| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 118.93805309734513, | |
| "eval_steps": 500, | |
| "global_step": 3360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 8.875021934509277, | |
| "learning_rate": 1.4880952380952381e-06, | |
| "loss": 2.2984, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 10.44489860534668, | |
| "learning_rate": 2.9761904761904763e-06, | |
| "loss": 2.302, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9911504424778761, | |
| "eval_accuracy": 0.1575, | |
| "eval_loss": 2.266624689102173, | |
| "eval_runtime": 7.4116, | |
| "eval_samples_per_second": 53.97, | |
| "eval_steps_per_second": 1.754, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0619469026548674, | |
| "grad_norm": 11.131507873535156, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 2.2827, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.415929203539823, | |
| "grad_norm": 11.846382141113281, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 2.2517, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "grad_norm": 8.047672271728516, | |
| "learning_rate": 7.4404761904761905e-06, | |
| "loss": 2.2226, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9823008849557522, | |
| "eval_accuracy": 0.315, | |
| "eval_loss": 2.1654422283172607, | |
| "eval_runtime": 5.7675, | |
| "eval_samples_per_second": 69.354, | |
| "eval_steps_per_second": 2.254, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.1238938053097347, | |
| "grad_norm": 7.9003190994262695, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 2.1848, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.47787610619469, | |
| "grad_norm": 14.892489433288574, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 2.1238, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.831858407079646, | |
| "grad_norm": 13.308501243591309, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 2.0639, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.9734513274336285, | |
| "eval_accuracy": 0.445, | |
| "eval_loss": 1.9969619512557983, | |
| "eval_runtime": 5.7209, | |
| "eval_samples_per_second": 69.919, | |
| "eval_steps_per_second": 2.272, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.185840707964602, | |
| "grad_norm": 16.818639755249023, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 2.005, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "grad_norm": 13.497295379638672, | |
| "learning_rate": 1.4880952380952381e-05, | |
| "loss": 1.9232, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.893805309734513, | |
| "grad_norm": 16.994335174560547, | |
| "learning_rate": 1.636904761904762e-05, | |
| "loss": 1.8559, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.56, | |
| "eval_loss": 1.7373383045196533, | |
| "eval_runtime": 5.8519, | |
| "eval_samples_per_second": 68.354, | |
| "eval_steps_per_second": 2.222, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.247787610619469, | |
| "grad_norm": 25.03571891784668, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.7371, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.601769911504425, | |
| "grad_norm": 27.68413734436035, | |
| "learning_rate": 1.9345238095238097e-05, | |
| "loss": 1.6511, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.95575221238938, | |
| "grad_norm": 21.25657081604004, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.5966, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.991150442477876, | |
| "eval_accuracy": 0.605, | |
| "eval_loss": 1.48233962059021, | |
| "eval_runtime": 5.791, | |
| "eval_samples_per_second": 69.072, | |
| "eval_steps_per_second": 2.245, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 5.3097345132743365, | |
| "grad_norm": 33.357173919677734, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 1.4767, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.663716814159292, | |
| "grad_norm": 21.856212615966797, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 1.3967, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.982300884955752, | |
| "eval_accuracy": 0.6125, | |
| "eval_loss": 1.2925149202346802, | |
| "eval_runtime": 5.8558, | |
| "eval_samples_per_second": 68.308, | |
| "eval_steps_per_second": 2.22, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 6.017699115044247, | |
| "grad_norm": 34.79808807373047, | |
| "learning_rate": 2.529761904761905e-05, | |
| "loss": 1.3046, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.371681415929204, | |
| "grad_norm": 27.307743072509766, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 1.2429, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.725663716814159, | |
| "grad_norm": 42.35990524291992, | |
| "learning_rate": 2.8273809523809523e-05, | |
| "loss": 1.204, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.9734513274336285, | |
| "eval_accuracy": 0.68, | |
| "eval_loss": 1.0512183904647827, | |
| "eval_runtime": 6.0412, | |
| "eval_samples_per_second": 66.212, | |
| "eval_steps_per_second": 2.152, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 7.079646017699115, | |
| "grad_norm": 35.177921295166016, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 1.1471, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.433628318584071, | |
| "grad_norm": 42.960391998291016, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.1087, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.787610619469026, | |
| "grad_norm": 28.81920623779297, | |
| "learning_rate": 3.273809523809524e-05, | |
| "loss": 1.0206, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7025, | |
| "eval_loss": 0.930656909942627, | |
| "eval_runtime": 5.9577, | |
| "eval_samples_per_second": 67.14, | |
| "eval_steps_per_second": 2.182, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 8.141592920353983, | |
| "grad_norm": 27.41872787475586, | |
| "learning_rate": 3.422619047619048e-05, | |
| "loss": 0.9949, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.495575221238939, | |
| "grad_norm": 30.292728424072266, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.9248, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.849557522123893, | |
| "grad_norm": 41.145442962646484, | |
| "learning_rate": 3.7202380952380956e-05, | |
| "loss": 0.9408, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.991150442477876, | |
| "eval_accuracy": 0.7425, | |
| "eval_loss": 0.8286006450653076, | |
| "eval_runtime": 6.1346, | |
| "eval_samples_per_second": 65.204, | |
| "eval_steps_per_second": 2.119, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 9.20353982300885, | |
| "grad_norm": 100.8248519897461, | |
| "learning_rate": 3.8690476190476195e-05, | |
| "loss": 0.9164, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.557522123893806, | |
| "grad_norm": 33.74626541137695, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 0.8392, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.91150442477876, | |
| "grad_norm": 23.371883392333984, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.8501, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.982300884955752, | |
| "eval_accuracy": 0.6975, | |
| "eval_loss": 0.8589820265769958, | |
| "eval_runtime": 5.7507, | |
| "eval_samples_per_second": 69.557, | |
| "eval_steps_per_second": 2.261, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 10.265486725663717, | |
| "grad_norm": 27.10701560974121, | |
| "learning_rate": 4.315476190476191e-05, | |
| "loss": 0.8698, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.619469026548673, | |
| "grad_norm": 31.57968521118164, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.7621, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 10.973451327433628, | |
| "grad_norm": 16.75609588623047, | |
| "learning_rate": 4.613095238095239e-05, | |
| "loss": 0.7545, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.973451327433628, | |
| "eval_accuracy": 0.7475, | |
| "eval_loss": 0.7702187895774841, | |
| "eval_runtime": 5.9752, | |
| "eval_samples_per_second": 66.944, | |
| "eval_steps_per_second": 2.176, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 11.327433628318584, | |
| "grad_norm": 38.2448616027832, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.7884, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 11.68141592920354, | |
| "grad_norm": 55.255489349365234, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 0.7484, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.765, | |
| "eval_loss": 0.7738745212554932, | |
| "eval_runtime": 5.7936, | |
| "eval_samples_per_second": 69.041, | |
| "eval_steps_per_second": 2.244, | |
| "step": 339 | |
| }, | |
| { | |
| "epoch": 12.035398230088495, | |
| "grad_norm": 48.28477478027344, | |
| "learning_rate": 4.993386243386244e-05, | |
| "loss": 0.7274, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.389380530973451, | |
| "grad_norm": 29.47681999206543, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 0.6793, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.743362831858407, | |
| "grad_norm": 25.996002197265625, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 0.6909, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 12.991150442477876, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7344464063644409, | |
| "eval_runtime": 5.7939, | |
| "eval_samples_per_second": 69.038, | |
| "eval_steps_per_second": 2.244, | |
| "step": 367 | |
| }, | |
| { | |
| "epoch": 13.097345132743364, | |
| "grad_norm": 17.77817153930664, | |
| "learning_rate": 4.943783068783069e-05, | |
| "loss": 0.6684, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.451327433628318, | |
| "grad_norm": 42.287635803222656, | |
| "learning_rate": 4.927248677248678e-05, | |
| "loss": 0.6309, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 13.805309734513274, | |
| "grad_norm": 23.582502365112305, | |
| "learning_rate": 4.910714285714286e-05, | |
| "loss": 0.6558, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 13.982300884955752, | |
| "eval_accuracy": 0.775, | |
| "eval_loss": 0.6874340772628784, | |
| "eval_runtime": 5.9948, | |
| "eval_samples_per_second": 66.725, | |
| "eval_steps_per_second": 2.169, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 14.15929203539823, | |
| "grad_norm": 23.51905059814453, | |
| "learning_rate": 4.894179894179895e-05, | |
| "loss": 0.5937, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.513274336283185, | |
| "grad_norm": 23.072551727294922, | |
| "learning_rate": 4.8776455026455034e-05, | |
| "loss": 0.6672, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 14.867256637168142, | |
| "grad_norm": 20.066308975219727, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.5923, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 14.973451327433628, | |
| "eval_accuracy": 0.7675, | |
| "eval_loss": 0.6640682816505432, | |
| "eval_runtime": 5.7834, | |
| "eval_samples_per_second": 69.163, | |
| "eval_steps_per_second": 2.248, | |
| "step": 423 | |
| }, | |
| { | |
| "epoch": 15.221238938053098, | |
| "grad_norm": 82.714599609375, | |
| "learning_rate": 4.84457671957672e-05, | |
| "loss": 0.6061, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 15.575221238938052, | |
| "grad_norm": 23.3370361328125, | |
| "learning_rate": 4.8280423280423284e-05, | |
| "loss": 0.5595, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 15.929203539823009, | |
| "grad_norm": 39.12126159667969, | |
| "learning_rate": 4.811507936507937e-05, | |
| "loss": 0.5764, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7925, | |
| "eval_loss": 0.6109621524810791, | |
| "eval_runtime": 5.8219, | |
| "eval_samples_per_second": 68.706, | |
| "eval_steps_per_second": 2.233, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 16.283185840707965, | |
| "grad_norm": 18.160350799560547, | |
| "learning_rate": 4.794973544973545e-05, | |
| "loss": 0.564, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 16.63716814159292, | |
| "grad_norm": 22.667816162109375, | |
| "learning_rate": 4.778439153439154e-05, | |
| "loss": 0.5416, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 16.991150442477878, | |
| "grad_norm": 43.60499954223633, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.5235, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 16.991150442477878, | |
| "eval_accuracy": 0.76, | |
| "eval_loss": 0.680573582649231, | |
| "eval_runtime": 6.0278, | |
| "eval_samples_per_second": 66.359, | |
| "eval_steps_per_second": 2.157, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 17.345132743362832, | |
| "grad_norm": 17.05644416809082, | |
| "learning_rate": 4.745370370370371e-05, | |
| "loss": 0.5301, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 17.699115044247787, | |
| "grad_norm": 14.023980140686035, | |
| "learning_rate": 4.7288359788359796e-05, | |
| "loss": 0.4883, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 17.98230088495575, | |
| "eval_accuracy": 0.76, | |
| "eval_loss": 0.7902570962905884, | |
| "eval_runtime": 5.8085, | |
| "eval_samples_per_second": 68.865, | |
| "eval_steps_per_second": 2.238, | |
| "step": 508 | |
| }, | |
| { | |
| "epoch": 18.053097345132745, | |
| "grad_norm": 14.874748229980469, | |
| "learning_rate": 4.7123015873015876e-05, | |
| "loss": 0.529, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 18.4070796460177, | |
| "grad_norm": 17.90399742126465, | |
| "learning_rate": 4.6957671957671964e-05, | |
| "loss": 0.455, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 18.761061946902654, | |
| "grad_norm": 17.8863468170166, | |
| "learning_rate": 4.6792328042328045e-05, | |
| "loss": 0.4682, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 18.97345132743363, | |
| "eval_accuracy": 0.7825, | |
| "eval_loss": 0.6469230055809021, | |
| "eval_runtime": 5.8117, | |
| "eval_samples_per_second": 68.826, | |
| "eval_steps_per_second": 2.237, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 19.115044247787612, | |
| "grad_norm": 25.429996490478516, | |
| "learning_rate": 4.662698412698413e-05, | |
| "loss": 0.4268, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 19.469026548672566, | |
| "grad_norm": 28.328603744506836, | |
| "learning_rate": 4.646164021164021e-05, | |
| "loss": 0.4233, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 19.82300884955752, | |
| "grad_norm": 18.160737991333008, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.441, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7825, | |
| "eval_loss": 0.6693841814994812, | |
| "eval_runtime": 5.8278, | |
| "eval_samples_per_second": 68.637, | |
| "eval_steps_per_second": 2.231, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 20.17699115044248, | |
| "grad_norm": 18.324892044067383, | |
| "learning_rate": 4.613095238095239e-05, | |
| "loss": 0.4421, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 20.530973451327434, | |
| "grad_norm": 27.71529769897461, | |
| "learning_rate": 4.596560846560847e-05, | |
| "loss": 0.3583, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 20.884955752212388, | |
| "grad_norm": 10.921418190002441, | |
| "learning_rate": 4.580026455026456e-05, | |
| "loss": 0.4201, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 20.991150442477878, | |
| "eval_accuracy": 0.7625, | |
| "eval_loss": 0.7144609689712524, | |
| "eval_runtime": 5.725, | |
| "eval_samples_per_second": 69.869, | |
| "eval_steps_per_second": 2.271, | |
| "step": 593 | |
| }, | |
| { | |
| "epoch": 21.238938053097346, | |
| "grad_norm": 27.364168167114258, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.3886, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 21.5929203539823, | |
| "grad_norm": 22.475797653198242, | |
| "learning_rate": 4.5469576719576725e-05, | |
| "loss": 0.366, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 21.946902654867255, | |
| "grad_norm": 14.525550842285156, | |
| "learning_rate": 4.5304232804232806e-05, | |
| "loss": 0.387, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 21.98230088495575, | |
| "eval_accuracy": 0.7775, | |
| "eval_loss": 0.6505405902862549, | |
| "eval_runtime": 6.0629, | |
| "eval_samples_per_second": 65.975, | |
| "eval_steps_per_second": 2.144, | |
| "step": 621 | |
| }, | |
| { | |
| "epoch": 22.300884955752213, | |
| "grad_norm": 15.278279304504395, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.3681, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 22.654867256637168, | |
| "grad_norm": 23.836380004882812, | |
| "learning_rate": 4.4973544973544974e-05, | |
| "loss": 0.4034, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 22.97345132743363, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.6168544292449951, | |
| "eval_runtime": 5.8701, | |
| "eval_samples_per_second": 68.143, | |
| "eval_steps_per_second": 2.215, | |
| "step": 649 | |
| }, | |
| { | |
| "epoch": 23.008849557522122, | |
| "grad_norm": 15.120473861694336, | |
| "learning_rate": 4.480820105820106e-05, | |
| "loss": 0.3424, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 23.36283185840708, | |
| "grad_norm": 12.84585189819336, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.3669, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 23.716814159292035, | |
| "grad_norm": 24.2416934967041, | |
| "learning_rate": 4.447751322751323e-05, | |
| "loss": 0.3041, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.795, | |
| "eval_loss": 0.6416059136390686, | |
| "eval_runtime": 5.7657, | |
| "eval_samples_per_second": 69.375, | |
| "eval_steps_per_second": 2.255, | |
| "step": 678 | |
| }, | |
| { | |
| "epoch": 24.07079646017699, | |
| "grad_norm": 17.557270050048828, | |
| "learning_rate": 4.431216931216932e-05, | |
| "loss": 0.3344, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 24.424778761061948, | |
| "grad_norm": 12.692173957824707, | |
| "learning_rate": 4.41468253968254e-05, | |
| "loss": 0.2928, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 24.778761061946902, | |
| "grad_norm": 15.073899269104004, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.3021, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 24.991150442477878, | |
| "eval_accuracy": 0.775, | |
| "eval_loss": 0.6992344856262207, | |
| "eval_runtime": 5.8878, | |
| "eval_samples_per_second": 67.937, | |
| "eval_steps_per_second": 2.208, | |
| "step": 706 | |
| }, | |
| { | |
| "epoch": 25.13274336283186, | |
| "grad_norm": 10.598650932312012, | |
| "learning_rate": 4.381613756613757e-05, | |
| "loss": 0.3147, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 25.486725663716815, | |
| "grad_norm": 17.572126388549805, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.302, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 25.84070796460177, | |
| "grad_norm": 30.176406860351562, | |
| "learning_rate": 4.3485449735449736e-05, | |
| "loss": 0.2853, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 25.98230088495575, | |
| "eval_accuracy": 0.7975, | |
| "eval_loss": 0.6566324830055237, | |
| "eval_runtime": 5.7476, | |
| "eval_samples_per_second": 69.594, | |
| "eval_steps_per_second": 2.262, | |
| "step": 734 | |
| }, | |
| { | |
| "epoch": 26.194690265486727, | |
| "grad_norm": 18.816850662231445, | |
| "learning_rate": 4.332010582010582e-05, | |
| "loss": 0.297, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 26.548672566371682, | |
| "grad_norm": 14.217058181762695, | |
| "learning_rate": 4.315476190476191e-05, | |
| "loss": 0.2849, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 26.902654867256636, | |
| "grad_norm": 17.27300453186035, | |
| "learning_rate": 4.298941798941799e-05, | |
| "loss": 0.27, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 26.97345132743363, | |
| "eval_accuracy": 0.7825, | |
| "eval_loss": 0.696976900100708, | |
| "eval_runtime": 6.0434, | |
| "eval_samples_per_second": 66.188, | |
| "eval_steps_per_second": 2.151, | |
| "step": 762 | |
| }, | |
| { | |
| "epoch": 27.256637168141594, | |
| "grad_norm": 10.05718994140625, | |
| "learning_rate": 4.282407407407408e-05, | |
| "loss": 0.2331, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 27.61061946902655, | |
| "grad_norm": 12.319819450378418, | |
| "learning_rate": 4.265873015873016e-05, | |
| "loss": 0.2462, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 27.964601769911503, | |
| "grad_norm": 14.027650833129883, | |
| "learning_rate": 4.249338624338625e-05, | |
| "loss": 0.2722, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.785, | |
| "eval_loss": 0.6862995624542236, | |
| "eval_runtime": 5.7837, | |
| "eval_samples_per_second": 69.16, | |
| "eval_steps_per_second": 2.248, | |
| "step": 791 | |
| }, | |
| { | |
| "epoch": 28.31858407079646, | |
| "grad_norm": 11.350334167480469, | |
| "learning_rate": 4.232804232804233e-05, | |
| "loss": 0.2366, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 28.672566371681416, | |
| "grad_norm": 8.22404670715332, | |
| "learning_rate": 4.2162698412698416e-05, | |
| "loss": 0.2143, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 28.991150442477878, | |
| "eval_accuracy": 0.795, | |
| "eval_loss": 0.679432213306427, | |
| "eval_runtime": 5.7657, | |
| "eval_samples_per_second": 69.376, | |
| "eval_steps_per_second": 2.255, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 29.02654867256637, | |
| "grad_norm": 11.07326889038086, | |
| "learning_rate": 4.1997354497354504e-05, | |
| "loss": 0.2443, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 29.38053097345133, | |
| "grad_norm": 28.747331619262695, | |
| "learning_rate": 4.1832010582010584e-05, | |
| "loss": 0.2424, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 29.734513274336283, | |
| "grad_norm": 12.262460708618164, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2238, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 29.98230088495575, | |
| "eval_accuracy": 0.7975, | |
| "eval_loss": 0.6781703233718872, | |
| "eval_runtime": 5.7261, | |
| "eval_samples_per_second": 69.856, | |
| "eval_steps_per_second": 2.27, | |
| "step": 847 | |
| }, | |
| { | |
| "epoch": 30.088495575221238, | |
| "grad_norm": 12.248478889465332, | |
| "learning_rate": 4.150132275132275e-05, | |
| "loss": 0.2414, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 30.442477876106196, | |
| "grad_norm": 26.66759490966797, | |
| "learning_rate": 4.133597883597884e-05, | |
| "loss": 0.222, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 30.79646017699115, | |
| "grad_norm": 17.73442268371582, | |
| "learning_rate": 4.117063492063492e-05, | |
| "loss": 0.2387, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 30.97345132743363, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.6944553256034851, | |
| "eval_runtime": 5.9266, | |
| "eval_samples_per_second": 67.492, | |
| "eval_steps_per_second": 2.193, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 31.150442477876105, | |
| "grad_norm": 10.954143524169922, | |
| "learning_rate": 4.100529100529101e-05, | |
| "loss": 0.1897, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 31.504424778761063, | |
| "grad_norm": 11.53641414642334, | |
| "learning_rate": 4.083994708994709e-05, | |
| "loss": 0.215, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 31.858407079646017, | |
| "grad_norm": 19.236658096313477, | |
| "learning_rate": 4.067460317460318e-05, | |
| "loss": 0.223, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7825, | |
| "eval_loss": 0.7377186417579651, | |
| "eval_runtime": 5.6881, | |
| "eval_samples_per_second": 70.322, | |
| "eval_steps_per_second": 2.285, | |
| "step": 904 | |
| }, | |
| { | |
| "epoch": 32.21238938053097, | |
| "grad_norm": 18.005434036254883, | |
| "learning_rate": 4.0509259259259265e-05, | |
| "loss": 0.1703, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 32.56637168141593, | |
| "grad_norm": 10.189879417419434, | |
| "learning_rate": 4.0343915343915346e-05, | |
| "loss": 0.1882, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 32.92035398230089, | |
| "grad_norm": 23.737546920776367, | |
| "learning_rate": 4.017857142857143e-05, | |
| "loss": 0.2211, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 32.991150442477874, | |
| "eval_accuracy": 0.7775, | |
| "eval_loss": 0.7430591583251953, | |
| "eval_runtime": 5.8045, | |
| "eval_samples_per_second": 68.911, | |
| "eval_steps_per_second": 2.24, | |
| "step": 932 | |
| }, | |
| { | |
| "epoch": 33.27433628318584, | |
| "grad_norm": 12.777314186096191, | |
| "learning_rate": 4.0013227513227514e-05, | |
| "loss": 0.2187, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 33.6283185840708, | |
| "grad_norm": 11.4727783203125, | |
| "learning_rate": 3.98478835978836e-05, | |
| "loss": 0.1952, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 33.982300884955755, | |
| "grad_norm": 9.033596992492676, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.1882, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 33.982300884955755, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.702938437461853, | |
| "eval_runtime": 5.7972, | |
| "eval_samples_per_second": 68.998, | |
| "eval_steps_per_second": 2.242, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 34.336283185840706, | |
| "grad_norm": 10.962681770324707, | |
| "learning_rate": 3.951719576719577e-05, | |
| "loss": 0.1788, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 34.690265486725664, | |
| "grad_norm": 9.644322395324707, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 0.1562, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 34.97345132743363, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.6886518597602844, | |
| "eval_runtime": 5.7736, | |
| "eval_samples_per_second": 69.281, | |
| "eval_steps_per_second": 2.252, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 35.04424778761062, | |
| "grad_norm": 7.208992004394531, | |
| "learning_rate": 3.918650793650794e-05, | |
| "loss": 0.1779, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 35.39823008849557, | |
| "grad_norm": 15.789375305175781, | |
| "learning_rate": 3.9021164021164026e-05, | |
| "loss": 0.1652, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 35.75221238938053, | |
| "grad_norm": 9.542279243469238, | |
| "learning_rate": 3.885582010582011e-05, | |
| "loss": 0.1689, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7975, | |
| "eval_loss": 0.7189816236495972, | |
| "eval_runtime": 5.7578, | |
| "eval_samples_per_second": 69.47, | |
| "eval_steps_per_second": 2.258, | |
| "step": 1017 | |
| }, | |
| { | |
| "epoch": 36.10619469026549, | |
| "grad_norm": 12.757901191711426, | |
| "learning_rate": 3.8690476190476195e-05, | |
| "loss": 0.178, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 36.46017699115044, | |
| "grad_norm": 8.646611213684082, | |
| "learning_rate": 3.8525132275132275e-05, | |
| "loss": 0.1613, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 36.8141592920354, | |
| "grad_norm": 12.245649337768555, | |
| "learning_rate": 3.835978835978836e-05, | |
| "loss": 0.1886, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 36.991150442477874, | |
| "eval_accuracy": 0.795, | |
| "eval_loss": 0.7677862048149109, | |
| "eval_runtime": 5.9107, | |
| "eval_samples_per_second": 67.674, | |
| "eval_steps_per_second": 2.199, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 37.16814159292036, | |
| "grad_norm": 16.76270866394043, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.1777, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 37.52212389380531, | |
| "grad_norm": 12.296770095825195, | |
| "learning_rate": 3.802910052910053e-05, | |
| "loss": 0.182, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 37.876106194690266, | |
| "grad_norm": 19.789642333984375, | |
| "learning_rate": 3.786375661375662e-05, | |
| "loss": 0.1887, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 37.982300884955755, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.7334153652191162, | |
| "eval_runtime": 5.7675, | |
| "eval_samples_per_second": 69.354, | |
| "eval_steps_per_second": 2.254, | |
| "step": 1073 | |
| }, | |
| { | |
| "epoch": 38.230088495575224, | |
| "grad_norm": 9.019023895263672, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.1377, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 38.584070796460175, | |
| "grad_norm": 8.896963119506836, | |
| "learning_rate": 3.753306878306879e-05, | |
| "loss": 0.1751, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 38.93805309734513, | |
| "grad_norm": 8.166861534118652, | |
| "learning_rate": 3.736772486772487e-05, | |
| "loss": 0.1531, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 38.97345132743363, | |
| "eval_accuracy": 0.7925, | |
| "eval_loss": 0.7359188199043274, | |
| "eval_runtime": 5.8237, | |
| "eval_samples_per_second": 68.685, | |
| "eval_steps_per_second": 2.232, | |
| "step": 1101 | |
| }, | |
| { | |
| "epoch": 39.29203539823009, | |
| "grad_norm": 11.483717918395996, | |
| "learning_rate": 3.7202380952380956e-05, | |
| "loss": 0.1703, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 39.64601769911504, | |
| "grad_norm": 7.077027797698975, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.1376, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 10.199649810791016, | |
| "learning_rate": 3.6871693121693124e-05, | |
| "loss": 0.1662, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.7593528032302856, | |
| "eval_runtime": 5.7627, | |
| "eval_samples_per_second": 69.411, | |
| "eval_steps_per_second": 2.256, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 40.35398230088496, | |
| "grad_norm": 12.18074893951416, | |
| "learning_rate": 3.6706349206349205e-05, | |
| "loss": 0.1502, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 40.70796460176991, | |
| "grad_norm": 13.122712135314941, | |
| "learning_rate": 3.654100529100529e-05, | |
| "loss": 0.1273, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 40.991150442477874, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.7341694831848145, | |
| "eval_runtime": 5.7253, | |
| "eval_samples_per_second": 69.866, | |
| "eval_steps_per_second": 2.271, | |
| "step": 1158 | |
| }, | |
| { | |
| "epoch": 41.06194690265487, | |
| "grad_norm": 12.024120330810547, | |
| "learning_rate": 3.637566137566138e-05, | |
| "loss": 0.1334, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 41.415929203539825, | |
| "grad_norm": 13.965011596679688, | |
| "learning_rate": 3.621031746031746e-05, | |
| "loss": 0.1394, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 41.769911504424776, | |
| "grad_norm": 17.121511459350586, | |
| "learning_rate": 3.604497354497355e-05, | |
| "loss": 0.1986, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 41.982300884955755, | |
| "eval_accuracy": 0.805, | |
| "eval_loss": 0.7780522108078003, | |
| "eval_runtime": 5.9228, | |
| "eval_samples_per_second": 67.536, | |
| "eval_steps_per_second": 2.195, | |
| "step": 1186 | |
| }, | |
| { | |
| "epoch": 42.123893805309734, | |
| "grad_norm": 11.281728744506836, | |
| "learning_rate": 3.587962962962963e-05, | |
| "loss": 0.1563, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 42.47787610619469, | |
| "grad_norm": 6.99351167678833, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.1541, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 42.83185840707964, | |
| "grad_norm": 12.189485549926758, | |
| "learning_rate": 3.55489417989418e-05, | |
| "loss": 0.1891, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 42.97345132743363, | |
| "eval_accuracy": 0.8225, | |
| "eval_loss": 0.7375677227973938, | |
| "eval_runtime": 5.736, | |
| "eval_samples_per_second": 69.735, | |
| "eval_steps_per_second": 2.266, | |
| "step": 1214 | |
| }, | |
| { | |
| "epoch": 43.1858407079646, | |
| "grad_norm": 7.004004955291748, | |
| "learning_rate": 3.5383597883597885e-05, | |
| "loss": 0.1332, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 43.53982300884956, | |
| "grad_norm": 15.816964149475098, | |
| "learning_rate": 3.521825396825397e-05, | |
| "loss": 0.1231, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 43.89380530973451, | |
| "grad_norm": 8.720746040344238, | |
| "learning_rate": 3.5052910052910054e-05, | |
| "loss": 0.1573, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.730354905128479, | |
| "eval_runtime": 5.7537, | |
| "eval_samples_per_second": 69.52, | |
| "eval_steps_per_second": 2.259, | |
| "step": 1243 | |
| }, | |
| { | |
| "epoch": 44.24778761061947, | |
| "grad_norm": 11.798288345336914, | |
| "learning_rate": 3.488756613756614e-05, | |
| "loss": 0.1337, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 44.60176991150443, | |
| "grad_norm": 8.698972702026367, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.1536, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 44.95575221238938, | |
| "grad_norm": 14.468975067138672, | |
| "learning_rate": 3.455687830687831e-05, | |
| "loss": 0.1536, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 44.991150442477874, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.777264416217804, | |
| "eval_runtime": 5.7994, | |
| "eval_samples_per_second": 68.972, | |
| "eval_steps_per_second": 2.242, | |
| "step": 1271 | |
| }, | |
| { | |
| "epoch": 45.309734513274336, | |
| "grad_norm": 5.724658966064453, | |
| "learning_rate": 3.439153439153439e-05, | |
| "loss": 0.1362, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 45.663716814159294, | |
| "grad_norm": 15.177201271057129, | |
| "learning_rate": 3.422619047619048e-05, | |
| "loss": 0.1562, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 45.982300884955755, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7622714042663574, | |
| "eval_runtime": 5.6562, | |
| "eval_samples_per_second": 70.719, | |
| "eval_steps_per_second": 2.298, | |
| "step": 1299 | |
| }, | |
| { | |
| "epoch": 46.017699115044245, | |
| "grad_norm": 14.262038230895996, | |
| "learning_rate": 3.406084656084656e-05, | |
| "loss": 0.1609, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 46.3716814159292, | |
| "grad_norm": 10.63355541229248, | |
| "learning_rate": 3.3895502645502647e-05, | |
| "loss": 0.1297, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 46.72566371681416, | |
| "grad_norm": 11.54215145111084, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.1264, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 46.97345132743363, | |
| "eval_accuracy": 0.7925, | |
| "eval_loss": 0.8314040899276733, | |
| "eval_runtime": 5.9374, | |
| "eval_samples_per_second": 67.369, | |
| "eval_steps_per_second": 2.189, | |
| "step": 1327 | |
| }, | |
| { | |
| "epoch": 47.07964601769911, | |
| "grad_norm": 7.798260688781738, | |
| "learning_rate": 3.3564814814814815e-05, | |
| "loss": 0.1552, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 47.43362831858407, | |
| "grad_norm": 10.773923873901367, | |
| "learning_rate": 3.33994708994709e-05, | |
| "loss": 0.1188, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 47.78761061946903, | |
| "grad_norm": 5.990432262420654, | |
| "learning_rate": 3.3234126984126983e-05, | |
| "loss": 0.1596, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8175, | |
| "eval_loss": 0.7830512523651123, | |
| "eval_runtime": 5.8001, | |
| "eval_samples_per_second": 68.964, | |
| "eval_steps_per_second": 2.241, | |
| "step": 1356 | |
| }, | |
| { | |
| "epoch": 48.14159292035398, | |
| "grad_norm": 11.099681854248047, | |
| "learning_rate": 3.306878306878307e-05, | |
| "loss": 0.1342, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 48.49557522123894, | |
| "grad_norm": 7.706020355224609, | |
| "learning_rate": 3.290343915343915e-05, | |
| "loss": 0.1382, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 48.849557522123895, | |
| "grad_norm": 5.685822010040283, | |
| "learning_rate": 3.273809523809524e-05, | |
| "loss": 0.1237, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 48.991150442477874, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.7949317097663879, | |
| "eval_runtime": 5.6938, | |
| "eval_samples_per_second": 70.252, | |
| "eval_steps_per_second": 2.283, | |
| "step": 1384 | |
| }, | |
| { | |
| "epoch": 49.203539823008846, | |
| "grad_norm": 10.283050537109375, | |
| "learning_rate": 3.257275132275133e-05, | |
| "loss": 0.1161, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 49.557522123893804, | |
| "grad_norm": 6.40301513671875, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.1363, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 49.91150442477876, | |
| "grad_norm": 17.212697982788086, | |
| "learning_rate": 3.2242063492063495e-05, | |
| "loss": 0.1355, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 49.982300884955755, | |
| "eval_accuracy": 0.795, | |
| "eval_loss": 0.781341552734375, | |
| "eval_runtime": 5.872, | |
| "eval_samples_per_second": 68.12, | |
| "eval_steps_per_second": 2.214, | |
| "step": 1412 | |
| }, | |
| { | |
| "epoch": 50.26548672566372, | |
| "grad_norm": 4.185158729553223, | |
| "learning_rate": 3.2076719576719576e-05, | |
| "loss": 0.1294, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 50.61946902654867, | |
| "grad_norm": 10.888021469116211, | |
| "learning_rate": 3.1911375661375664e-05, | |
| "loss": 0.163, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 50.97345132743363, | |
| "grad_norm": 8.630338668823242, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.1251, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 50.97345132743363, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.7647480964660645, | |
| "eval_runtime": 5.8333, | |
| "eval_samples_per_second": 68.571, | |
| "eval_steps_per_second": 2.229, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 51.32743362831859, | |
| "grad_norm": 19.542953491210938, | |
| "learning_rate": 3.158068783068783e-05, | |
| "loss": 0.1289, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 51.68141592920354, | |
| "grad_norm": 14.15263843536377, | |
| "learning_rate": 3.141534391534391e-05, | |
| "loss": 0.1181, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8175, | |
| "eval_loss": 0.7552239894866943, | |
| "eval_runtime": 6.0471, | |
| "eval_samples_per_second": 66.148, | |
| "eval_steps_per_second": 2.15, | |
| "step": 1469 | |
| }, | |
| { | |
| "epoch": 52.0353982300885, | |
| "grad_norm": 8.696867942810059, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.1232, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 52.389380530973455, | |
| "grad_norm": 15.115714073181152, | |
| "learning_rate": 3.108465608465609e-05, | |
| "loss": 0.128, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 52.743362831858406, | |
| "grad_norm": 26.04839515686035, | |
| "learning_rate": 3.091931216931217e-05, | |
| "loss": 0.1224, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 52.991150442477874, | |
| "eval_accuracy": 0.795, | |
| "eval_loss": 0.8345909118652344, | |
| "eval_runtime": 5.7656, | |
| "eval_samples_per_second": 69.377, | |
| "eval_steps_per_second": 2.255, | |
| "step": 1497 | |
| }, | |
| { | |
| "epoch": 53.097345132743364, | |
| "grad_norm": 6.820947647094727, | |
| "learning_rate": 3.075396825396826e-05, | |
| "loss": 0.1088, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 53.45132743362832, | |
| "grad_norm": 10.287090301513672, | |
| "learning_rate": 3.058862433862434e-05, | |
| "loss": 0.1094, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 53.80530973451327, | |
| "grad_norm": 12.619726181030273, | |
| "learning_rate": 3.0423280423280425e-05, | |
| "loss": 0.1201, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 53.982300884955755, | |
| "eval_accuracy": 0.7975, | |
| "eval_loss": 0.7740535140037537, | |
| "eval_runtime": 5.8431, | |
| "eval_samples_per_second": 68.457, | |
| "eval_steps_per_second": 2.225, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 54.15929203539823, | |
| "grad_norm": 3.6249613761901855, | |
| "learning_rate": 3.0257936507936506e-05, | |
| "loss": 0.1278, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 54.51327433628319, | |
| "grad_norm": 8.490986824035645, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.1359, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 54.86725663716814, | |
| "grad_norm": 8.873104095458984, | |
| "learning_rate": 2.9927248677248678e-05, | |
| "loss": 0.1109, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 54.97345132743363, | |
| "eval_accuracy": 0.785, | |
| "eval_loss": 0.772351861000061, | |
| "eval_runtime": 5.8419, | |
| "eval_samples_per_second": 68.47, | |
| "eval_steps_per_second": 2.225, | |
| "step": 1553 | |
| }, | |
| { | |
| "epoch": 55.2212389380531, | |
| "grad_norm": 3.0876898765563965, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.1108, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 55.575221238938056, | |
| "grad_norm": 6.528329372406006, | |
| "learning_rate": 2.959656084656085e-05, | |
| "loss": 0.1236, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 55.92920353982301, | |
| "grad_norm": 11.930560111999512, | |
| "learning_rate": 2.943121693121693e-05, | |
| "loss": 0.1084, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.805, | |
| "eval_loss": 0.7904237508773804, | |
| "eval_runtime": 5.7739, | |
| "eval_samples_per_second": 69.277, | |
| "eval_steps_per_second": 2.252, | |
| "step": 1582 | |
| }, | |
| { | |
| "epoch": 56.283185840707965, | |
| "grad_norm": 6.980546951293945, | |
| "learning_rate": 2.9265873015873018e-05, | |
| "loss": 0.1144, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 56.63716814159292, | |
| "grad_norm": 11.986266136169434, | |
| "learning_rate": 2.91005291005291e-05, | |
| "loss": 0.1084, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 56.991150442477874, | |
| "grad_norm": 6.527588367462158, | |
| "learning_rate": 2.8935185185185186e-05, | |
| "loss": 0.1187, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 56.991150442477874, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.7423826456069946, | |
| "eval_runtime": 5.9051, | |
| "eval_samples_per_second": 67.738, | |
| "eval_steps_per_second": 2.201, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 57.34513274336283, | |
| "grad_norm": 6.542221546173096, | |
| "learning_rate": 2.876984126984127e-05, | |
| "loss": 0.1191, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 57.69911504424779, | |
| "grad_norm": 6.406174659729004, | |
| "learning_rate": 2.8604497354497355e-05, | |
| "loss": 0.0935, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 57.982300884955755, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.7410502433776855, | |
| "eval_runtime": 6.0733, | |
| "eval_samples_per_second": 65.862, | |
| "eval_steps_per_second": 2.141, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 58.05309734513274, | |
| "grad_norm": 5.907907485961914, | |
| "learning_rate": 2.8439153439153442e-05, | |
| "loss": 0.0776, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 58.4070796460177, | |
| "grad_norm": 7.151067733764648, | |
| "learning_rate": 2.8273809523809523e-05, | |
| "loss": 0.1022, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 58.76106194690266, | |
| "grad_norm": 6.808224201202393, | |
| "learning_rate": 2.810846560846561e-05, | |
| "loss": 0.1023, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 58.97345132743363, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.7475782632827759, | |
| "eval_runtime": 5.7079, | |
| "eval_samples_per_second": 70.078, | |
| "eval_steps_per_second": 2.278, | |
| "step": 1666 | |
| }, | |
| { | |
| "epoch": 59.11504424778761, | |
| "grad_norm": 7.42250919342041, | |
| "learning_rate": 2.7943121693121695e-05, | |
| "loss": 0.1206, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 59.469026548672566, | |
| "grad_norm": 11.038393020629883, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1148, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 59.823008849557525, | |
| "grad_norm": 6.563925266265869, | |
| "learning_rate": 2.7612433862433863e-05, | |
| "loss": 0.1166, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8175, | |
| "eval_loss": 0.7742089033126831, | |
| "eval_runtime": 5.6896, | |
| "eval_samples_per_second": 70.304, | |
| "eval_steps_per_second": 2.285, | |
| "step": 1695 | |
| }, | |
| { | |
| "epoch": 60.176991150442475, | |
| "grad_norm": 8.62690544128418, | |
| "learning_rate": 2.7447089947089948e-05, | |
| "loss": 0.1133, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 60.530973451327434, | |
| "grad_norm": 8.701354026794434, | |
| "learning_rate": 2.7281746031746032e-05, | |
| "loss": 0.0961, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 60.88495575221239, | |
| "grad_norm": 5.880581855773926, | |
| "learning_rate": 2.7116402116402116e-05, | |
| "loss": 0.099, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 60.991150442477874, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.7696540951728821, | |
| "eval_runtime": 5.8242, | |
| "eval_samples_per_second": 68.679, | |
| "eval_steps_per_second": 2.232, | |
| "step": 1723 | |
| }, | |
| { | |
| "epoch": 61.23893805309734, | |
| "grad_norm": 6.893290996551514, | |
| "learning_rate": 2.6951058201058204e-05, | |
| "loss": 0.0968, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 61.5929203539823, | |
| "grad_norm": 9.383642196655273, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.0927, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 61.94690265486726, | |
| "grad_norm": 10.903635025024414, | |
| "learning_rate": 2.6620370370370372e-05, | |
| "loss": 0.1157, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 61.982300884955755, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8537997603416443, | |
| "eval_runtime": 5.862, | |
| "eval_samples_per_second": 68.236, | |
| "eval_steps_per_second": 2.218, | |
| "step": 1751 | |
| }, | |
| { | |
| "epoch": 62.30088495575221, | |
| "grad_norm": 11.495246887207031, | |
| "learning_rate": 2.6455026455026456e-05, | |
| "loss": 0.1071, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 62.65486725663717, | |
| "grad_norm": 8.735861778259277, | |
| "learning_rate": 2.628968253968254e-05, | |
| "loss": 0.1137, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 62.97345132743363, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.8544909954071045, | |
| "eval_runtime": 5.719, | |
| "eval_samples_per_second": 69.942, | |
| "eval_steps_per_second": 2.273, | |
| "step": 1779 | |
| }, | |
| { | |
| "epoch": 63.008849557522126, | |
| "grad_norm": 9.511134147644043, | |
| "learning_rate": 2.6124338624338625e-05, | |
| "loss": 0.0962, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 63.36283185840708, | |
| "grad_norm": 9.274235725402832, | |
| "learning_rate": 2.5958994708994712e-05, | |
| "loss": 0.09, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 63.716814159292035, | |
| "grad_norm": 17.82819175720215, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.094, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7925, | |
| "eval_loss": 0.8463137745857239, | |
| "eval_runtime": 5.7366, | |
| "eval_samples_per_second": 69.727, | |
| "eval_steps_per_second": 2.266, | |
| "step": 1808 | |
| }, | |
| { | |
| "epoch": 64.070796460177, | |
| "grad_norm": 13.545799255371094, | |
| "learning_rate": 2.562830687830688e-05, | |
| "loss": 0.1186, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 64.42477876106194, | |
| "grad_norm": 8.884902954101562, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.1027, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 64.77876106194691, | |
| "grad_norm": 9.229199409484863, | |
| "learning_rate": 2.529761904761905e-05, | |
| "loss": 0.1161, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 64.99115044247787, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.8350917100906372, | |
| "eval_runtime": 5.7053, | |
| "eval_samples_per_second": 70.11, | |
| "eval_steps_per_second": 2.279, | |
| "step": 1836 | |
| }, | |
| { | |
| "epoch": 65.13274336283186, | |
| "grad_norm": 4.687648296356201, | |
| "learning_rate": 2.5132275132275137e-05, | |
| "loss": 0.0954, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 65.48672566371681, | |
| "grad_norm": 8.967106819152832, | |
| "learning_rate": 2.496693121693122e-05, | |
| "loss": 0.1102, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 65.84070796460178, | |
| "grad_norm": 7.873806953430176, | |
| "learning_rate": 2.4801587301587305e-05, | |
| "loss": 0.08, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 65.98230088495575, | |
| "eval_accuracy": 0.7925, | |
| "eval_loss": 0.8610497117042542, | |
| "eval_runtime": 5.7201, | |
| "eval_samples_per_second": 69.929, | |
| "eval_steps_per_second": 2.273, | |
| "step": 1864 | |
| }, | |
| { | |
| "epoch": 66.19469026548673, | |
| "grad_norm": 7.836842060089111, | |
| "learning_rate": 2.463624338624339e-05, | |
| "loss": 0.1068, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 66.54867256637168, | |
| "grad_norm": 8.14975357055664, | |
| "learning_rate": 2.4470899470899473e-05, | |
| "loss": 0.0787, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 66.90265486725664, | |
| "grad_norm": 5.568120956420898, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.0799, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 66.97345132743362, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.8592771887779236, | |
| "eval_runtime": 5.7815, | |
| "eval_samples_per_second": 69.186, | |
| "eval_steps_per_second": 2.249, | |
| "step": 1892 | |
| }, | |
| { | |
| "epoch": 67.2566371681416, | |
| "grad_norm": 5.271720886230469, | |
| "learning_rate": 2.4140211640211642e-05, | |
| "loss": 0.0781, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 67.61061946902655, | |
| "grad_norm": 3.000176429748535, | |
| "learning_rate": 2.3974867724867726e-05, | |
| "loss": 0.0927, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 67.96460176991151, | |
| "grad_norm": 1.825054407119751, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.0783, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.8423022627830505, | |
| "eval_runtime": 5.6528, | |
| "eval_samples_per_second": 70.762, | |
| "eval_steps_per_second": 2.3, | |
| "step": 1921 | |
| }, | |
| { | |
| "epoch": 68.31858407079646, | |
| "grad_norm": 9.14376449584961, | |
| "learning_rate": 2.3644179894179898e-05, | |
| "loss": 0.1037, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 68.67256637168141, | |
| "grad_norm": 11.533547401428223, | |
| "learning_rate": 2.3478835978835982e-05, | |
| "loss": 0.0851, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 68.99115044247787, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.8265037536621094, | |
| "eval_runtime": 5.9247, | |
| "eval_samples_per_second": 67.514, | |
| "eval_steps_per_second": 2.194, | |
| "step": 1949 | |
| }, | |
| { | |
| "epoch": 69.02654867256638, | |
| "grad_norm": 8.8108549118042, | |
| "learning_rate": 2.3313492063492066e-05, | |
| "loss": 0.1005, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 69.38053097345133, | |
| "grad_norm": 7.721718788146973, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.0839, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 69.73451327433628, | |
| "grad_norm": 9.032380104064941, | |
| "learning_rate": 2.2982804232804235e-05, | |
| "loss": 0.0775, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 69.98230088495575, | |
| "eval_accuracy": 0.805, | |
| "eval_loss": 0.8707697987556458, | |
| "eval_runtime": 5.6882, | |
| "eval_samples_per_second": 70.321, | |
| "eval_steps_per_second": 2.285, | |
| "step": 1977 | |
| }, | |
| { | |
| "epoch": 70.08849557522124, | |
| "grad_norm": 7.697175979614258, | |
| "learning_rate": 2.281746031746032e-05, | |
| "loss": 0.1005, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 70.4424778761062, | |
| "grad_norm": 11.258265495300293, | |
| "learning_rate": 2.2652116402116403e-05, | |
| "loss": 0.1152, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 70.79646017699115, | |
| "grad_norm": 6.715971946716309, | |
| "learning_rate": 2.2486772486772487e-05, | |
| "loss": 0.0902, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 70.97345132743362, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.8180540204048157, | |
| "eval_runtime": 5.7974, | |
| "eval_samples_per_second": 68.996, | |
| "eval_steps_per_second": 2.242, | |
| "step": 2005 | |
| }, | |
| { | |
| "epoch": 71.15044247787611, | |
| "grad_norm": 8.447669982910156, | |
| "learning_rate": 2.2321428571428575e-05, | |
| "loss": 0.0664, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 71.50442477876106, | |
| "grad_norm": 10.469520568847656, | |
| "learning_rate": 2.215608465608466e-05, | |
| "loss": 0.0893, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 71.85840707964601, | |
| "grad_norm": 6.971242904663086, | |
| "learning_rate": 2.1990740740740743e-05, | |
| "loss": 0.0904, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.8296825885772705, | |
| "eval_runtime": 5.7586, | |
| "eval_samples_per_second": 69.462, | |
| "eval_steps_per_second": 2.258, | |
| "step": 2034 | |
| }, | |
| { | |
| "epoch": 72.21238938053098, | |
| "grad_norm": 6.059171199798584, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.0767, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 72.56637168141593, | |
| "grad_norm": 9.134629249572754, | |
| "learning_rate": 2.166005291005291e-05, | |
| "loss": 0.0897, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 72.92035398230088, | |
| "grad_norm": 6.5583415031433105, | |
| "learning_rate": 2.1494708994708996e-05, | |
| "loss": 0.0898, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 72.99115044247787, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.8464323282241821, | |
| "eval_runtime": 5.7076, | |
| "eval_samples_per_second": 70.082, | |
| "eval_steps_per_second": 2.278, | |
| "step": 2062 | |
| }, | |
| { | |
| "epoch": 73.27433628318585, | |
| "grad_norm": 12.021257400512695, | |
| "learning_rate": 2.132936507936508e-05, | |
| "loss": 0.1061, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 73.6283185840708, | |
| "grad_norm": 4.3469367027282715, | |
| "learning_rate": 2.1164021164021164e-05, | |
| "loss": 0.0744, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 73.98230088495575, | |
| "grad_norm": 9.244330406188965, | |
| "learning_rate": 2.0998677248677252e-05, | |
| "loss": 0.1013, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 73.98230088495575, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.8324652314186096, | |
| "eval_runtime": 5.8164, | |
| "eval_samples_per_second": 68.771, | |
| "eval_steps_per_second": 2.235, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 74.33628318584071, | |
| "grad_norm": 4.678669452667236, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.1001, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 74.69026548672566, | |
| "grad_norm": 9.78695297241211, | |
| "learning_rate": 2.066798941798942e-05, | |
| "loss": 0.0726, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 74.97345132743362, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.8771929144859314, | |
| "eval_runtime": 5.7555, | |
| "eval_samples_per_second": 69.499, | |
| "eval_steps_per_second": 2.259, | |
| "step": 2118 | |
| }, | |
| { | |
| "epoch": 75.04424778761062, | |
| "grad_norm": 3.598501443862915, | |
| "learning_rate": 2.0502645502645504e-05, | |
| "loss": 0.0975, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 75.39823008849558, | |
| "grad_norm": 10.049243927001953, | |
| "learning_rate": 2.033730158730159e-05, | |
| "loss": 0.0858, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 75.75221238938053, | |
| "grad_norm": 10.887565612792969, | |
| "learning_rate": 2.0171957671957673e-05, | |
| "loss": 0.0745, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.8505265712738037, | |
| "eval_runtime": 5.6825, | |
| "eval_samples_per_second": 70.391, | |
| "eval_steps_per_second": 2.288, | |
| "step": 2147 | |
| }, | |
| { | |
| "epoch": 76.10619469026548, | |
| "grad_norm": 9.399345397949219, | |
| "learning_rate": 2.0006613756613757e-05, | |
| "loss": 0.0909, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 76.46017699115045, | |
| "grad_norm": 3.6555662155151367, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.0799, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 76.8141592920354, | |
| "grad_norm": 5.8622894287109375, | |
| "learning_rate": 1.967592592592593e-05, | |
| "loss": 0.0891, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 76.99115044247787, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.8693811297416687, | |
| "eval_runtime": 5.7342, | |
| "eval_samples_per_second": 69.757, | |
| "eval_steps_per_second": 2.267, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 77.16814159292035, | |
| "grad_norm": 3.6685290336608887, | |
| "learning_rate": 1.9510582010582013e-05, | |
| "loss": 0.0909, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 77.52212389380531, | |
| "grad_norm": 9.96608829498291, | |
| "learning_rate": 1.9345238095238097e-05, | |
| "loss": 0.0962, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 77.87610619469027, | |
| "grad_norm": 7.857000827789307, | |
| "learning_rate": 1.917989417989418e-05, | |
| "loss": 0.0791, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 77.98230088495575, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.8765752911567688, | |
| "eval_runtime": 5.7248, | |
| "eval_samples_per_second": 69.872, | |
| "eval_steps_per_second": 2.271, | |
| "step": 2203 | |
| }, | |
| { | |
| "epoch": 78.23008849557522, | |
| "grad_norm": 10.403280258178711, | |
| "learning_rate": 1.9014550264550266e-05, | |
| "loss": 0.0622, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 78.58407079646018, | |
| "grad_norm": 6.770401477813721, | |
| "learning_rate": 1.884920634920635e-05, | |
| "loss": 0.0689, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 78.93805309734513, | |
| "grad_norm": 10.228433609008789, | |
| "learning_rate": 1.8683862433862434e-05, | |
| "loss": 0.0639, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 78.97345132743362, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.8461715579032898, | |
| "eval_runtime": 6.0121, | |
| "eval_samples_per_second": 66.532, | |
| "eval_steps_per_second": 2.162, | |
| "step": 2231 | |
| }, | |
| { | |
| "epoch": 79.29203539823008, | |
| "grad_norm": 6.610928535461426, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0705, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 79.64601769911505, | |
| "grad_norm": 9.066596984863281, | |
| "learning_rate": 1.8353174603174602e-05, | |
| "loss": 0.0795, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 11.537944793701172, | |
| "learning_rate": 1.818783068783069e-05, | |
| "loss": 0.0676, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.8991250395774841, | |
| "eval_runtime": 5.7259, | |
| "eval_samples_per_second": 69.858, | |
| "eval_steps_per_second": 2.27, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 80.35398230088495, | |
| "grad_norm": 5.221861839294434, | |
| "learning_rate": 1.8022486772486774e-05, | |
| "loss": 0.0932, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 80.70796460176992, | |
| "grad_norm": 6.904067039489746, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.0904, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 80.99115044247787, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.8550813794136047, | |
| "eval_runtime": 5.7321, | |
| "eval_samples_per_second": 69.782, | |
| "eval_steps_per_second": 2.268, | |
| "step": 2288 | |
| }, | |
| { | |
| "epoch": 81.06194690265487, | |
| "grad_norm": 10.940287590026855, | |
| "learning_rate": 1.7691798941798943e-05, | |
| "loss": 0.0942, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 81.41592920353982, | |
| "grad_norm": 4.109130382537842, | |
| "learning_rate": 1.7526455026455027e-05, | |
| "loss": 0.0791, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 81.76991150442478, | |
| "grad_norm": 10.844812393188477, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.0788, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 81.98230088495575, | |
| "eval_accuracy": 0.795, | |
| "eval_loss": 0.9301651120185852, | |
| "eval_runtime": 5.6474, | |
| "eval_samples_per_second": 70.829, | |
| "eval_steps_per_second": 2.302, | |
| "step": 2316 | |
| }, | |
| { | |
| "epoch": 82.12389380530973, | |
| "grad_norm": 3.5678551197052, | |
| "learning_rate": 1.7195767195767195e-05, | |
| "loss": 0.0603, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 82.47787610619469, | |
| "grad_norm": 7.562065124511719, | |
| "learning_rate": 1.703042328042328e-05, | |
| "loss": 0.093, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 82.83185840707965, | |
| "grad_norm": 10.092254638671875, | |
| "learning_rate": 1.6865079365079367e-05, | |
| "loss": 0.0787, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 82.97345132743362, | |
| "eval_accuracy": 0.8025, | |
| "eval_loss": 0.8706057071685791, | |
| "eval_runtime": 5.8345, | |
| "eval_samples_per_second": 68.558, | |
| "eval_steps_per_second": 2.228, | |
| "step": 2344 | |
| }, | |
| { | |
| "epoch": 83.1858407079646, | |
| "grad_norm": 4.765111923217773, | |
| "learning_rate": 1.669973544973545e-05, | |
| "loss": 0.0579, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 83.53982300884955, | |
| "grad_norm": 6.426796913146973, | |
| "learning_rate": 1.6534391534391536e-05, | |
| "loss": 0.0697, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 83.89380530973452, | |
| "grad_norm": 7.382542610168457, | |
| "learning_rate": 1.636904761904762e-05, | |
| "loss": 0.0918, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.805, | |
| "eval_loss": 0.868044912815094, | |
| "eval_runtime": 5.7723, | |
| "eval_samples_per_second": 69.297, | |
| "eval_steps_per_second": 2.252, | |
| "step": 2373 | |
| }, | |
| { | |
| "epoch": 84.24778761061947, | |
| "grad_norm": 5.388473987579346, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.0752, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 84.60176991150442, | |
| "grad_norm": 6.751432418823242, | |
| "learning_rate": 1.6038359788359788e-05, | |
| "loss": 0.0671, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 84.95575221238938, | |
| "grad_norm": 8.372601509094238, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.0681, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 84.99115044247787, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.8481296300888062, | |
| "eval_runtime": 5.711, | |
| "eval_samples_per_second": 70.04, | |
| "eval_steps_per_second": 2.276, | |
| "step": 2401 | |
| }, | |
| { | |
| "epoch": 85.30973451327434, | |
| "grad_norm": 6.79984712600708, | |
| "learning_rate": 1.5707671957671957e-05, | |
| "loss": 0.0634, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 85.66371681415929, | |
| "grad_norm": 9.60888671875, | |
| "learning_rate": 1.5542328042328044e-05, | |
| "loss": 0.115, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 85.98230088495575, | |
| "eval_accuracy": 0.8025, | |
| "eval_loss": 0.8552606105804443, | |
| "eval_runtime": 6.0227, | |
| "eval_samples_per_second": 66.415, | |
| "eval_steps_per_second": 2.158, | |
| "step": 2429 | |
| }, | |
| { | |
| "epoch": 86.01769911504425, | |
| "grad_norm": 9.006217956542969, | |
| "learning_rate": 1.537698412698413e-05, | |
| "loss": 0.0741, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 86.3716814159292, | |
| "grad_norm": 8.767806053161621, | |
| "learning_rate": 1.5211640211640213e-05, | |
| "loss": 0.0652, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 86.72566371681415, | |
| "grad_norm": 6.8285675048828125, | |
| "learning_rate": 1.5046296296296297e-05, | |
| "loss": 0.0599, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 86.97345132743362, | |
| "eval_accuracy": 0.805, | |
| "eval_loss": 0.8886809349060059, | |
| "eval_runtime": 5.6679, | |
| "eval_samples_per_second": 70.573, | |
| "eval_steps_per_second": 2.294, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 87.07964601769912, | |
| "grad_norm": 7.630321502685547, | |
| "learning_rate": 1.4880952380952381e-05, | |
| "loss": 0.0805, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 87.43362831858407, | |
| "grad_norm": 8.364428520202637, | |
| "learning_rate": 1.4715608465608465e-05, | |
| "loss": 0.0743, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 87.78761061946902, | |
| "grad_norm": 8.15882682800293, | |
| "learning_rate": 1.455026455026455e-05, | |
| "loss": 0.0774, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.9255210161209106, | |
| "eval_runtime": 5.684, | |
| "eval_samples_per_second": 70.374, | |
| "eval_steps_per_second": 2.287, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 88.14159292035399, | |
| "grad_norm": 3.7120983600616455, | |
| "learning_rate": 1.4384920634920635e-05, | |
| "loss": 0.0635, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 88.49557522123894, | |
| "grad_norm": 3.117091655731201, | |
| "learning_rate": 1.4219576719576721e-05, | |
| "loss": 0.0722, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 88.84955752212389, | |
| "grad_norm": 4.884605407714844, | |
| "learning_rate": 1.4054232804232805e-05, | |
| "loss": 0.0701, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 88.99115044247787, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.8794758319854736, | |
| "eval_runtime": 5.697, | |
| "eval_samples_per_second": 70.213, | |
| "eval_steps_per_second": 2.282, | |
| "step": 2514 | |
| }, | |
| { | |
| "epoch": 89.20353982300885, | |
| "grad_norm": 6.620547294616699, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0626, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 89.5575221238938, | |
| "grad_norm": 6.801345348358154, | |
| "learning_rate": 1.3723544973544974e-05, | |
| "loss": 0.0682, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 89.91150442477876, | |
| "grad_norm": 5.619492053985596, | |
| "learning_rate": 1.3558201058201058e-05, | |
| "loss": 0.074, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 89.98230088495575, | |
| "eval_accuracy": 0.8175, | |
| "eval_loss": 0.8634124994277954, | |
| "eval_runtime": 5.7593, | |
| "eval_samples_per_second": 69.453, | |
| "eval_steps_per_second": 2.257, | |
| "step": 2542 | |
| }, | |
| { | |
| "epoch": 90.26548672566372, | |
| "grad_norm": 10.153610229492188, | |
| "learning_rate": 1.3392857142857144e-05, | |
| "loss": 0.0794, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 90.61946902654867, | |
| "grad_norm": 5.089029312133789, | |
| "learning_rate": 1.3227513227513228e-05, | |
| "loss": 0.0611, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 90.97345132743362, | |
| "grad_norm": 6.003979206085205, | |
| "learning_rate": 1.3062169312169312e-05, | |
| "loss": 0.0497, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 90.97345132743362, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.8793442249298096, | |
| "eval_runtime": 6.0414, | |
| "eval_samples_per_second": 66.21, | |
| "eval_steps_per_second": 2.152, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 91.32743362831859, | |
| "grad_norm": 6.3248090744018555, | |
| "learning_rate": 1.2896825396825398e-05, | |
| "loss": 0.0557, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 91.68141592920354, | |
| "grad_norm": 4.216904640197754, | |
| "learning_rate": 1.2731481481481482e-05, | |
| "loss": 0.0569, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.7925, | |
| "eval_loss": 0.9006764888763428, | |
| "eval_runtime": 5.7399, | |
| "eval_samples_per_second": 69.688, | |
| "eval_steps_per_second": 2.265, | |
| "step": 2599 | |
| }, | |
| { | |
| "epoch": 92.03539823008849, | |
| "grad_norm": 6.895070552825928, | |
| "learning_rate": 1.2566137566137568e-05, | |
| "loss": 0.0769, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 92.38938053097345, | |
| "grad_norm": 3.7500102519989014, | |
| "learning_rate": 1.2400793650793652e-05, | |
| "loss": 0.0543, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 92.7433628318584, | |
| "grad_norm": 10.693458557128906, | |
| "learning_rate": 1.2235449735449737e-05, | |
| "loss": 0.0722, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 92.99115044247787, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.8700942993164062, | |
| "eval_runtime": 5.6593, | |
| "eval_samples_per_second": 70.681, | |
| "eval_steps_per_second": 2.297, | |
| "step": 2627 | |
| }, | |
| { | |
| "epoch": 93.09734513274336, | |
| "grad_norm": 5.407220363616943, | |
| "learning_rate": 1.2070105820105821e-05, | |
| "loss": 0.0822, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 93.45132743362832, | |
| "grad_norm": 2.9195971488952637, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.0624, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 93.80530973451327, | |
| "grad_norm": 7.752827167510986, | |
| "learning_rate": 1.1739417989417991e-05, | |
| "loss": 0.0674, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 93.98230088495575, | |
| "eval_accuracy": 0.8225, | |
| "eval_loss": 0.8879609704017639, | |
| "eval_runtime": 5.7797, | |
| "eval_samples_per_second": 69.208, | |
| "eval_steps_per_second": 2.249, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 94.15929203539822, | |
| "grad_norm": 6.314310550689697, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.0685, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 94.51327433628319, | |
| "grad_norm": 9.070263862609863, | |
| "learning_rate": 1.140873015873016e-05, | |
| "loss": 0.0649, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 94.86725663716814, | |
| "grad_norm": 4.211071968078613, | |
| "learning_rate": 1.1243386243386244e-05, | |
| "loss": 0.0643, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 94.97345132743362, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.8854994177818298, | |
| "eval_runtime": 5.6723, | |
| "eval_samples_per_second": 70.518, | |
| "eval_steps_per_second": 2.292, | |
| "step": 2683 | |
| }, | |
| { | |
| "epoch": 95.22123893805309, | |
| "grad_norm": 6.701349258422852, | |
| "learning_rate": 1.107804232804233e-05, | |
| "loss": 0.054, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 95.57522123893806, | |
| "grad_norm": 9.44869327545166, | |
| "learning_rate": 1.0912698412698414e-05, | |
| "loss": 0.0594, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 95.929203539823, | |
| "grad_norm": 5.749889373779297, | |
| "learning_rate": 1.0747354497354498e-05, | |
| "loss": 0.0583, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.8918322920799255, | |
| "eval_runtime": 5.8266, | |
| "eval_samples_per_second": 68.651, | |
| "eval_steps_per_second": 2.231, | |
| "step": 2712 | |
| }, | |
| { | |
| "epoch": 96.28318584070796, | |
| "grad_norm": 9.790497779846191, | |
| "learning_rate": 1.0582010582010582e-05, | |
| "loss": 0.0731, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 96.63716814159292, | |
| "grad_norm": 10.20504379272461, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.067, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 96.99115044247787, | |
| "grad_norm": 9.304971694946289, | |
| "learning_rate": 1.0251322751322752e-05, | |
| "loss": 0.0558, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 96.99115044247787, | |
| "eval_accuracy": 0.8275, | |
| "eval_loss": 0.8735535144805908, | |
| "eval_runtime": 5.748, | |
| "eval_samples_per_second": 69.59, | |
| "eval_steps_per_second": 2.262, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 97.34513274336283, | |
| "grad_norm": 2.155658483505249, | |
| "learning_rate": 1.0085978835978836e-05, | |
| "loss": 0.0952, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 97.69911504424779, | |
| "grad_norm": 4.216080665588379, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.0622, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 97.98230088495575, | |
| "eval_accuracy": 0.815, | |
| "eval_loss": 0.905790388584137, | |
| "eval_runtime": 6.0004, | |
| "eval_samples_per_second": 66.662, | |
| "eval_steps_per_second": 2.167, | |
| "step": 2768 | |
| }, | |
| { | |
| "epoch": 98.05309734513274, | |
| "grad_norm": 5.462803840637207, | |
| "learning_rate": 9.755291005291007e-06, | |
| "loss": 0.0576, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 98.40707964601769, | |
| "grad_norm": 6.204135894775391, | |
| "learning_rate": 9.58994708994709e-06, | |
| "loss": 0.0871, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 98.76106194690266, | |
| "grad_norm": 7.025479316711426, | |
| "learning_rate": 9.424603174603175e-06, | |
| "loss": 0.0689, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 98.97345132743362, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.9006683230400085, | |
| "eval_runtime": 5.6446, | |
| "eval_samples_per_second": 70.864, | |
| "eval_steps_per_second": 2.303, | |
| "step": 2796 | |
| }, | |
| { | |
| "epoch": 99.11504424778761, | |
| "grad_norm": 12.175108909606934, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0653, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 99.46902654867256, | |
| "grad_norm": 2.277102470397949, | |
| "learning_rate": 9.093915343915345e-06, | |
| "loss": 0.0533, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 99.82300884955752, | |
| "grad_norm": 3.374624252319336, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.0782, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8025, | |
| "eval_loss": 0.9216282367706299, | |
| "eval_runtime": 5.6935, | |
| "eval_samples_per_second": 70.255, | |
| "eval_steps_per_second": 2.283, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 100.17699115044248, | |
| "grad_norm": 4.110525131225586, | |
| "learning_rate": 8.763227513227513e-06, | |
| "loss": 0.0678, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 100.53097345132744, | |
| "grad_norm": 1.8253668546676636, | |
| "learning_rate": 8.597883597883598e-06, | |
| "loss": 0.0506, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 100.88495575221239, | |
| "grad_norm": 5.427482604980469, | |
| "learning_rate": 8.432539682539684e-06, | |
| "loss": 0.0696, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 100.99115044247787, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.9158985614776611, | |
| "eval_runtime": 5.7392, | |
| "eval_samples_per_second": 69.696, | |
| "eval_steps_per_second": 2.265, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 101.23893805309734, | |
| "grad_norm": 11.669960975646973, | |
| "learning_rate": 8.267195767195768e-06, | |
| "loss": 0.0424, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 101.59292035398231, | |
| "grad_norm": 3.504544496536255, | |
| "learning_rate": 8.101851851851852e-06, | |
| "loss": 0.0676, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 101.94690265486726, | |
| "grad_norm": 4.86131477355957, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.0554, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 101.98230088495575, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9194761514663696, | |
| "eval_runtime": 5.7011, | |
| "eval_samples_per_second": 70.162, | |
| "eval_steps_per_second": 2.28, | |
| "step": 2881 | |
| }, | |
| { | |
| "epoch": 102.30088495575221, | |
| "grad_norm": 10.070198059082031, | |
| "learning_rate": 7.771164021164022e-06, | |
| "loss": 0.0784, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 102.65486725663717, | |
| "grad_norm": 6.48520040512085, | |
| "learning_rate": 7.605820105820106e-06, | |
| "loss": 0.0585, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 102.97345132743362, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9314340949058533, | |
| "eval_runtime": 5.7273, | |
| "eval_samples_per_second": 69.841, | |
| "eval_steps_per_second": 2.27, | |
| "step": 2909 | |
| }, | |
| { | |
| "epoch": 103.00884955752213, | |
| "grad_norm": 3.9195964336395264, | |
| "learning_rate": 7.4404761904761905e-06, | |
| "loss": 0.068, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 103.36283185840708, | |
| "grad_norm": 7.709494590759277, | |
| "learning_rate": 7.275132275132275e-06, | |
| "loss": 0.0555, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 103.71681415929204, | |
| "grad_norm": 5.181375026702881, | |
| "learning_rate": 7.1097883597883606e-06, | |
| "loss": 0.0541, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "eval_accuracy": 0.825, | |
| "eval_loss": 0.893872082233429, | |
| "eval_runtime": 5.7025, | |
| "eval_samples_per_second": 70.145, | |
| "eval_steps_per_second": 2.28, | |
| "step": 2938 | |
| }, | |
| { | |
| "epoch": 104.070796460177, | |
| "grad_norm": 11.922988891601562, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.0656, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 104.42477876106194, | |
| "grad_norm": 6.153034687042236, | |
| "learning_rate": 6.779100529100529e-06, | |
| "loss": 0.0668, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 104.77876106194691, | |
| "grad_norm": 7.586311340332031, | |
| "learning_rate": 6.613756613756614e-06, | |
| "loss": 0.0636, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 104.99115044247787, | |
| "eval_accuracy": 0.8025, | |
| "eval_loss": 0.9045028686523438, | |
| "eval_runtime": 5.7995, | |
| "eval_samples_per_second": 68.972, | |
| "eval_steps_per_second": 2.242, | |
| "step": 2966 | |
| }, | |
| { | |
| "epoch": 105.13274336283186, | |
| "grad_norm": 7.499364852905273, | |
| "learning_rate": 6.448412698412699e-06, | |
| "loss": 0.0578, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 105.48672566371681, | |
| "grad_norm": 5.914554595947266, | |
| "learning_rate": 6.283068783068784e-06, | |
| "loss": 0.0683, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 105.84070796460178, | |
| "grad_norm": 5.446691513061523, | |
| "learning_rate": 6.117724867724868e-06, | |
| "loss": 0.0684, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 105.98230088495575, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.8892062306404114, | |
| "eval_runtime": 5.7229, | |
| "eval_samples_per_second": 69.895, | |
| "eval_steps_per_second": 2.272, | |
| "step": 2994 | |
| }, | |
| { | |
| "epoch": 106.19469026548673, | |
| "grad_norm": 5.555883407592773, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.0629, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 106.54867256637168, | |
| "grad_norm": 6.0038042068481445, | |
| "learning_rate": 5.787037037037038e-06, | |
| "loss": 0.0637, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 106.90265486725664, | |
| "grad_norm": 4.660765647888184, | |
| "learning_rate": 5.621693121693122e-06, | |
| "loss": 0.0608, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 106.97345132743362, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.8998861908912659, | |
| "eval_runtime": 5.6896, | |
| "eval_samples_per_second": 70.303, | |
| "eval_steps_per_second": 2.285, | |
| "step": 3022 | |
| }, | |
| { | |
| "epoch": 107.2566371681416, | |
| "grad_norm": 2.3250956535339355, | |
| "learning_rate": 5.456349206349207e-06, | |
| "loss": 0.065, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 107.61061946902655, | |
| "grad_norm": 3.4695749282836914, | |
| "learning_rate": 5.291005291005291e-06, | |
| "loss": 0.0614, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 107.96460176991151, | |
| "grad_norm": 1.7842631340026855, | |
| "learning_rate": 5.125661375661376e-06, | |
| "loss": 0.0663, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.9033371210098267, | |
| "eval_runtime": 6.1003, | |
| "eval_samples_per_second": 65.571, | |
| "eval_steps_per_second": 2.131, | |
| "step": 3051 | |
| }, | |
| { | |
| "epoch": 108.31858407079646, | |
| "grad_norm": 4.970130920410156, | |
| "learning_rate": 4.96031746031746e-06, | |
| "loss": 0.0466, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 108.67256637168141, | |
| "grad_norm": 6.935737609863281, | |
| "learning_rate": 4.794973544973545e-06, | |
| "loss": 0.054, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 108.99115044247787, | |
| "eval_accuracy": 0.805, | |
| "eval_loss": 0.9248512387275696, | |
| "eval_runtime": 5.6759, | |
| "eval_samples_per_second": 70.473, | |
| "eval_steps_per_second": 2.29, | |
| "step": 3079 | |
| }, | |
| { | |
| "epoch": 109.02654867256638, | |
| "grad_norm": 1.8442103862762451, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.0681, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 109.38053097345133, | |
| "grad_norm": 6.584358215332031, | |
| "learning_rate": 4.464285714285715e-06, | |
| "loss": 0.0551, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 109.73451327433628, | |
| "grad_norm": 2.5174403190612793, | |
| "learning_rate": 4.298941798941799e-06, | |
| "loss": 0.0538, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 109.98230088495575, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.9065310955047607, | |
| "eval_runtime": 5.7721, | |
| "eval_samples_per_second": 69.299, | |
| "eval_steps_per_second": 2.252, | |
| "step": 3107 | |
| }, | |
| { | |
| "epoch": 110.08849557522124, | |
| "grad_norm": 3.406076669692993, | |
| "learning_rate": 4.133597883597884e-06, | |
| "loss": 0.0768, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 110.4424778761062, | |
| "grad_norm": 3.832432746887207, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.0674, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 110.79646017699115, | |
| "grad_norm": 2.3623971939086914, | |
| "learning_rate": 3.802910052910053e-06, | |
| "loss": 0.0696, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 110.97345132743362, | |
| "eval_accuracy": 0.8175, | |
| "eval_loss": 0.9002352952957153, | |
| "eval_runtime": 5.823, | |
| "eval_samples_per_second": 68.693, | |
| "eval_steps_per_second": 2.233, | |
| "step": 3135 | |
| }, | |
| { | |
| "epoch": 111.15044247787611, | |
| "grad_norm": 8.914955139160156, | |
| "learning_rate": 3.6375661375661373e-06, | |
| "loss": 0.059, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 111.50442477876106, | |
| "grad_norm": 6.295753479003906, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.0496, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 111.85840707964601, | |
| "grad_norm": 4.604882717132568, | |
| "learning_rate": 3.306878306878307e-06, | |
| "loss": 0.0585, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 112.0, | |
| "eval_accuracy": 0.8025, | |
| "eval_loss": 0.9105786085128784, | |
| "eval_runtime": 5.6843, | |
| "eval_samples_per_second": 70.369, | |
| "eval_steps_per_second": 2.287, | |
| "step": 3164 | |
| }, | |
| { | |
| "epoch": 112.21238938053098, | |
| "grad_norm": 6.111964702606201, | |
| "learning_rate": 3.141534391534392e-06, | |
| "loss": 0.0387, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 112.56637168141593, | |
| "grad_norm": 9.29433536529541, | |
| "learning_rate": 2.9761904761904763e-06, | |
| "loss": 0.0623, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 112.92035398230088, | |
| "grad_norm": 10.9270601272583, | |
| "learning_rate": 2.810846560846561e-06, | |
| "loss": 0.0641, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 112.99115044247787, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.9088242053985596, | |
| "eval_runtime": 5.7312, | |
| "eval_samples_per_second": 69.794, | |
| "eval_steps_per_second": 2.268, | |
| "step": 3192 | |
| }, | |
| { | |
| "epoch": 113.27433628318585, | |
| "grad_norm": 9.549851417541504, | |
| "learning_rate": 2.6455026455026455e-06, | |
| "loss": 0.0555, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 113.6283185840708, | |
| "grad_norm": 7.800489902496338, | |
| "learning_rate": 2.48015873015873e-06, | |
| "loss": 0.0527, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 113.98230088495575, | |
| "grad_norm": 4.22268533706665, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.0611, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 113.98230088495575, | |
| "eval_accuracy": 0.8075, | |
| "eval_loss": 0.915170431137085, | |
| "eval_runtime": 5.6521, | |
| "eval_samples_per_second": 70.77, | |
| "eval_steps_per_second": 2.3, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 114.33628318584071, | |
| "grad_norm": 6.939643383026123, | |
| "learning_rate": 2.1494708994708994e-06, | |
| "loss": 0.092, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 114.69026548672566, | |
| "grad_norm": 4.18217658996582, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.0528, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 114.97345132743362, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9140109419822693, | |
| "eval_runtime": 5.9093, | |
| "eval_samples_per_second": 67.69, | |
| "eval_steps_per_second": 2.2, | |
| "step": 3248 | |
| }, | |
| { | |
| "epoch": 115.04424778761062, | |
| "grad_norm": 2.470015287399292, | |
| "learning_rate": 1.8187830687830687e-06, | |
| "loss": 0.0664, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 115.39823008849558, | |
| "grad_norm": 5.625057220458984, | |
| "learning_rate": 1.6534391534391535e-06, | |
| "loss": 0.0567, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 115.75221238938053, | |
| "grad_norm": 8.182403564453125, | |
| "learning_rate": 1.4880952380952381e-06, | |
| "loss": 0.0631, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 116.0, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.9184489250183105, | |
| "eval_runtime": 5.7237, | |
| "eval_samples_per_second": 69.885, | |
| "eval_steps_per_second": 2.271, | |
| "step": 3277 | |
| }, | |
| { | |
| "epoch": 116.10619469026548, | |
| "grad_norm": 3.7127795219421387, | |
| "learning_rate": 1.3227513227513228e-06, | |
| "loss": 0.0667, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 116.46017699115045, | |
| "grad_norm": 9.592253684997559, | |
| "learning_rate": 1.1574074074074074e-06, | |
| "loss": 0.0559, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 116.8141592920354, | |
| "grad_norm": 12.001859664916992, | |
| "learning_rate": 9.92063492063492e-07, | |
| "loss": 0.0744, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 116.99115044247787, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9216000437736511, | |
| "eval_runtime": 5.7416, | |
| "eval_samples_per_second": 69.667, | |
| "eval_steps_per_second": 2.264, | |
| "step": 3305 | |
| }, | |
| { | |
| "epoch": 117.16814159292035, | |
| "grad_norm": 5.99811315536499, | |
| "learning_rate": 8.267195767195768e-07, | |
| "loss": 0.051, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 117.52212389380531, | |
| "grad_norm": 4.772040367126465, | |
| "learning_rate": 6.613756613756614e-07, | |
| "loss": 0.0705, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 117.87610619469027, | |
| "grad_norm": 3.2538766860961914, | |
| "learning_rate": 4.96031746031746e-07, | |
| "loss": 0.0407, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 117.98230088495575, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9210975766181946, | |
| "eval_runtime": 5.6807, | |
| "eval_samples_per_second": 70.414, | |
| "eval_steps_per_second": 2.288, | |
| "step": 3333 | |
| }, | |
| { | |
| "epoch": 118.23008849557522, | |
| "grad_norm": 5.481634140014648, | |
| "learning_rate": 3.306878306878307e-07, | |
| "loss": 0.0682, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 118.58407079646018, | |
| "grad_norm": 7.475958824157715, | |
| "learning_rate": 1.6534391534391535e-07, | |
| "loss": 0.058, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 118.93805309734513, | |
| "grad_norm": 12.294817924499512, | |
| "learning_rate": 0.0, | |
| "loss": 0.0573, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 118.93805309734513, | |
| "eval_accuracy": 0.81, | |
| "eval_loss": 0.92020583152771, | |
| "eval_runtime": 5.7978, | |
| "eval_samples_per_second": 68.992, | |
| "eval_steps_per_second": 2.242, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 118.93805309734513, | |
| "step": 3360, | |
| "total_flos": 1.0779764781475824e+19, | |
| "train_loss": 0.2818299961143306, | |
| "train_runtime": 9749.4057, | |
| "train_samples_per_second": 44.31, | |
| "train_steps_per_second": 0.345 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3360, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 120, | |
| "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": 1.0779764781475824e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |