| { |
| "best_global_step": 5760, |
| "best_metric": 0.05086889490485191, |
| "best_model_checkpoint": "./results/checkpoint-5760", |
| "epoch": 25.0, |
| "eval_steps": 500, |
| "global_step": 14400, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.08680555555555555, |
| "grad_norm": 3.413226842880249, |
| "learning_rate": 1.9931944444444447e-05, |
| "loss": 1.8426, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.1736111111111111, |
| "grad_norm": 19.604671478271484, |
| "learning_rate": 1.98625e-05, |
| "loss": 1.2984, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.2604166666666667, |
| "grad_norm": 73.20538330078125, |
| "learning_rate": 1.979305555555556e-05, |
| "loss": 0.9999, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.3472222222222222, |
| "grad_norm": 4.863669395446777, |
| "learning_rate": 1.972361111111111e-05, |
| "loss": 0.8302, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.4340277777777778, |
| "grad_norm": 17.846479415893555, |
| "learning_rate": 1.965416666666667e-05, |
| "loss": 0.6309, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.5208333333333334, |
| "grad_norm": 11.574165344238281, |
| "learning_rate": 1.9584722222222224e-05, |
| "loss": 0.4987, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.6076388888888888, |
| "grad_norm": 31.07317352294922, |
| "learning_rate": 1.9515277777777778e-05, |
| "loss": 0.3274, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.6944444444444444, |
| "grad_norm": 3.164802074432373, |
| "learning_rate": 1.9445833333333336e-05, |
| "loss": 0.3305, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.78125, |
| "grad_norm": 2.5798959732055664, |
| "learning_rate": 1.937638888888889e-05, |
| "loss": 0.2624, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.8680555555555556, |
| "grad_norm": 12.735908508300781, |
| "learning_rate": 1.9306944444444445e-05, |
| "loss": 0.3035, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.9548611111111112, |
| "grad_norm": 17.255126953125, |
| "learning_rate": 1.9237500000000003e-05, |
| "loss": 0.2323, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.17769542336463928, |
| "eval_runtime": 2.138, |
| "eval_samples_per_second": 478.963, |
| "eval_steps_per_second": 29.935, |
| "step": 576 |
| }, |
| { |
| "epoch": 1.0416666666666667, |
| "grad_norm": 32.9837646484375, |
| "learning_rate": 1.9168055555555558e-05, |
| "loss": 0.1779, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.1284722222222223, |
| "grad_norm": 3.967150926589966, |
| "learning_rate": 1.9098611111111113e-05, |
| "loss": 0.145, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.2152777777777777, |
| "grad_norm": 12.614818572998047, |
| "learning_rate": 1.9029166666666667e-05, |
| "loss": 0.1287, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.3020833333333333, |
| "grad_norm": 26.14882469177246, |
| "learning_rate": 1.8959722222222222e-05, |
| "loss": 0.1397, |
| "step": 750 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 39.56580352783203, |
| "learning_rate": 1.889027777777778e-05, |
| "loss": 0.1194, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4756944444444444, |
| "grad_norm": 26.76836395263672, |
| "learning_rate": 1.8820833333333335e-05, |
| "loss": 0.1274, |
| "step": 850 |
| }, |
| { |
| "epoch": 1.5625, |
| "grad_norm": 26.581575393676758, |
| "learning_rate": 1.875138888888889e-05, |
| "loss": 0.1633, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.6493055555555556, |
| "grad_norm": 0.08225402981042862, |
| "learning_rate": 1.8681944444444447e-05, |
| "loss": 0.0979, |
| "step": 950 |
| }, |
| { |
| "epoch": 1.7361111111111112, |
| "grad_norm": 0.38262873888015747, |
| "learning_rate": 1.8612500000000002e-05, |
| "loss": 0.1227, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.8229166666666665, |
| "grad_norm": 1.3281371593475342, |
| "learning_rate": 1.8543055555555556e-05, |
| "loss": 0.0741, |
| "step": 1050 |
| }, |
| { |
| "epoch": 1.9097222222222223, |
| "grad_norm": 48.16311264038086, |
| "learning_rate": 1.847361111111111e-05, |
| "loss": 0.1376, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9965277777777777, |
| "grad_norm": 0.10062026977539062, |
| "learning_rate": 1.840416666666667e-05, |
| "loss": 0.1454, |
| "step": 1150 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.08461333066225052, |
| "eval_runtime": 2.1421, |
| "eval_samples_per_second": 478.037, |
| "eval_steps_per_second": 29.877, |
| "step": 1152 |
| }, |
| { |
| "epoch": 2.0833333333333335, |
| "grad_norm": 0.03965457156300545, |
| "learning_rate": 1.8334722222222224e-05, |
| "loss": 0.0425, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.170138888888889, |
| "grad_norm": 0.06563384085893631, |
| "learning_rate": 1.8265277777777778e-05, |
| "loss": 0.0516, |
| "step": 1250 |
| }, |
| { |
| "epoch": 2.2569444444444446, |
| "grad_norm": 1.1895874738693237, |
| "learning_rate": 1.8195833333333336e-05, |
| "loss": 0.0601, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.34375, |
| "grad_norm": 0.5909057259559631, |
| "learning_rate": 1.812638888888889e-05, |
| "loss": 0.0484, |
| "step": 1350 |
| }, |
| { |
| "epoch": 2.4305555555555554, |
| "grad_norm": 0.06617053598165512, |
| "learning_rate": 1.8056944444444446e-05, |
| "loss": 0.093, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.517361111111111, |
| "grad_norm": 0.03617614507675171, |
| "learning_rate": 1.7987500000000004e-05, |
| "loss": 0.0812, |
| "step": 1450 |
| }, |
| { |
| "epoch": 2.6041666666666665, |
| "grad_norm": 1.4127732515335083, |
| "learning_rate": 1.7918055555555558e-05, |
| "loss": 0.0834, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6909722222222223, |
| "grad_norm": 0.03689781203866005, |
| "learning_rate": 1.7848611111111113e-05, |
| "loss": 0.0581, |
| "step": 1550 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 82.96894073486328, |
| "learning_rate": 1.7779166666666667e-05, |
| "loss": 0.0607, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8645833333333335, |
| "grad_norm": 20.012392044067383, |
| "learning_rate": 1.7709722222222222e-05, |
| "loss": 0.0886, |
| "step": 1650 |
| }, |
| { |
| "epoch": 2.951388888888889, |
| "grad_norm": 0.08668874949216843, |
| "learning_rate": 1.764027777777778e-05, |
| "loss": 0.1223, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.08079180121421814, |
| "eval_runtime": 2.156, |
| "eval_samples_per_second": 474.96, |
| "eval_steps_per_second": 29.685, |
| "step": 1728 |
| }, |
| { |
| "epoch": 3.0381944444444446, |
| "grad_norm": 0.5279293656349182, |
| "learning_rate": 1.7570833333333335e-05, |
| "loss": 0.0713, |
| "step": 1750 |
| }, |
| { |
| "epoch": 3.125, |
| "grad_norm": 31.233030319213867, |
| "learning_rate": 1.750138888888889e-05, |
| "loss": 0.0746, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2118055555555554, |
| "grad_norm": 0.0305875763297081, |
| "learning_rate": 1.7431944444444447e-05, |
| "loss": 0.0742, |
| "step": 1850 |
| }, |
| { |
| "epoch": 3.298611111111111, |
| "grad_norm": 43.64688491821289, |
| "learning_rate": 1.7362500000000002e-05, |
| "loss": 0.0365, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3854166666666665, |
| "grad_norm": 0.034333955496549606, |
| "learning_rate": 1.7293055555555557e-05, |
| "loss": 0.058, |
| "step": 1950 |
| }, |
| { |
| "epoch": 3.4722222222222223, |
| "grad_norm": 86.51885986328125, |
| "learning_rate": 1.722361111111111e-05, |
| "loss": 0.0559, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590277777777777, |
| "grad_norm": 71.18854522705078, |
| "learning_rate": 1.7154166666666666e-05, |
| "loss": 0.0511, |
| "step": 2050 |
| }, |
| { |
| "epoch": 3.6458333333333335, |
| "grad_norm": 37.69118881225586, |
| "learning_rate": 1.7084722222222224e-05, |
| "loss": 0.0393, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.732638888888889, |
| "grad_norm": 0.011939575895667076, |
| "learning_rate": 1.701527777777778e-05, |
| "loss": 0.0394, |
| "step": 2150 |
| }, |
| { |
| "epoch": 3.8194444444444446, |
| "grad_norm": 0.23084449768066406, |
| "learning_rate": 1.6945833333333333e-05, |
| "loss": 0.02, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.90625, |
| "grad_norm": 0.22158057987689972, |
| "learning_rate": 1.687638888888889e-05, |
| "loss": 0.0248, |
| "step": 2250 |
| }, |
| { |
| "epoch": 3.9930555555555554, |
| "grad_norm": 0.06880240887403488, |
| "learning_rate": 1.6806944444444446e-05, |
| "loss": 0.054, |
| "step": 2300 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.07962702214717865, |
| "eval_runtime": 2.1469, |
| "eval_samples_per_second": 476.958, |
| "eval_steps_per_second": 29.81, |
| "step": 2304 |
| }, |
| { |
| "epoch": 4.079861111111111, |
| "grad_norm": 0.0237971693277359, |
| "learning_rate": 1.6737500000000004e-05, |
| "loss": 0.0151, |
| "step": 2350 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.012101912871003151, |
| "learning_rate": 1.6668055555555558e-05, |
| "loss": 0.0195, |
| "step": 2400 |
| }, |
| { |
| "epoch": 4.253472222222222, |
| "grad_norm": 0.010883960872888565, |
| "learning_rate": 1.6598611111111113e-05, |
| "loss": 0.0415, |
| "step": 2450 |
| }, |
| { |
| "epoch": 4.340277777777778, |
| "grad_norm": 0.021902142092585564, |
| "learning_rate": 1.6529166666666668e-05, |
| "loss": 0.0426, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.427083333333333, |
| "grad_norm": 0.04989313334226608, |
| "learning_rate": 1.6459722222222222e-05, |
| "loss": 0.0136, |
| "step": 2550 |
| }, |
| { |
| "epoch": 4.513888888888889, |
| "grad_norm": 0.22097109258174896, |
| "learning_rate": 1.639027777777778e-05, |
| "loss": 0.0465, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.600694444444445, |
| "grad_norm": 0.010039995424449444, |
| "learning_rate": 1.6320833333333335e-05, |
| "loss": 0.028, |
| "step": 2650 |
| }, |
| { |
| "epoch": 4.6875, |
| "grad_norm": 0.013202900998294353, |
| "learning_rate": 1.625138888888889e-05, |
| "loss": 0.0461, |
| "step": 2700 |
| }, |
| { |
| "epoch": 4.774305555555555, |
| "grad_norm": 0.015310668386518955, |
| "learning_rate": 1.6181944444444447e-05, |
| "loss": 0.0278, |
| "step": 2750 |
| }, |
| { |
| "epoch": 4.861111111111111, |
| "grad_norm": 0.008756318129599094, |
| "learning_rate": 1.6112500000000002e-05, |
| "loss": 0.0218, |
| "step": 2800 |
| }, |
| { |
| "epoch": 4.947916666666667, |
| "grad_norm": 0.015537718310952187, |
| "learning_rate": 1.6043055555555557e-05, |
| "loss": 0.0525, |
| "step": 2850 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.07712521404027939, |
| "eval_runtime": 2.1488, |
| "eval_samples_per_second": 476.534, |
| "eval_steps_per_second": 29.783, |
| "step": 2880 |
| }, |
| { |
| "epoch": 5.034722222222222, |
| "grad_norm": 0.011155390180647373, |
| "learning_rate": 1.597361111111111e-05, |
| "loss": 0.0136, |
| "step": 2900 |
| }, |
| { |
| "epoch": 5.121527777777778, |
| "grad_norm": 0.013077272102236748, |
| "learning_rate": 1.5904166666666666e-05, |
| "loss": 0.038, |
| "step": 2950 |
| }, |
| { |
| "epoch": 5.208333333333333, |
| "grad_norm": 0.009396527893841267, |
| "learning_rate": 1.5834722222222224e-05, |
| "loss": 0.0007, |
| "step": 3000 |
| }, |
| { |
| "epoch": 5.295138888888889, |
| "grad_norm": 0.03254946321249008, |
| "learning_rate": 1.576527777777778e-05, |
| "loss": 0.0218, |
| "step": 3050 |
| }, |
| { |
| "epoch": 5.381944444444445, |
| "grad_norm": 0.005586525425314903, |
| "learning_rate": 1.5695833333333333e-05, |
| "loss": 0.0203, |
| "step": 3100 |
| }, |
| { |
| "epoch": 5.46875, |
| "grad_norm": 0.009627276100218296, |
| "learning_rate": 1.562638888888889e-05, |
| "loss": 0.0668, |
| "step": 3150 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.0053014811128377914, |
| "learning_rate": 1.5556944444444446e-05, |
| "loss": 0.0119, |
| "step": 3200 |
| }, |
| { |
| "epoch": 5.642361111111111, |
| "grad_norm": 0.038525983691215515, |
| "learning_rate": 1.54875e-05, |
| "loss": 0.0255, |
| "step": 3250 |
| }, |
| { |
| "epoch": 5.729166666666667, |
| "grad_norm": 0.04675915837287903, |
| "learning_rate": 1.541805555555556e-05, |
| "loss": 0.0644, |
| "step": 3300 |
| }, |
| { |
| "epoch": 5.815972222222222, |
| "grad_norm": 0.04972713068127632, |
| "learning_rate": 1.5348611111111113e-05, |
| "loss": 0.0225, |
| "step": 3350 |
| }, |
| { |
| "epoch": 5.902777777777778, |
| "grad_norm": 0.004159330390393734, |
| "learning_rate": 1.5279166666666668e-05, |
| "loss": 0.0151, |
| "step": 3400 |
| }, |
| { |
| "epoch": 5.989583333333333, |
| "grad_norm": 0.01926554925739765, |
| "learning_rate": 1.5209722222222222e-05, |
| "loss": 0.0205, |
| "step": 3450 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.07027073949575424, |
| "eval_runtime": 2.1711, |
| "eval_samples_per_second": 471.645, |
| "eval_steps_per_second": 29.478, |
| "step": 3456 |
| }, |
| { |
| "epoch": 6.076388888888889, |
| "grad_norm": 0.09293384104967117, |
| "learning_rate": 1.5140277777777779e-05, |
| "loss": 0.0077, |
| "step": 3500 |
| }, |
| { |
| "epoch": 6.163194444444445, |
| "grad_norm": 0.010024973191320896, |
| "learning_rate": 1.5070833333333335e-05, |
| "loss": 0.0423, |
| "step": 3550 |
| }, |
| { |
| "epoch": 6.25, |
| "grad_norm": 0.029724491760134697, |
| "learning_rate": 1.500138888888889e-05, |
| "loss": 0.0199, |
| "step": 3600 |
| }, |
| { |
| "epoch": 6.336805555555555, |
| "grad_norm": 0.015937596559524536, |
| "learning_rate": 1.4931944444444446e-05, |
| "loss": 0.017, |
| "step": 3650 |
| }, |
| { |
| "epoch": 6.423611111111111, |
| "grad_norm": 0.009556828066706657, |
| "learning_rate": 1.4862500000000002e-05, |
| "loss": 0.0004, |
| "step": 3700 |
| }, |
| { |
| "epoch": 6.510416666666667, |
| "grad_norm": 0.032106589525938034, |
| "learning_rate": 1.4793055555555558e-05, |
| "loss": 0.0147, |
| "step": 3750 |
| }, |
| { |
| "epoch": 6.597222222222222, |
| "grad_norm": 0.005125823896378279, |
| "learning_rate": 1.4723611111111111e-05, |
| "loss": 0.0056, |
| "step": 3800 |
| }, |
| { |
| "epoch": 6.684027777777778, |
| "grad_norm": 0.018202291801571846, |
| "learning_rate": 1.4654166666666668e-05, |
| "loss": 0.0264, |
| "step": 3850 |
| }, |
| { |
| "epoch": 6.770833333333333, |
| "grad_norm": 31.37969970703125, |
| "learning_rate": 1.4584722222222222e-05, |
| "loss": 0.0119, |
| "step": 3900 |
| }, |
| { |
| "epoch": 6.857638888888889, |
| "grad_norm": 0.037925682961940765, |
| "learning_rate": 1.4515277777777779e-05, |
| "loss": 0.0203, |
| "step": 3950 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.1194610670208931, |
| "learning_rate": 1.4445833333333335e-05, |
| "loss": 0.0158, |
| "step": 4000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.0702182799577713, |
| "eval_runtime": 2.1701, |
| "eval_samples_per_second": 471.867, |
| "eval_steps_per_second": 29.492, |
| "step": 4032 |
| }, |
| { |
| "epoch": 7.03125, |
| "grad_norm": 0.0035032748710364103, |
| "learning_rate": 1.437638888888889e-05, |
| "loss": 0.0065, |
| "step": 4050 |
| }, |
| { |
| "epoch": 7.118055555555555, |
| "grad_norm": 0.0025438766460865736, |
| "learning_rate": 1.4306944444444446e-05, |
| "loss": 0.0003, |
| "step": 4100 |
| }, |
| { |
| "epoch": 7.204861111111111, |
| "grad_norm": 4.609809875488281, |
| "learning_rate": 1.4237500000000002e-05, |
| "loss": 0.02, |
| "step": 4150 |
| }, |
| { |
| "epoch": 7.291666666666667, |
| "grad_norm": 0.0041930885054171085, |
| "learning_rate": 1.4168055555555558e-05, |
| "loss": 0.0496, |
| "step": 4200 |
| }, |
| { |
| "epoch": 7.378472222222222, |
| "grad_norm": 0.3876318037509918, |
| "learning_rate": 1.4098611111111111e-05, |
| "loss": 0.0078, |
| "step": 4250 |
| }, |
| { |
| "epoch": 7.465277777777778, |
| "grad_norm": 0.003726869821548462, |
| "learning_rate": 1.4029166666666668e-05, |
| "loss": 0.0122, |
| "step": 4300 |
| }, |
| { |
| "epoch": 7.552083333333333, |
| "grad_norm": 0.00823493953794241, |
| "learning_rate": 1.3959722222222222e-05, |
| "loss": 0.0136, |
| "step": 4350 |
| }, |
| { |
| "epoch": 7.638888888888889, |
| "grad_norm": 0.004769823048263788, |
| "learning_rate": 1.3890277777777779e-05, |
| "loss": 0.0132, |
| "step": 4400 |
| }, |
| { |
| "epoch": 7.725694444444445, |
| "grad_norm": 45.848167419433594, |
| "learning_rate": 1.3820833333333335e-05, |
| "loss": 0.0116, |
| "step": 4450 |
| }, |
| { |
| "epoch": 7.8125, |
| "grad_norm": 0.0101216621696949, |
| "learning_rate": 1.375138888888889e-05, |
| "loss": 0.0275, |
| "step": 4500 |
| }, |
| { |
| "epoch": 7.899305555555555, |
| "grad_norm": 0.009534699842333794, |
| "learning_rate": 1.3681944444444446e-05, |
| "loss": 0.0073, |
| "step": 4550 |
| }, |
| { |
| "epoch": 7.986111111111111, |
| "grad_norm": 0.003179518273100257, |
| "learning_rate": 1.3612500000000002e-05, |
| "loss": 0.0128, |
| "step": 4600 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.08423544466495514, |
| "eval_runtime": 2.1727, |
| "eval_samples_per_second": 471.311, |
| "eval_steps_per_second": 29.457, |
| "step": 4608 |
| }, |
| { |
| "epoch": 8.072916666666666, |
| "grad_norm": 0.0026477461215108633, |
| "learning_rate": 1.3543055555555557e-05, |
| "loss": 0.0261, |
| "step": 4650 |
| }, |
| { |
| "epoch": 8.159722222222221, |
| "grad_norm": 0.019464371725916862, |
| "learning_rate": 1.3473611111111111e-05, |
| "loss": 0.0015, |
| "step": 4700 |
| }, |
| { |
| "epoch": 8.246527777777779, |
| "grad_norm": 0.0027786209248006344, |
| "learning_rate": 1.3404166666666668e-05, |
| "loss": 0.0125, |
| "step": 4750 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.0063923560082912445, |
| "learning_rate": 1.3334722222222222e-05, |
| "loss": 0.0107, |
| "step": 4800 |
| }, |
| { |
| "epoch": 8.42013888888889, |
| "grad_norm": 0.0071762921288609505, |
| "learning_rate": 1.3265277777777779e-05, |
| "loss": 0.01, |
| "step": 4850 |
| }, |
| { |
| "epoch": 8.506944444444445, |
| "grad_norm": 0.003973743878304958, |
| "learning_rate": 1.3195833333333335e-05, |
| "loss": 0.0262, |
| "step": 4900 |
| }, |
| { |
| "epoch": 8.59375, |
| "grad_norm": 0.015603139996528625, |
| "learning_rate": 1.312638888888889e-05, |
| "loss": 0.0213, |
| "step": 4950 |
| }, |
| { |
| "epoch": 8.680555555555555, |
| "grad_norm": 0.002647754270583391, |
| "learning_rate": 1.3056944444444446e-05, |
| "loss": 0.01, |
| "step": 5000 |
| }, |
| { |
| "epoch": 8.76736111111111, |
| "grad_norm": 14.067428588867188, |
| "learning_rate": 1.2987500000000002e-05, |
| "loss": 0.0324, |
| "step": 5050 |
| }, |
| { |
| "epoch": 8.854166666666666, |
| "grad_norm": 0.004289311822503805, |
| "learning_rate": 1.2918055555555557e-05, |
| "loss": 0.0099, |
| "step": 5100 |
| }, |
| { |
| "epoch": 8.940972222222221, |
| "grad_norm": 0.01604336127638817, |
| "learning_rate": 1.2848611111111112e-05, |
| "loss": 0.0226, |
| "step": 5150 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.08734692633152008, |
| "eval_runtime": 2.1655, |
| "eval_samples_per_second": 472.877, |
| "eval_steps_per_second": 29.555, |
| "step": 5184 |
| }, |
| { |
| "epoch": 9.027777777777779, |
| "grad_norm": 21.617219924926758, |
| "learning_rate": 1.2779166666666666e-05, |
| "loss": 0.0284, |
| "step": 5200 |
| }, |
| { |
| "epoch": 9.114583333333334, |
| "grad_norm": 0.0036318660713732243, |
| "learning_rate": 1.2709722222222222e-05, |
| "loss": 0.0242, |
| "step": 5250 |
| }, |
| { |
| "epoch": 9.20138888888889, |
| "grad_norm": 1.550016164779663, |
| "learning_rate": 1.2640277777777779e-05, |
| "loss": 0.0195, |
| "step": 5300 |
| }, |
| { |
| "epoch": 9.288194444444445, |
| "grad_norm": 0.0063721900805830956, |
| "learning_rate": 1.2570833333333335e-05, |
| "loss": 0.0326, |
| "step": 5350 |
| }, |
| { |
| "epoch": 9.375, |
| "grad_norm": 6.12042236328125, |
| "learning_rate": 1.250138888888889e-05, |
| "loss": 0.0153, |
| "step": 5400 |
| }, |
| { |
| "epoch": 9.461805555555555, |
| "grad_norm": 0.0024620750918984413, |
| "learning_rate": 1.2431944444444446e-05, |
| "loss": 0.0126, |
| "step": 5450 |
| }, |
| { |
| "epoch": 9.54861111111111, |
| "grad_norm": 0.008434666320681572, |
| "learning_rate": 1.2362500000000002e-05, |
| "loss": 0.0168, |
| "step": 5500 |
| }, |
| { |
| "epoch": 9.635416666666666, |
| "grad_norm": 24.517446517944336, |
| "learning_rate": 1.2293055555555557e-05, |
| "loss": 0.0236, |
| "step": 5550 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.0033647818490862846, |
| "learning_rate": 1.2223611111111112e-05, |
| "loss": 0.0128, |
| "step": 5600 |
| }, |
| { |
| "epoch": 9.809027777777779, |
| "grad_norm": 0.008964108303189278, |
| "learning_rate": 1.2154166666666666e-05, |
| "loss": 0.0064, |
| "step": 5650 |
| }, |
| { |
| "epoch": 9.895833333333334, |
| "grad_norm": 0.00735318660736084, |
| "learning_rate": 1.2084722222222223e-05, |
| "loss": 0.0069, |
| "step": 5700 |
| }, |
| { |
| "epoch": 9.98263888888889, |
| "grad_norm": 68.18215942382812, |
| "learning_rate": 1.2015277777777779e-05, |
| "loss": 0.0278, |
| "step": 5750 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.05086889490485191, |
| "eval_runtime": 2.1757, |
| "eval_samples_per_second": 470.664, |
| "eval_steps_per_second": 29.416, |
| "step": 5760 |
| }, |
| { |
| "epoch": 10.069444444444445, |
| "grad_norm": 0.008082223124802113, |
| "learning_rate": 1.1945833333333333e-05, |
| "loss": 0.0001, |
| "step": 5800 |
| }, |
| { |
| "epoch": 10.15625, |
| "grad_norm": 0.0015207990072667599, |
| "learning_rate": 1.187638888888889e-05, |
| "loss": 0.0023, |
| "step": 5850 |
| }, |
| { |
| "epoch": 10.243055555555555, |
| "grad_norm": 0.003384481882676482, |
| "learning_rate": 1.1806944444444446e-05, |
| "loss": 0.0039, |
| "step": 5900 |
| }, |
| { |
| "epoch": 10.32986111111111, |
| "grad_norm": 0.0017064920393750072, |
| "learning_rate": 1.1737500000000002e-05, |
| "loss": 0.0092, |
| "step": 5950 |
| }, |
| { |
| "epoch": 10.416666666666666, |
| "grad_norm": 0.0034557634498924017, |
| "learning_rate": 1.1668055555555557e-05, |
| "loss": 0.0291, |
| "step": 6000 |
| }, |
| { |
| "epoch": 10.503472222222221, |
| "grad_norm": 0.005785680841654539, |
| "learning_rate": 1.1598611111111112e-05, |
| "loss": 0.0002, |
| "step": 6050 |
| }, |
| { |
| "epoch": 10.590277777777779, |
| "grad_norm": 0.005150977522134781, |
| "learning_rate": 1.1529166666666666e-05, |
| "loss": 0.0209, |
| "step": 6100 |
| }, |
| { |
| "epoch": 10.677083333333334, |
| "grad_norm": 0.0023805610835552216, |
| "learning_rate": 1.1459722222222223e-05, |
| "loss": 0.0133, |
| "step": 6150 |
| }, |
| { |
| "epoch": 10.76388888888889, |
| "grad_norm": 0.0017482911935076118, |
| "learning_rate": 1.1390277777777779e-05, |
| "loss": 0.0034, |
| "step": 6200 |
| }, |
| { |
| "epoch": 10.850694444444445, |
| "grad_norm": 0.16960591077804565, |
| "learning_rate": 1.1320833333333334e-05, |
| "loss": 0.0023, |
| "step": 6250 |
| }, |
| { |
| "epoch": 10.9375, |
| "grad_norm": 0.005295192822813988, |
| "learning_rate": 1.125138888888889e-05, |
| "loss": 0.0018, |
| "step": 6300 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 0.09595341980457306, |
| "eval_runtime": 2.1636, |
| "eval_samples_per_second": 473.292, |
| "eval_steps_per_second": 29.581, |
| "step": 6336 |
| }, |
| { |
| "epoch": 11.024305555555555, |
| "grad_norm": 0.0025636740028858185, |
| "learning_rate": 1.1181944444444446e-05, |
| "loss": 0.0003, |
| "step": 6350 |
| }, |
| { |
| "epoch": 11.11111111111111, |
| "grad_norm": 0.001272220746614039, |
| "learning_rate": 1.1112500000000002e-05, |
| "loss": 0.0071, |
| "step": 6400 |
| }, |
| { |
| "epoch": 11.197916666666666, |
| "grad_norm": 0.00381342857144773, |
| "learning_rate": 1.1043055555555557e-05, |
| "loss": 0.0146, |
| "step": 6450 |
| }, |
| { |
| "epoch": 11.284722222222221, |
| "grad_norm": 0.001548461732454598, |
| "learning_rate": 1.0973611111111112e-05, |
| "loss": 0.012, |
| "step": 6500 |
| }, |
| { |
| "epoch": 11.371527777777779, |
| "grad_norm": 0.0014353245496749878, |
| "learning_rate": 1.0904166666666666e-05, |
| "loss": 0.0001, |
| "step": 6550 |
| }, |
| { |
| "epoch": 11.458333333333334, |
| "grad_norm": 0.0014444834087044, |
| "learning_rate": 1.0834722222222223e-05, |
| "loss": 0.02, |
| "step": 6600 |
| }, |
| { |
| "epoch": 11.54513888888889, |
| "grad_norm": 0.002948822919279337, |
| "learning_rate": 1.0765277777777779e-05, |
| "loss": 0.0005, |
| "step": 6650 |
| }, |
| { |
| "epoch": 11.631944444444445, |
| "grad_norm": 0.001776829012669623, |
| "learning_rate": 1.0695833333333334e-05, |
| "loss": 0.0138, |
| "step": 6700 |
| }, |
| { |
| "epoch": 11.71875, |
| "grad_norm": 0.0016048089601099491, |
| "learning_rate": 1.062638888888889e-05, |
| "loss": 0.0229, |
| "step": 6750 |
| }, |
| { |
| "epoch": 11.805555555555555, |
| "grad_norm": 0.0019577471539378166, |
| "learning_rate": 1.0556944444444446e-05, |
| "loss": 0.0318, |
| "step": 6800 |
| }, |
| { |
| "epoch": 11.89236111111111, |
| "grad_norm": 0.02152959071099758, |
| "learning_rate": 1.04875e-05, |
| "loss": 0.0147, |
| "step": 6850 |
| }, |
| { |
| "epoch": 11.979166666666666, |
| "grad_norm": 0.003166797338053584, |
| "learning_rate": 1.0418055555555557e-05, |
| "loss": 0.0002, |
| "step": 6900 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 0.0941697284579277, |
| "eval_runtime": 2.1655, |
| "eval_samples_per_second": 472.877, |
| "eval_steps_per_second": 29.555, |
| "step": 6912 |
| }, |
| { |
| "epoch": 12.065972222222221, |
| "grad_norm": 0.0014085551956668496, |
| "learning_rate": 1.0348611111111112e-05, |
| "loss": 0.0127, |
| "step": 6950 |
| }, |
| { |
| "epoch": 12.152777777777779, |
| "grad_norm": 0.0020881230011582375, |
| "learning_rate": 1.0279166666666666e-05, |
| "loss": 0.0056, |
| "step": 7000 |
| }, |
| { |
| "epoch": 12.239583333333334, |
| "grad_norm": 0.0022252153139561415, |
| "learning_rate": 1.0209722222222223e-05, |
| "loss": 0.0001, |
| "step": 7050 |
| }, |
| { |
| "epoch": 12.32638888888889, |
| "grad_norm": 0.014756396412849426, |
| "learning_rate": 1.0140277777777779e-05, |
| "loss": 0.0021, |
| "step": 7100 |
| }, |
| { |
| "epoch": 12.413194444444445, |
| "grad_norm": 0.0016215373761951923, |
| "learning_rate": 1.0070833333333334e-05, |
| "loss": 0.0017, |
| "step": 7150 |
| }, |
| { |
| "epoch": 12.5, |
| "grad_norm": 8.211623191833496, |
| "learning_rate": 1.000138888888889e-05, |
| "loss": 0.0081, |
| "step": 7200 |
| }, |
| { |
| "epoch": 12.586805555555555, |
| "grad_norm": 0.001198956393636763, |
| "learning_rate": 9.931944444444446e-06, |
| "loss": 0.0066, |
| "step": 7250 |
| }, |
| { |
| "epoch": 12.67361111111111, |
| "grad_norm": 0.09066519886255264, |
| "learning_rate": 9.862500000000001e-06, |
| "loss": 0.0109, |
| "step": 7300 |
| }, |
| { |
| "epoch": 12.760416666666666, |
| "grad_norm": 0.006189523730427027, |
| "learning_rate": 9.793055555555555e-06, |
| "loss": 0.0133, |
| "step": 7350 |
| }, |
| { |
| "epoch": 12.847222222222221, |
| "grad_norm": 0.0009115805733017623, |
| "learning_rate": 9.723611111111112e-06, |
| "loss": 0.0002, |
| "step": 7400 |
| }, |
| { |
| "epoch": 12.934027777777779, |
| "grad_norm": 0.001321441144682467, |
| "learning_rate": 9.654166666666668e-06, |
| "loss": 0.002, |
| "step": 7450 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_loss": 0.07037966698408127, |
| "eval_runtime": 2.158, |
| "eval_samples_per_second": 474.51, |
| "eval_steps_per_second": 29.657, |
| "step": 7488 |
| }, |
| { |
| "epoch": 13.020833333333334, |
| "grad_norm": 0.001222062623128295, |
| "learning_rate": 9.584722222222223e-06, |
| "loss": 0.0059, |
| "step": 7500 |
| }, |
| { |
| "epoch": 13.10763888888889, |
| "grad_norm": 0.0009936870774254203, |
| "learning_rate": 9.515277777777779e-06, |
| "loss": 0.0147, |
| "step": 7550 |
| }, |
| { |
| "epoch": 13.194444444444445, |
| "grad_norm": 0.0007664511213079095, |
| "learning_rate": 9.445833333333334e-06, |
| "loss": 0.0009, |
| "step": 7600 |
| }, |
| { |
| "epoch": 13.28125, |
| "grad_norm": 0.0019496489549055696, |
| "learning_rate": 9.37638888888889e-06, |
| "loss": 0.0034, |
| "step": 7650 |
| }, |
| { |
| "epoch": 13.368055555555555, |
| "grad_norm": 0.0010263716103509068, |
| "learning_rate": 9.306944444444446e-06, |
| "loss": 0.0157, |
| "step": 7700 |
| }, |
| { |
| "epoch": 13.45486111111111, |
| "grad_norm": 0.0018417923711240292, |
| "learning_rate": 9.237500000000001e-06, |
| "loss": 0.0191, |
| "step": 7750 |
| }, |
| { |
| "epoch": 13.541666666666666, |
| "grad_norm": 0.009017466567456722, |
| "learning_rate": 9.168055555555556e-06, |
| "loss": 0.0142, |
| "step": 7800 |
| }, |
| { |
| "epoch": 13.628472222222221, |
| "grad_norm": 0.00774754025042057, |
| "learning_rate": 9.098611111111112e-06, |
| "loss": 0.0213, |
| "step": 7850 |
| }, |
| { |
| "epoch": 13.715277777777779, |
| "grad_norm": 0.0038997160736471415, |
| "learning_rate": 9.029166666666668e-06, |
| "loss": 0.0006, |
| "step": 7900 |
| }, |
| { |
| "epoch": 13.802083333333334, |
| "grad_norm": 0.001874367124401033, |
| "learning_rate": 8.959722222222223e-06, |
| "loss": 0.0135, |
| "step": 7950 |
| }, |
| { |
| "epoch": 13.88888888888889, |
| "grad_norm": 0.0013082403456792235, |
| "learning_rate": 8.890277777777777e-06, |
| "loss": 0.005, |
| "step": 8000 |
| }, |
| { |
| "epoch": 13.975694444444445, |
| "grad_norm": 0.005922501441091299, |
| "learning_rate": 8.820833333333334e-06, |
| "loss": 0.008, |
| "step": 8050 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_loss": 0.11005319654941559, |
| "eval_runtime": 2.1582, |
| "eval_samples_per_second": 474.459, |
| "eval_steps_per_second": 29.654, |
| "step": 8064 |
| }, |
| { |
| "epoch": 14.0625, |
| "grad_norm": 0.000921198632568121, |
| "learning_rate": 8.75138888888889e-06, |
| "loss": 0.0178, |
| "step": 8100 |
| }, |
| { |
| "epoch": 14.149305555555555, |
| "grad_norm": 0.0010431046830490232, |
| "learning_rate": 8.681944444444446e-06, |
| "loss": 0.0002, |
| "step": 8150 |
| }, |
| { |
| "epoch": 14.23611111111111, |
| "grad_norm": 0.0012669609859585762, |
| "learning_rate": 8.612500000000001e-06, |
| "loss": 0.0031, |
| "step": 8200 |
| }, |
| { |
| "epoch": 14.322916666666666, |
| "grad_norm": 0.0029816783498972654, |
| "learning_rate": 8.543055555555556e-06, |
| "loss": 0.025, |
| "step": 8250 |
| }, |
| { |
| "epoch": 14.409722222222221, |
| "grad_norm": 0.005197410471737385, |
| "learning_rate": 8.473611111111112e-06, |
| "loss": 0.0016, |
| "step": 8300 |
| }, |
| { |
| "epoch": 14.496527777777779, |
| "grad_norm": 0.0018497951095923781, |
| "learning_rate": 8.404166666666668e-06, |
| "loss": 0.0002, |
| "step": 8350 |
| }, |
| { |
| "epoch": 14.583333333333334, |
| "grad_norm": 0.0028671796899288893, |
| "learning_rate": 8.334722222222223e-06, |
| "loss": 0.0012, |
| "step": 8400 |
| }, |
| { |
| "epoch": 14.67013888888889, |
| "grad_norm": 0.010854833759367466, |
| "learning_rate": 8.265277777777777e-06, |
| "loss": 0.0027, |
| "step": 8450 |
| }, |
| { |
| "epoch": 14.756944444444445, |
| "grad_norm": 0.011619815602898598, |
| "learning_rate": 8.195833333333334e-06, |
| "loss": 0.0051, |
| "step": 8500 |
| }, |
| { |
| "epoch": 14.84375, |
| "grad_norm": 0.004773242399096489, |
| "learning_rate": 8.12638888888889e-06, |
| "loss": 0.0203, |
| "step": 8550 |
| }, |
| { |
| "epoch": 14.930555555555555, |
| "grad_norm": 0.0017284239875152707, |
| "learning_rate": 8.056944444444446e-06, |
| "loss": 0.0084, |
| "step": 8600 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.09546509385108948, |
| "eval_runtime": 2.159, |
| "eval_samples_per_second": 474.301, |
| "eval_steps_per_second": 29.644, |
| "step": 8640 |
| }, |
| { |
| "epoch": 15.01736111111111, |
| "grad_norm": 0.003186532063409686, |
| "learning_rate": 7.987500000000001e-06, |
| "loss": 0.0044, |
| "step": 8650 |
| }, |
| { |
| "epoch": 15.104166666666666, |
| "grad_norm": 0.0046905651688575745, |
| "learning_rate": 7.918055555555556e-06, |
| "loss": 0.0003, |
| "step": 8700 |
| }, |
| { |
| "epoch": 15.190972222222221, |
| "grad_norm": 0.001585672376677394, |
| "learning_rate": 7.848611111111112e-06, |
| "loss": 0.0014, |
| "step": 8750 |
| }, |
| { |
| "epoch": 15.277777777777779, |
| "grad_norm": 0.001381274894811213, |
| "learning_rate": 7.779166666666668e-06, |
| "loss": 0.0002, |
| "step": 8800 |
| }, |
| { |
| "epoch": 15.364583333333334, |
| "grad_norm": 0.0018836313392966986, |
| "learning_rate": 7.709722222222223e-06, |
| "loss": 0.0002, |
| "step": 8850 |
| }, |
| { |
| "epoch": 15.45138888888889, |
| "grad_norm": 0.0009360660915262997, |
| "learning_rate": 7.640277777777778e-06, |
| "loss": 0.0009, |
| "step": 8900 |
| }, |
| { |
| "epoch": 15.538194444444445, |
| "grad_norm": 0.0008543449803255498, |
| "learning_rate": 7.570833333333334e-06, |
| "loss": 0.0001, |
| "step": 8950 |
| }, |
| { |
| "epoch": 15.625, |
| "grad_norm": 0.38564541935920715, |
| "learning_rate": 7.501388888888889e-06, |
| "loss": 0.0148, |
| "step": 9000 |
| }, |
| { |
| "epoch": 15.711805555555555, |
| "grad_norm": 0.0016453195130452514, |
| "learning_rate": 7.431944444444446e-06, |
| "loss": 0.0103, |
| "step": 9050 |
| }, |
| { |
| "epoch": 15.79861111111111, |
| "grad_norm": 0.001604390563443303, |
| "learning_rate": 7.3625e-06, |
| "loss": 0.0039, |
| "step": 9100 |
| }, |
| { |
| "epoch": 15.885416666666666, |
| "grad_norm": 0.0020437692292034626, |
| "learning_rate": 7.293055555555556e-06, |
| "loss": 0.0001, |
| "step": 9150 |
| }, |
| { |
| "epoch": 15.972222222222221, |
| "grad_norm": 0.004224094562232494, |
| "learning_rate": 7.223611111111112e-06, |
| "loss": 0.0024, |
| "step": 9200 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_loss": 0.057210568338632584, |
| "eval_runtime": 2.1916, |
| "eval_samples_per_second": 467.23, |
| "eval_steps_per_second": 29.202, |
| "step": 9216 |
| }, |
| { |
| "epoch": 16.05902777777778, |
| "grad_norm": 0.008274748921394348, |
| "learning_rate": 7.1541666666666675e-06, |
| "loss": 0.0002, |
| "step": 9250 |
| }, |
| { |
| "epoch": 16.145833333333332, |
| "grad_norm": 0.0013211554614827037, |
| "learning_rate": 7.084722222222222e-06, |
| "loss": 0.0051, |
| "step": 9300 |
| }, |
| { |
| "epoch": 16.23263888888889, |
| "grad_norm": 0.0025199875235557556, |
| "learning_rate": 7.015277777777778e-06, |
| "loss": 0.0001, |
| "step": 9350 |
| }, |
| { |
| "epoch": 16.319444444444443, |
| "grad_norm": 0.00225885515101254, |
| "learning_rate": 6.945833333333334e-06, |
| "loss": 0.0144, |
| "step": 9400 |
| }, |
| { |
| "epoch": 16.40625, |
| "grad_norm": 8.875648498535156, |
| "learning_rate": 6.876388888888889e-06, |
| "loss": 0.0097, |
| "step": 9450 |
| }, |
| { |
| "epoch": 16.493055555555557, |
| "grad_norm": 0.0012961571337655187, |
| "learning_rate": 6.806944444444446e-06, |
| "loss": 0.0008, |
| "step": 9500 |
| }, |
| { |
| "epoch": 16.57986111111111, |
| "grad_norm": 0.0071833692491054535, |
| "learning_rate": 6.7375e-06, |
| "loss": 0.0107, |
| "step": 9550 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 0.0016137856291607022, |
| "learning_rate": 6.668055555555556e-06, |
| "loss": 0.011, |
| "step": 9600 |
| }, |
| { |
| "epoch": 16.75347222222222, |
| "grad_norm": 0.002170548541471362, |
| "learning_rate": 6.598611111111112e-06, |
| "loss": 0.0002, |
| "step": 9650 |
| }, |
| { |
| "epoch": 16.84027777777778, |
| "grad_norm": 0.15523186326026917, |
| "learning_rate": 6.5291666666666675e-06, |
| "loss": 0.0001, |
| "step": 9700 |
| }, |
| { |
| "epoch": 16.927083333333332, |
| "grad_norm": 0.0006897877901792526, |
| "learning_rate": 6.459722222222222e-06, |
| "loss": 0.0, |
| "step": 9750 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_loss": 0.07502129673957825, |
| "eval_runtime": 2.1485, |
| "eval_samples_per_second": 476.611, |
| "eval_steps_per_second": 29.788, |
| "step": 9792 |
| }, |
| { |
| "epoch": 17.01388888888889, |
| "grad_norm": 0.007862378843128681, |
| "learning_rate": 6.390277777777778e-06, |
| "loss": 0.0072, |
| "step": 9800 |
| }, |
| { |
| "epoch": 17.100694444444443, |
| "grad_norm": 0.0048260875046253204, |
| "learning_rate": 6.320833333333334e-06, |
| "loss": 0.0001, |
| "step": 9850 |
| }, |
| { |
| "epoch": 17.1875, |
| "grad_norm": 0.001168318442068994, |
| "learning_rate": 6.251388888888889e-06, |
| "loss": 0.0001, |
| "step": 9900 |
| }, |
| { |
| "epoch": 17.274305555555557, |
| "grad_norm": 0.0010404903441667557, |
| "learning_rate": 6.181944444444446e-06, |
| "loss": 0.0, |
| "step": 9950 |
| }, |
| { |
| "epoch": 17.36111111111111, |
| "grad_norm": 0.0008959461702033877, |
| "learning_rate": 6.1125e-06, |
| "loss": 0.0, |
| "step": 10000 |
| }, |
| { |
| "epoch": 17.447916666666668, |
| "grad_norm": 0.0006551714614033699, |
| "learning_rate": 6.043055555555556e-06, |
| "loss": 0.0083, |
| "step": 10050 |
| }, |
| { |
| "epoch": 17.53472222222222, |
| "grad_norm": 0.004672654904425144, |
| "learning_rate": 5.973611111111111e-06, |
| "loss": 0.0053, |
| "step": 10100 |
| }, |
| { |
| "epoch": 17.62152777777778, |
| "grad_norm": 0.012874463573098183, |
| "learning_rate": 5.904166666666668e-06, |
| "loss": 0.0024, |
| "step": 10150 |
| }, |
| { |
| "epoch": 17.708333333333332, |
| "grad_norm": 0.0007252011564560235, |
| "learning_rate": 5.834722222222222e-06, |
| "loss": 0.0081, |
| "step": 10200 |
| }, |
| { |
| "epoch": 17.79513888888889, |
| "grad_norm": 0.003966485150158405, |
| "learning_rate": 5.765277777777778e-06, |
| "loss": 0.0001, |
| "step": 10250 |
| }, |
| { |
| "epoch": 17.881944444444443, |
| "grad_norm": 0.004643677733838558, |
| "learning_rate": 5.695833333333334e-06, |
| "loss": 0.0016, |
| "step": 10300 |
| }, |
| { |
| "epoch": 17.96875, |
| "grad_norm": 0.0022119064815342426, |
| "learning_rate": 5.6263888888888895e-06, |
| "loss": 0.0, |
| "step": 10350 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_loss": 0.081531822681427, |
| "eval_runtime": 2.1481, |
| "eval_samples_per_second": 476.707, |
| "eval_steps_per_second": 29.794, |
| "step": 10368 |
| }, |
| { |
| "epoch": 18.055555555555557, |
| "grad_norm": 0.0007135890191420913, |
| "learning_rate": 5.556944444444446e-06, |
| "loss": 0.0008, |
| "step": 10400 |
| }, |
| { |
| "epoch": 18.14236111111111, |
| "grad_norm": 0.0011914368951693177, |
| "learning_rate": 5.4875e-06, |
| "loss": 0.0029, |
| "step": 10450 |
| }, |
| { |
| "epoch": 18.229166666666668, |
| "grad_norm": 0.000494650739710778, |
| "learning_rate": 5.418055555555556e-06, |
| "loss": 0.0, |
| "step": 10500 |
| }, |
| { |
| "epoch": 18.31597222222222, |
| "grad_norm": 0.0006326820584945381, |
| "learning_rate": 5.348611111111111e-06, |
| "loss": 0.0005, |
| "step": 10550 |
| }, |
| { |
| "epoch": 18.40277777777778, |
| "grad_norm": 0.0005152356461621821, |
| "learning_rate": 5.279166666666668e-06, |
| "loss": 0.0, |
| "step": 10600 |
| }, |
| { |
| "epoch": 18.489583333333332, |
| "grad_norm": 0.001848602551035583, |
| "learning_rate": 5.209722222222222e-06, |
| "loss": 0.0001, |
| "step": 10650 |
| }, |
| { |
| "epoch": 18.57638888888889, |
| "grad_norm": 0.00047650947817601264, |
| "learning_rate": 5.140277777777778e-06, |
| "loss": 0.014, |
| "step": 10700 |
| }, |
| { |
| "epoch": 18.663194444444443, |
| "grad_norm": 0.0004927313420921564, |
| "learning_rate": 5.070833333333334e-06, |
| "loss": 0.0101, |
| "step": 10750 |
| }, |
| { |
| "epoch": 18.75, |
| "grad_norm": 0.0006253819447010756, |
| "learning_rate": 5.0013888888888895e-06, |
| "loss": 0.0, |
| "step": 10800 |
| }, |
| { |
| "epoch": 18.836805555555557, |
| "grad_norm": 0.0005877416697330773, |
| "learning_rate": 4.931944444444445e-06, |
| "loss": 0.0, |
| "step": 10850 |
| }, |
| { |
| "epoch": 18.92361111111111, |
| "grad_norm": 0.000749260769225657, |
| "learning_rate": 4.8625000000000005e-06, |
| "loss": 0.0034, |
| "step": 10900 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_loss": 0.08709079027175903, |
| "eval_runtime": 2.1461, |
| "eval_samples_per_second": 477.135, |
| "eval_steps_per_second": 29.821, |
| "step": 10944 |
| }, |
| { |
| "epoch": 19.010416666666668, |
| "grad_norm": 0.0033058973494917154, |
| "learning_rate": 4.793055555555556e-06, |
| "loss": 0.0028, |
| "step": 10950 |
| }, |
| { |
| "epoch": 19.09722222222222, |
| "grad_norm": 0.0005344200180843472, |
| "learning_rate": 4.723611111111111e-06, |
| "loss": 0.0012, |
| "step": 11000 |
| }, |
| { |
| "epoch": 19.18402777777778, |
| "grad_norm": 0.0011495526414364576, |
| "learning_rate": 4.654166666666667e-06, |
| "loss": 0.0, |
| "step": 11050 |
| }, |
| { |
| "epoch": 19.270833333333332, |
| "grad_norm": 0.0007638471433892846, |
| "learning_rate": 4.584722222222222e-06, |
| "loss": 0.0008, |
| "step": 11100 |
| }, |
| { |
| "epoch": 19.35763888888889, |
| "grad_norm": 0.0016245943261310458, |
| "learning_rate": 4.515277777777778e-06, |
| "loss": 0.0, |
| "step": 11150 |
| }, |
| { |
| "epoch": 19.444444444444443, |
| "grad_norm": 0.001350992708466947, |
| "learning_rate": 4.445833333333333e-06, |
| "loss": 0.0006, |
| "step": 11200 |
| }, |
| { |
| "epoch": 19.53125, |
| "grad_norm": 0.0007238370017148554, |
| "learning_rate": 4.3763888888888896e-06, |
| "loss": 0.0, |
| "step": 11250 |
| }, |
| { |
| "epoch": 19.618055555555557, |
| "grad_norm": 0.0009173430735245347, |
| "learning_rate": 4.306944444444445e-06, |
| "loss": 0.005, |
| "step": 11300 |
| }, |
| { |
| "epoch": 19.70486111111111, |
| "grad_norm": 0.002093716524541378, |
| "learning_rate": 4.2375000000000005e-06, |
| "loss": 0.0, |
| "step": 11350 |
| }, |
| { |
| "epoch": 19.791666666666668, |
| "grad_norm": 0.0008814275497570634, |
| "learning_rate": 4.168055555555556e-06, |
| "loss": 0.0004, |
| "step": 11400 |
| }, |
| { |
| "epoch": 19.87847222222222, |
| "grad_norm": 0.000544120033737272, |
| "learning_rate": 4.0986111111111114e-06, |
| "loss": 0.0001, |
| "step": 11450 |
| }, |
| { |
| "epoch": 19.96527777777778, |
| "grad_norm": 0.0006753376801498234, |
| "learning_rate": 4.029166666666667e-06, |
| "loss": 0.0078, |
| "step": 11500 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.13657595217227936, |
| "eval_runtime": 2.1501, |
| "eval_samples_per_second": 476.265, |
| "eval_steps_per_second": 29.767, |
| "step": 11520 |
| }, |
| { |
| "epoch": 20.052083333333332, |
| "grad_norm": 0.000810706231277436, |
| "learning_rate": 3.959722222222222e-06, |
| "loss": 0.0, |
| "step": 11550 |
| }, |
| { |
| "epoch": 20.13888888888889, |
| "grad_norm": 0.0008722566999495029, |
| "learning_rate": 3.890277777777778e-06, |
| "loss": 0.002, |
| "step": 11600 |
| }, |
| { |
| "epoch": 20.225694444444443, |
| "grad_norm": 0.0010332430247217417, |
| "learning_rate": 3.820833333333333e-06, |
| "loss": 0.0, |
| "step": 11650 |
| }, |
| { |
| "epoch": 20.3125, |
| "grad_norm": 0.0003171579446643591, |
| "learning_rate": 3.751388888888889e-06, |
| "loss": 0.0001, |
| "step": 11700 |
| }, |
| { |
| "epoch": 20.399305555555557, |
| "grad_norm": 0.0003055994166061282, |
| "learning_rate": 3.6819444444444447e-06, |
| "loss": 0.0, |
| "step": 11750 |
| }, |
| { |
| "epoch": 20.48611111111111, |
| "grad_norm": 0.0005060540861450136, |
| "learning_rate": 3.6125000000000006e-06, |
| "loss": 0.0, |
| "step": 11800 |
| }, |
| { |
| "epoch": 20.572916666666668, |
| "grad_norm": 0.0005677157896570861, |
| "learning_rate": 3.5430555555555556e-06, |
| "loss": 0.0, |
| "step": 11850 |
| }, |
| { |
| "epoch": 20.65972222222222, |
| "grad_norm": 0.000569184310734272, |
| "learning_rate": 3.4736111111111115e-06, |
| "loss": 0.0011, |
| "step": 11900 |
| }, |
| { |
| "epoch": 20.74652777777778, |
| "grad_norm": 0.0008372145821340382, |
| "learning_rate": 3.4041666666666665e-06, |
| "loss": 0.0, |
| "step": 11950 |
| }, |
| { |
| "epoch": 20.833333333333332, |
| "grad_norm": 0.0009117226582020521, |
| "learning_rate": 3.3347222222222224e-06, |
| "loss": 0.0027, |
| "step": 12000 |
| }, |
| { |
| "epoch": 20.92013888888889, |
| "grad_norm": 0.0006149787222966552, |
| "learning_rate": 3.265277777777778e-06, |
| "loss": 0.0009, |
| "step": 12050 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_loss": 0.07403512299060822, |
| "eval_runtime": 2.1707, |
| "eval_samples_per_second": 471.738, |
| "eval_steps_per_second": 29.484, |
| "step": 12096 |
| }, |
| { |
| "epoch": 21.006944444444443, |
| "grad_norm": 0.0009870927315205336, |
| "learning_rate": 3.1958333333333334e-06, |
| "loss": 0.0133, |
| "step": 12100 |
| }, |
| { |
| "epoch": 21.09375, |
| "grad_norm": 0.008030719123780727, |
| "learning_rate": 3.1263888888888893e-06, |
| "loss": 0.0, |
| "step": 12150 |
| }, |
| { |
| "epoch": 21.180555555555557, |
| "grad_norm": 0.0010351603850722313, |
| "learning_rate": 3.0569444444444447e-06, |
| "loss": 0.0025, |
| "step": 12200 |
| }, |
| { |
| "epoch": 21.26736111111111, |
| "grad_norm": 0.004166141152381897, |
| "learning_rate": 2.9875e-06, |
| "loss": 0.0023, |
| "step": 12250 |
| }, |
| { |
| "epoch": 21.354166666666668, |
| "grad_norm": 0.0021667364053428173, |
| "learning_rate": 2.9180555555555557e-06, |
| "loss": 0.005, |
| "step": 12300 |
| }, |
| { |
| "epoch": 21.44097222222222, |
| "grad_norm": 0.005213022232055664, |
| "learning_rate": 2.8486111111111116e-06, |
| "loss": 0.0001, |
| "step": 12350 |
| }, |
| { |
| "epoch": 21.52777777777778, |
| "grad_norm": 0.0016998907085508108, |
| "learning_rate": 2.7791666666666666e-06, |
| "loss": 0.0, |
| "step": 12400 |
| }, |
| { |
| "epoch": 21.614583333333332, |
| "grad_norm": 0.0011233899276703596, |
| "learning_rate": 2.7097222222222225e-06, |
| "loss": 0.0013, |
| "step": 12450 |
| }, |
| { |
| "epoch": 21.70138888888889, |
| "grad_norm": 0.00541352853178978, |
| "learning_rate": 2.6402777777777775e-06, |
| "loss": 0.0, |
| "step": 12500 |
| }, |
| { |
| "epoch": 21.788194444444443, |
| "grad_norm": 0.0006274741608649492, |
| "learning_rate": 2.5708333333333334e-06, |
| "loss": 0.0007, |
| "step": 12550 |
| }, |
| { |
| "epoch": 21.875, |
| "grad_norm": 0.000486269302200526, |
| "learning_rate": 2.5013888888888893e-06, |
| "loss": 0.0, |
| "step": 12600 |
| }, |
| { |
| "epoch": 21.961805555555557, |
| "grad_norm": 0.000523523660376668, |
| "learning_rate": 2.4319444444444444e-06, |
| "loss": 0.0005, |
| "step": 12650 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_loss": 0.09000910818576813, |
| "eval_runtime": 2.1668, |
| "eval_samples_per_second": 472.584, |
| "eval_steps_per_second": 29.537, |
| "step": 12672 |
| }, |
| { |
| "epoch": 22.04861111111111, |
| "grad_norm": 0.0002973914088215679, |
| "learning_rate": 2.3625000000000003e-06, |
| "loss": 0.0, |
| "step": 12700 |
| }, |
| { |
| "epoch": 22.135416666666668, |
| "grad_norm": 0.00037905474891886115, |
| "learning_rate": 2.2930555555555557e-06, |
| "loss": 0.0003, |
| "step": 12750 |
| }, |
| { |
| "epoch": 22.22222222222222, |
| "grad_norm": 0.0003634969179984182, |
| "learning_rate": 2.223611111111111e-06, |
| "loss": 0.0, |
| "step": 12800 |
| }, |
| { |
| "epoch": 22.30902777777778, |
| "grad_norm": 0.0003171905700583011, |
| "learning_rate": 2.154166666666667e-06, |
| "loss": 0.0, |
| "step": 12850 |
| }, |
| { |
| "epoch": 22.395833333333332, |
| "grad_norm": 0.0005504607688635588, |
| "learning_rate": 2.0847222222222225e-06, |
| "loss": 0.0037, |
| "step": 12900 |
| }, |
| { |
| "epoch": 22.48263888888889, |
| "grad_norm": 0.001255808281712234, |
| "learning_rate": 2.015277777777778e-06, |
| "loss": 0.0025, |
| "step": 12950 |
| }, |
| { |
| "epoch": 22.569444444444443, |
| "grad_norm": 0.0004096803313586861, |
| "learning_rate": 1.9458333333333335e-06, |
| "loss": 0.0021, |
| "step": 13000 |
| }, |
| { |
| "epoch": 22.65625, |
| "grad_norm": 0.001167616923339665, |
| "learning_rate": 1.876388888888889e-06, |
| "loss": 0.0, |
| "step": 13050 |
| }, |
| { |
| "epoch": 22.743055555555557, |
| "grad_norm": 0.0003252147580496967, |
| "learning_rate": 1.8069444444444444e-06, |
| "loss": 0.0006, |
| "step": 13100 |
| }, |
| { |
| "epoch": 22.82986111111111, |
| "grad_norm": 0.0006310039316304028, |
| "learning_rate": 1.7375e-06, |
| "loss": 0.0001, |
| "step": 13150 |
| }, |
| { |
| "epoch": 22.916666666666668, |
| "grad_norm": 0.00035475249751470983, |
| "learning_rate": 1.6680555555555558e-06, |
| "loss": 0.0026, |
| "step": 13200 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_loss": 0.0978098213672638, |
| "eval_runtime": 2.1724, |
| "eval_samples_per_second": 471.367, |
| "eval_steps_per_second": 29.46, |
| "step": 13248 |
| }, |
| { |
| "epoch": 23.00347222222222, |
| "grad_norm": 0.0005200915038585663, |
| "learning_rate": 1.5986111111111112e-06, |
| "loss": 0.0, |
| "step": 13250 |
| }, |
| { |
| "epoch": 23.09027777777778, |
| "grad_norm": 0.622466504573822, |
| "learning_rate": 1.529166666666667e-06, |
| "loss": 0.0003, |
| "step": 13300 |
| }, |
| { |
| "epoch": 23.177083333333332, |
| "grad_norm": 0.0003293896734248847, |
| "learning_rate": 1.4597222222222224e-06, |
| "loss": 0.0024, |
| "step": 13350 |
| }, |
| { |
| "epoch": 23.26388888888889, |
| "grad_norm": 0.00028986833058297634, |
| "learning_rate": 1.3902777777777779e-06, |
| "loss": 0.0, |
| "step": 13400 |
| }, |
| { |
| "epoch": 23.350694444444443, |
| "grad_norm": 0.00039684175862930715, |
| "learning_rate": 1.3208333333333333e-06, |
| "loss": 0.0004, |
| "step": 13450 |
| }, |
| { |
| "epoch": 23.4375, |
| "grad_norm": 0.0009431451908312738, |
| "learning_rate": 1.251388888888889e-06, |
| "loss": 0.0, |
| "step": 13500 |
| }, |
| { |
| "epoch": 23.524305555555557, |
| "grad_norm": 0.0009446613257750869, |
| "learning_rate": 1.1819444444444447e-06, |
| "loss": 0.0, |
| "step": 13550 |
| }, |
| { |
| "epoch": 23.61111111111111, |
| "grad_norm": 0.000527396856341511, |
| "learning_rate": 1.1125000000000001e-06, |
| "loss": 0.0, |
| "step": 13600 |
| }, |
| { |
| "epoch": 23.697916666666668, |
| "grad_norm": 0.0008534564403817058, |
| "learning_rate": 1.0430555555555556e-06, |
| "loss": 0.0, |
| "step": 13650 |
| }, |
| { |
| "epoch": 23.78472222222222, |
| "grad_norm": 0.000303189066471532, |
| "learning_rate": 9.73611111111111e-07, |
| "loss": 0.0, |
| "step": 13700 |
| }, |
| { |
| "epoch": 23.87152777777778, |
| "grad_norm": 0.0006793912034481764, |
| "learning_rate": 9.041666666666668e-07, |
| "loss": 0.0069, |
| "step": 13750 |
| }, |
| { |
| "epoch": 23.958333333333332, |
| "grad_norm": 0.00035732006654143333, |
| "learning_rate": 8.347222222222223e-07, |
| "loss": 0.0, |
| "step": 13800 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_loss": 0.09862537682056427, |
| "eval_runtime": 2.1709, |
| "eval_samples_per_second": 471.693, |
| "eval_steps_per_second": 29.481, |
| "step": 13824 |
| }, |
| { |
| "epoch": 24.04513888888889, |
| "grad_norm": 0.000883078551851213, |
| "learning_rate": 7.652777777777778e-07, |
| "loss": 0.0, |
| "step": 13850 |
| }, |
| { |
| "epoch": 24.131944444444443, |
| "grad_norm": 0.00040300763794220984, |
| "learning_rate": 6.958333333333334e-07, |
| "loss": 0.0005, |
| "step": 13900 |
| }, |
| { |
| "epoch": 24.21875, |
| "grad_norm": 0.001412940793670714, |
| "learning_rate": 6.263888888888888e-07, |
| "loss": 0.0017, |
| "step": 13950 |
| }, |
| { |
| "epoch": 24.305555555555557, |
| "grad_norm": 0.000599319173488766, |
| "learning_rate": 5.569444444444444e-07, |
| "loss": 0.0, |
| "step": 14000 |
| }, |
| { |
| "epoch": 24.39236111111111, |
| "grad_norm": 0.0007638942333869636, |
| "learning_rate": 4.875000000000001e-07, |
| "loss": 0.0, |
| "step": 14050 |
| }, |
| { |
| "epoch": 24.479166666666668, |
| "grad_norm": 0.00034263054840266705, |
| "learning_rate": 4.1805555555555556e-07, |
| "loss": 0.0009, |
| "step": 14100 |
| }, |
| { |
| "epoch": 24.56597222222222, |
| "grad_norm": 0.0004652358475141227, |
| "learning_rate": 3.4861111111111114e-07, |
| "loss": 0.0015, |
| "step": 14150 |
| }, |
| { |
| "epoch": 24.65277777777778, |
| "grad_norm": 0.00027831687475554645, |
| "learning_rate": 2.7916666666666666e-07, |
| "loss": 0.0, |
| "step": 14200 |
| }, |
| { |
| "epoch": 24.739583333333332, |
| "grad_norm": 0.0002827314310707152, |
| "learning_rate": 2.0972222222222223e-07, |
| "loss": 0.0015, |
| "step": 14250 |
| }, |
| { |
| "epoch": 24.82638888888889, |
| "grad_norm": 0.00031116604804992676, |
| "learning_rate": 1.4027777777777778e-07, |
| "loss": 0.0, |
| "step": 14300 |
| }, |
| { |
| "epoch": 24.913194444444443, |
| "grad_norm": 0.00027885418967343867, |
| "learning_rate": 7.083333333333334e-08, |
| "loss": 0.0018, |
| "step": 14350 |
| }, |
| { |
| "epoch": 25.0, |
| "grad_norm": 0.0005058420938439667, |
| "learning_rate": 1.388888888888889e-09, |
| "loss": 0.0005, |
| "step": 14400 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_loss": 0.09845860302448273, |
| "eval_runtime": 2.17, |
| "eval_samples_per_second": 471.882, |
| "eval_steps_per_second": 29.493, |
| "step": 14400 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 14400, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 25, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1377061586400000.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
