| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9974102848686645, | |
| "eval_steps": 100, | |
| "global_step": 337, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014798372179060304, | |
| "grad_norm": 4.499125891117204, | |
| "learning_rate": 2.9411764705882355e-06, | |
| "loss": 1.0817, | |
| "mean_token_accuracy": 0.7186651544542142, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.029596744358120607, | |
| "grad_norm": 2.028064542942808, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 1.0291, | |
| "mean_token_accuracy": 0.7241585050542387, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04439511653718091, | |
| "grad_norm": 1.195820001347328, | |
| "learning_rate": 8.823529411764707e-06, | |
| "loss": 0.9251, | |
| "mean_token_accuracy": 0.7413986712612658, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.059193488716241215, | |
| "grad_norm": 0.8696007046749331, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.8867, | |
| "mean_token_accuracy": 0.7452784136287651, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07399186089530152, | |
| "grad_norm": 0.7500875079543033, | |
| "learning_rate": 1.4705882352941179e-05, | |
| "loss": 0.8208, | |
| "mean_token_accuracy": 0.759392804158606, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08879023307436182, | |
| "grad_norm": 0.7362563546206198, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 0.8106, | |
| "mean_token_accuracy": 0.7598172702415025, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10358860525342212, | |
| "grad_norm": 0.5815757891473115, | |
| "learning_rate": 1.9999462497359468e-05, | |
| "loss": 0.773, | |
| "mean_token_accuracy": 0.7692409359846121, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.11838697743248243, | |
| "grad_norm": 0.5865978124387222, | |
| "learning_rate": 1.9980655971335944e-05, | |
| "loss": 0.7514, | |
| "mean_token_accuracy": 0.774326789041432, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13318534961154274, | |
| "grad_norm": 0.5621508587552188, | |
| "learning_rate": 1.993503206718859e-05, | |
| "loss": 0.7507, | |
| "mean_token_accuracy": 0.7731476189941919, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.14798372179060304, | |
| "grad_norm": 0.588509404275968, | |
| "learning_rate": 1.986271337340182e-05, | |
| "loss": 0.7336, | |
| "mean_token_accuracy": 0.7776760801058202, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16278209396966334, | |
| "grad_norm": 0.5438119771027593, | |
| "learning_rate": 1.976389420563607e-05, | |
| "loss": 0.7413, | |
| "mean_token_accuracy": 0.7752871810804264, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.17758046614872364, | |
| "grad_norm": 0.5105261291977568, | |
| "learning_rate": 1.9638840084614182e-05, | |
| "loss": 0.7299, | |
| "mean_token_accuracy": 0.7780510166069364, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.19237883832778394, | |
| "grad_norm": 0.4755545789127681, | |
| "learning_rate": 1.9487887022684336e-05, | |
| "loss": 0.7209, | |
| "mean_token_accuracy": 0.7794192467152918, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.20717721050684423, | |
| "grad_norm": 0.5035052967946048, | |
| "learning_rate": 1.9311440620976597e-05, | |
| "loss": 0.7235, | |
| "mean_token_accuracy": 0.7790004699092896, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22197558268590456, | |
| "grad_norm": 0.5225934980325596, | |
| "learning_rate": 1.9109974979578852e-05, | |
| "loss": 0.7064, | |
| "mean_token_accuracy": 0.7835738862803112, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.23677395486496486, | |
| "grad_norm": 0.5273120144064899, | |
| "learning_rate": 1.8884031423660492e-05, | |
| "loss": 0.7149, | |
| "mean_token_accuracy": 0.7813292441170601, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.25157232704402516, | |
| "grad_norm": 0.48729913827356003, | |
| "learning_rate": 1.8634217048966638e-05, | |
| "loss": 0.7189, | |
| "mean_token_accuracy": 0.7797419058095878, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.2663706992230855, | |
| "grad_norm": 0.5058775655815102, | |
| "learning_rate": 1.836120309059107e-05, | |
| "loss": 0.694, | |
