,
"worse": 100,
"equal": 5330,
"p": 9.348895347113809e-88,
"ci_low": 0.07462686567164178,
"ci_high": 0.09137645107794362,
"superior": true,
"p_holm": 1.4023343020670715e-86
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.7859038142620233,
"rate_b": 0.8306799336650083,
"delta": 0.04477611940298507,
"better": 442,
"worse": 172,
"equal": 5416,
"p": 2.414329124019748e-28,
"ci_low": 0.03681592039800995,
"ci_high": 0.052570480928689885,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 3.1386278612256722e-27
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9620232172470978,
"rate_b": 0.9718076285240465,
"delta": 0.00978441127694859,
"better": 82,
"worse": 23,
"equal": 5925,
"p": 5.956502026439826e-09,
"ci_low": 0.006467661691542288,
"ci_high": 0.013101160862354893,
"superior": true,
"p_holm": 5.956502026439826e-08
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.8993366500829187,
"rate_b": 0.9212271973466003,
"delta": 0.021890547263681594,
"better": 296,
"worse": 164,
"equal": 5570,
"p": 7.782239199341524e-10,
"ci_low": 0.015091210613598673,
"ci_high": 0.028855721393034824,
"superior": true,
"p_holm": 9.338687039209828e-09
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.36965174129353234,
"rate_b": 0.4537313432835821,
"delta": 0.08407960199004975,
"better": 608,
"worse": 101,
"equal": 5321,
"p": 4.35933007640598e-89,
"ci_low": 0.075787728026534,
"ci_high": 0.09253731343283582,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 6.974928122249568e-88
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9902155887230514,
"rate_b": 0.9971807628524046,
"delta": 0.006965174129353234,
"better": 59,
"worse": 17,
"equal": 5954,
"p": 1.3968718113140024e-06,
"ci_low": 0.0041459369817578775,
"ci_high": 0.00978441127694859,
"superior": true,
"p_holm": 1.1174974490512019e-05
}
],
"groups": {
"grammar": {
"n": 3220,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 3220,
"lower_is_better": false,
"mean_a": 3.0270186335403726,
"mean_b": 3.0596273291925464,
"delta": 0.03260869565217391,
"ci_low": -0.04503105590062112,
"ci_high": 0.11055900621118013,
"superior": false
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 3220,
"lower_is_better": false,
"mean_a": 0.6330357142857143,
"mean_b": 0.6410397204968944,
"delta": 0.008004006211180125,
"ci_low": 0.0015777158385093183,
"ci_high": 0.014488523291925461,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.37577639751552794,
"rate_b": 0.38354037267080743,
"delta": 0.007763975155279503,
"better": 95,
"worse": 70,
"equal": 3055,
"p": 0.06137832710948886,
"ci_low": 0.0,
"ci_high": 0.015838509316770187,
"superior": false,
"p_holm": 0.24551330843795544
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.7897515527950311,
"rate_b": 0.8,
"delta": 0.010248447204968944,
"better": 112,
"worse": 79,
"equal": 3029,
"p": 0.020347077476115385,
"ci_low": 0.0021661490683230504,
"ci_high": 0.018944099378881987,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 0.10173538738057693
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.9968944099378882,
"rate_b": 0.9981366459627329,
"delta": 0.0012422360248447205,
"better": 10,
"worse": 6,
"equal": 3204,
"p": 0.454498291015625,
"ci_low": -0.0012422360248447205,
"ci_high": 0.0037267080745341614,
"superior": false,
"p_holm": 0.775390625
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.927639751552795,
"rate_b": 0.9472049689440993,
"delta": 0.01956521739130435,
"better": 115,
"worse": 52,
"equal": 3053,
"p": 1.2145886270456976e-06,
"ci_low": 0.011490683229813664,
"ci_high": 0.02732919254658385,
"superior": true,
"p_holm": 1.0931297643411278e-05
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.3760869565217391,
"rate_b": 0.38354037267080743,
"delta": 0.007453416149068323,
"better": 95,
"worse": 71,
"equal": 3054,
"p": 0.0739190926121619,
"ci_low": -0.00031055900621118014,
"ci_high": 0.015527950310559006,
"superior": false,
"margin": 0.02,
"noninferior": true,
"p_holm": 0.24551330843795544
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.9975155279503105,
"rate_b": 0.9987577639751553,
"delta": 0.0012422360248447205,
"better": 8,
"worse": 4,
"equal": 3208,
"p": 0.3876953125,
"ci_low": -0.0009316770186335404,
"ci_high": 0.0034161490683229812,
"superior": false,
"p_holm": 0.775390625
}
]
},
"llm": {
"n": 2810,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 2810,
"lower_is_better": false,
"mean_a": 2.516370106761566,
"mean_b": 4.8612099644128115,
"delta": 2.3448398576512455,
"ci_low": 2.126325622775801,
"ci_high": 2.5722419928825624,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 2810,
"lower_is_better": false,
"mean_a": 0.6024896797153025,
"mean_b": 0.7699590391459075,
"delta": 0.16746935943060498,
"ci_low": 0.15663636298932385,
"ci_high": 0.17816107740213522,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.3604982206405694,
"rate_b": 0.5295373665480427,
"delta": 0.16903914590747332,
"better": 505,
"worse": 30,
"equal": 2275,
"p": 2.2061014161757547e-112,
"ci_low": 0.15409252669039145,
"ci_high": 0.18398576512455517,
"superior": true,
"p_holm": 3.750372407498783e-111
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.7814946619217081,
"rate_b": 0.8658362989323843,
"delta": 0.08434163701067615,
"better": 330,
"worse": 93,
"equal": 2387,
"p": 3.4779869095107494e-32,
"ci_low": 0.0701067615658363,
"ci_high": 0.09822064056939502,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 4.869181673315049e-31
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.9220640569395018,
"rate_b": 0.9416370106761566,
"delta": 0.019572953736654804,
"better": 72,
"worse": 17,
"equal": 2721,
"p": 3.1727390190435945e-09,
"ci_low": 0.013167259786476869,
"ci_high": 0.02597864768683274,
"superior": true,
"p_holm": 3.490012920947954e-08
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.8669039145907473,
"rate_b": 0.891459074733096,
"delta": 0.024555160142348754,
"better": 181,
"worse": 112,
"equal": 2517,
"p": 6.617443552871609e-05,
"ci_low": 0.012455516014234875,
"ci_high": 0.036298932384341634,
"superior": true,
"p_holm": 0.00039704661317229654
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.3622775800711744,
"rate_b": 0.5341637010676157,
"delta": 0.1718861209964413,
"better": 513,
"worse": 30,
"equal": 2267,
"p": 1.3606411844347843e-114,
"ci_low": 0.15693950177935942,
"ci_high": 0.18683274021352314,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 2.4491541319826116e-113
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.9818505338078292,
"rate_b": 0.995373665480427,
"delta": 0.013523131672597865,
"better": 51,
"worse": 13,
"equal": 2746,
"p": 1.8809621564294015e-06,
"ci_low": 0.007829181494661922,
"ci_high": 0.019217081850533807,
"superior": true,
"p_holm": 1.316673509500581e-05
}
]
}
},
"groups_skipped": {}
}
},
"t1_v3_vs_base_1.5b": {
"context_kind": {
"n_aligned": 6030,
"overall": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 2.2487562189054726,
"mean_b": 3.8991708126036486,
"delta": 1.6504145936981758,
"ci_low": 1.5303482587064676,
"ci_high": 1.7753026533996683,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 0.37318669983416247,
"mean_b": 0.7011165505804311,
"delta": 0.3279298507462686,
"ci_low": 0.3182463428689884,
"ci_high": 0.3378565907960199,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.1855721393034826,
"rate_b": 0.451575456053068,
"delta": 0.2660033167495854,
"better": 1686,
"worse": 82,
"equal": 4262,
"p": 0.0,
"ci_low": 0.25422885572139303,
"ci_high": 0.2779436152570481,
"superior": true,
"p_holm": 0.0
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.49154228855721394,
"rate_b": 0.8306799336650083,
"delta": 0.33913764510779437,
"better": 2251,
"worse": 206,
"equal": 3573,
"p": 0.0,
"ci_low": 0.32553897180762853,
"ci_high": 0.3529021558872305,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 0.0
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6048092868988392,
"rate_b": 0.9718076285240465,
"delta": 0.3669983416252073,
"better": 2225,
"worse": 12,
"equal": 3793,
"p": 0.0,
"ci_low": 0.35505804311774464,
"ci_high": 0.37960199004975126,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9182421227197347,
"rate_b": 0.9212271973466003,
"delta": 0.0029850746268656717,
"better": 331,
"worse": 313,
"equal": 5386,
"p": 0.5029560522139299,
"ci_low": -0.005306799336650083,
"ci_high": 0.011111111111111112,
"superior": false,
"p_holm": 1.0
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.18739635157545606,
"rate_b": 0.4537313432835821,
"delta": 0.266334991708126,
"better": 1687,
"worse": 81,
"equal": 4262,
"p": 0.0,
"ci_low": 0.2543946932006633,
"ci_high": 0.2782794361525704,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 0.0
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6774461028192371,
"rate_b": 0.9971807628524046,
"delta": 0.3197346600331675,
"better": 1939,
"worse": 11,
"equal": 4080,
"p": 0.0,
"ci_low": 0.3081260364842454,
"ci_high": 0.3320066334991708,
"superior": true,
"p_holm": 0.0
}
],
"groups": {
"full": {
"n": 3720,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 3720,
"lower_is_better": false,
"mean_a": 2.150537634408602,
"mean_b": 3.7911290322580644,
"delta": 1.6405913978494624,
"ci_low": 1.4833333333333334,
"ci_high": 1.8016129032258064,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 3720,
"lower_is_better": false,
"mean_a": 0.3220927956989247,
"mean_b": 0.6471357795698924,
"delta": 0.32504298387096775,
"ci_low": 0.31256386626344085,
"ci_high": 0.33756841397849463,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.139247311827957,
"rate_b": 0.3811827956989247,
"delta": 0.24193548387096775,
"better": 938,
"worse": 38,
"equal": 2744,
"p": 1.1949364281529616e-225,
"ci_low": 0.22768817204301076,
"ci_high": 0.2564516129032258,
"superior": true,
"p_holm": 1.7924046422294423e-224
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.45,
"rate_b": 0.8204301075268817,
"delta": 0.3704301075268817,
"better": 1533,
"worse": 155,
"equal": 2032,
"p": 4.001190348593105e-285,
"ci_low": 0.3521505376344086,
"ci_high": 0.38844086021505375,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 6.802023592608279e-284
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.5567204301075269,
"rate_b": 0.9704301075268817,
"delta": 0.4137096774193548,
"better": 1545,
"worse": 6,
"equal": 2169,
"p": 0.0,
"ci_low": 0.39731182795698927,
"ci_high": 0.42983870967741933,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.928494623655914,
"rate_b": 0.925,
"delta": -0.0034946236559139786,
"better": 194,
"worse": 207,
"equal": 3319,
"p": 0.5490587682630458,
"ci_low": -0.013978494623655914,
"ci_high": 0.006989247311827957,
"superior": false,
"p_holm": 1.0
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.13978494623655913,
"rate_b": 0.3827956989247312,
"delta": 0.24301075268817204,
"better": 942,
"worse": 38,
"equal": 2740,
"p": 8.748941920297441e-227,
"ci_low": 0.22849462365591397,
"ci_high": 0.2575268817204301,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.3998307072475906e-225
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 3720,
"lower_is_better": false,
"rate_a": 0.6231182795698925,
"rate_b": 0.996505376344086,
"delta": 0.37338709677419357,
"better": 1396,
"worse": 7,
"equal": 2317,
"p": 0.0,
"ci_low": 0.357258064516129,
"ci_high": 0.38924731182795697,
"superior": true,
"p_holm": 0.0
}
]
},
"irrelevant": {
"n": 1155,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 2.3445887445887448,
"mean_b": 3.895238095238095,
"delta": 1.5506493506493506,
"ci_low": 1.3142857142857143,
"ci_high": 1.7948051948051948,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 0.4300482251082251,
"mean_b": 0.7820035497835498,
"delta": 0.35195532467532464,
"ci_low": 0.32793832034632037,
"ci_high": 0.3758575714285714,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.24415584415584415,
"rate_b": 0.5523809523809524,
"delta": 0.30822510822510824,
"better": 375,
"worse": 19,
"equal": 761,
"p": 5.694924653737641e-87,
"ci_low": 0.2804978354978357,
"ci_high": 0.3367965367965368,
"superior": true,
"p_holm": 6.83390958448517e-86
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.5341991341991342,
"rate_b": 0.8510822510822511,
"delta": 0.3168831168831169,
"better": 392,
"worse": 26,
"equal": 737,
"p": 5.0158096301156385e-85,
"ci_low": 0.2883116883116883,
"ci_high": 0.3463203463203463,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 4.514228667104075e-84
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.6441558441558441,
"rate_b": 0.974025974025974,
"delta": 0.32987012987012987,
"better": 386,
"worse": 5,
"equal": 764,
"p": 2.981856573542977e-107,
"ci_low": 0.30303030303030304,
"ci_high": 0.3575757575757576,
"superior": true,
"p_holm": 4.174599202960168e-106
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.9038961038961039,
"rate_b": 0.9177489177489178,
"delta": 0.013852813852813853,
"better": 74,
"worse": 58,
"equal": 1023,
"p": 0.19149877187347097,
"ci_low": -0.005194805194805195,
"ci_high": 0.033766233766233764,
"superior": false,
"p_holm": 0.7346286876070992
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.24848484848484848,
"rate_b": 0.554978354978355,
"delta": 0.3064935064935065,
"better": 373,
"worse": 19,
"equal": 763,
"p": 2.0638822530610425e-86,
"ci_low": 0.2779220779220779,
"ci_high": 0.33506493506493507,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 2.0638822530610427e-85
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.7238095238095238,
"rate_b": 0.9965367965367965,
"delta": 0.2727272727272727,
"better": 319,
"worse": 4,
"equal": 832,
"p": 5.2758083314590226e-89,
"ci_low": 0.24675324675324675,
"ci_high": 0.2987012987012987,
"superior": true,
"p_holm": 6.858550830896729e-88
}
]
},
"none": {
"n": 1155,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 2.469264069264069,
"mean_b": 4.251082251082251,
"delta": 1.7818181818181817,
"ci_low": 1.4987012987012986,
"ci_high": 2.084848484848485,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 1155,
"lower_is_better": false,
"mean_a": 0.48088735930735926,
"mean_b": 0.7940896969696969,
"delta": 0.3132023376623376,
"ci_low": 0.290192645021645,
"ci_high": 0.33632416017316014,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.2761904761904762,
"rate_b": 0.5774891774891775,
"delta": 0.3012987012987013,
"better": 373,
"worse": 25,
"equal": 757,
"p": 9.844313337086202e-81,
"ci_low": 0.2727272727272727,
"ci_high": 0.32987012987012987,
"superior": true,
"p_holm": 6.891019335960342e-80
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.5826839826839827,
"rate_b": 0.8432900432900433,
"delta": 0.2606060606060606,
"better": 326,
"worse": 25,
"equal": 804,
"p": 5.442582279000005e-68,
"ci_low": 0.2329004329004329,
"ci_high": 0.2891774891774892,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 3.265549367400003e-67
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.7203463203463204,
"rate_b": 0.974025974025974,
"delta": 0.25367965367965367,
"better": 294,
"worse": 1,
"equal": 860,
"p": 9.299786660660013e-87,
"ci_low": 0.22857142857142856,
"ci_high": 0.27965367965367965,
"superior": true,
"p_holm": 1.0229765326726015e-85
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.8995670995670996,
"rate_b": 0.9125541125541126,
"delta": 0.012987012987012988,
"better": 63,
"worse": 48,
"equal": 1044,
"p": 0.1836571719017748,
"ci_low": -0.004329004329004329,
"ci_high": 0.03116883116883117,
"superior": false,
"p_holm": 0.7346286876070992
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.27965367965367965,
"rate_b": 0.580952380952381,
"delta": 0.3012987012987013,
"better": 372,
"worse": 24,
"equal": 759,
"p": 2.3177249009102542e-81,
"ci_low": 0.2727272727272727,
"ci_high": 0.32987012987012987,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 1.8541799207282034e-80
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 1155,
"lower_is_better": false,
"rate_a": 0.806060606060606,
"rate_b": 1.0,
"delta": 0.19393939393939394,
"better": 224,
"worse": 0,
"equal": 931,
"p": 7.418412301374843e-68,
"ci_low": 0.17227272727272747,
"ci_high": 0.21731601731601732,
"superior": true,
"p_holm": 3.7092061506874214e-67
}
]
}
},
"groups_skipped": {}
},
"author": {
"n_aligned": 6030,
"overall": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 2.2487562189054726,
"mean_b": 3.8991708126036486,
"delta": 1.6504145936981758,
"ci_low": 1.5303482587064676,
"ci_high": 1.7753026533996683,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 6030,
"lower_is_better": false,
"mean_a": 0.37318669983416247,
"mean_b": 0.7011165505804311,
"delta": 0.3279298507462686,
"ci_low": 0.3182463428689884,
"ci_high": 0.3378565907960199,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.1855721393034826,
"rate_b": 0.451575456053068,
"delta": 0.2660033167495854,
"better": 1686,
"worse": 82,
"equal": 4262,
"p": 0.0,
"ci_low": 0.25422885572139303,
"ci_high": 0.2779436152570481,
"superior": true,
"p_holm": 0.0
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.49154228855721394,
"rate_b": 0.8306799336650083,
"delta": 0.33913764510779437,
"better": 2251,
"worse": 206,
"equal": 3573,
"p": 0.0,
"ci_low": 0.32553897180762853,
"ci_high": 0.3529021558872305,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 0.0
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6048092868988392,
"rate_b": 0.9718076285240465,
"delta": 0.3669983416252073,
"better": 2225,
"worse": 12,
"equal": 3793,
"p": 0.0,
"ci_low": 0.35505804311774464,
"ci_high": 0.37960199004975126,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.9182421227197347,
"rate_b": 0.9212271973466003,
"delta": 0.0029850746268656717,
"better": 331,
"worse": 313,
"equal": 5386,
"p": 0.5029560522139299,
"ci_low": -0.005306799336650083,
"ci_high": 0.011111111111111112,
"superior": false,
"p_holm": 0.5029560522139299
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.18739635157545606,
"rate_b": 0.4537313432835821,
"delta": 0.266334991708126,
"better": 1687,
"worse": 81,
"equal": 4262,
"p": 0.0,
"ci_low": 0.2543946932006633,
"ci_high": 0.2782794361525704,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 0.0
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 6030,
"lower_is_better": false,
"rate_a": 0.6774461028192371,
"rate_b": 0.9971807628524046,
"delta": 0.3197346600331675,
"better": 1939,
"worse": 11,
"equal": 4080,
"p": 0.0,
"ci_low": 0.3081260364842454,
"ci_high": 0.3320066334991708,
"superior": true,
"p_holm": 0.0
}
],
"groups": {
"grammar": {
"n": 3220,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 3220,
"lower_is_better": false,
"mean_a": 1.8161490683229813,
"mean_b": 3.0596273291925464,
"delta": 1.2434782608695651,
"ci_low": 1.1235947204968944,
"ci_high": 1.3670885093167702,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 3220,
"lower_is_better": false,
"mean_a": 0.2908079192546584,
"mean_b": 0.6410397204968944,
"delta": 0.350231801242236,
"ci_low": 0.3362704782608696,
"ci_high": 0.36421224689440995,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.12608695652173912,
"rate_b": 0.38354037267080743,
"delta": 0.2574534161490683,
"better": 864,
"worse": 35,
"equal": 2321,
"p": 5.876926180008631e-208,
"ci_low": 0.24161490683229814,
"ci_high": 0.2732919254658385,
"superior": true,
"p_holm": 5.2892335620077676e-207
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.3829192546583851,
"rate_b": 0.8,
"delta": 0.41708074534161493,
"better": 1421,
"worse": 78,
"equal": 1721,
"p": 7.1556e-320,
"ci_low": 0.3984472049689441,
"ci_high": 0.4360248447204969,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 7.8711e-319
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.5509316770186335,
"rate_b": 0.9981366459627329,
"delta": 0.4472049689440994,
"better": 1442,
"worse": 2,
"equal": 1776,
"p": 0.0,
"ci_low": 0.4295031055900621,
"ci_high": 0.4649068322981366,
"superior": true,
"p_holm": 0.0
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.9239130434782609,
"rate_b": 0.9472049689440993,
"delta": 0.023291925465838508,
"better": 196,
"worse": 121,
"equal": 2903,
"p": 2.9877855768950636e-05,
"ci_low": 0.012732919254658385,
"ci_high": 0.033850931677018636,
"superior": true,
"p_holm": 8.96335673068519e-05
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.12608695652173912,
"rate_b": 0.38354037267080743,
"delta": 0.2574534161490683,
"better": 864,
"worse": 35,
"equal": 2321,
"p": 5.876926180008631e-208,
"ci_low": 0.24161490683229814,
"ci_high": 0.2732919254658385,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 5.2892335620077676e-207
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 3220,
"lower_is_better": false,
"rate_a": 0.5894409937888199,
"rate_b": 0.9987577639751553,
"delta": 0.4093167701863354,
"better": 1320,
"worse": 2,
"equal": 1898,
"p": 0.0,
"ci_low": 0.3919254658385093,
"ci_high": 0.4267080745341615,
"superior": true,
"p_holm": 0.0
}
]
},
"llm": {
"n": 2810,
"metrics": [
{
"metric": "accepted_prefix_tokens",
"kind": "continuous",
"n": 2810,
"lower_is_better": false,
"mean_a": 2.7444839857651244,
"mean_b": 4.8612099644128115,
"delta": 2.1167259786476866,
"ci_low": 1.903202846975089,
"ci_high": 2.3387900355871887,
"superior": true
},
{
"metric": "edit_similarity",
"kind": "continuous",
"n": 2810,
"lower_is_better": false,
"mean_a": 0.46758516014234874,
"mean_b": 0.7699590391459075,
"delta": 0.3023738790035587,
"ci_low": 0.28847639679715303,
"ci_high": 0.31637307918149465,
"superior": true
},
{
"metric": "exact_match",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.2537366548042705,
"rate_b": 0.5295373665480427,
"delta": 0.27580071174377224,
"better": 822,
"worse": 47,
"equal": 1941,
"p": 7.990635342893872e-184,
"ci_low": 0.25836298932384344,
"ci_high": 0.29323843416370104,
"superior": true,
"p_holm": 4.794381205736324e-183
},
{
"metric": "exec_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.6160142348754448,
"rate_b": 0.8658362989323843,
"delta": 0.2498220640569395,
"better": 830,
"worse": 128,
"equal": 1852,
"p": 1.4303288564350014e-126,
"ci_low": 0.2302491103202847,
"ci_high": 0.2701067615658363,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 5.7213154257400056e-126
},
{
"metric": "mongo_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.6665480427046263,
"rate_b": 0.9416370106761566,
"delta": 0.27508896797153026,
"better": 783,
"worse": 10,
"equal": 2017,
"p": 9.954887411232013e-217,
"ci_low": 0.2580071174377224,
"ci_high": 0.29145907473309607,
"superior": true,
"p_holm": 9.954887411232013e-216
},
{
"metric": "no_nonexistent_field",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.9117437722419929,
"rate_b": 0.891459074733096,
"delta": -0.020284697508896797,
"better": 135,
"worse": 192,
"equal": 2483,
"p": 0.0019114402812567247,
"ci_low": -0.03309608540925267,
"ci_high": -0.007829181494661922,
"superior": false,
"p_holm": 0.0038228805625134493
},
{
"metric": "normalized_match",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.2576512455516014,
"rate_b": 0.5341637010676157,
"delta": 0.2765124555160142,
"better": 823,
"worse": 46,
"equal": 1941,
"p": 4.5571131842281796e-185,
"ci_low": 0.2587188612099644,
"ci_high": 0.29359430604982206,
"superior": true,
"margin": 0.02,
"noninferior": true,
"p_holm": 3.1899792289597256e-184
},
{
"metric": "syntax_valid",
"kind": "binary",
"n": 2810,
"lower_is_better": false,
"rate_a": 0.7782918149466193,
"rate_b": 0.995373665480427,
"delta": 0.21708185053380782,
"better": 619,
"worse": 9,
"equal": 2182,
"p": 7.201076790162522e-170,
"ci_low": 0.201423487544484,
"ci_high": 0.23274021352313168,
"superior": true,
"p_holm": 3.600538395081261e-169
}
]
}
},
"groups_skipped": {}
}
}
},
"status_distribution": {
"base_1.5b": {
"ok": 3404,
"syntax": 1442,
"error": 1027,
"no_exec": 26,
"empty": 85,
"static": 45,
"timeout": 1
},
"t1_distill": {
"ok": 5328,
"error": 520,
"static": 98,
"syntax": 49,
"no_exec": 25,
"empty": 10
},
"t1_v3": {
"ok": 5478,
"error": 453,
"static": 82,
"syntax": 16,
"timeout": 1
},
"_reference": {
"ok": 5612,
"error": 418
}
},
"readonly_proof": {
"base_1.5b": {
"dbHash_before_equals_after": true,
"dbs_checked": 28,
"proxy_write_commands_forwarded": {},
"maxTimeMS_injected": 3000
},
"t1_distill": {
"dbHash_before_equals_after": true,
"dbs_checked": 28,
"proxy_write_commands_forwarded": {},
"maxTimeMS_injected": 3000
},
"t1_v3": {
"dbHash_before_equals_after": true,
"dbs_checked": 28,
"proxy_write_commands_forwarded": {},
"maxTimeMS_injected": 3000
},
"_reference": {
"dbHash_before_equals_after": true,
"dbs_checked": 28,
"proxy_write_commands_forwarded": {},
"maxTimeMS_injected": 3000
}
},
"definitions": {
"normalized_match": "pred == middle com espacos normalizados",
"exec_valid": "prefix+pred+suffix executa (mongosh via proxy RO, status ok, comando de leitura enviado) sem campo/colecao inexistente no plano capturado e pred nao vazia",
"ceiling_exec_valid": "mesma validacao sobre prefix+middle+suffix",
"exec_over_ceiling": "exec_valid / ceiling_exec_valid (razao de medias)",
"nonexistent_fields_rate": "itens cujo plano capturado cita campo/colecao fora do schema, sobre todos os itens",
"json_dialect": "dialect json executa como db.getCollection(c).aggregate(EJSON.parse(texto))"
}
}