| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9998665421059656, | |
| "global_step": 3746, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0002, | |
| "loss": 10.6012, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0004, | |
| "loss": 3.651, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00059, | |
| "loss": 3.3911, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0007491827839825661, | |
| "loss": 3.0481, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0007453010078997548, | |
| "loss": 3.4039, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1260.1628, | |
| "eval_samples_per_second": 21.123, | |
| "eval_steps_per_second": 0.66, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0007412149278125851, | |
| "loss": 3.0553, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0007371288477254155, | |
| "loss": 3.2191, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0007332470716426042, | |
| "loss": 3.4094, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0007291609915554345, | |
| "loss": 3.0607, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0007252792154726233, | |
| "loss": 3.4396, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1235.1686, | |
| "eval_samples_per_second": 21.55, | |
| "eval_steps_per_second": 0.674, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0007211931353854536, | |
| "loss": 3.1198, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0007171070552982838, | |
| "loss": 3.1801, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0007132252792154726, | |
| "loss": 3.3954, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.000709139199128303, | |
| "loss": 3.0566, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0007052574230454917, | |
| "loss": 3.483, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1262.2253, | |
| "eval_samples_per_second": 21.088, | |
| "eval_steps_per_second": 0.659, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.000701171342958322, | |
| "loss": 3.1616, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0006970852628711522, | |
| "loss": 3.2017, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0006932034867883411, | |
| "loss": 3.4474, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0006891174067011714, | |
| "loss": 3.0656, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0006852356306183601, | |
| "loss": 3.5014, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1255.7751, | |
| "eval_samples_per_second": 21.196, | |
| "eval_steps_per_second": 0.663, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0006811495505311904, | |
| "loss": 3.1757, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0006770634704440206, | |
| "loss": 3.2038, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0006731816943612095, | |
| "loss": 3.4072, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0006690956142740398, | |
| "loss": 3.0498, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00066500953418687, | |
| "loss": 3.331, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1239.2922, | |
| "eval_samples_per_second": 21.478, | |
| "eval_steps_per_second": 0.671, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0006609234540997004, | |
| "loss": 3.0833, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0006568373740125306, | |
| "loss": 3.1989, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0006531599019340779, | |
| "loss": 3.5001, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0006490738218469082, | |
| "loss": 3.0685, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.000645192045764097, | |
| "loss": 3.4809, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1232.3119, | |
| "eval_samples_per_second": 21.6, | |
| "eval_steps_per_second": 0.675, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0006411059656769273, | |
| "loss": 3.1877, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0006370198855897576, | |
| "loss": 3.2365, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0006333424135113049, | |
| "loss": 3.67, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0006292563334241351, | |
| "loss": 3.086, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0006253745573413239, | |
| "loss": 3.4678, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.8236, | |
| "eval_samples_per_second": 21.521, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0006212884772541542, | |
| "loss": 3.2344, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0006172023971669845, | |
| "loss": 3.2507, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0006131163170798147, | |
| "loss": 3.4283, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0006090302369926451, | |
| "loss": 3.0802, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0006051484609098339, | |
| "loss": 3.4596, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1233.785, | |
| "eval_samples_per_second": 21.574, | |
| "eval_steps_per_second": 0.674, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0006010623808226641, | |
| "loss": 3.2227, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0005969763007354945, | |
| "loss": 3.2357, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0005930945246526832, | |
| "loss": 3.5563, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0005890084445655135, | |
| "loss": 3.0845, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0005851266684827023, | |
| "loss": 3.4644, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1235.1488, | |
| "eval_samples_per_second": 21.55, | |
| "eval_steps_per_second": 0.674, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0005810405883955325, | |
| "loss": 3.2092, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0005769545083083629, | |
| "loss": 3.2157, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0005730727322255517, | |
| "loss": 3.4679, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0005689866521383819, | |
| "loss": 3.0662, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0005651048760555707, | |
| "loss": 3.4671, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1241.2249, | |
| "eval_samples_per_second": 21.445, | |
| "eval_steps_per_second": 0.67, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0005610187959684009, | |
| "loss": 3.2428, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0005569327158812314, | |
| "loss": 3.2165, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0005530509397984201, | |
| "loss": 3.5106, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0005489648597112503, | |
| "loss": 3.1137, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0005452873876327976, | |
| "loss": 3.6005, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.2955, | |
| "eval_samples_per_second": 21.53, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0005412013075456279, | |
| "loss": 3.2702, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0005371152274584582, | |
| "loss": 3.6768, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.000533233451375647, | |
| "loss": 3.6313, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0005291473712884773, | |
| "loss": 3.2456, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.000525265595205666, | |
| "loss": 3.9182, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1229.4316, | |
| "eval_samples_per_second": 21.651, | |
| "eval_steps_per_second": 0.677, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0005211795151184963, | |
| "loss": 3.3805, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0005170934350313267, | |
| "loss": 3.4687, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.000513007354944157, | |
| "loss": 3.4473, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0005089212748569872, | |
| "loss": 3.2232, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.000505039498774176, | |
| "loss": 3.6466, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1237.3739, | |
| "eval_samples_per_second": 21.512, | |
| "eval_steps_per_second": 0.672, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0005009534186870062, | |
| "loss": 3.3496, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0004968673385998365, | |
| "loss": 3.3918, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0004927812585126669, | |
| "loss": 3.3232, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0004886951784254972, | |
| "loss": 3.288, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00048481340234268587, | |
| "loss": 3.6932, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.694, | |
| "eval_samples_per_second": 21.524, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00048072732225551624, | |
| "loss": 3.5411, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0004766412421683465, | |
| "loss": 3.5658, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00047275946608553524, | |
| "loss": 3.7666, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0004686733859983656, | |
| "loss": 3.7564, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00046479160991555436, | |
| "loss": 3.7939, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1235.7295, | |
| "eval_samples_per_second": 21.54, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00046070552982838467, | |
| "loss": 3.7471, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00045661944974121493, | |
| "loss": 3.7457, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0004527376736584037, | |
| "loss": 3.798, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00044865159357123405, | |
| "loss": 3.7611, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0004447698174884228, | |
| "loss": 3.9284, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1251.8158, | |
| "eval_samples_per_second": 21.264, | |
| "eval_steps_per_second": 0.665, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0004406837374012531, | |
| "loss": 3.7416, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00043659765731408337, | |
| "loss": 3.7416, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0004327158812312721, | |
| "loss": 3.8024, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0004286298011441025, | |
| "loss": 3.7578, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0004247480250612912, | |
| "loss": 3.7859, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1235.734, | |
| "eval_samples_per_second": 21.54, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0004206619449741215, | |
| "loss": 3.7427, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0004165758648869518, | |
| "loss": 3.751, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00041248978479978206, | |
| "loss": 3.6701, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0004084037047126123, | |
| "loss": 3.7613, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00040472623263415966, | |
| "loss": 3.9363, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.9158, | |
| "eval_samples_per_second": 21.52, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0004006401525469899, | |
| "loss": 3.7464, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00039655407245982023, | |
| "loss": 3.7414, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.000392672296377009, | |
| "loss": 3.8036, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00038879052029419783, | |
| "loss": 3.8768, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.0003847044402070281, | |
| "loss": 3.7573, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1232.3535, | |
| "eval_samples_per_second": 21.599, | |
| "eval_steps_per_second": 0.675, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00038061836011985835, | |
| "loss": 3.6959, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00037653228003268867, | |
| "loss": 3.7313, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.00037244619994551893, | |
| "loss": 3.7676, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0003685644238627077, | |
| "loss": 3.7872, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.000364478343775538, | |
| "loss": 3.7553, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1239.2247, | |
| "eval_samples_per_second": 21.48, | |
| "eval_steps_per_second": 0.671, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0003603922636883683, | |
| "loss": 3.694, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0003565104876055571, | |
| "loss": 3.8389, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.00035242440751838736, | |
| "loss": 3.7573, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0003483383274312177, | |
| "loss": 3.6641, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00034425224734404794, | |
| "loss": 3.7606, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1237.7997, | |
| "eval_samples_per_second": 21.504, | |
| "eval_steps_per_second": 0.672, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00034016616725687825, | |
| "loss": 3.6827, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.000336284391174067, | |
| "loss": 3.8375, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0003321983110868973, | |
| "loss": 3.7527, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0003283165350040861, | |
| "loss": 3.7881, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0003242304549169164, | |
| "loss": 3.7514, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1237.076, | |
| "eval_samples_per_second": 21.517, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0003201443748297467, | |
| "loss": 3.6727, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00031626259874693543, | |
| "loss": 3.8424, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00031217651865976574, | |
| "loss": 3.7594, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00030829474257695454, | |
| "loss": 3.7862, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0003042086624897848, | |
| "loss": 3.7472, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.6319, | |
| "eval_samples_per_second": 21.525, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0003001225824026151, | |
| "loss": 3.6956, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0002960365023154454, | |
| "loss": 3.7248, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.0002919504222282757, | |
| "loss": 3.7531, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.0002882729501498229, | |
| "loss": 3.9245, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00028418687006265324, | |
| "loss": 3.7478, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.7725, | |
| "eval_samples_per_second": 21.522, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00028010078997548355, | |
| "loss": 3.6826, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0002760147098883138, | |
| "loss": 3.7236, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00027192862980114413, | |
| "loss": 3.7609, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00026804685371833287, | |
| "loss": 3.7846, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002639607736311632, | |
| "loss": 3.7496, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1233.5496, | |
| "eval_samples_per_second": 21.578, | |
| "eval_steps_per_second": 0.674, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00025987469354399345, | |
| "loss": 3.6785, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00025578861345682376, | |
| "loss": 3.7212, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0002517025333696541, | |
| "loss": 3.7637, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00024761645328248434, | |
| "loss": 3.6513, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00024353037319531465, | |
| "loss": 3.7513, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1231.8132, | |
| "eval_samples_per_second": 21.609, | |
| "eval_steps_per_second": 0.675, | |
| "eval_wer": 1.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00023944429310814494, | |
| "loss": 3.6818, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00023556251702533368, | |
| "loss": 3.8417, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.000231476436938164, | |
| "loss": 3.7639, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00022759466085535277, | |
| "loss": 3.7842, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00022350858076818309, | |
| "loss": 3.7497, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1234.8437, | |
| "eval_samples_per_second": 21.556, | |
| "eval_steps_per_second": 0.674, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00021942250068101335, | |
| "loss": 3.6795, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00021554072459820212, | |
| "loss": 3.8421, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00021145464451103243, | |
| "loss": 3.7614, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.0002075728684282212, | |
| "loss": 3.9259, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00020348678834105147, | |
| "loss": 3.7539, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1240.1103, | |
| "eval_samples_per_second": 21.464, | |
| "eval_steps_per_second": 0.671, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00019940070825388178, | |
| "loss": 3.6889, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00019551893217107055, | |
| "loss": 3.843, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00019143285208390087, | |
| "loss": 3.7666, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00018755107600108964, | |
| "loss": 3.786, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0001834649959139199, | |
| "loss": 3.7581, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1231.1699, | |
| "eval_samples_per_second": 21.62, | |
| "eval_steps_per_second": 0.676, | |
| "eval_wer": 1.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00017937891582675021, | |
| "loss": 3.672, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00017549713974393899, | |
| "loss": 3.8539, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00017141105965676927, | |
| "loss": 3.7652, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00016752928357395807, | |
| "loss": 3.7825, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00016344320348678833, | |
| "loss": 3.7572, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1241.8499, | |
| "eval_samples_per_second": 21.434, | |
| "eval_steps_per_second": 0.67, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00015935712339961862, | |
| "loss": 3.6906, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00015547534731680742, | |
| "loss": 3.845, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0001513892672296377, | |
| "loss": 3.763, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00014750749114682648, | |
| "loss": 3.779, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0001434214110596568, | |
| "loss": 3.7589, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1236.916, | |
| "eval_samples_per_second": 21.52, | |
| "eval_steps_per_second": 0.673, | |
| "eval_wer": 1.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00013933533097248705, | |
| "loss": 3.6948, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00013524925088531734, | |
| "loss": 3.7211, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00013116317079814763, | |
| "loss": 3.7588, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00012728139471533643, | |
| "loss": 3.7898, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00012319531462816672, | |
| "loss": 3.7592, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1232.1567, | |
| "eval_samples_per_second": 21.603, | |
| "eval_steps_per_second": 0.675, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.000119109234540997, | |
| "loss": 3.688, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00011522745845818579, | |
| "loss": 3.8419, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00011114137837101608, | |
| "loss": 3.7578, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00010705529828384636, | |
| "loss": 3.6534, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00010296921819667667, | |
| "loss": 3.7531, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1237.4955, | |
| "eval_samples_per_second": 21.51, | |
| "eval_steps_per_second": 0.672, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 9.888313810950695e-05, | |
| "loss": 3.691, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 9.500136202669572e-05, | |
| "loss": 3.8424, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 9.091528193952601e-05, | |
| "loss": 3.7698, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 8.70335058567148e-05, | |
| "loss": 3.7836, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 8.294742576954509e-05, | |
| "loss": 3.7567, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1242.7888, | |
| "eval_samples_per_second": 21.418, | |
| "eval_steps_per_second": 0.669, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.886134568237537e-05, | |
| "loss": 3.6741, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 7.497956959956416e-05, | |
| "loss": 3.8442, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 7.089348951239445e-05, | |
| "loss": 3.7663, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.701171342958322e-05, | |
| "loss": 3.7789, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 6.292563334241352e-05, | |
| "loss": 3.7613, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1244.1733, | |
| "eval_samples_per_second": 21.394, | |
| "eval_steps_per_second": 0.669, | |
| "eval_wer": 1.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 5.88395532552438e-05, | |
| "loss": 3.6884, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5.4957777172432585e-05, | |
| "loss": 3.8268, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 5.0871697085262866e-05, | |
| "loss": 3.7517, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 4.6989921002451645e-05, | |
| "loss": 3.7802, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 4.290384091528194e-05, | |
| "loss": 3.7516, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1253.9236, | |
| "eval_samples_per_second": 21.228, | |
| "eval_steps_per_second": 0.664, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 3.8817760828112234e-05, | |
| "loss": 3.6719, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 3.4935984745301006e-05, | |
| "loss": 3.8471, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 3.08499046581313e-05, | |
| "loss": 3.7568, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 2.6968128575320075e-05, | |
| "loss": 3.7887, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 2.2882048488150366e-05, | |
| "loss": 3.7581, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1242.0033, | |
| "eval_samples_per_second": 21.432, | |
| "eval_steps_per_second": 0.67, | |
| "eval_wer": 1.0, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.879596840098066e-05, | |
| "loss": 3.6785, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.4914192318169436e-05, | |
| "loss": 3.8524, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 3746, | |
| "total_flos": 7.191358836520801e+19, | |
| "train_loss": 3.6355768978691203, | |
| "train_runtime": 71670.0084, | |
| "train_samples_per_second": 6.691, | |
| "train_steps_per_second": 0.052 | |
| } | |
| ], | |
| "max_steps": 3746, | |
| "num_train_epochs": 2, | |
| "total_flos": 7.191358836520801e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |