| { | |
| "best_metric": 0.09356310187531677, | |
| "best_model_checkpoint": "wav2vec2-xls-r-slavic-pomak/checkpoint-27580", | |
| "epoch": 34.99936588459099, | |
| "global_step": 27580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 5.3636, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 1.2489, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0002988921713441654, | |
| "loss": 0.8015, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.07991897799971923, | |
| "eval_loss": 0.34495434165000916, | |
| "eval_runtime": 40.7653, | |
| "eval_samples_per_second": 38.685, | |
| "eval_steps_per_second": 4.857, | |
| "eval_wer": 0.33421186011150533, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00029667651403249626, | |
| "loss": 0.6801, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00029446085672082717, | |
| "loss": 0.6282, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00029224519940915803, | |
| "loss": 0.5794, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0002900295420974889, | |
| "loss": 0.5688, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.05737721356516856, | |
| "eval_loss": 0.2721518874168396, | |
| "eval_runtime": 42.4122, | |
| "eval_samples_per_second": 37.183, | |
| "eval_steps_per_second": 4.668, | |
| "eval_wer": 0.22199695894576787, | |
| "step": 1576 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.0002878138847858198, | |
| "loss": 0.5417, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.00028559822747415065, | |
| "loss": 0.4888, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.0002833825701624815, | |
| "loss": 0.5066, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.0002811669128508124, | |
| "loss": 0.4869, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.053346168501694644, | |
| "eval_loss": 0.2440285086631775, | |
| "eval_runtime": 40.5216, | |
| "eval_samples_per_second": 38.917, | |
| "eval_steps_per_second": 4.886, | |
| "eval_wer": 0.20293968575772936, | |
| "step": 2364 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.0002789512555391432, | |
| "loss": 0.4609, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.00027673559822747414, | |
| "loss": 0.4369, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.000274519940915805, | |
| "loss": 0.4398, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.00027230428360413585, | |
| "loss": 0.4494, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 0.04845276056394521, | |
| "eval_loss": 0.2370358109474182, | |
| "eval_runtime": 40.4374, | |
| "eval_samples_per_second": 38.999, | |
| "eval_steps_per_second": 4.896, | |
| "eval_wer": 0.18459199189052206, | |
| "step": 3152 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 0.00027008862629246676, | |
| "loss": 0.433, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.0002678729689807976, | |
| "loss": 0.4144, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.0002656573116691285, | |
| "loss": 0.4018, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.0002634416543574594, | |
| "loss": 0.3969, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 0.04522391352305317, | |
| "eval_loss": 0.23047970235347748, | |
| "eval_runtime": 41.6141, | |
| "eval_samples_per_second": 37.896, | |
| "eval_steps_per_second": 4.758, | |
| "eval_wer": 0.1727318803852002, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.0002612259970457902, | |
| "loss": 0.3934, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0002590103397341211, | |
| "loss": 0.3676, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 0.00025679468242245196, | |
| "loss": 0.3725, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 0.0002545790251107828, | |
| "loss": 0.373, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.0472494635300724, | |
| "eval_loss": 0.21691733598709106, | |
| "eval_runtime": 40.268, | |
| "eval_samples_per_second": 39.163, | |
| "eval_steps_per_second": 4.917, | |
| "eval_wer": 0.17465788139888494, | |
| "step": 4728 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 0.0002523633677991137, | |
| "loss": 0.3554, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 0.0002501477104874446, | |
| "loss": 0.332, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 0.00024793205317577544, | |
| "loss": 0.3462, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 0.00024571639586410635, | |
| "loss": 0.3483, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.04143352786635381, | |
| "eval_loss": 0.22267436981201172, | |
| "eval_runtime": 40.4667, | |
| "eval_samples_per_second": 38.97, | |
| "eval_steps_per_second": 4.893, | |
| "eval_wer": 0.15701976685250887, | |
| "step": 5516 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 0.0002435007385524372, | |
| "loss": 0.3532, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 0.00024128508124076806, | |
| "loss": 0.3249, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "learning_rate": 0.00023906942392909895, | |
| "loss": 0.3436, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 0.00023685376661742983, | |
| "loss": 0.31, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.0421153961855484, | |
| "eval_loss": 0.21147780120372772, | |
| "eval_runtime": 40.4158, | |
| "eval_samples_per_second": 39.019, | |
| "eval_steps_per_second": 4.899, | |
| "eval_wer": 0.15306639635073493, | |
| "step": 6304 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0002346381093057607, | |
| "loss": 0.3175, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.00023242245199409157, | |
| "loss": 0.2886, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 0.00023020679468242246, | |
| "loss": 0.2919, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.0002279911373707533, | |
| "loss": 0.3127, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.04123297836070834, | |
| "eval_loss": 0.2211967408657074, | |
| "eval_runtime": 40.3337, | |
| "eval_samples_per_second": 39.099, | |
| "eval_steps_per_second": 4.909, | |
| "eval_wer": 0.1559047136340598, | |
| "step": 7092 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 0.00022577548005908417, | |
| "loss": 0.3022, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 0.00022355982274741506, | |
| "loss": 0.2737, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 0.0002213441654357459, | |
| "loss": 0.2681, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 0.0002191285081240768, | |
| "loss": 0.2795, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.038124461023203576, | |
| "eval_loss": 0.21809443831443787, | |
| "eval_runtime": 40.3044, | |
| "eval_samples_per_second": 39.127, | |
| "eval_steps_per_second": 4.913, | |
| "eval_wer": 0.1395843892549417, | |
| "step": 7880 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 0.00021691285081240765, | |
| "loss": 0.2714, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 0.00021469719350073854, | |
| "loss": 0.2575, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 10.66, | |
| "learning_rate": 0.00021248153618906942, | |
| "loss": 0.2758, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 0.00021026587887740025, | |
| "loss": 0.2676, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.03782363676473537, | |
| "eval_loss": 0.22525641322135925, | |
| "eval_runtime": 40.3665, | |
| "eval_samples_per_second": 39.067, | |
| "eval_steps_per_second": 4.905, | |
| "eval_wer": 0.1375570197668525, | |
| "step": 8668 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 0.00020805022156573114, | |
| "loss": 0.2398, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 0.00020583456425406202, | |
| "loss": 0.2476, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 0.00020361890694239288, | |
| "loss": 0.2491, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 0.00020140324963072376, | |
| "loss": 0.2385, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_cer": 0.038024186270380844, | |
| "eval_loss": 0.24045319855213165, | |
| "eval_runtime": 40.4554, | |
| "eval_samples_per_second": 38.981, | |
| "eval_steps_per_second": 4.894, | |
| "eval_wer": 0.1352255448555499, | |
| "step": 9456 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "learning_rate": 0.00019918759231905465, | |
| "loss": 0.2425, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 0.0001969719350073855, | |
| "loss": 0.2223, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 0.0001947562776957164, | |
| "loss": 0.2448, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 0.00019254062038404727, | |
| "loss": 0.216, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_cer": 0.035657702103764316, | |
| "eval_loss": 0.2175922691822052, | |
| "eval_runtime": 40.4091, | |
| "eval_samples_per_second": 39.026, | |
| "eval_steps_per_second": 4.9, | |
| "eval_wer": 0.12579827673593513, | |
| "step": 10244 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "learning_rate": 0.0001903249630723781, | |
| "loss": 0.2153, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.00018810930576070899, | |
| "loss": 0.2215, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 0.00018589364844903987, | |
| "loss": 0.2102, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "learning_rate": 0.00018367799113737073, | |
| "loss": 0.2197, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_cer": 0.03569781200489341, | |
| "eval_loss": 0.21506036818027496, | |
| "eval_runtime": 40.487, | |
| "eval_samples_per_second": 38.951, | |
| "eval_steps_per_second": 4.89, | |
| "eval_wer": 0.12630511910795741, | |
| "step": 11032 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 0.0001814623338257016, | |
| "loss": 0.1977, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 0.0001792466765140325, | |
| "loss": 0.1966, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 0.00017703101920236335, | |
| "loss": 0.1966, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 0.00017481536189069424, | |
| "loss": 0.2036, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_cer": 0.036981328841024404, | |
| "eval_loss": 0.21347209811210632, | |
| "eval_runtime": 40.2901, | |
| "eval_samples_per_second": 39.141, | |
| "eval_steps_per_second": 4.914, | |
| "eval_wer": 0.12965027876330462, | |
| "step": 11820 | |
| }, | |
| { | |
| "epoch": 15.23, | |
| "learning_rate": 0.00017259970457902512, | |
| "loss": 0.178, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 0.00017038404726735595, | |
| "loss": 0.1952, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 15.74, | |
| "learning_rate": 0.00016816838995568683, | |
| "loss": 0.1791, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "learning_rate": 0.0001659527326440177, | |
| "loss": 0.1902, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_cer": 0.037302208050057156, | |
| "eval_loss": 0.22892530262470245, | |
| "eval_runtime": 40.2075, | |
| "eval_samples_per_second": 39.222, | |
| "eval_steps_per_second": 4.924, | |
| "eval_wer": 0.13127217435377597, | |
| "step": 12608 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 0.00016373707533234858, | |
| "loss": 0.1775, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "learning_rate": 0.00016152141802067946, | |
| "loss": 0.1704, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 16.75, | |
| "learning_rate": 0.00015930576070901032, | |
| "loss": 0.1753, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_cer": 0.03613902091731344, | |
| "eval_loss": 0.23591692745685577, | |
| "eval_runtime": 41.9181, | |
| "eval_samples_per_second": 37.621, | |
| "eval_steps_per_second": 4.723, | |
| "eval_wer": 0.12083122149011657, | |
| "step": 13396 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 0.0001570901033973412, | |
| "loss": 0.1875, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 0.00015487444608567209, | |
| "loss": 0.1639, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 17.51, | |
| "learning_rate": 0.00015265878877400292, | |
| "loss": 0.1637, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 17.77, | |
| "learning_rate": 0.0001504431314623338, | |
| "loss": 0.1679, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_cer": 0.03790385656699356, | |
| "eval_loss": 0.22805896401405334, | |
| "eval_runtime": 40.624, | |
| "eval_samples_per_second": 38.819, | |
| "eval_steps_per_second": 4.874, | |
| "eval_wer": 0.12103395843892549, | |
| "step": 14184 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 0.00014822747415066468, | |
| "loss": 0.1639, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 18.27, | |
| "learning_rate": 0.00014601181683899557, | |
| "loss": 0.1567, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 18.53, | |
| "learning_rate": 0.00014379615952732642, | |
| "loss": 0.1548, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 18.78, | |
| "learning_rate": 0.00014158050221565728, | |
| "loss": 0.1558, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_cer": 0.035376932795860655, | |
| "eval_loss": 0.23306047916412354, | |
| "eval_runtime": 40.2559, | |
| "eval_samples_per_second": 39.174, | |
| "eval_steps_per_second": 4.919, | |
| "eval_wer": 0.11910795742524075, | |
| "step": 14972 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 0.00013936484490398817, | |
| "loss": 0.1471, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "learning_rate": 0.00013714918759231905, | |
| "loss": 0.1493, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 19.54, | |
| "learning_rate": 0.0001349335302806499, | |
| "loss": 0.1462, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 0.0001327178729689808, | |
| "loss": 0.1411, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.03383270160239055, | |
| "eval_loss": 0.23826096951961517, | |
| "eval_runtime": 40.4774, | |
| "eval_samples_per_second": 38.96, | |
| "eval_steps_per_second": 4.892, | |
| "eval_wer": 0.11525595539787126, | |
| "step": 15760 | |
| }, | |
| { | |
| "epoch": 20.05, | |
| "learning_rate": 0.00013050221565731165, | |
| "loss": 0.1495, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 20.3, | |
| "learning_rate": 0.00012828655834564253, | |
| "loss": 0.141, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 20.56, | |
| "learning_rate": 0.00012607090103397342, | |
| "loss": 0.1279, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 20.81, | |
| "learning_rate": 0.00012385524372230427, | |
| "loss": 0.1438, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_cer": 0.033652207047309625, | |
| "eval_loss": 0.2469082772731781, | |
| "eval_runtime": 40.7985, | |
| "eval_samples_per_second": 38.653, | |
| "eval_steps_per_second": 4.853, | |
| "eval_wer": 0.11373542828180436, | |
| "step": 16548 | |
| }, | |
| { | |
| "epoch": 21.07, | |
| "learning_rate": 0.00012163958641063514, | |
| "loss": 0.1426, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 21.32, | |
| "learning_rate": 0.00011942392909896601, | |
| "loss": 0.1313, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 21.57, | |
| "learning_rate": 0.0001172082717872969, | |
| "loss": 0.1349, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 21.83, | |
| "learning_rate": 0.00011499261447562776, | |
| "loss": 0.1262, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_cer": 0.03353187734392235, | |
| "eval_loss": 0.25218185782432556, | |
| "eval_runtime": 40.3798, | |
| "eval_samples_per_second": 39.054, | |
| "eval_steps_per_second": 4.903, | |
| "eval_wer": 0.11211353269133299, | |
| "step": 17336 | |
| }, | |
| { | |
| "epoch": 22.08, | |
| "learning_rate": 0.00011277695716395863, | |
| "loss": 0.1223, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 22.33, | |
| "learning_rate": 0.00011056129985228951, | |
| "loss": 0.1162, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 22.59, | |
| "learning_rate": 0.00010834564254062038, | |
| "loss": 0.1181, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 22.84, | |
| "learning_rate": 0.00010612998522895124, | |
| "loss": 0.1109, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.032007701101016785, | |
| "eval_loss": 0.2507190406322479, | |
| "eval_runtime": 40.4365, | |
| "eval_samples_per_second": 38.999, | |
| "eval_steps_per_second": 4.897, | |
| "eval_wer": 0.10562595032944753, | |
| "step": 18124 | |
| }, | |
| { | |
| "epoch": 23.1, | |
| "learning_rate": 0.00010391432791728212, | |
| "loss": 0.1212, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 23.35, | |
| "learning_rate": 0.00010169867060561299, | |
| "loss": 0.1089, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 9.948301329394386e-05, | |
| "loss": 0.1055, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 9.726735598227472e-05, | |
| "loss": 0.122, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_cer": 0.03224836050779135, | |
| "eval_loss": 0.2498846799135208, | |
| "eval_runtime": 40.3499, | |
| "eval_samples_per_second": 39.083, | |
| "eval_steps_per_second": 4.907, | |
| "eval_wer": 0.10876837303598581, | |
| "step": 18912 | |
| }, | |
| { | |
| "epoch": 24.11, | |
| "learning_rate": 9.50516986706056e-05, | |
| "loss": 0.1089, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 24.37, | |
| "learning_rate": 9.283604135893648e-05, | |
| "loss": 0.1053, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 9.062038404726735e-05, | |
| "loss": 0.1104, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 24.87, | |
| "learning_rate": 8.840472673559823e-05, | |
| "loss": 0.1012, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.031045063473918538, | |
| "eval_loss": 0.24705451726913452, | |
| "eval_runtime": 40.2856, | |
| "eval_samples_per_second": 39.146, | |
| "eval_steps_per_second": 4.915, | |
| "eval_wer": 0.10258489609731374, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 25.13, | |
| "learning_rate": 8.618906942392909e-05, | |
| "loss": 0.1187, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 25.38, | |
| "learning_rate": 8.397341211225996e-05, | |
| "loss": 0.1013, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 25.63, | |
| "learning_rate": 8.175775480059084e-05, | |
| "loss": 0.1013, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 25.89, | |
| "learning_rate": 7.954209748892171e-05, | |
| "loss": 0.1005, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_cer": 0.03116539317730582, | |
| "eval_loss": 0.25858306884765625, | |
| "eval_runtime": 40.7124, | |
| "eval_samples_per_second": 38.735, | |
| "eval_steps_per_second": 4.863, | |
| "eval_wer": 0.1042067916877851, | |
| "step": 20488 | |
| }, | |
| { | |
| "epoch": 26.14, | |
| "learning_rate": 7.732644017725257e-05, | |
| "loss": 0.0904, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 26.4, | |
| "learning_rate": 7.511078286558345e-05, | |
| "loss": 0.0923, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 26.65, | |
| "learning_rate": 7.289512555391432e-05, | |
| "loss": 0.0956, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 26.9, | |
| "learning_rate": 7.06794682422452e-05, | |
| "loss": 0.0902, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_cer": 0.030563744660369412, | |
| "eval_loss": 0.25710704922676086, | |
| "eval_runtime": 40.3222, | |
| "eval_samples_per_second": 39.11, | |
| "eval_steps_per_second": 4.91, | |
| "eval_wer": 0.10146984287886468, | |
| "step": 21276 | |
| }, | |
| { | |
| "epoch": 27.16, | |
| "learning_rate": 6.846381093057607e-05, | |
| "loss": 0.0812, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 27.41, | |
| "learning_rate": 6.624815361890694e-05, | |
| "loss": 0.0866, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 27.66, | |
| "learning_rate": 6.40324963072378e-05, | |
| "loss": 0.0907, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 27.92, | |
| "learning_rate": 6.181683899556868e-05, | |
| "loss": 0.0829, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_cer": 0.03272967932134047, | |
| "eval_loss": 0.25568026304244995, | |
| "eval_runtime": 40.5085, | |
| "eval_samples_per_second": 38.93, | |
| "eval_steps_per_second": 4.888, | |
| "eval_wer": 0.1036999493157628, | |
| "step": 22064 | |
| }, | |
| { | |
| "epoch": 28.17, | |
| "learning_rate": 5.960118168389955e-05, | |
| "loss": 0.0857, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 28.43, | |
| "learning_rate": 5.7385524372230425e-05, | |
| "loss": 0.0826, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 28.68, | |
| "learning_rate": 5.5169867060561295e-05, | |
| "loss": 0.0817, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 28.93, | |
| "learning_rate": 5.2954209748892166e-05, | |
| "loss": 0.0773, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_cer": 0.03084451396827307, | |
| "eval_loss": 0.25824475288391113, | |
| "eval_runtime": 40.434, | |
| "eval_samples_per_second": 39.002, | |
| "eval_steps_per_second": 4.897, | |
| "eval_wer": 0.09761784085149519, | |
| "step": 22852 | |
| }, | |
| { | |
| "epoch": 29.19, | |
| "learning_rate": 5.073855243722304e-05, | |
| "loss": 0.0888, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 29.44, | |
| "learning_rate": 4.852289512555391e-05, | |
| "loss": 0.0782, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 29.69, | |
| "learning_rate": 4.6307237813884784e-05, | |
| "loss": 0.0757, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 29.95, | |
| "learning_rate": 4.4091580502215655e-05, | |
| "loss": 0.0795, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_cer": 0.02968132683552935, | |
| "eval_loss": 0.25763553380966187, | |
| "eval_runtime": 40.3436, | |
| "eval_samples_per_second": 39.089, | |
| "eval_steps_per_second": 4.908, | |
| "eval_wer": 0.09670552458185504, | |
| "step": 23640 | |
| }, | |
| { | |
| "epoch": 30.2, | |
| "learning_rate": 4.1875923190546526e-05, | |
| "loss": 0.0678, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 30.46, | |
| "learning_rate": 3.9660265878877396e-05, | |
| "loss": 0.0708, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 30.71, | |
| "learning_rate": 3.744460856720827e-05, | |
| "loss": 0.0764, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 30.96, | |
| "learning_rate": 3.522895125553914e-05, | |
| "loss": 0.0737, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_cer": 0.02891923871407657, | |
| "eval_loss": 0.2654246985912323, | |
| "eval_runtime": 40.3564, | |
| "eval_samples_per_second": 39.077, | |
| "eval_steps_per_second": 4.906, | |
| "eval_wer": 0.09508362899138367, | |
| "step": 24428 | |
| }, | |
| { | |
| "epoch": 31.22, | |
| "learning_rate": 3.301329394387001e-05, | |
| "loss": 0.0688, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 31.47, | |
| "learning_rate": 3.0797636632200885e-05, | |
| "loss": 0.0669, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 31.73, | |
| "learning_rate": 2.8581979320531756e-05, | |
| "loss": 0.0679, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 31.98, | |
| "learning_rate": 2.6366322008862626e-05, | |
| "loss": 0.0744, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_cer": 0.029741491687222992, | |
| "eval_loss": 0.26390764117240906, | |
| "eval_runtime": 40.4656, | |
| "eval_samples_per_second": 38.971, | |
| "eval_steps_per_second": 4.893, | |
| "eval_wer": 0.09640141915864166, | |
| "step": 25216 | |
| }, | |
| { | |
| "epoch": 32.23, | |
| "learning_rate": 2.4150664697193497e-05, | |
| "loss": 0.0618, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 32.49, | |
| "learning_rate": 2.1935007385524368e-05, | |
| "loss": 0.0675, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 32.74, | |
| "learning_rate": 1.971935007385524e-05, | |
| "loss": 0.0637, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 32.99, | |
| "learning_rate": 1.7503692762186116e-05, | |
| "loss": 0.0656, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_cer": 0.029300282774802962, | |
| "eval_loss": 0.26851221919059753, | |
| "eval_runtime": 40.4956, | |
| "eval_samples_per_second": 38.943, | |
| "eval_steps_per_second": 4.889, | |
| "eval_wer": 0.09488089204257476, | |
| "step": 26004 | |
| }, | |
| { | |
| "epoch": 33.25, | |
| "learning_rate": 1.5288035450516986e-05, | |
| "loss": 0.0639, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 33.5, | |
| "learning_rate": 1.3072378138847857e-05, | |
| "loss": 0.0615, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 33.76, | |
| "learning_rate": 1.0856720827178729e-05, | |
| "loss": 0.0661, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_cer": 0.029079678318592945, | |
| "eval_loss": 0.2691967487335205, | |
| "eval_runtime": 40.2387, | |
| "eval_samples_per_second": 39.191, | |
| "eval_steps_per_second": 4.921, | |
| "eval_wer": 0.0937658388241257, | |
| "step": 26792 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 8.6410635155096e-06, | |
| "loss": 0.0608, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 34.26, | |
| "learning_rate": 6.425406203840472e-06, | |
| "loss": 0.0637, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 34.52, | |
| "learning_rate": 4.209748892171343e-06, | |
| "loss": 0.057, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 34.77, | |
| "learning_rate": 1.9940915805022156e-06, | |
| "loss": 0.0603, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_cer": 0.028899183763512023, | |
| "eval_loss": 0.26925861835479736, | |
| "eval_runtime": 40.4005, | |
| "eval_samples_per_second": 39.034, | |
| "eval_steps_per_second": 4.901, | |
| "eval_wer": 0.09356310187531677, | |
| "step": 27580 | |
| } | |
| ], | |
| "max_steps": 27580, | |
| "num_train_epochs": 35, | |
| "total_flos": 3.481960163555136e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |