| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.2776, | |
| "eval_steps": 1000, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 88.54288482666016, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 2.5968, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 71.25262451171875, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 2.3534, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 80.52667999267578, | |
| "learning_rate": 1.42e-06, | |
| "loss": 1.9642, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 57.064151763916016, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.6321, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 50.983245849609375, | |
| "learning_rate": 2.42e-06, | |
| "loss": 1.5842, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 52.22635269165039, | |
| "learning_rate": 2.92e-06, | |
| "loss": 1.4106, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.035, | |
| "grad_norm": 53.9218864440918, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 1.417, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 48.82598876953125, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 1.4496, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.045, | |
| "grad_norm": 63.08296585083008, | |
| "learning_rate": 4.42e-06, | |
| "loss": 1.3191, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 41.98011016845703, | |
| "learning_rate": 4.92e-06, | |
| "loss": 1.0756, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.055, | |
| "grad_norm": 53.777217864990234, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 1.2538, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 57.49655532836914, | |
| "learning_rate": 5.92e-06, | |
| "loss": 1.2225, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.065, | |
| "grad_norm": 52.660003662109375, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 1.2031, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 45.03782272338867, | |
| "learning_rate": 6.9e-06, | |
| "loss": 1.2953, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 49.226261138916016, | |
| "learning_rate": 7.4e-06, | |
| "loss": 1.0642, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 33.70058822631836, | |
| "learning_rate": 7.9e-06, | |
| "loss": 1.0962, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.085, | |
| "grad_norm": 47.71772766113281, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 1.3042, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 55.22367477416992, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 1.3794, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.095, | |
| "grad_norm": 62.34690475463867, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.1516, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 84.5876693725586, | |
| "learning_rate": 9.9e-06, | |
| "loss": 1.2527, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.105, | |
| "grad_norm": 53.2620735168457, | |
| "learning_rate": 9.955555555555556e-06, | |
| "loss": 1.1349, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 54.91654586791992, | |
| "learning_rate": 9.9e-06, | |
| "loss": 1.2107, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.115, | |
| "grad_norm": 43.851173400878906, | |
| "learning_rate": 9.844444444444446e-06, | |
| "loss": 1.4065, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 48.8751335144043, | |
| "learning_rate": 9.78888888888889e-06, | |
| "loss": 1.2255, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 58.15187454223633, | |
| "learning_rate": 9.733333333333334e-06, | |
| "loss": 1.1948, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 53.069602966308594, | |
| "learning_rate": 9.677777777777778e-06, | |
| "loss": 1.2878, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.135, | |
| "grad_norm": 45.524715423583984, | |
| "learning_rate": 9.622222222222222e-06, | |
| "loss": 1.1325, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 26.86398696899414, | |
| "learning_rate": 9.566666666666668e-06, | |
| "loss": 1.0762, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.145, | |
| "grad_norm": 48.63008499145508, | |
| "learning_rate": 9.511111111111112e-06, | |
| "loss": 1.279, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 50.58299255371094, | |
| "learning_rate": 9.455555555555557e-06, | |
| "loss": 1.0627, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.155, | |
| "grad_norm": 32.71992492675781, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.0438, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 42.12641143798828, | |
| "learning_rate": 9.344444444444446e-06, | |
| "loss": 1.1204, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.165, | |
| "grad_norm": 44.01963806152344, | |
| "learning_rate": 9.28888888888889e-06, | |
| "loss": 0.908, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 39.93177795410156, | |
| "learning_rate": 9.233333333333334e-06, | |
| "loss": 1.158, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 38.166259765625, | |
| "learning_rate": 9.17777777777778e-06, | |
| "loss": 1.2021, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 40.696678161621094, | |
| "learning_rate": 9.122222222222223e-06, | |
| "loss": 1.0054, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.185, | |
| "grad_norm": 48.79116439819336, | |
| "learning_rate": 9.066666666666667e-06, | |
| "loss": 0.9587, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 39.57270431518555, | |
| "learning_rate": 9.011111111111111e-06, | |
| "loss": 1.1084, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.195, | |
| "grad_norm": 54.647216796875, | |
| "learning_rate": 8.955555555555555e-06, | |
| "loss": 1.0416, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 42.69215774536133, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 0.97, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_cer": 38.476190476190474, | |
| "eval_loss": 0.7355929613113403, | |
| "eval_model_preparation_time": 0.0121, | |
| "eval_runtime": 116.5648, | |
| "eval_samples_per_second": 2.574, | |
| "eval_steps_per_second": 1.287, | |
| "eval_wer": 38.17307692307692, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.205, | |
| "grad_norm": 53.78874206542969, | |
| "learning_rate": 8.844444444444445e-06, | |
| "loss": 1.2455, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 34.4659423828125, | |
| "learning_rate": 8.788888888888891e-06, | |
| "loss": 1.0515, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.215, | |
| "grad_norm": 51.5352668762207, | |
| "learning_rate": 8.733333333333333e-06, | |
| "loss": 0.9412, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 41.004310607910156, | |
| "learning_rate": 8.677777777777779e-06, | |
| "loss": 1.0831, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 43.79686737060547, | |
| "learning_rate": 8.622222222222223e-06, | |
| "loss": 0.8895, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 32.330204010009766, | |
| "learning_rate": 8.566666666666667e-06, | |
| "loss": 0.8984, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.235, | |
| "grad_norm": 43.38914108276367, | |
| "learning_rate": 8.511111111111113e-06, | |
| "loss": 1.1215, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 62.52465057373047, | |
| "learning_rate": 8.455555555555555e-06, | |
| "loss": 0.9912, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.245, | |
| "grad_norm": 39.1291618347168, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 1.0422, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 53.39470291137695, | |
| "learning_rate": 8.344444444444445e-06, | |
| "loss": 1.1502, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.255, | |
| "grad_norm": 42.28481674194336, | |
| "learning_rate": 8.288888888888889e-06, | |
| "loss": 1.112, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 37.374542236328125, | |
| "learning_rate": 8.233333333333335e-06, | |
| "loss": 1.0152, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.265, | |
| "grad_norm": 52.94354248046875, | |
| "learning_rate": 8.177777777777779e-06, | |
| "loss": 0.9147, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 46.23354721069336, | |
| "learning_rate": 8.122222222222223e-06, | |
| "loss": 0.8429, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 40.089210510253906, | |
| "learning_rate": 8.066666666666667e-06, | |
| "loss": 0.8026, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 42.95212173461914, | |
| "learning_rate": 8.011111111111113e-06, | |
| "loss": 0.7804, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.285, | |
| "grad_norm": 56.216678619384766, | |
| "learning_rate": 7.955555555555557e-06, | |
| "loss": 0.8045, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 45.01803207397461, | |
| "learning_rate": 7.902222222222223e-06, | |
| "loss": 0.9807, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.295, | |
| "grad_norm": 63.81135940551758, | |
| "learning_rate": 7.846666666666667e-06, | |
| "loss": 0.9447, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 38.958457946777344, | |
| "learning_rate": 7.791111111111111e-06, | |
| "loss": 1.0144, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.305, | |
| "grad_norm": 48.87187957763672, | |
| "learning_rate": 7.735555555555557e-06, | |
| "loss": 1.0261, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 34.80329513549805, | |
| "learning_rate": 7.680000000000001e-06, | |
| "loss": 0.8152, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.315, | |
| "grad_norm": 45.12413024902344, | |
| "learning_rate": 7.624444444444445e-06, | |
| "loss": 0.8943, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 38.103729248046875, | |
| "learning_rate": 7.56888888888889e-06, | |
| "loss": 1.085, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 38.715492248535156, | |
| "learning_rate": 7.513333333333334e-06, | |
| "loss": 0.9853, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 42.01364517211914, | |
| "learning_rate": 7.457777777777778e-06, | |
| "loss": 0.8214, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.335, | |
| "grad_norm": 37.475799560546875, | |
| "learning_rate": 7.402222222222223e-06, | |
| "loss": 0.7706, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 40.387210845947266, | |
| "learning_rate": 7.346666666666668e-06, | |
| "loss": 1.081, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.345, | |
| "grad_norm": 21.215946197509766, | |
| "learning_rate": 7.291111111111112e-06, | |
| "loss": 0.8047, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 21.454147338867188, | |
| "learning_rate": 7.235555555555556e-06, | |
| "loss": 0.836, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.355, | |
| "grad_norm": 46.05310821533203, | |
| "learning_rate": 7.180000000000001e-06, | |
| "loss": 0.8883, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 61.988643646240234, | |
| "learning_rate": 7.124444444444445e-06, | |
| "loss": 0.8254, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.0038, | |
| "grad_norm": 34.00222396850586, | |
| "learning_rate": 7.06888888888889e-06, | |
| "loss": 0.5202, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 1.0088, | |
| "grad_norm": 23.966150283813477, | |
| "learning_rate": 7.0133333333333345e-06, | |
| "loss": 0.362, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.0138, | |
| "grad_norm": 19.050518035888672, | |
| "learning_rate": 6.9577777777777785e-06, | |
| "loss": 0.2891, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 1.0188, | |
| "grad_norm": 34.59785079956055, | |
| "learning_rate": 6.902222222222223e-06, | |
| "loss": 0.3438, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.0238, | |
| "grad_norm": 29.40850257873535, | |
| "learning_rate": 6.846666666666667e-06, | |
| "loss": 0.2822, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 1.0288, | |
| "grad_norm": 15.476716041564941, | |
| "learning_rate": 6.7911111111111115e-06, | |
| "loss": 0.4843, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.0338, | |
| "grad_norm": 33.912174224853516, | |
| "learning_rate": 6.735555555555556e-06, | |
| "loss": 0.3741, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 1.0388, | |
| "grad_norm": 30.411020278930664, | |
| "learning_rate": 6.680000000000001e-06, | |
| "loss": 0.3044, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0388, | |
| "eval_cer": 23.904761904761905, | |
| "eval_loss": 0.309874564409256, | |
| "eval_model_preparation_time": 0.0121, | |
| "eval_runtime": 119.6464, | |
| "eval_samples_per_second": 2.507, | |
| "eval_steps_per_second": 1.254, | |
| "eval_wer": 23.46153846153846, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0438, | |
| "grad_norm": 19.959854125976562, | |
| "learning_rate": 6.6244444444444445e-06, | |
| "loss": 0.2255, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 1.0488, | |
| "grad_norm": 20.751602172851562, | |
| "learning_rate": 6.568888888888889e-06, | |
| "loss": 0.3341, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.0538, | |
| "grad_norm": 24.52460479736328, | |
| "learning_rate": 6.513333333333333e-06, | |
| "loss": 0.4861, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.0588, | |
| "grad_norm": 5.740904331207275, | |
| "learning_rate": 6.457777777777778e-06, | |
| "loss": 0.4165, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.0638, | |
| "grad_norm": 24.452116012573242, | |
| "learning_rate": 6.402222222222223e-06, | |
| "loss": 0.4478, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.0688, | |
| "grad_norm": 29.230716705322266, | |
| "learning_rate": 6.346666666666668e-06, | |
| "loss": 0.4387, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.0738, | |
| "grad_norm": 41.07571792602539, | |
| "learning_rate": 6.291111111111111e-06, | |
| "loss": 0.2466, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.0788, | |
| "grad_norm": 19.89525032043457, | |
| "learning_rate": 6.235555555555556e-06, | |
| "loss": 0.3156, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.0838, | |
| "grad_norm": 33.628971099853516, | |
| "learning_rate": 6.18e-06, | |
| "loss": 0.3624, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.0888, | |
| "grad_norm": 23.34870147705078, | |
| "learning_rate": 6.124444444444445e-06, | |
| "loss": 0.4263, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.0937999999999999, | |
| "grad_norm": 30.75408172607422, | |
| "learning_rate": 6.06888888888889e-06, | |
| "loss": 0.4153, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.0988, | |
| "grad_norm": 7.229944705963135, | |
| "learning_rate": 6.013333333333335e-06, | |
| "loss": 0.3068, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.1038000000000001, | |
| "grad_norm": 37.87436294555664, | |
| "learning_rate": 5.957777777777778e-06, | |
| "loss": 0.3049, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 1.1088, | |
| "grad_norm": 33.34481430053711, | |
| "learning_rate": 5.902222222222223e-06, | |
| "loss": 0.2759, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.1138, | |
| "grad_norm": 24.211904525756836, | |
| "learning_rate": 5.846666666666667e-06, | |
| "loss": 0.35, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.1188, | |
| "grad_norm": 41.01383590698242, | |
| "learning_rate": 5.791111111111112e-06, | |
| "loss": 0.3992, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.1238, | |
| "grad_norm": 6.543262004852295, | |
| "learning_rate": 5.735555555555557e-06, | |
| "loss": 0.3356, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 1.1288, | |
| "grad_norm": 34.344913482666016, | |
| "learning_rate": 5.68e-06, | |
| "loss": 0.3364, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.1338, | |
| "grad_norm": 9.35561752319336, | |
| "learning_rate": 5.624444444444445e-06, | |
| "loss": 0.3606, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 1.1388, | |
| "grad_norm": 21.762096405029297, | |
| "learning_rate": 5.56888888888889e-06, | |
| "loss": 0.3437, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.1438, | |
| "grad_norm": 29.119796752929688, | |
| "learning_rate": 5.513333333333334e-06, | |
| "loss": 0.428, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 1.1488, | |
| "grad_norm": 46.66371536254883, | |
| "learning_rate": 5.4577777777777785e-06, | |
| "loss": 0.4566, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.1538, | |
| "grad_norm": 15.0108642578125, | |
| "learning_rate": 5.402222222222223e-06, | |
| "loss": 0.3162, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 1.1588, | |
| "grad_norm": 60.40862274169922, | |
| "learning_rate": 5.346666666666667e-06, | |
| "loss": 0.3696, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.1638, | |
| "grad_norm": 26.4654598236084, | |
| "learning_rate": 5.2911111111111115e-06, | |
| "loss": 0.2831, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.1688, | |
| "grad_norm": 23.651691436767578, | |
| "learning_rate": 5.235555555555556e-06, | |
| "loss": 0.4484, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.1738, | |
| "grad_norm": 33.675167083740234, | |
| "learning_rate": 5.18e-06, | |
| "loss": 0.3062, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 1.1788, | |
| "grad_norm": 27.336896896362305, | |
| "learning_rate": 5.124444444444445e-06, | |
| "loss": 0.3738, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.1838, | |
| "grad_norm": 17.323768615722656, | |
| "learning_rate": 5.06888888888889e-06, | |
| "loss": 0.3182, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 1.1888, | |
| "grad_norm": 18.657014846801758, | |
| "learning_rate": 5.013333333333333e-06, | |
| "loss": 0.3589, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.1938, | |
| "grad_norm": 39.54949188232422, | |
| "learning_rate": 4.957777777777778e-06, | |
| "loss": 0.4187, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 1.1988, | |
| "grad_norm": 57.646602630615234, | |
| "learning_rate": 4.902222222222222e-06, | |
| "loss": 0.3374, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.2038, | |
| "grad_norm": 24.029294967651367, | |
| "learning_rate": 4.846666666666667e-06, | |
| "loss": 0.2373, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 1.2088, | |
| "grad_norm": 34.673709869384766, | |
| "learning_rate": 4.791111111111111e-06, | |
| "loss": 0.3274, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.2138, | |
| "grad_norm": 40.28670120239258, | |
| "learning_rate": 4.735555555555556e-06, | |
| "loss": 0.3105, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 1.2187999999999999, | |
| "grad_norm": 22.120824813842773, | |
| "learning_rate": 4.680000000000001e-06, | |
| "loss": 0.2678, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.2238, | |
| "grad_norm": 21.942718505859375, | |
| "learning_rate": 4.624444444444445e-06, | |
| "loss": 0.5025, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 1.2288000000000001, | |
| "grad_norm": 31.645906448364258, | |
| "learning_rate": 4.568888888888889e-06, | |
| "loss": 0.3753, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.2338, | |
| "grad_norm": 19.00498390197754, | |
| "learning_rate": 4.513333333333333e-06, | |
| "loss": 0.3421, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 1.2388, | |
| "grad_norm": 42.73635482788086, | |
| "learning_rate": 4.457777777777778e-06, | |
| "loss": 0.3108, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2388, | |
| "eval_cer": 7.7142857142857135, | |
| "eval_loss": 0.11533673852682114, | |
| "eval_model_preparation_time": 0.0121, | |
| "eval_runtime": 120.7728, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 1.242, | |
| "eval_wer": 7.5, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2438, | |
| "grad_norm": 20.023395538330078, | |
| "learning_rate": 4.402222222222223e-06, | |
| "loss": 0.3144, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 1.2488, | |
| "grad_norm": 32.98371505737305, | |
| "learning_rate": 4.346666666666667e-06, | |
| "loss": 0.2618, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.2538, | |
| "grad_norm": 42.568119049072266, | |
| "learning_rate": 4.291111111111112e-06, | |
| "loss": 0.331, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 1.2588, | |
| "grad_norm": 0.5769469738006592, | |
| "learning_rate": 4.235555555555556e-06, | |
| "loss": 0.3391, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.2638, | |
| "grad_norm": 20.518579483032227, | |
| "learning_rate": 4.18e-06, | |
| "loss": 0.2234, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 1.2688, | |
| "grad_norm": 27.69402313232422, | |
| "learning_rate": 4.124444444444445e-06, | |
| "loss": 0.295, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.2738, | |
| "grad_norm": 31.121999740600586, | |
| "learning_rate": 4.0688888888888896e-06, | |
| "loss": 0.3753, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 1.2788, | |
| "grad_norm": 14.77844524383545, | |
| "learning_rate": 4.013333333333334e-06, | |
| "loss": 0.3706, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.2838, | |
| "grad_norm": 27.002138137817383, | |
| "learning_rate": 3.9577777777777785e-06, | |
| "loss": 0.2957, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 1.2888, | |
| "grad_norm": 13.426039695739746, | |
| "learning_rate": 3.9022222222222225e-06, | |
| "loss": 0.2902, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.2938, | |
| "grad_norm": 41.75296401977539, | |
| "learning_rate": 3.8466666666666665e-06, | |
| "loss": 0.2924, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 1.2988, | |
| "grad_norm": 52.886409759521484, | |
| "learning_rate": 3.7911111111111114e-06, | |
| "loss": 0.2519, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.3038, | |
| "grad_norm": 33.73487854003906, | |
| "learning_rate": 3.7355555555555555e-06, | |
| "loss": 0.2966, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 1.3088, | |
| "grad_norm": 27.994157791137695, | |
| "learning_rate": 3.6800000000000003e-06, | |
| "loss": 0.2887, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.3138, | |
| "grad_norm": 32.397579193115234, | |
| "learning_rate": 3.624444444444445e-06, | |
| "loss": 0.2368, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 1.3188, | |
| "grad_norm": 25.96181869506836, | |
| "learning_rate": 3.568888888888889e-06, | |
| "loss": 0.2911, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.3237999999999999, | |
| "grad_norm": 7.705018520355225, | |
| "learning_rate": 3.5133333333333337e-06, | |
| "loss": 0.2647, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 1.3288, | |
| "grad_norm": 28.499221801757812, | |
| "learning_rate": 3.457777777777778e-06, | |
| "loss": 0.4149, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 1.3338, | |
| "grad_norm": 40.06334686279297, | |
| "learning_rate": 3.4022222222222222e-06, | |
| "loss": 0.2952, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 1.3388, | |
| "grad_norm": 1.474075198173523, | |
| "learning_rate": 3.346666666666667e-06, | |
| "loss": 0.207, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3437999999999999, | |
| "grad_norm": 27.121145248413086, | |
| "learning_rate": 3.2911111111111116e-06, | |
| "loss": 0.3432, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 1.3488, | |
| "grad_norm": 22.38213539123535, | |
| "learning_rate": 3.2355555555555556e-06, | |
| "loss": 0.3522, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 1.3538000000000001, | |
| "grad_norm": 42.13764190673828, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 0.3729, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 1.3588, | |
| "grad_norm": 27.436649322509766, | |
| "learning_rate": 3.124444444444445e-06, | |
| "loss": 0.1482, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.0026, | |
| "grad_norm": 16.43301010131836, | |
| "learning_rate": 3.068888888888889e-06, | |
| "loss": 0.1802, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 2.0076, | |
| "grad_norm": 11.994711875915527, | |
| "learning_rate": 3.013333333333334e-06, | |
| "loss": 0.0922, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.0126, | |
| "grad_norm": 7.826560020446777, | |
| "learning_rate": 2.957777777777778e-06, | |
| "loss": 0.0312, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 2.0176, | |
| "grad_norm": 6.457350730895996, | |
| "learning_rate": 2.9022222222222223e-06, | |
| "loss": 0.0343, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.0226, | |
| "grad_norm": 1.624599575996399, | |
| "learning_rate": 2.8466666666666672e-06, | |
| "loss": 0.0751, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 2.0276, | |
| "grad_norm": 4.553808212280273, | |
| "learning_rate": 2.7911111111111113e-06, | |
| "loss": 0.0905, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.0326, | |
| "grad_norm": 4.33929967880249, | |
| "learning_rate": 2.7355555555555557e-06, | |
| "loss": 0.079, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 2.0376, | |
| "grad_norm": 11.390565872192383, | |
| "learning_rate": 2.68e-06, | |
| "loss": 0.064, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.0426, | |
| "grad_norm": 0.5454270243644714, | |
| "learning_rate": 2.6244444444444446e-06, | |
| "loss": 0.0351, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 2.0476, | |
| "grad_norm": 25.50840950012207, | |
| "learning_rate": 2.568888888888889e-06, | |
| "loss": 0.0726, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 2.0526, | |
| "grad_norm": 5.380075931549072, | |
| "learning_rate": 2.5133333333333336e-06, | |
| "loss": 0.0405, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 2.0576, | |
| "grad_norm": 1.0126858949661255, | |
| "learning_rate": 2.457777777777778e-06, | |
| "loss": 0.0858, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.0626, | |
| "grad_norm": 1.7211323976516724, | |
| "learning_rate": 2.4022222222222225e-06, | |
| "loss": 0.0431, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 2.0676, | |
| "grad_norm": 2.049405574798584, | |
| "learning_rate": 2.346666666666667e-06, | |
| "loss": 0.0127, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 2.0726, | |
| "grad_norm": 1.535748839378357, | |
| "learning_rate": 2.2911111111111114e-06, | |
| "loss": 0.0559, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 2.0776, | |
| "grad_norm": 19.566328048706055, | |
| "learning_rate": 2.235555555555556e-06, | |
| "loss": 0.0544, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0776, | |
| "eval_cer": 2.2857142857142856, | |
| "eval_loss": 0.029534637928009033, | |
| "eval_model_preparation_time": 0.0121, | |
| "eval_runtime": 119.7419, | |
| "eval_samples_per_second": 2.505, | |
| "eval_steps_per_second": 1.253, | |
| "eval_wer": 2.307692307692308, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0826000000000002, | |
| "grad_norm": 0.4915294945240021, | |
| "learning_rate": 2.1800000000000003e-06, | |
| "loss": 0.0556, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 2.0876, | |
| "grad_norm": 13.67375659942627, | |
| "learning_rate": 2.1244444444444443e-06, | |
| "loss": 0.0377, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 2.0926, | |
| "grad_norm": 0.06471225619316101, | |
| "learning_rate": 2.0688888888888892e-06, | |
| "loss": 0.0535, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 2.0976, | |
| "grad_norm": 0.1591552495956421, | |
| "learning_rate": 2.0133333333333337e-06, | |
| "loss": 0.037, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.1026, | |
| "grad_norm": 1.6957018375396729, | |
| "learning_rate": 1.9577777777777777e-06, | |
| "loss": 0.0548, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 2.1076, | |
| "grad_norm": 7.906589508056641, | |
| "learning_rate": 1.9022222222222222e-06, | |
| "loss": 0.066, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 2.1126, | |
| "grad_norm": 0.4227987825870514, | |
| "learning_rate": 1.8466666666666668e-06, | |
| "loss": 0.0687, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 2.1176, | |
| "grad_norm": 0.3206275999546051, | |
| "learning_rate": 1.7911111111111113e-06, | |
| "loss": 0.0263, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.1226, | |
| "grad_norm": 0.16232123970985413, | |
| "learning_rate": 1.7355555555555555e-06, | |
| "loss": 0.0257, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 2.1276, | |
| "grad_norm": 21.80668830871582, | |
| "learning_rate": 1.6800000000000002e-06, | |
| "loss": 0.0346, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.1326, | |
| "grad_norm": 3.021885871887207, | |
| "learning_rate": 1.6244444444444447e-06, | |
| "loss": 0.0715, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 2.1376, | |
| "grad_norm": 9.51052188873291, | |
| "learning_rate": 1.568888888888889e-06, | |
| "loss": 0.0686, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.1426, | |
| "grad_norm": 7.770538806915283, | |
| "learning_rate": 1.5133333333333334e-06, | |
| "loss": 0.0681, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 2.1476, | |
| "grad_norm": 24.541362762451172, | |
| "learning_rate": 1.457777777777778e-06, | |
| "loss": 0.0589, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 2.1526, | |
| "grad_norm": 0.2108163684606552, | |
| "learning_rate": 1.4022222222222223e-06, | |
| "loss": 0.0407, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 2.1576, | |
| "grad_norm": 39.73221206665039, | |
| "learning_rate": 1.3466666666666668e-06, | |
| "loss": 0.0895, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.1626, | |
| "grad_norm": 4.3332438468933105, | |
| "learning_rate": 1.2911111111111112e-06, | |
| "loss": 0.0645, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 2.1676, | |
| "grad_norm": 1.142255425453186, | |
| "learning_rate": 1.2355555555555557e-06, | |
| "loss": 0.0592, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 2.1726, | |
| "grad_norm": 0.046843066811561584, | |
| "learning_rate": 1.1800000000000001e-06, | |
| "loss": 0.034, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 2.1776, | |
| "grad_norm": 14.69985580444336, | |
| "learning_rate": 1.1244444444444446e-06, | |
| "loss": 0.0704, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.1826, | |
| "grad_norm": 11.02341079711914, | |
| "learning_rate": 1.068888888888889e-06, | |
| "loss": 0.0621, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 2.1875999999999998, | |
| "grad_norm": 0.05967080965638161, | |
| "learning_rate": 1.0133333333333333e-06, | |
| "loss": 0.0331, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 2.1926, | |
| "grad_norm": 9.536999702453613, | |
| "learning_rate": 9.57777777777778e-07, | |
| "loss": 0.0762, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 2.1976, | |
| "grad_norm": 3.740366220474243, | |
| "learning_rate": 9.022222222222222e-07, | |
| "loss": 0.0346, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.2026, | |
| "grad_norm": 1.1851881742477417, | |
| "learning_rate": 8.466666666666668e-07, | |
| "loss": 0.0502, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 2.2076000000000002, | |
| "grad_norm": 23.711048126220703, | |
| "learning_rate": 7.911111111111111e-07, | |
| "loss": 0.0661, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 2.2126, | |
| "grad_norm": 1.0020112991333008, | |
| "learning_rate": 7.355555555555556e-07, | |
| "loss": 0.0252, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 2.2176, | |
| "grad_norm": 21.021774291992188, | |
| "learning_rate": 6.800000000000001e-07, | |
| "loss": 0.0415, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.2226, | |
| "grad_norm": 0.8725367784500122, | |
| "learning_rate": 6.244444444444445e-07, | |
| "loss": 0.0293, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 2.2276, | |
| "grad_norm": 0.3742901086807251, | |
| "learning_rate": 5.68888888888889e-07, | |
| "loss": 0.0146, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.2326, | |
| "grad_norm": 12.225566864013672, | |
| "learning_rate": 5.133333333333334e-07, | |
| "loss": 0.0775, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 2.2376, | |
| "grad_norm": 0.5857837200164795, | |
| "learning_rate": 4.5777777777777784e-07, | |
| "loss": 0.0335, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.2426, | |
| "grad_norm": 2.0213797092437744, | |
| "learning_rate": 4.0222222222222224e-07, | |
| "loss": 0.06, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 2.2476, | |
| "grad_norm": 0.22115977108478546, | |
| "learning_rate": 3.466666666666667e-07, | |
| "loss": 0.0437, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 2.2526, | |
| "grad_norm": 11.783333778381348, | |
| "learning_rate": 2.9111111111111116e-07, | |
| "loss": 0.0454, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 2.2576, | |
| "grad_norm": 2.67048978805542, | |
| "learning_rate": 2.3555555555555556e-07, | |
| "loss": 0.093, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.2626, | |
| "grad_norm": 0.16820040345191956, | |
| "learning_rate": 1.8e-07, | |
| "loss": 0.1193, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 2.2676, | |
| "grad_norm": 1.518930435180664, | |
| "learning_rate": 1.2444444444444446e-07, | |
| "loss": 0.0493, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 2.2726, | |
| "grad_norm": 0.10118613392114639, | |
| "learning_rate": 6.888888888888889e-08, | |
| "loss": 0.0538, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 2.2776, | |
| "grad_norm": 0.5119066834449768, | |
| "learning_rate": 1.3333333333333334e-08, | |
| "loss": 0.0678, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.2776, | |
| "eval_cer": 0.9523809523809524, | |
| "eval_loss": 0.014132725074887276, | |
| "eval_model_preparation_time": 0.0121, | |
| "eval_runtime": 119.856, | |
| "eval_samples_per_second": 2.503, | |
| "eval_steps_per_second": 1.252, | |
| "eval_wer": 0.9615384615384616, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.2776, | |
| "step": 5000, | |
| "total_flos": 1.02060490752e+19, | |
| "train_loss": 0.5477467903137208, | |
| "train_runtime": 5409.143, | |
| "train_samples_per_second": 1.849, | |
| "train_steps_per_second": 0.924 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.02060490752e+19, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |