| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.984025559105431, | |
| "eval_steps": 500, | |
| "global_step": 780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06389776357827476, | |
| "grad_norm": 130.31602916853151, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 3.6603, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12779552715654952, | |
| "grad_norm": 14.781977919866863, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.3776, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19169329073482427, | |
| "grad_norm": 4.3629613510157546, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.1595, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25559105431309903, | |
| "grad_norm": 3.870157217981069, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 0.1568, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3194888178913738, | |
| "grad_norm": 5.134748713124099, | |
| "learning_rate": 6.410256410256412e-06, | |
| "loss": 0.1566, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.38338658146964855, | |
| "grad_norm": 2.9173590197272645, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.146, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4472843450479233, | |
| "grad_norm": 7.85067184702089, | |
| "learning_rate": 8.974358974358976e-06, | |
| "loss": 0.1465, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5111821086261981, | |
| "grad_norm": 4.0072039812523155, | |
| "learning_rate": 9.999799726899261e-06, | |
| "loss": 0.152, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5750798722044729, | |
| "grad_norm": 4.781263018607648, | |
| "learning_rate": 9.992791852820709e-06, | |
| "loss": 0.1531, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6389776357827476, | |
| "grad_norm": 2.108926556744404, | |
| "learning_rate": 9.975786361654959e-06, | |
| "loss": 0.1298, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7028753993610224, | |
| "grad_norm": 2.8153254735688513, | |
| "learning_rate": 9.948817305370145e-06, | |
| "loss": 0.1161, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7667731629392971, | |
| "grad_norm": 4.0418418716369535, | |
| "learning_rate": 9.911938687078324e-06, | |
| "loss": 0.1011, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8306709265175719, | |
| "grad_norm": 3.9843806569017146, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.0841, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8945686900958466, | |
| "grad_norm": 1.3291937913520984, | |
| "learning_rate": 9.80876784408948e-06, | |
| "loss": 0.0698, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9584664536741214, | |
| "grad_norm": 2.7308366191283513, | |
| "learning_rate": 9.742682209735727e-06, | |
| "loss": 0.0553, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0223642172523961, | |
| "grad_norm": 1.1100404502138488, | |
| "learning_rate": 9.66709978038292e-06, | |
| "loss": 0.0516, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0862619808306708, | |
| "grad_norm": 1.2597692857731515, | |
| "learning_rate": 9.582171903054815e-06, | |
| "loss": 0.0454, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1501597444089458, | |
| "grad_norm": 1.1022185938066056, | |
| "learning_rate": 9.488068638195072e-06, | |
| "loss": 0.0387, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2140575079872205, | |
| "grad_norm": 1.7172177840149363, | |
| "learning_rate": 9.384978419136469e-06, | |
| "loss": 0.0302, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.2779552715654952, | |
| "grad_norm": 1.5637027754240362, | |
| "learning_rate": 9.273107674780102e-06, | |
| "loss": 0.0325, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.34185303514377, | |
| "grad_norm": 2.783634090632637, | |
| "learning_rate": 9.152680416240059e-06, | |
| "loss": 0.0223, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4057507987220448, | |
| "grad_norm": 1.3608054132422451, | |
| "learning_rate": 9.023937788281278e-06, | |
| "loss": 0.0195, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.4696485623003195, | |
| "grad_norm": 0.9992709605849684, | |
| "learning_rate": 8.88713758644883e-06, | |
| "loss": 0.0129, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5335463258785942, | |
| "grad_norm": 1.3478860777735155, | |
| "learning_rate": 8.742553740855507e-06, | |
| "loss": 0.0143, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.5974440894568689, | |
| "grad_norm": 0.5160241577185327, | |
| "learning_rate": 8.590475767661371e-06, | |
| "loss": 0.012, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6613418530351438, | |
| "grad_norm": 1.0951030534716653, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.0072, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.7252396166134185, | |
| "grad_norm": 1.087985002445579, | |
| "learning_rate": 8.265069924917925e-06, | |
| "loss": 0.0095, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.7891373801916934, | |
| "grad_norm": 1.7998564363536036, | |
| "learning_rate": 8.092393651331275e-06, | |
| "loss": 0.006, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.8530351437699681, | |
| "grad_norm": 0.18457425441572486, | |
| "learning_rate": 7.913525137306756e-06, | |
| "loss": 0.0062, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.9169329073482428, | |
| "grad_norm": 0.6609117232245165, | |
| "learning_rate": 7.728822550972523e-06, | |
| "loss": 0.0058, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.9808306709265175, | |
| "grad_norm": 0.4879357476412688, | |
| "learning_rate": 7.53865574266234e-06, | |
| "loss": 0.0034, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.0447284345047922, | |
| "grad_norm": 0.5473744973987851, | |
| "learning_rate": 7.343405504323519e-06, | |
| "loss": 0.0055, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.108626198083067, | |
| "grad_norm": 0.20699630966461133, | |
| "learning_rate": 7.143462807015271e-06, | |
| "loss": 0.0023, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.1725239616613417, | |
| "grad_norm": 0.5079647163308747, | |
| "learning_rate": 6.939228018024275e-06, | |
| "loss": 0.0016, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.236421725239617, | |
| "grad_norm": 0.06394720110875801, | |
| "learning_rate": 6.731110099165165e-06, | |
| "loss": 0.0006, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3003194888178915, | |
| "grad_norm": 0.043984983795711925, | |
| "learning_rate": 6.519525787871235e-06, | |
| "loss": 0.0006, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.364217252396166, | |
| "grad_norm": 0.08347825436904602, | |
| "learning_rate": 6.304898762715187e-06, | |
| "loss": 0.0002, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.428115015974441, | |
| "grad_norm": 0.00554983020767827, | |
| "learning_rate": 6.087658795030838e-06, | |
| "loss": 0.0001, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.4920127795527156, | |
| "grad_norm": 0.002074949463970475, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.0001, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.5559105431309903, | |
| "grad_norm": 0.00091993210892952, | |
| "learning_rate": 5.647084407270277e-06, | |
| "loss": 0.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.619808306709265, | |
| "grad_norm": 0.001187914956817107, | |
| "learning_rate": 5.424632197820325e-06, | |
| "loss": 0.0004, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.68370607028754, | |
| "grad_norm": 0.04613840276160312, | |
| "learning_rate": 5.201329700547077e-06, | |
| "loss": 0.0006, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.747603833865815, | |
| "grad_norm": 0.07217118291405861, | |
| "learning_rate": 4.977624058637783e-06, | |
| "loss": 0.0004, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.8115015974440896, | |
| "grad_norm": 0.019680043938271063, | |
| "learning_rate": 4.75396322254061e-06, | |
| "loss": 0.0015, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.8753993610223643, | |
| "grad_norm": 0.17125091829931163, | |
| "learning_rate": 4.530795052984104e-06, | |
| "loss": 0.0012, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.939297124600639, | |
| "grad_norm": 0.011882926064553328, | |
| "learning_rate": 4.308566424176336e-06, | |
| "loss": 0.0005, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.0031948881789137, | |
| "grad_norm": 0.020395494744052244, | |
| "learning_rate": 4.087722328979437e-06, | |
| "loss": 0.0001, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.0670926517571884, | |
| "grad_norm": 0.47336792227463537, | |
| "learning_rate": 3.86870498785139e-06, | |
| "loss": 0.0002, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.130990415335463, | |
| "grad_norm": 0.04112389438112822, | |
| "learning_rate": 3.6519529633392825e-06, | |
| "loss": 0.0006, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.194888178913738, | |
| "grad_norm": 0.9883623259675256, | |
| "learning_rate": 3.4379002818972122e-06, | |
| "loss": 0.0002, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.258785942492013, | |
| "grad_norm": 0.11942001202877613, | |
| "learning_rate": 3.226975564787322e-06, | |
| "loss": 0.0002, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.3226837060702876, | |
| "grad_norm": 0.014912857922381487, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.0005, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.3865814696485623, | |
| "grad_norm": 0.005218117954201568, | |
| "learning_rate": 2.816192345541437e-06, | |
| "loss": 0.0001, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.450479233226837, | |
| "grad_norm": 0.003478383092221035, | |
| "learning_rate": 2.6171563998934605e-06, | |
| "loss": 0.0, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.5143769968051117, | |
| "grad_norm": 0.002282251966100028, | |
| "learning_rate": 2.422891884458241e-06, | |
| "loss": 0.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.5782747603833864, | |
| "grad_norm": 0.0013298086133440442, | |
| "learning_rate": 2.2337877964734324e-06, | |
| "loss": 0.0, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.642172523961661, | |
| "grad_norm": 0.001191917339054205, | |
| "learning_rate": 2.050222799884387e-06, | |
| "loss": 0.0, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.7060702875399363, | |
| "grad_norm": 0.0019405632722846752, | |
| "learning_rate": 1.8725644671036125e-06, | |
| "loss": 0.0, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.769968051118211, | |
| "grad_norm": 0.001688089251520516, | |
| "learning_rate": 1.7011685429800596e-06, | |
| "loss": 0.0, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.8338658146964857, | |
| "grad_norm": 0.0013767767393313205, | |
| "learning_rate": 1.5363782324520033e-06, | |
| "loss": 0.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.8977635782747604, | |
| "grad_norm": 0.0015121179776807685, | |
| "learning_rate": 1.3785235133100088e-06, | |
| "loss": 0.0, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.961661341853035, | |
| "grad_norm": 0.0010094220393594992, | |
| "learning_rate": 1.2279204754460494e-06, | |
| "loss": 0.0, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.02555910543131, | |
| "grad_norm": 0.0008270190129985694, | |
| "learning_rate": 1.0848706879118893e-06, | |
| "loss": 0.0, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.0894568690095845, | |
| "grad_norm": 0.0009842840206759395, | |
| "learning_rate": 9.496605950541676e-07, | |
| "loss": 0.0, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.15335463258786, | |
| "grad_norm": 0.0011455384204312185, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.217252396166134, | |
| "grad_norm": 0.0009744990691985237, | |
| "learning_rate": 7.03826237188916e-07, | |
| "loss": 0.0, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.281150159744409, | |
| "grad_norm": 0.0010987588457946568, | |
| "learning_rate": 5.936942333950063e-07, | |
| "loss": 0.0, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.345047923322683, | |
| "grad_norm": 0.0007905637486817264, | |
| "learning_rate": 4.9238546099592e-07, | |
| "loss": 0.0, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.4089456869009584, | |
| "grad_norm": 0.0009142256811990887, | |
| "learning_rate": 4.001027817058789e-07, | |
| "loss": 0.0, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.472843450479234, | |
| "grad_norm": 0.0011330787800452912, | |
| "learning_rate": 3.1703098329864237e-07, | |
| "loss": 0.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.536741214057508, | |
| "grad_norm": 0.0007900970323302085, | |
| "learning_rate": 2.4333640958659144e-07, | |
| "loss": 0.0, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.600638977635783, | |
| "grad_norm": 0.0005185271373596653, | |
| "learning_rate": 1.7916662733218848e-07, | |
| "loss": 0.0, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.664536741214057, | |
| "grad_norm": 0.0007954554045334525, | |
| "learning_rate": 1.2465013075879884e-07, | |
| "loss": 0.0, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.728434504792332, | |
| "grad_norm": 0.001375306774010583, | |
| "learning_rate": 7.989608425254924e-08, | |
| "loss": 0.0, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.792332268370607, | |
| "grad_norm": 0.0009216072149331596, | |
| "learning_rate": 4.499410377045765e-08, | |
| "loss": 0.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.856230031948882, | |
| "grad_norm": 0.0008157911744637327, | |
| "learning_rate": 2.0014077392525035e-08, | |
| "loss": 0.0, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.920127795527156, | |
| "grad_norm": 0.0009134259574951338, | |
| "learning_rate": 5.006025377138901e-09, | |
| "loss": 0.0, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.984025559105431, | |
| "grad_norm": 0.0008560713988891386, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.984025559105431, | |
| "step": 780, | |
| "total_flos": 216765195878400.0, | |
| "train_loss": 0.07692194245522935, | |
| "train_runtime": 12037.1174, | |
| "train_samples_per_second": 16.615, | |
| "train_steps_per_second": 0.065 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 780, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 216765195878400.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |