| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 820, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24691358024691357, | |
| "grad_norm": 3.9386653900146484, | |
| "learning_rate": 4.945121951219512e-05, | |
| "loss": 2.6292, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.49382716049382713, | |
| "grad_norm": 2.194047451019287, | |
| "learning_rate": 4.884146341463415e-05, | |
| "loss": 1.7588, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7407407407407407, | |
| "grad_norm": 3.3464252948760986, | |
| "learning_rate": 4.823170731707317e-05, | |
| "loss": 1.2321, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 4.9754157066345215, | |
| "learning_rate": 4.76219512195122e-05, | |
| "loss": 1.1094, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2222222222222223, | |
| "grad_norm": 1.6857168674468994, | |
| "learning_rate": 4.701219512195122e-05, | |
| "loss": 0.8016, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4691358024691357, | |
| "grad_norm": 1.711855173110962, | |
| "learning_rate": 4.640243902439025e-05, | |
| "loss": 0.6045, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7160493827160495, | |
| "grad_norm": 3.110994338989258, | |
| "learning_rate": 4.5792682926829275e-05, | |
| "loss": 0.611, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9629629629629628, | |
| "grad_norm": 2.2832303047180176, | |
| "learning_rate": 4.5182926829268296e-05, | |
| "loss": 0.4029, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.197530864197531, | |
| "grad_norm": 2.3041841983795166, | |
| "learning_rate": 4.457317073170732e-05, | |
| "loss": 0.4292, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4444444444444446, | |
| "grad_norm": 1.5648462772369385, | |
| "learning_rate": 4.3963414634146346e-05, | |
| "loss": 0.3602, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.691358024691358, | |
| "grad_norm": 0.8374887704849243, | |
| "learning_rate": 4.335365853658537e-05, | |
| "loss": 0.309, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.9382716049382713, | |
| "grad_norm": 0.9409658908843994, | |
| "learning_rate": 4.2743902439024395e-05, | |
| "loss": 0.2612, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.1728395061728394, | |
| "grad_norm": 3.718695878982544, | |
| "learning_rate": 4.2134146341463416e-05, | |
| "loss": 0.2936, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.419753086419753, | |
| "grad_norm": 3.2449347972869873, | |
| "learning_rate": 4.152439024390244e-05, | |
| "loss": 0.1758, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.6666666666666665, | |
| "grad_norm": 1.2654057741165161, | |
| "learning_rate": 4.0914634146341465e-05, | |
| "loss": 0.2181, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.9135802469135803, | |
| "grad_norm": 0.7008330225944519, | |
| "learning_rate": 4.030487804878049e-05, | |
| "loss": 0.1788, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.927452802658081, | |
| "learning_rate": 3.9695121951219514e-05, | |
| "loss": 0.2255, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.395061728395062, | |
| "grad_norm": 0.9944292902946472, | |
| "learning_rate": 3.908536585365854e-05, | |
| "loss": 0.2719, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.6419753086419755, | |
| "grad_norm": 1.1106724739074707, | |
| "learning_rate": 3.847560975609756e-05, | |
| "loss": 0.1458, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.888888888888889, | |
| "grad_norm": 3.588369369506836, | |
| "learning_rate": 3.786585365853659e-05, | |
| "loss": 0.2225, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.1234567901234565, | |
| "grad_norm": 14.498685836791992, | |
| "learning_rate": 3.725609756097561e-05, | |
| "loss": 0.1737, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.37037037037037, | |
| "grad_norm": 15.858694076538086, | |
| "learning_rate": 3.664634146341463e-05, | |
| "loss": 0.1728, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.617283950617284, | |
| "grad_norm": 3.0714364051818848, | |
| "learning_rate": 3.603658536585366e-05, | |
| "loss": 0.0774, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.864197530864198, | |
| "grad_norm": 0.775265634059906, | |
| "learning_rate": 3.542682926829268e-05, | |
| "loss": 0.2107, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.098765432098766, | |
| "grad_norm": 1.262907862663269, | |
| "learning_rate": 3.48170731707317e-05, | |
| "loss": 0.1312, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.345679012345679, | |
| "grad_norm": 0.7409253120422363, | |
| "learning_rate": 3.420731707317074e-05, | |
| "loss": 0.1074, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.592592592592593, | |
| "grad_norm": 2.9899144172668457, | |
| "learning_rate": 3.359756097560976e-05, | |
| "loss": 0.2688, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.839506172839506, | |
| "grad_norm": 0.6718606352806091, | |
| "learning_rate": 3.298780487804878e-05, | |
| "loss": 0.0848, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.074074074074074, | |
| "grad_norm": 0.3312334716320038, | |
| "learning_rate": 3.237804878048781e-05, | |
| "loss": 0.0607, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.320987654320987, | |
| "grad_norm": 2.2608187198638916, | |
| "learning_rate": 3.176829268292683e-05, | |
| "loss": 0.1274, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.567901234567901, | |
| "grad_norm": 0.7812129855155945, | |
| "learning_rate": 3.115853658536586e-05, | |
| "loss": 0.0824, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.814814814814815, | |
| "grad_norm": 0.8673564195632935, | |
| "learning_rate": 3.054878048780488e-05, | |
| "loss": 0.1774, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.049382716049383, | |
| "grad_norm": 1.7585481405258179, | |
| "learning_rate": 2.9939024390243903e-05, | |
| "loss": 0.1415, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.296296296296296, | |
| "grad_norm": 2.8376216888427734, | |
| "learning_rate": 2.9329268292682927e-05, | |
| "loss": 0.1173, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.54320987654321, | |
| "grad_norm": 1.6730660200119019, | |
| "learning_rate": 2.8719512195121952e-05, | |
| "loss": 0.1585, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.790123456790123, | |
| "grad_norm": 0.30232733488082886, | |
| "learning_rate": 2.810975609756098e-05, | |
| "loss": 0.0377, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.024691358024691, | |
| "grad_norm": 2.1726369857788086, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0568, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.271604938271604, | |
| "grad_norm": 0.7406997084617615, | |
| "learning_rate": 2.689024390243903e-05, | |
| "loss": 0.105, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.518518518518519, | |
| "grad_norm": 0.2634631097316742, | |
| "learning_rate": 2.628048780487805e-05, | |
| "loss": 0.0328, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.765432098765432, | |
| "grad_norm": 0.15690064430236816, | |
| "learning_rate": 2.5670731707317075e-05, | |
| "loss": 0.069, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.09763717651367188, | |
| "learning_rate": 2.50609756097561e-05, | |
| "loss": 0.1503, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.246913580246913, | |
| "grad_norm": 0.9397181272506714, | |
| "learning_rate": 2.4451219512195124e-05, | |
| "loss": 0.1123, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.493827160493828, | |
| "grad_norm": 0.1911344975233078, | |
| "learning_rate": 2.3841463414634148e-05, | |
| "loss": 0.0439, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.74074074074074, | |
| "grad_norm": 2.133268117904663, | |
| "learning_rate": 2.3231707317073173e-05, | |
| "loss": 0.0466, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.987654320987655, | |
| "grad_norm": 3.681490898132324, | |
| "learning_rate": 2.2621951219512197e-05, | |
| "loss": 0.083, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.222222222222221, | |
| "grad_norm": 0.09520290791988373, | |
| "learning_rate": 2.2012195121951222e-05, | |
| "loss": 0.1275, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.469135802469136, | |
| "grad_norm": 0.18467915058135986, | |
| "learning_rate": 2.1402439024390243e-05, | |
| "loss": 0.0206, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 11.716049382716049, | |
| "grad_norm": 0.1692470908164978, | |
| "learning_rate": 2.0792682926829267e-05, | |
| "loss": 0.0694, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 11.962962962962964, | |
| "grad_norm": 0.21938467025756836, | |
| "learning_rate": 2.0182926829268295e-05, | |
| "loss": 0.1137, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.197530864197532, | |
| "grad_norm": 0.6872503161430359, | |
| "learning_rate": 1.957317073170732e-05, | |
| "loss": 0.0909, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.444444444444445, | |
| "grad_norm": 0.18933941423892975, | |
| "learning_rate": 1.896341463414634e-05, | |
| "loss": 0.0255, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.691358024691358, | |
| "grad_norm": 0.2120848298072815, | |
| "learning_rate": 1.8353658536585365e-05, | |
| "loss": 0.0383, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.938271604938272, | |
| "grad_norm": 0.12255409359931946, | |
| "learning_rate": 1.774390243902439e-05, | |
| "loss": 0.0941, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.17283950617284, | |
| "grad_norm": 0.10398901998996735, | |
| "learning_rate": 1.7134146341463418e-05, | |
| "loss": 0.0191, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 13.419753086419753, | |
| "grad_norm": 0.31278660893440247, | |
| "learning_rate": 1.652439024390244e-05, | |
| "loss": 0.0281, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 13.666666666666666, | |
| "grad_norm": 1.0805507898330688, | |
| "learning_rate": 1.5914634146341464e-05, | |
| "loss": 0.1726, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 13.91358024691358, | |
| "grad_norm": 0.4380980432033539, | |
| "learning_rate": 1.5304878048780488e-05, | |
| "loss": 0.0186, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.148148148148149, | |
| "grad_norm": 0.07801090180873871, | |
| "learning_rate": 1.4695121951219513e-05, | |
| "loss": 0.0473, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 14.395061728395062, | |
| "grad_norm": 0.10922332108020782, | |
| "learning_rate": 1.4085365853658535e-05, | |
| "loss": 0.023, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 14.641975308641975, | |
| "grad_norm": 0.2793682813644409, | |
| "learning_rate": 1.3475609756097562e-05, | |
| "loss": 0.0149, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.88888888888889, | |
| "grad_norm": 1.3744605779647827, | |
| "learning_rate": 1.2865853658536586e-05, | |
| "loss": 0.1422, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 15.123456790123457, | |
| "grad_norm": 0.1263941377401352, | |
| "learning_rate": 1.225609756097561e-05, | |
| "loss": 0.0124, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 15.37037037037037, | |
| "grad_norm": 0.13304823637008667, | |
| "learning_rate": 1.1646341463414635e-05, | |
| "loss": 0.0317, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.617283950617283, | |
| "grad_norm": 1.772745132446289, | |
| "learning_rate": 1.103658536585366e-05, | |
| "loss": 0.0443, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 15.864197530864198, | |
| "grad_norm": 0.06537698954343796, | |
| "learning_rate": 1.0426829268292683e-05, | |
| "loss": 0.0918, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.098765432098766, | |
| "grad_norm": 0.07521393895149231, | |
| "learning_rate": 9.817073170731709e-06, | |
| "loss": 0.0395, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 16.34567901234568, | |
| "grad_norm": 0.16562214493751526, | |
| "learning_rate": 9.207317073170732e-06, | |
| "loss": 0.1106, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 16.59259259259259, | |
| "grad_norm": 0.24475818872451782, | |
| "learning_rate": 8.597560975609756e-06, | |
| "loss": 0.019, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.839506172839506, | |
| "grad_norm": 0.32736074924468994, | |
| "learning_rate": 7.98780487804878e-06, | |
| "loss": 0.0326, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 17.074074074074073, | |
| "grad_norm": 0.05227570980787277, | |
| "learning_rate": 7.378048780487805e-06, | |
| "loss": 0.0103, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.320987654320987, | |
| "grad_norm": 0.09976907074451447, | |
| "learning_rate": 6.76829268292683e-06, | |
| "loss": 0.0864, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 17.567901234567902, | |
| "grad_norm": 0.04490213841199875, | |
| "learning_rate": 6.1585365853658535e-06, | |
| "loss": 0.0178, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 17.814814814814813, | |
| "grad_norm": 0.11479590833187103, | |
| "learning_rate": 5.548780487804878e-06, | |
| "loss": 0.0236, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 18.049382716049383, | |
| "grad_norm": 0.16496780514717102, | |
| "learning_rate": 4.9390243902439025e-06, | |
| "loss": 0.0481, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 18.296296296296298, | |
| "grad_norm": 6.131174564361572, | |
| "learning_rate": 4.329268292682927e-06, | |
| "loss": 0.0696, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.54320987654321, | |
| "grad_norm": 0.612872838973999, | |
| "learning_rate": 3.719512195121951e-06, | |
| "loss": 0.0194, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 18.790123456790123, | |
| "grad_norm": 1.3853808641433716, | |
| "learning_rate": 3.1097560975609757e-06, | |
| "loss": 0.037, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 19.02469135802469, | |
| "grad_norm": 0.06428851932287216, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0269, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 19.271604938271604, | |
| "grad_norm": 0.15213467180728912, | |
| "learning_rate": 1.8902439024390245e-06, | |
| "loss": 0.017, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 19.51851851851852, | |
| "grad_norm": 2.0686116218566895, | |
| "learning_rate": 1.2804878048780488e-06, | |
| "loss": 0.088, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.765432098765434, | |
| "grad_norm": 0.05781777948141098, | |
| "learning_rate": 6.707317073170731e-07, | |
| "loss": 0.0367, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.05907975882291794, | |
| "learning_rate": 6.097560975609757e-08, | |
| "loss": 0.0093, | |
| "step": 820 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 820, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 430125247180800.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |