| { |
| "format": "d1a-torch", |
| "format_version": 1, |
| "base": "google/gemma-4-E4B", |
| "base_revision": "411aa17b749aa952df1359d2dcea73917a544d9a", |
| "lora": 16, |
| "head_dim": 256, |
| "option_isolation": false, |
| "special_embeddings": false, |
| "weights_dtype": "bf16", |
| "temperature": 1.9543199368684914, |
| "holdout": [], |
| "weights": "lora", |
| "head": "head.safetensors", |
| "extra": { |
| "args": { |
| "base": "google/gemma-4-E4B", |
| "n_per_source": 1000, |
| "epochs": 1, |
| "lr": 2e-05, |
| "head_lr": 0.0, |
| "weight_decay": 0.01, |
| "lora": 16, |
| "accum": 4, |
| "holdout": "", |
| "suite": "evals/v7/decision-v7", |
| "train_sources": "", |
| "device": "mps", |
| "batch": 2, |
| "dtype": "fp32", |
| "weights_dtype": "bf16", |
| "checkpointing": 1, |
| "head_dim": 256, |
| "lora_targets": "all", |
| "base_revision": "411aa17b749aa952df1359d2dcea73917a544d9a", |
| "p_none": 0.1, |
| "p_none_distract": 0.12, |
| "p_distract": 0.15, |
| "p_none_pair": 0.0, |
| "none_pair_max_state": null, |
| "synthetic_repeat": 1, |
| "public_frac": 1.0, |
| "out": "d1", |
| "data": "train.jsonl", |
| "max_state": 5120, |
| "extra_suites": "", |
| "replay": 0, |
| "init_from": "JohnP1/d1a-e4b@v0.5", |
| "row_budget": 0, |
| "shared_prefix": 0, |
| "length_sort": 0, |
| "pass_tokens_max": 0, |
| "max_steps": 0, |
| "save_every_steps": 0, |
| "save_every_minutes": 30.0, |
| "resume": 1, |
| "stop_after": 0, |
| "seed": 0 |
| }, |
| "suite_sha256": "a8f50e481b7d90b97da049e0ff6a01cee2f1ed204aed61a8265af0edbb5514d2", |
| "init_source": { |
| "init_from": "JohnP1/d1a-e4b@v0.5", |
| "resolved": "JohnP1/d1a-e4b@e7db32f2254b6feb6ec830ae98662755dce39504", |
| "weights_sha256": "314242334d9483563d015b7dfb5971e774d03df075bdcb4da7f270cf2e356a1e", |
| "head_sha256": "4391749cd23185ea1dcd9039cd767aeaf3db915ff0db238ed485b4103bd9930c", |
| "tensors": 516 |
| }, |
| "temperature_fit": { |
| "rows": [ |
| "cal/rows.json", |
| "calpr/rows.json" |
| ], |
| "n": 3178, |
| "method": "min micro mean NLL over a 121-point log grid 0.25..4", |
| "cross_validation": { |
| "raw": { |
| "n": 3178, |
| "ece": 0.09340278932221213, |
| "brier": 0.2839500311089837, |
| "nll": 0.596971700327815, |
| "confident_error_rate": 0.05506607929515418, |
| "coverage_at_5pct_error": 0.6145374449339207 |
| }, |
| "out_of_fold": { |
| "n": 3178, |
| "ece": 0.01534796432900617, |
| "brier": 0.26274562369701476, |
| "nll": 0.4838656395724584, |
| "confident_error_rate": 0.015418502202643172, |
| "coverage_at_5pct_error": 0.5928256765261171 |
| }, |
| "ece_ci95": { |
| "raw": [ |
| 0.08070170353656396, |
| 0.1056609387588288 |
| ], |
| "out_of_fold": [ |
| 0.011641786092636034, |
| 0.03051411919309515 |
| ], |
| "delta": [ |
| -0.08459541487179573, |
| -0.05775058610041101 |
| ] |
| }, |
| "separated": true, |
| "temperatures": [ |
| 1.9543199368684914, |
| 1.9999999999999998, |
| 1.9543199368684914, |
| 1.9999999999999998, |
| 1.909683207820833 |
| ], |
| "folds": 5, |
| "seed": 0, |
| "samples": 1000, |
| "groups": 1670, |
| "unit": "source-stratified (source, group); sibling questions and variants stay together" |
| }, |
| "fit_rows": [ |
| { |
| "rows": "cal/rows.json", |
| "suite": "evals/v7/decision-v7", |
| "split": "calibration", |
| "questions": 1148 |
| }, |
| { |
| "rows": "calpr/rows.json", |
| "suite": null, |
| "split": "custom", |
| "questions": 2030 |
| } |
| ], |
| "training_suites": [ |
| "evals/v7/decision-v7" |
| ], |
| "in_distribution": { |
| "allowed": true, |
| "problems": [ |
| "calpr/rows.json: cannot tell which suite these rows were scored on (no d1a.eval.benchmark report.json with a suite of this checkout beside them)", |
| "cal/rows.json: evals/v7/decision-v7 is training data of the checkpoint", |
| "cal/rows.json: it pools 12 training source(s) ['agnews', 'amazon', 'banking77', 'boolq', 'compositional', 'dbpedia14', 'imdb', 'legacy_policy'] ...", |
| "cal/rows.json: it reads the calibration partition of evals/v7/decision-v7, a training corpus", |
| "cannot list the sources of --data train.jsonl (outside evals/), training data of d1" |
| ] |
| } |
| }, |
| "use_case_temperatures": { |
| "routing": 1.1755479062836087 |
| }, |
| "use_case_temperature_fits": { |
| "routing": { |
| "rows": [ |
| "phase-d3/gate/suites/routing_generic-development/head/rows.json", |
| "phase-d3/gate/suites/routing_handlabelled-45/head/rows.json" |
| ], |
| "n": 315, |
| "method": "min micro mean NLL over a 121-point log grid 0.25..4", |
| "cross_validation": { |
| "raw": { |
| "n": 315, |
| "ece": 0.005376748980220513, |
| "brier": 0.02213792789102334, |
| "nll": 0.037784826756766074, |
| "confident_error_rate": 0.0031746031746031746, |
| "coverage_at_5pct_error": 1.0 |
| }, |
| "out_of_fold": { |
| "n": 315, |
| "ece": 0.008392239537266366, |
| "brier": 0.021994761576082954, |
| "nll": 0.0385561510340784, |
| "confident_error_rate": 0.0031746031746031746, |
| "coverage_at_5pct_error": 1.0 |
| }, |
| "ece_ci95": { |
| "raw": [ |
| 0.003303220868466807, |
| 0.018871250910842064 |
| ], |
| "out_of_fold": [ |
| 0.0058853472476678124, |
| 0.022409565392811873 |
| ], |
| "delta": [ |
| -0.001788803325547818, |
| 0.007955193851064562 |
| ] |
| }, |
| "separated": false, |
| "temperatures": [ |
| 1.2311444133449163, |
| 1.148698354997035, |
| 1.2893703084395791, |
| 1.2311444133449163, |
| 1.0233738919967748 |
| ], |
| "folds": 5, |
| "seed": 0, |
| "samples": 1000, |
| "groups": 315, |
| "unit": "source-stratified (source, group); sibling questions and variants stay together" |
| }, |
| "interval": { |
| "level": 0.9, |
| "samples": 2000, |
| "seed": 0, |
| "lower": 0.6744039221992122, |
| "upper": 1.5157165665103978, |
| "questions": 315 |
| }, |
| "fit_rows": [ |
| { |
| "rows": "phase-d3/gate/suites/routing_generic-development/head/rows.json", |
| "suite": "evals/d1a/routing", |
| "split": "generic-development", |
| "questions": 270 |
| }, |
| { |
| "rows": "phase-d3/gate/suites/routing_handlabelled-45/head/rows.json", |
| "suite": "evals/d1a/routing", |
| "split": "handlabelled-45", |
| "questions": 45 |
| } |
| ], |
| "training_suites": [ |
| "evals/v7/decision-v7" |
| ], |
| "training_partitions": [], |
| "in_distribution": { |
| "allowed": true, |
| "problems": [ |
| "cannot list the sources of --data train.jsonl (outside evals/, no sidecar train.jsonl.json naming its sources), training data of routing-temp-v061/stub/d1a-e4b" |
| ] |
| } |
| } |
| } |
| } |
| } |
|
|
