keppy's picture
lane encoder v1: ModernBERT-small-v2, 9 lanes; eval 19 real rows 63.2%; gonogo: opt-in (tie vs rules+llm); public training rows only
ae0b2f6 verified
Raw History Blame Contribute Delete
9.96 kB
{
"seed": 7,
"eval_frac": 0.3,
"min_real": 8,
"train_ids": [
"04c4dca5901141279bfcaf2ffa5c1320",
"0f236c8908a044f499695294b558d1e3",
"1d95083ed6734be88d9f3476dd7a6cd5",
"25aae6dee8164785826606d3897f0855",
"26297e7ef510437298063c071b038193",
"26297e7ef510437298063c071b038193#2",
"352d6017328448d5bc9bad8b2632dfd9",
"36896635fd224b8e8fbf2dbf843ef8a2",
"3fe27fd9b3ca41d882c2b9ca39f23b9a",
"53bac0c48c75427192bc8a93718f7440",
"54e90646a98e4e928b04ee6163fad855",
"617222d6cc814fbea2d35c012c17a853",
"621c6ca023c44c30b2086c36a2727985",
"6366775c3e4c462bb3095556bc9bf3a8",
"668056b8a5d94c61abb59ad7d3f0993f",
"7d0dfd15e2af42ca9d3b04fa0d71fd5b",
"7d47724d8236437785bf89fc04197a76",
"a4509525b10043d4917a2beecb2b6771",
"aaf9f1eb734e4a669ee6777e8172db84",
"aug:alignment-reasoning:100",
"aug:alignment-reasoning:101",
"aug:alignment-reasoning:102",
"aug:alignment-reasoning:103",
"aug:alignment-reasoning:104",
"aug:alignment-reasoning:105",
"aug:alignment-reasoning:106",
"aug:alignment-reasoning:107",
"aug:alignment-reasoning:108",
"aug:alignment-reasoning:109",
"aug:alignment-reasoning:110",
"aug:alignment-reasoning:111",
"aug:alignment-reasoning:112",
"aug:alignment-reasoning:113",
"aug:alignment-reasoning:114",
"aug:alignment-reasoning:115",
"aug:alignment-reasoning:116",
"aug:alignment-reasoning:117",
"aug:alignment-reasoning:98",
"aug:alignment-reasoning:99",
"aug:dl-ml-research-engineering:118",
"aug:dl-ml-research-engineering:119",
"aug:dl-ml-research-engineering:120",
"aug:dl-ml-research-engineering:121",
"aug:dl-ml-research-engineering:122",
"aug:dl-ml-research-engineering:123",
"aug:dl-ml-research-engineering:124",
"aug:dl-ml-research-engineering:125",
"aug:dl-ml-research-engineering:126",
"aug:dl-ml-research-engineering:127",
"aug:dl-ml-research-engineering:128",
"aug:dl-ml-research-engineering:129",
"aug:dl-ml-research-engineering:130",
"aug:dl-ml-research-engineering:131",
"aug:dl-ml-research-engineering:132",
"aug:dl-ml-research-engineering:133",
"aug:dl-ml-research-engineering:134",
"aug:hard-agentic-coding:1",
"aug:hard-agentic-coding:10",
"aug:hard-agentic-coding:11",
"aug:hard-agentic-coding:12",
"aug:hard-agentic-coding:13",
"aug:hard-agentic-coding:14",
"aug:hard-agentic-coding:15",
"aug:hard-agentic-coding:16",
"aug:hard-agentic-coding:17",
"aug:hard-agentic-coding:18",
"aug:hard-agentic-coding:19",
"aug:hard-agentic-coding:2",
"aug:hard-agentic-coding:20",
"aug:hard-agentic-coding:21",
"aug:hard-agentic-coding:22",
"aug:hard-agentic-coding:23",
"aug:hard-agentic-coding:24",
"aug:hard-agentic-coding:25",
"aug:hard-agentic-coding:26",
"aug:hard-agentic-coding:27",
"aug:hard-agentic-coding:28",
"aug:hard-agentic-coding:29",
"aug:hard-agentic-coding:3",
"aug:hard-agentic-coding:30",
"aug:hard-agentic-coding:4",
"aug:hard-agentic-coding:5",
"aug:hard-agentic-coding:6",
"aug:hard-agentic-coding:7",
"aug:hard-agentic-coding:8",
"aug:hard-agentic-coding:9",
"aug:long-doc-reading:135",
"aug:long-doc-reading:136",
"aug:long-doc-reading:137",
"aug:long-doc-reading:138",
"aug:long-doc-reading:139",
"aug:long-doc-reading:140",
"aug:long-doc-reading:141",
"aug:long-doc-reading:142",
"aug:long-doc-reading:143",
"aug:long-doc-reading:144",
"aug:long-doc-reading:31",
"aug:long-doc-reading:32",
"aug:long-doc-reading:33",
"aug:long-doc-reading:34",
"aug:long-doc-reading:35",
"aug:long-doc-reading:36",
"aug:long-doc-reading:37",
"aug:long-doc-reading:38",
"aug:long-doc-reading:39",
"aug:long-doc-reading:40",
"aug:math-first-principles:156",
"aug:math-first-principles:157",
"aug:math-first-principles:158",
"aug:math-first-principles:159",
"aug:math-first-principles:160",
"aug:math-first-principles:161",
"aug:math-first-principles:162",
"aug:math-first-principles:163",
"aug:math-first-principles:164",
"aug:math-first-principles:165",
"aug:math-first-principles:166",
"aug:math-first-principles:167",
"aug:math-first-principles:168",
"aug:math-first-principles:169",
"aug:math-first-principles:170",
"aug:math-first-principles:171",
"aug:math-first-principles:172",
"aug:math-first-principles:173",
"aug:math-first-principles:174",
"aug:math-first-principles:175",
"aug:orchestration:185",
"aug:orchestration:186",
"aug:orchestration:187",
"aug:orchestration:188",
"aug:orchestration:189",
"aug:orchestration:190",
"aug:orchestration:191",
"aug:orchestration:75",
"aug:orchestration:76",
"aug:orchestration:77",
"aug:orchestration:78",
"aug:orchestration:79",
"aug:orchestration:80",
"aug:orchestration:81",
"aug:orchestration:82",
"aug:orchestration:83",
"aug:orchestration:84",
"aug:orchestration:85",
"aug:orchestration:86",
"aug:orchestration:87",
"aug:orchestration:88",
"aug:orchestration:89",
"aug:orchestration:90",
"aug:orchestration:91",
"aug:orchestration:92",
"aug:orchestration:93",
"aug:orchestration:94",
"aug:orchestration:95",
"aug:orchestration:96",
"aug:orchestration:97",
"aug:prose:176",
"aug:prose:177",
"aug:prose:178",
"aug:prose:179",
"aug:prose:180",
"aug:prose:181",
"aug:prose:182",
"aug:prose:183",
"aug:prose:184",
"aug:prose:60",
"aug:prose:61",
"aug:prose:62",
"aug:prose:63",
"aug:prose:64",
"aug:prose:65",
"aug:prose:66",
"aug:prose:67",
"aug:prose:68",
"aug:prose:69",
"aug:prose:70",
"aug:prose:71",
"aug:prose:72",
"aug:prose:73",
"aug:prose:74",
"aug:web-research:145",
"aug:web-research:146",
"aug:web-research:147",
"aug:web-research:148",
"aug:web-research:149",
"aug:web-research:150",
"aug:web-research:151",
"aug:web-research:152",
"aug:web-research:153",
"aug:web-research:154",
"aug:web-research:155",
"aug:web-research:41",
"aug:web-research:42",
"aug:web-research:43",
"aug:web-research:44",
"aug:web-research:45",
"aug:web-research:46",
"aug:web-research:47",
"aug:web-research:48",
"aug:web-research:49",
"aug:web-research:50",
"aug:web-research:51",
"aug:web-research:52",
"aug:web-research:53",
"aug:web-research:54",
"aug:web-research:55",
"aug:web-research:56",
"aug:web-research:57",
"aug:web-research:58",
"aug:web-research:59",
"b5cd3815b87047bc9f185bc916d7e61b",
"bec22730d7684657830d14a85a99c311",
"bf6b4f88ce1b43c9b686d6bb1e6fe58e",
"c307a180a73644c7b14aa11e415081ff",
"c515f2fc15da422fb0e5fc704f4ef25b",
"cdd9e3b9e93c48178a6df7f043763d36",
"d550be61d4c34980805717b2b8aafd88",
"ed46e2ac8a3444d9897b98e2af1f7ad8",
"f6a94e3627f44af59ed55a2819f94d6a",
"fe2e145c7d9249f5b0ea2cabe2d131ff",
"seed:alignment-reasoning:hint",
"seed:alignment-reasoning:keywords",
"seed:dl-ml-research-engineering:hint",
"seed:dl-ml-research-engineering:keywords",
"seed:hard-agentic-coding:hint",
"seed:hard-agentic-coding:keywords",
"seed:long-doc-reading:hint",
"seed:long-doc-reading:keywords",
"seed:math-first-principles:hint",
"seed:math-first-principles:keywords",
"seed:orchestration:hint",
"seed:orchestration:keywords",
"seed:prose:hint",
"seed:prose:keywords",
"seed:routine-coding:hint",
"seed:routine-coding:keywords",
"seed:web-research:hint",
"seed:web-research:keywords",
"taskset:align-conjunctive-goals",
"taskset:align-goodhart-proxy",
"taskset:align-oversight-vs-capability",
"taskset:align-rm-optimum",
"taskset:align-rrm-vs-rlhf",
"taskset:align-scalable-oversight",
"taskset:align-sycophancy-rlhf",
"taskset:dl-cosine-warmup",
"taskset:dl-grpo-adv-norm",
"taskset:dl-mixed-precision",
"taskset:dl-sgd-momentum",
"taskset:dl-tokenizer-bpe",
"taskset:dl-topo-sort-dag",
"taskset:dl-weight-decay",
"taskset:rc-dict-flatten",
"taskset:rc-edit-script",
"taskset:rc-roman",
"taskset:rc-slugify",
"taskset:rc-template-render",
"taskset:rc-token-bucket"
],
"eval_ids": [
"1a84b311446b4571969af0f31c7ffd70",
"540704307436479487184f6b019ee4bd",
"6292a60a9a2241e08603b294ae393b3c",
"72c3a76e6eb04afdba944b8cbd8bf9b0",
"a9f795fd0bb5419fb634cd98cb9b07b2",
"e025c31944f446de9e69bb969adad8b1",
"e47fca02cf0542eab1b25ef8398b80c5",
"ece1863753294f4bbbaa36527cc662a5",
"fa85d94d4b0f4f59bb3d334dcc27004d",
"taskset:align-emergent-deception",
"taskset:align-interpretability-circuits",
"taskset:align-shard-theory",
"taskset:dl-early-stop-patience",
"taskset:dl-grad-accum",
"taskset:dl-train-loop-epochs",
"taskset:rc-backoff-schedule",
"taskset:rc-csv-group-count",
"taskset:rc-iso-dates",
"taskset:rc-log-parse"
],
"per_lane": {
"alignment-reasoning": {
"train": 29,
"eval": 3,
"real": 10,
"aug": 20,
"seed": 2
},
"dl-ml-research-engineering": {
"train": 28,
"eval": 4,
"real": 13,
"aug": 17,
"seed": 2
},
"hard-agentic-coding": {
"train": 32,
"eval": 0,
"real": 0,
"aug": 30,
"seed": 2
},
"long-doc-reading": {
"train": 22,
"eval": 0,
"real": 0,
"aug": 20,
"seed": 2
},
"math-first-principles": {
"train": 22,
"eval": 0,
"real": 0,
"aug": 20,
"seed": 2
},
"orchestration": {
"train": 32,
"eval": 0,
"real": 0,
"aug": 30,
"seed": 2
},
"prose": {
"train": 32,
"eval": 0,
"real": 6,
"aug": 24,
"seed": 2
},
"routine-coding": {
"train": 29,
"eval": 12,
"real": 39,
"aug": 0,
"seed": 2
},
"web-research": {
"train": 32,
"eval": 0,
"real": 0,
"aug": 30,
"seed": 2
}
},
"leaked_dropped": 0
}