| "mean_token_accuracy": 0.7865881005836394, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.28116907140214575, | |
| "grad_norm": 0.521123665431936, | |
| "learning_rate": 1.8065723119410885e-05, | |
| "loss": 0.703, | |
| "mean_token_accuracy": 0.783876903449907, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.2959674435812061, | |
| "grad_norm": 0.5237502769789281, | |
| "learning_rate": 1.77485710710289e-05, | |
| "loss": 0.7068, | |
| "mean_token_accuracy": 0.7823339948300922, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2959674435812061, | |
| "eval_loss": 0.7265122532844543, | |
| "eval_mean_token_accuracy": 0.7651502703273343, | |
| "eval_runtime": 11.3095, | |
| "eval_samples_per_second": 11.406, | |
| "eval_steps_per_second": 0.796, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31076581576026635, | |
| "grad_norm": 0.5150330824058433, | |
| "learning_rate": 1.741059911251997e-05, | |
| "loss": 0.7042, | |
| "mean_token_accuracy": 0.7831868045021135, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.3255641879393267, | |
| "grad_norm": 0.5021097759573874, | |
| "learning_rate": 1.7052715352713076e-05, | |
| "loss": 0.7002, | |
| "mean_token_accuracy": 0.7842077707353533, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.340362560118387, | |
| "grad_norm": 0.5246539797859209, | |
| "learning_rate": 1.667588140216154e-05, | |
| "loss": 0.7116, | |
| "mean_token_accuracy": 0.7816215675612058, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.3551609322974473, | |
| "grad_norm": 0.5257256878255661, | |
| "learning_rate": 1.628110978935756e-05, | |
| "loss": 0.7143, | |
| "mean_token_accuracy": 0.77973606197895, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3699593044765076, | |
| "grad_norm": 0.5043321185039642, | |
| "learning_rate": 1.586946124013354e-05, | |
| "loss": 0.7017, | |
| "mean_token_accuracy": 0.7831448532655912, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.38475767665556787, | |
| "grad_norm": 0.4939932678973825, | |
| "learning_rate": 1.5442041827560274e-05, | |
| "loss": 0.6991, | |
| "mean_token_accuracy": 0.7838827441771967, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3995560488346282, | |
| "grad_norm": 0.5004307223258988, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.7087, | |
| "mean_token_accuracy": 0.7809473288391601, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.41435442101368847, | |
| "grad_norm": 0.5696905296498661, | |
| "learning_rate": 1.4544523495299843e-05, | |
| "loss": 0.6784, | |
| "mean_token_accuracy": 0.7901156549726547, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4291527931927488, | |
| "grad_norm": 0.4613731794628925, | |
| "learning_rate": 1.4076836149416889e-05, | |
| "loss": 0.6931, | |
| "mean_token_accuracy": 0.7849786012574935, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.4439511653718091, | |
| "grad_norm": 0.4755166889453361, | |
| "learning_rate": 1.3598194608050011e-05, | |
| "loss": 0.6898, | |
| "mean_token_accuracy": 0.7873076132415269, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4587495375508694, | |
| "grad_norm": 0.5445199974148243, | |
| "learning_rate": 1.3109884950114007e-05, | |
| "loss": 0.6785, | |
| "mean_token_accuracy": 0.7890940163476381, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.4735479097299297, | |
| "grad_norm": 0.49722511848702816, | |
| "learning_rate": 1.2613219232128608e-05, | |
| "loss": 0.708, | |
| "mean_token_accuracy": 0.781048528219191, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.48834628190899, | |
| "grad_norm": 0.5334948317427054, | |
| "learning_rate": 1.2109531962807333e-05, | |
| "loss": 0.6893, | |
| "mean_token_accuracy": 0.78680416542453, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.5031446540880503, | |
| "grad_norm": 0.5131283864518137, | |
| "learning_rate": 1.1600176517318742e-05, | |
| "loss": 0.699, | |
| "mean_token_accuracy": 0.7831559400639638, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5179430262671106, | |
| "grad_norm": 0.5402575210828245, | |
| "learning_rate": 1.1086521500854746e-05, | |
| "loss": 0.6937, | |
| "mean_token_accuracy": 0.7847521205834683, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.532741398446171, | |
| "grad_norm": 0.46083681222368844, | |
| "learning_rate": 1.0569947071276847e-05, | |
| "loss": 0.6817, | |
| "mean_token_accuracy": 0.7891778585833855, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5475397706252312, | |
| "grad_norm": 0.5044221043841943, | |
| "learning_rate": 1.0051841230721065e-05, | |
| "loss": 0.6875, | |
| "mean_token_accuracy": 0.7867798023285817, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.5623381428042915, | |
| "grad_norm": 0.5700621888786264, | |
| "learning_rate": 9.533596096125826e-06, | |
| "loss": 0.6778, | |
| "mean_token_accuracy": 0.7891581715783473, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5771365149833518, | |
| "grad_norm": 0.49429508373470876, | |
| "learning_rate": 9.016604158703654e-06, | |
| "loss": 0.6686, | |
| "mean_token_accuracy": 0.7916483148675258, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.5919348871624122, | |
| "grad_norm": 0.5343207458940201, | |
| "learning_rate": 8.502254542407186e-06, | |
| "loss": 0.6825, | |
| "mean_token_accuracy": 0.7879164138334223, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5919348871624122, | |
| "eval_loss": 0.702451765537262, | |
| "eval_mean_token_accuracy": 0.7704640433711472, | |
| "eval_runtime": 10.3965, | |
| "eval_samples_per_second": 12.408, | |
| "eval_steps_per_second": 0.866, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6067332593414725, | |
| "grad_norm": 0.4912916019075468, | |
| "learning_rate": 7.991929271442817e-06, | |
| "loss": 0.6714, | |
| "mean_token_accuracy": 0.790756132895696, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.6215316315205327, | |
| "grad_norm": 0.44436620085337536, | |
| "learning_rate": 7.48699955686089e-06, | |
| "loss": 0.6777, | |
| "mean_token_accuracy": 0.7891227156405589, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.636330003699593, | |
| "grad_norm": 0.5226248071336361, | |
| "learning_rate": 6.988822112200157e-06, | |
| "loss": 0.7035, | |
| "mean_token_accuracy": 0.7816691567131292, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.6511283758786534, | |
| "grad_norm": 0.4439324753603916, | |
| "learning_rate": 6.498735508086094e-06, | |
| "loss": 0.6793, | |
| "mean_token_accuracy": 0.7888931945862602, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6659267480577137, | |
| "grad_norm": 0.4856547535868286, | |
| "learning_rate": 6.018056575578075e-06, | |
| "loss": 0.6881, | |
| "mean_token_accuracy": 0.7862525869831556, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.680725120236774, | |
| "grad_norm": 0.4529074271087974, | |
| "learning_rate": 5.548076867929331e-06, | |
| "loss": 0.6683, | |
| "mean_token_accuracy": 0.7922489034733624, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6955234924158342, | |
| "grad_norm": 0.44201312752808586, | |
| "learning_rate": 5.090059190266779e-06, | |
| "loss": 0.6714, | |
| "mean_token_accuracy": 0.7916120356352796, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.7103218645948945, | |
| "grad_norm": 0.4666939989516639, | |
| "learning_rate": 4.645234206515171e-06, | |
| "loss": 0.6637, | |
| "mean_token_accuracy": 0.7934646732927882, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7251202367739549, | |
| "grad_norm": 0.4323711823594102, | |
| "learning_rate": 4.214797132682597e-06, | |
| "loss": 0.6752, | |
| "mean_token_accuracy": 0.7900097765606582, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.7399186089530152, | |
| "grad_norm": 0.46893763760514173, | |
| "learning_rate": 3.799904525392251e-06, | |
| "loss": 0.6741, | |
| "mean_token_accuracy": 0.78994434647701, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "grad_norm": 0.4694142891909629, | |
| "learning_rate": 3.401671174289469e-06, | |
| "loss": 0.6788, | |
| "mean_token_accuracy": 0.7890182387790959, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.7695153533111357, | |
| "grad_norm": 0.469045682451544, | |
| "learning_rate": 3.021167106673928e-06, | |
| "loss": 0.6749, | |
| "mean_token_accuracy": 0.7897524888092856, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.7843137254901961, | |
| "grad_norm": 0.44594623491417784, | |
| "learning_rate": 2.6594147124053983e-06, | |
| "loss": 0.6914, | |
| "mean_token_accuracy": 0.7845483341547375, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.7991120976692564, | |
| "grad_norm": 0.433648253477015, | |
| "learning_rate": 2.317385996808195e-06, | |
| "loss": 0.6765, | |
| "mean_token_accuracy": 0.7901010520726933, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8139104698483167, | |
| "grad_norm": 0.46203654818236867, | |
| "learning_rate": 1.9959999689556407e-06, | |
| "loss": 0.6748, | |
| "mean_token_accuracy": 0.7899316717241394, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.8287088420273769, | |
| "grad_norm": 0.42103491997104836, | |
| "learning_rate": 1.6961201723520248e-06, | |
| "loss": 0.6753, | |
| "mean_token_accuracy": 0.7905570099684078, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8435072142064373, | |
| "grad_norm": 0.415975878998559, | |
| "learning_rate": 1.4185523646469822e-06, | |
| "loss": 0.6759, | |
| "mean_token_accuracy": 0.7900502935103929, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.8583055863854976, | |
| "grad_norm": 0.42758539576739607, | |
| "learning_rate": 1.1640423526166987e-06, | |
| "loss": 0.6742, | |
| "mean_token_accuracy": 0.7901006355952095, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8731039585645579, | |
| "grad_norm": 0.43594185823345216, | |
| "learning_rate": 9.332739882292752e-07, | |
| "loss": 0.6686, | |
| "mean_token_accuracy": 0.7915432863960005, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.8879023307436182, | |
| "grad_norm": 0.4262849779970212, | |
| "learning_rate": 7.268673311786378e-07, | |
| "loss": 0.663, | |
| "mean_token_accuracy": 0.7938472235979525, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8879023307436182, | |
| "eval_loss": 0.694024384021759, | |
| "eval_mean_token_accuracy": 0.7736351693291722, | |
| "eval_runtime": 10.4423, | |
| "eval_samples_per_second": 12.354, | |
| "eval_steps_per_second": 0.862, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9027007029226785, | |
| "grad_norm": 0.41940512769726646, | |
| "learning_rate": 5.453769828241872e-07, | |
| "loss": 0.6705, | |
| "mean_token_accuracy": 0.7914737299412992, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.9174990751017388, | |
| "grad_norm": 0.4029118184925996, | |
| "learning_rate": 3.8929059601275463e-07, | |
| "loss": 0.6446, | |
| "mean_token_accuracy": 0.7988981996676228, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9322974472807991, | |
| "grad_norm": 0.41171354636720464, | |
| "learning_rate": 2.5902756478688674e-07, | |
| "loss": 0.671, | |
| "mean_token_accuracy": 0.7913782021428799, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.9470958194598594, | |
| "grad_norm": 0.40299640725435004, | |
| "learning_rate": 1.5493789750014032e-07, | |
| "loss": 0.6681, | |
| "mean_token_accuracy": 0.7920879136935925, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9618941916389198, | |
| "grad_norm": 0.40540893101591874, | |
| "learning_rate": 7.730127636723539e-08, | |
| "loss": 0.655, | |
| "mean_token_accuracy": 0.7955860193987451, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.97669256381798, | |
| "grad_norm": 0.3925894396480368, | |
| "learning_rate": 2.6326305976001054e-08, | |
| "loss": 0.6696, | |
| "mean_token_accuracy": 0.791397284333766, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9914909359970403, | |
| "grad_norm": 0.42664846663701095, | |
| "learning_rate": 2.149952780321485e-09, | |
| "loss": 0.6674, | |
| "mean_token_accuracy": 0.7917057423636786, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.9974102848686645, | |
| "mean_token_accuracy": 0.7845949683152492, | |
| "step": 337, | |
| "total_flos": 128047739043840.0, | |
| "train_loss": 0.7127808205098002, | |
| "train_runtime": 4896.2549, | |
| "train_samples_per_second": 4.416, | |
| "train_steps_per_second": 0.069 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 337, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 128047739043840.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |