| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.65345170671973, | |
| "eval_steps": 1000, | |
| "global_step": 200000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019133629266799325, | |
| "grad_norm": 3.534731864929199, | |
| "learning_rate": 0.0002982, | |
| "loss": 2.0385, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03826725853359865, | |
| "grad_norm": 3.997960090637207, | |
| "learning_rate": 0.00029925413533834583, | |
| "loss": 1.5221, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03826725853359865, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3189, | |
| "eval_samples_per_second": 22.199, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.9139914321980114, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.05740088780039798, | |
| "grad_norm": 2.7378287315368652, | |
| "learning_rate": 0.0002985022556390977, | |
| "loss": 1.4633, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.0765345170671973, | |
| "grad_norm": 3.522592782974243, | |
| "learning_rate": 0.0002977503759398496, | |
| "loss": 1.4078, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.0765345170671973, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2822, | |
| "eval_samples_per_second": 22.202, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.9065607150412524, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.09566814633399663, | |
| "grad_norm": 8.136958122253418, | |
| "learning_rate": 0.0002969984962406015, | |
| "loss": 1.3589, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.11480177560079596, | |
| "grad_norm": 2.9784021377563477, | |
| "learning_rate": 0.00029624661654135335, | |
| "loss": 1.336, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.11480177560079596, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7333, | |
| "eval_samples_per_second": 22.24, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.8949386503067485, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.13393540486759528, | |
| "grad_norm": 3.5435404777526855, | |
| "learning_rate": 0.0002954947368421052, | |
| "loss": 1.3258, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.1530690341343946, | |
| "grad_norm": 2.8903892040252686, | |
| "learning_rate": 0.0002947443609022556, | |
| "loss": 1.2887, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1530690341343946, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2047, | |
| "eval_samples_per_second": 22.207, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.8745901205838799, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.17220266340119394, | |
| "grad_norm": 2.647843360900879, | |
| "learning_rate": 0.000293993984962406, | |
| "loss": 1.2713, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.19133629266799326, | |
| "grad_norm": 3.8844950199127197, | |
| "learning_rate": 0.00029324210526315787, | |
| "loss": 1.26, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.19133629266799326, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.2217, | |
| "eval_samples_per_second": 22.276, | |
| "eval_steps_per_second": 2.785, | |
| "eval_wer": 0.8671197376771737, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.21046992193479258, | |
| "grad_norm": 3.8498334884643555, | |
| "learning_rate": 0.00029249022556390974, | |
| "loss": 1.231, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.22960355120159193, | |
| "grad_norm": 2.3221595287323, | |
| "learning_rate": 0.0002917383458646616, | |
| "loss": 1.2188, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.22960355120159193, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.8377, | |
| "eval_samples_per_second": 22.233, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 0.8699756716733658, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24873718046839124, | |
| "grad_norm": 2.9522998332977295, | |
| "learning_rate": 0.00029098646616541353, | |
| "loss": 1.1974, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.26787080973519056, | |
| "grad_norm": 5.0502214431762695, | |
| "learning_rate": 0.0002902345864661654, | |
| "loss": 1.1992, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.26787080973519056, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6699, | |
| "eval_samples_per_second": 22.245, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.8537920456949439, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.2870044390019899, | |
| "grad_norm": 3.480316400527954, | |
| "learning_rate": 0.00028948270676691727, | |
| "loss": 1.1786, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.3061380682687892, | |
| "grad_norm": 3.4583587646484375, | |
| "learning_rate": 0.00028873082706766913, | |
| "loss": 1.1773, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.3061380682687892, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6868, | |
| "eval_samples_per_second": 22.244, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.8329278612227629, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.32527169753558854, | |
| "grad_norm": 3.2121310234069824, | |
| "learning_rate": 0.00028797894736842106, | |
| "loss": 1.1475, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.3444053268023879, | |
| "grad_norm": 6.1687846183776855, | |
| "learning_rate": 0.0002872285714285714, | |
| "loss": 1.1602, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.3444053268023879, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.9185, | |
| "eval_samples_per_second": 22.227, | |
| "eval_steps_per_second": 2.779, | |
| "eval_wer": 0.8200497144066003, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.3635389560691872, | |
| "grad_norm": 2.9368505477905273, | |
| "learning_rate": 0.0002864766917293233, | |
| "loss": 1.1233, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.3826725853359865, | |
| "grad_norm": 2.0912511348724365, | |
| "learning_rate": 0.00028572481203007513, | |
| "loss": 1.1128, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.3826725853359865, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2709, | |
| "eval_samples_per_second": 22.202, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.8059287074254284, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.40180621460278587, | |
| "grad_norm": 2.822795867919922, | |
| "learning_rate": 0.00028497293233082705, | |
| "loss": 1.096, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.42093984386958516, | |
| "grad_norm": 2.4663002490997314, | |
| "learning_rate": 0.0002842210526315789, | |
| "loss": 1.0893, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.42093984386958516, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.2266, | |
| "eval_samples_per_second": 22.136, | |
| "eval_steps_per_second": 2.768, | |
| "eval_wer": 0.827678760313095, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.4400734731363845, | |
| "grad_norm": 1.9610426425933838, | |
| "learning_rate": 0.0002834691729323308, | |
| "loss": 1.0702, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.45920710240318385, | |
| "grad_norm": 3.893796682357788, | |
| "learning_rate": 0.00028271729323308266, | |
| "loss": 1.0809, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.45920710240318385, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2707, | |
| "eval_samples_per_second": 22.202, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.8028480008462027, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.47834073166998314, | |
| "grad_norm": 3.08317494392395, | |
| "learning_rate": 0.0002819669172932331, | |
| "loss": 1.0616, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.4974743609367825, | |
| "grad_norm": 1.9941602945327759, | |
| "learning_rate": 0.0002812150375939849, | |
| "loss": 1.0426, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.4974743609367825, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.9265, | |
| "eval_samples_per_second": 22.297, | |
| "eval_steps_per_second": 2.788, | |
| "eval_wer": 0.8004416120160779, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.5166079902035818, | |
| "grad_norm": 2.075686454772949, | |
| "learning_rate": 0.0002804646616541353, | |
| "loss": 1.0316, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.5357416194703811, | |
| "grad_norm": 1.7053288221359253, | |
| "learning_rate": 0.0002797127819548872, | |
| "loss": 1.0202, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.5357416194703811, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6762, | |
| "eval_samples_per_second": 22.244, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.7824598053733869, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.5548752487371804, | |
| "grad_norm": 2.8614988327026367, | |
| "learning_rate": 0.00027896240601503757, | |
| "loss": 1.0398, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.5740088780039798, | |
| "grad_norm": 2.7677505016326904, | |
| "learning_rate": 0.0002782105263157895, | |
| "loss": 1.0005, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5740088780039798, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.6237, | |
| "eval_samples_per_second": 22.178, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.7785725618785699, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5931425072707791, | |
| "grad_norm": 1.8297598361968994, | |
| "learning_rate": 0.0002774586466165413, | |
| "loss": 1.0069, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.6122761365375784, | |
| "grad_norm": 3.259901285171509, | |
| "learning_rate": 0.0002767067669172932, | |
| "loss": 0.9987, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6122761365375784, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.076, | |
| "eval_samples_per_second": 22.146, | |
| "eval_steps_per_second": 2.769, | |
| "eval_wer": 0.752102284747197, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6314097658043778, | |
| "grad_norm": 2.1204450130462646, | |
| "learning_rate": 0.0002759548872180451, | |
| "loss": 0.9862, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.6505433950711771, | |
| "grad_norm": 4.308948993682861, | |
| "learning_rate": 0.00027520300751879696, | |
| "loss": 0.9705, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.6505433950711771, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.4489, | |
| "eval_samples_per_second": 22.19, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.7592817854876243, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.6696770243379764, | |
| "grad_norm": 3.809417247772217, | |
| "learning_rate": 0.00027445112781954883, | |
| "loss": 0.9836, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.6888106536047758, | |
| "grad_norm": 8.3826904296875, | |
| "learning_rate": 0.00027369924812030075, | |
| "loss": 0.9884, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.6888106536047758, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.2775, | |
| "eval_samples_per_second": 22.132, | |
| "eval_steps_per_second": 2.767, | |
| "eval_wer": 0.7380606092659192, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.7079442828715751, | |
| "grad_norm": 2.119809627532959, | |
| "learning_rate": 0.0002729473684210526, | |
| "loss": 0.9554, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.7270779121383744, | |
| "grad_norm": 2.714183807373047, | |
| "learning_rate": 0.0002721954887218045, | |
| "loss": 0.9618, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.7270779121383744, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8028, | |
| "eval_samples_per_second": 22.165, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.7364739792680347, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.7462115414051738, | |
| "grad_norm": 4.593650817871094, | |
| "learning_rate": 0.00027144360902255635, | |
| "loss": 0.9389, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.765345170671973, | |
| "grad_norm": 1.7419074773788452, | |
| "learning_rate": 0.0002706917293233083, | |
| "loss": 0.9374, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.765345170671973, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3558, | |
| "eval_samples_per_second": 22.197, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.7415644171779141, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.7844787999387723, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002699413533834586, | |
| "loss": 0.9241, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.8036124292055717, | |
| "grad_norm": 2.104843854904175, | |
| "learning_rate": 0.0002691894736842105, | |
| "loss": 0.9175, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.8036124292055717, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.3422, | |
| "eval_samples_per_second": 22.268, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.7299291305267611, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.822746058472371, | |
| "grad_norm": 3.1283345222473145, | |
| "learning_rate": 0.0002684390977443609, | |
| "loss": 0.9193, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.8418796877391703, | |
| "grad_norm": 3.1702754497528076, | |
| "learning_rate": 0.00026768721804511274, | |
| "loss": 0.9247, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.8418796877391703, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0504, | |
| "eval_samples_per_second": 22.218, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.7260154431986461, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.8610133170059697, | |
| "grad_norm": 4.537879943847656, | |
| "learning_rate": 0.00026693533834586466, | |
| "loss": 0.9154, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.880146946272769, | |
| "grad_norm": 3.9466328620910645, | |
| "learning_rate": 0.00026618345864661653, | |
| "loss": 0.9001, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.880146946272769, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.881, | |
| "eval_samples_per_second": 22.23, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 0.72349005711868, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.8992805755395683, | |
| "grad_norm": 5.256113052368164, | |
| "learning_rate": 0.0002654315789473684, | |
| "loss": 0.9027, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.9184142048063677, | |
| "grad_norm": 2.1492791175842285, | |
| "learning_rate": 0.00026467969924812027, | |
| "loss": 0.8836, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.9184142048063677, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.4361, | |
| "eval_samples_per_second": 22.261, | |
| "eval_steps_per_second": 2.783, | |
| "eval_wer": 0.7085757351385656, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.937547834073167, | |
| "grad_norm": 3.541306734085083, | |
| "learning_rate": 0.0002639278195488722, | |
| "loss": 0.8782, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.9566814633399663, | |
| "grad_norm": 1.7392828464508057, | |
| "learning_rate": 0.000263175939849624, | |
| "loss": 0.8789, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.9566814633399663, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1779, | |
| "eval_samples_per_second": 22.209, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.7144859318806854, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.9758150926067657, | |
| "grad_norm": 2.769277572631836, | |
| "learning_rate": 0.0002624240601503759, | |
| "loss": 0.8776, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.994948721873565, | |
| "grad_norm": 3.246126413345337, | |
| "learning_rate": 0.0002616721804511278, | |
| "loss": 0.8734, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.994948721873565, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.5548, | |
| "eval_samples_per_second": 22.253, | |
| "eval_steps_per_second": 2.782, | |
| "eval_wer": 0.7195102602073197, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.0140823511403643, | |
| "grad_norm": 1.4256001710891724, | |
| "learning_rate": 0.00026092030075187966, | |
| "loss": 0.8471, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.0332159804071637, | |
| "grad_norm": 1.2950644493103027, | |
| "learning_rate": 0.00026016992481203005, | |
| "loss": 0.8398, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.0332159804071637, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.7022, | |
| "eval_samples_per_second": 22.313, | |
| "eval_steps_per_second": 2.79, | |
| "eval_wer": 0.6837978633382695, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.0523496096739628, | |
| "grad_norm": 0.8874345421791077, | |
| "learning_rate": 0.0002594180451127819, | |
| "loss": 0.8403, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.0714832389407623, | |
| "grad_norm": 1.146148920059204, | |
| "learning_rate": 0.0002586661654135338, | |
| "loss": 0.8268, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.0714832389407623, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.7884, | |
| "eval_samples_per_second": 22.378, | |
| "eval_steps_per_second": 2.798, | |
| "eval_wer": 0.6793288555108948, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.0906168682075617, | |
| "grad_norm": 0.687147319316864, | |
| "learning_rate": 0.0002579142857142857, | |
| "loss": 0.8228, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.1097504974743608, | |
| "grad_norm": 0.614025890827179, | |
| "learning_rate": 0.00025716390977443605, | |
| "loss": 0.8196, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.1097504974743608, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.3179, | |
| "eval_samples_per_second": 22.27, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.66387243494817, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.1288841267411602, | |
| "grad_norm": 0.7900418043136597, | |
| "learning_rate": 0.0002564120300751879, | |
| "loss": 0.8262, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 1.1480177560079596, | |
| "grad_norm": 0.9414839148521423, | |
| "learning_rate": 0.00025566015037593984, | |
| "loss": 0.8124, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.1480177560079596, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.1972, | |
| "eval_samples_per_second": 22.349, | |
| "eval_steps_per_second": 2.794, | |
| "eval_wer": 0.6615453776179395, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.1671513852747588, | |
| "grad_norm": 0.9619298577308655, | |
| "learning_rate": 0.0002549082706766917, | |
| "loss": 0.8052, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 1.1862850145415582, | |
| "grad_norm": 1.3762531280517578, | |
| "learning_rate": 0.0002541563909774436, | |
| "loss": 0.7935, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.1862850145415582, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.3254, | |
| "eval_samples_per_second": 22.269, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.6607520626189972, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.2054186438083576, | |
| "grad_norm": 1.0670289993286133, | |
| "learning_rate": 0.00025340451127819544, | |
| "loss": 0.799, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 1.2245522730751568, | |
| "grad_norm": 1.5455262660980225, | |
| "learning_rate": 0.00025265263157894736, | |
| "loss": 0.817, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.2245522730751568, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.3401, | |
| "eval_samples_per_second": 22.339, | |
| "eval_steps_per_second": 2.793, | |
| "eval_wer": 0.6709726041887032, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.2436859023419562, | |
| "grad_norm": 1.185735821723938, | |
| "learning_rate": 0.00025190075187969923, | |
| "loss": 0.7835, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 1.2628195316087556, | |
| "grad_norm": 0.8058122396469116, | |
| "learning_rate": 0.0002511503759398496, | |
| "loss": 0.7975, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.2628195316087556, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7071, | |
| "eval_samples_per_second": 22.242, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.66950497144066, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.2819531608755548, | |
| "grad_norm": 0.6225046515464783, | |
| "learning_rate": 0.0002503984962406015, | |
| "loss": 0.8023, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 1.3010867901423542, | |
| "grad_norm": 0.8266538381576538, | |
| "learning_rate": 0.00024964661654135336, | |
| "loss": 0.7746, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.3010867901423542, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3157, | |
| "eval_samples_per_second": 22.199, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.6674423524434102, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.3202204194091536, | |
| "grad_norm": 0.8520437479019165, | |
| "learning_rate": 0.00024889624060150375, | |
| "loss": 0.7828, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 1.3393540486759528, | |
| "grad_norm": 0.9565845131874084, | |
| "learning_rate": 0.0002481443609022556, | |
| "loss": 0.8013, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.3393540486759528, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.1504, | |
| "eval_samples_per_second": 22.281, | |
| "eval_steps_per_second": 2.786, | |
| "eval_wer": 0.6586762217050983, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.3584876779427522, | |
| "grad_norm": 1.2239874601364136, | |
| "learning_rate": 0.0002473924812030075, | |
| "loss": 0.7765, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 1.3776213072095516, | |
| "grad_norm": 0.6564140319824219, | |
| "learning_rate": 0.00024664060150375935, | |
| "loss": 0.7703, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.3776213072095516, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0508, | |
| "eval_samples_per_second": 22.218, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.6388301248148932, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.3967549364763507, | |
| "grad_norm": 0.7060734629631042, | |
| "learning_rate": 0.0002458887218045113, | |
| "loss": 0.7685, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 1.4158885657431501, | |
| "grad_norm": 1.3065494298934937, | |
| "learning_rate": 0.00024513684210526314, | |
| "loss": 0.7581, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.4158885657431501, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.8618, | |
| "eval_samples_per_second": 22.231, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 0.6461021789718637, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.4350221950099495, | |
| "grad_norm": 1.282378911972046, | |
| "learning_rate": 0.00024438646616541354, | |
| "loss": 0.7693, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 1.4541558242767487, | |
| "grad_norm": 1.1911076307296753, | |
| "learning_rate": 0.00024363458646616538, | |
| "loss": 0.7468, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.4541558242767487, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.6111, | |
| "eval_samples_per_second": 22.179, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.6333959170721388, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.4732894535435481, | |
| "grad_norm": 0.7600271105766296, | |
| "learning_rate": 0.00024288270676691727, | |
| "loss": 0.7582, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 1.4924230828103475, | |
| "grad_norm": 0.6814852356910706, | |
| "learning_rate": 0.00024213082706766914, | |
| "loss": 0.7534, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.4924230828103475, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0864, | |
| "eval_samples_per_second": 22.215, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.6300639940765813, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.5115567120771467, | |
| "grad_norm": 0.7314792275428772, | |
| "learning_rate": 0.00024137894736842104, | |
| "loss": 0.7336, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 1.530690341343946, | |
| "grad_norm": 1.1860034465789795, | |
| "learning_rate": 0.0002406270676691729, | |
| "loss": 0.752, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.530690341343946, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.462, | |
| "eval_samples_per_second": 22.05, | |
| "eval_steps_per_second": 2.757, | |
| "eval_wer": 0.6221969536704041, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.5498239706107455, | |
| "grad_norm": 1.3469467163085938, | |
| "learning_rate": 0.0002398766917293233, | |
| "loss": 0.7508, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 1.5689575998775447, | |
| "grad_norm": 0.7789831757545471, | |
| "learning_rate": 0.00023912481203007516, | |
| "loss": 0.736, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.5689575998775447, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8112, | |
| "eval_samples_per_second": 22.165, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.6203326634228897, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.588091229144344, | |
| "grad_norm": 1.3181277513504028, | |
| "learning_rate": 0.00023837443609022553, | |
| "loss": 0.752, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 1.6072248584111435, | |
| "grad_norm": 0.6017114520072937, | |
| "learning_rate": 0.00023762255639097742, | |
| "loss": 0.7188, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.6072248584111435, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.1913, | |
| "eval_samples_per_second": 22.279, | |
| "eval_steps_per_second": 2.786, | |
| "eval_wer": 0.620795430505606, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.6263584876779427, | |
| "grad_norm": 0.7502321600914001, | |
| "learning_rate": 0.0002368706766917293, | |
| "loss": 0.7138, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 1.645492116944742, | |
| "grad_norm": 0.6769944429397583, | |
| "learning_rate": 0.00023611879699248119, | |
| "loss": 0.7308, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.645492116944742, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0035, | |
| "eval_samples_per_second": 22.221, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.6056695578591073, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.6646257462115415, | |
| "grad_norm": 1.050374984741211, | |
| "learning_rate": 0.00023536691729323305, | |
| "loss": 0.7145, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 1.6837593754783406, | |
| "grad_norm": 1.2912209033966064, | |
| "learning_rate": 0.00023461503759398495, | |
| "loss": 0.7179, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.6837593754783406, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6998, | |
| "eval_samples_per_second": 22.243, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.6291649037444468, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.70289300474514, | |
| "grad_norm": 0.8525875210762024, | |
| "learning_rate": 0.00023386315789473682, | |
| "loss": 0.7237, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 1.7220266340119394, | |
| "grad_norm": 1.1356332302093506, | |
| "learning_rate": 0.0002331112781954887, | |
| "loss": 0.7341, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.7220266340119394, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.323, | |
| "eval_samples_per_second": 22.199, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.6034218320287709, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.7411602632787386, | |
| "grad_norm": 0.6900098323822021, | |
| "learning_rate": 0.00023235939849624058, | |
| "loss": 0.7166, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 1.760293892545538, | |
| "grad_norm": 0.7305801510810852, | |
| "learning_rate": 0.00023160902255639097, | |
| "loss": 0.7061, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.760293892545538, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.9517, | |
| "eval_samples_per_second": 22.225, | |
| "eval_steps_per_second": 2.779, | |
| "eval_wer": 0.6136555955151258, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.7794275218123374, | |
| "grad_norm": 1.960204839706421, | |
| "learning_rate": 0.00023085714285714284, | |
| "loss": 0.688, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 1.7985611510791366, | |
| "grad_norm": 1.001535415649414, | |
| "learning_rate": 0.00023010526315789473, | |
| "loss": 0.7081, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.7985611510791366, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.7057, | |
| "eval_samples_per_second": 22.172, | |
| "eval_steps_per_second": 2.772, | |
| "eval_wer": 0.6123334038502222, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.817694780345936, | |
| "grad_norm": 0.8360883593559265, | |
| "learning_rate": 0.0002293533834586466, | |
| "loss": 0.704, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 1.8368284096127354, | |
| "grad_norm": 1.349821925163269, | |
| "learning_rate": 0.0002286015037593985, | |
| "loss": 0.6957, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.8368284096127354, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3207, | |
| "eval_samples_per_second": 22.199, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.6053522318595304, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.8559620388795346, | |
| "grad_norm": 0.9751301407814026, | |
| "learning_rate": 0.00022785112781954886, | |
| "loss": 0.6974, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 1.875095668146334, | |
| "grad_norm": 0.7246661186218262, | |
| "learning_rate": 0.00022710075187969923, | |
| "loss": 0.7052, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.875095668146334, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.447, | |
| "eval_samples_per_second": 22.26, | |
| "eval_steps_per_second": 2.783, | |
| "eval_wer": 0.616498307594669, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.8942292974131334, | |
| "grad_norm": 0.716491162776947, | |
| "learning_rate": 0.00022634887218045112, | |
| "loss": 0.6898, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 1.9133629266799326, | |
| "grad_norm": 0.7438942193984985, | |
| "learning_rate": 0.000225596992481203, | |
| "loss": 0.6833, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.9133629266799326, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1363, | |
| "eval_samples_per_second": 22.212, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.5887455045483393, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.932496555946732, | |
| "grad_norm": 0.8066436648368835, | |
| "learning_rate": 0.00022484511278195488, | |
| "loss": 0.6755, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 1.9516301852135314, | |
| "grad_norm": 0.6446587443351746, | |
| "learning_rate": 0.00022409323308270672, | |
| "loss": 0.6995, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.9516301852135314, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7016, | |
| "eval_samples_per_second": 22.243, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.5871456526338058, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.9707638144803306, | |
| "grad_norm": 0.8823833465576172, | |
| "learning_rate": 0.00022334135338345862, | |
| "loss": 0.6909, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 1.98989744374713, | |
| "grad_norm": 0.8129588961601257, | |
| "learning_rate": 0.0002225894736842105, | |
| "loss": 0.6703, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.98989744374713, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.397, | |
| "eval_samples_per_second": 22.194, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.5954357943727523, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.0090310730139294, | |
| "grad_norm": 0.9364180564880371, | |
| "learning_rate": 0.0002218390977443609, | |
| "loss": 0.6462, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 2.0281647022807285, | |
| "grad_norm": 0.5429893732070923, | |
| "learning_rate": 0.00022108721804511275, | |
| "loss": 0.6265, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.0281647022807285, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6741, | |
| "eval_samples_per_second": 22.244, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.5791728368944362, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.0472983315475277, | |
| "grad_norm": 0.7964287996292114, | |
| "learning_rate": 0.00022033533834586464, | |
| "loss": 0.6283, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 2.0664319608143273, | |
| "grad_norm": 0.928997278213501, | |
| "learning_rate": 0.0002195834586466165, | |
| "loss": 0.633, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.0664319608143273, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8143, | |
| "eval_samples_per_second": 22.164, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.5696001692405331, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.0855655900811265, | |
| "grad_norm": 0.9358041882514954, | |
| "learning_rate": 0.0002188315789473684, | |
| "loss": 0.6608, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 2.1046992193479257, | |
| "grad_norm": 0.49673086404800415, | |
| "learning_rate": 0.00021807969924812027, | |
| "loss": 0.6399, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.1046992193479257, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.4793, | |
| "eval_samples_per_second": 22.188, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.5717685635709753, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.1238328486147253, | |
| "grad_norm": 0.824380099773407, | |
| "learning_rate": 0.00021732932330827064, | |
| "loss": 0.6317, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 2.1429664778815245, | |
| "grad_norm": 0.784487783908844, | |
| "learning_rate": 0.00021657744360902253, | |
| "loss": 0.6165, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.1429664778815245, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.9557, | |
| "eval_samples_per_second": 22.295, | |
| "eval_steps_per_second": 2.788, | |
| "eval_wer": 0.5836550666384599, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.1621001071483237, | |
| "grad_norm": 1.1931605339050293, | |
| "learning_rate": 0.0002158270676691729, | |
| "loss": 0.6268, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 2.1812337364151233, | |
| "grad_norm": 1.4376397132873535, | |
| "learning_rate": 0.0002150751879699248, | |
| "loss": 0.6148, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 2.1812337364151233, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.0058, | |
| "eval_samples_per_second": 22.363, | |
| "eval_steps_per_second": 2.796, | |
| "eval_wer": 0.5597233975037021, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 2.2003673656819225, | |
| "grad_norm": 1.6363264322280884, | |
| "learning_rate": 0.00021432330827067666, | |
| "loss": 0.6377, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 2.2195009949487217, | |
| "grad_norm": 4.29092264175415, | |
| "learning_rate": 0.00021357142857142855, | |
| "loss": 0.6228, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.2195009949487217, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.1699, | |
| "eval_samples_per_second": 22.351, | |
| "eval_steps_per_second": 2.795, | |
| "eval_wer": 0.5706843664057542, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.2386346242155213, | |
| "grad_norm": 1.1523572206497192, | |
| "learning_rate": 0.00021281954887218042, | |
| "loss": 0.6243, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 2.2577682534823205, | |
| "grad_norm": 2.0131170749664307, | |
| "learning_rate": 0.00021206766917293232, | |
| "loss": 0.6302, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.2577682534823205, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.1147, | |
| "eval_samples_per_second": 22.284, | |
| "eval_steps_per_second": 2.786, | |
| "eval_wer": 0.5717685635709753, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.2769018827491196, | |
| "grad_norm": 1.358688235282898, | |
| "learning_rate": 0.00021131578947368419, | |
| "loss": 0.612, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 2.2960355120159193, | |
| "grad_norm": 0.5921105742454529, | |
| "learning_rate": 0.00021056390977443608, | |
| "loss": 0.6035, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.2960355120159193, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.6822, | |
| "eval_samples_per_second": 22.315, | |
| "eval_steps_per_second": 2.79, | |
| "eval_wer": 0.5638883012481489, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.3151691412827184, | |
| "grad_norm": 0.600351095199585, | |
| "learning_rate": 0.00020981203007518795, | |
| "loss": 0.6157, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 2.3343027705495176, | |
| "grad_norm": 1.6874371767044067, | |
| "learning_rate": 0.00020906015037593984, | |
| "loss": 0.602, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.3343027705495176, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.103, | |
| "eval_samples_per_second": 22.214, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.5633197588322403, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.3534363998163172, | |
| "grad_norm": 0.5364500284194946, | |
| "learning_rate": 0.0002083082706766917, | |
| "loss": 0.6057, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 2.3725700290831164, | |
| "grad_norm": 1.9607787132263184, | |
| "learning_rate": 0.0002075563909774436, | |
| "loss": 0.6023, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.3725700290831164, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.4546, | |
| "eval_samples_per_second": 22.331, | |
| "eval_steps_per_second": 2.792, | |
| "eval_wer": 0.5581103236725196, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.3917036583499156, | |
| "grad_norm": 1.1173665523529053, | |
| "learning_rate": 0.00020680451127819547, | |
| "loss": 0.6097, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 2.4108372876167152, | |
| "grad_norm": 1.7585097551345825, | |
| "learning_rate": 0.00020605563909774434, | |
| "loss": 0.5924, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.4108372876167152, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.7721, | |
| "eval_samples_per_second": 22.308, | |
| "eval_steps_per_second": 2.789, | |
| "eval_wer": 0.5511820393484239, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.4299709168835144, | |
| "grad_norm": 0.9437362551689148, | |
| "learning_rate": 0.00020530375939849623, | |
| "loss": 0.5968, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 2.4491045461503136, | |
| "grad_norm": 0.6077060103416443, | |
| "learning_rate": 0.0002045518796992481, | |
| "loss": 0.5969, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.4491045461503136, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.4786, | |
| "eval_samples_per_second": 22.329, | |
| "eval_steps_per_second": 2.792, | |
| "eval_wer": 0.5489739792680347, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.468238175417113, | |
| "grad_norm": 1.9816350936889648, | |
| "learning_rate": 0.0002038, | |
| "loss": 0.5978, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 2.4873718046839124, | |
| "grad_norm": 1.865081787109375, | |
| "learning_rate": 0.00020304812030075186, | |
| "loss": 0.6029, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.4873718046839124, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.4282, | |
| "eval_samples_per_second": 22.333, | |
| "eval_steps_per_second": 2.792, | |
| "eval_wer": 0.5444124180241168, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.5065054339507116, | |
| "grad_norm": 0.9471901059150696, | |
| "learning_rate": 0.00020229624060150376, | |
| "loss": 0.6068, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 2.525639063217511, | |
| "grad_norm": 1.5803519487380981, | |
| "learning_rate": 0.0002015443609022556, | |
| "loss": 0.6046, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.525639063217511, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3481, | |
| "eval_samples_per_second": 22.197, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.5460519356885974, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.5447726924843104, | |
| "grad_norm": 0.7890714406967163, | |
| "learning_rate": 0.0002007924812030075, | |
| "loss": 0.5865, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 2.5639063217511096, | |
| "grad_norm": 0.7467088103294373, | |
| "learning_rate": 0.00020004060150375936, | |
| "loss": 0.6095, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.5639063217511096, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.5673, | |
| "eval_samples_per_second": 22.182, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.5476253437698329, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.583039951017909, | |
| "grad_norm": 0.7762987613677979, | |
| "learning_rate": 0.00019928872180451126, | |
| "loss": 0.5869, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 2.6021735802847084, | |
| "grad_norm": 0.5018890500068665, | |
| "learning_rate": 0.00019853684210526312, | |
| "loss": 0.598, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.6021735802847084, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3997, | |
| "eval_samples_per_second": 22.193, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.532129257457161, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.6213072095515075, | |
| "grad_norm": 1.0056216716766357, | |
| "learning_rate": 0.00019778496240601502, | |
| "loss": 0.5821, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 2.640440838818307, | |
| "grad_norm": 1.3957178592681885, | |
| "learning_rate": 0.00019703308270676689, | |
| "loss": 0.5812, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.640440838818307, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6807, | |
| "eval_samples_per_second": 22.244, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.5357388407023482, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 0.6841593384742737, | |
| "learning_rate": 0.00019628120300751878, | |
| "loss": 0.5802, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 2.6787080973519055, | |
| "grad_norm": 0.8700592517852783, | |
| "learning_rate": 0.00019552932330827065, | |
| "loss": 0.5957, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.6787080973519055, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.4344, | |
| "eval_samples_per_second": 22.191, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.5368230378675692, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.697841726618705, | |
| "grad_norm": 1.0066908597946167, | |
| "learning_rate": 0.00019477894736842104, | |
| "loss": 0.5817, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 2.7169753558855043, | |
| "grad_norm": 0.8253029584884644, | |
| "learning_rate": 0.0001940270676691729, | |
| "loss": 0.5909, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.7169753558855043, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.9129, | |
| "eval_samples_per_second": 22.019, | |
| "eval_steps_per_second": 2.753, | |
| "eval_wer": 0.5239448910514068, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.7361089851523035, | |
| "grad_norm": 0.9649154543876648, | |
| "learning_rate": 0.0001932751879699248, | |
| "loss": 0.5877, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 2.755242614419103, | |
| "grad_norm": 1.6835025548934937, | |
| "learning_rate": 0.00019252330827067667, | |
| "loss": 0.5953, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.755242614419103, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.62, | |
| "eval_samples_per_second": 22.178, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.5421911360270785, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.7743762436859023, | |
| "grad_norm": 1.4002177715301514, | |
| "learning_rate": 0.00019177443609022553, | |
| "loss": 0.5811, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 2.7935098729527015, | |
| "grad_norm": 1.1865595579147339, | |
| "learning_rate": 0.00019102255639097743, | |
| "loss": 0.5702, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.7935098729527015, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.9208, | |
| "eval_samples_per_second": 22.227, | |
| "eval_steps_per_second": 2.779, | |
| "eval_wer": 0.5225698117199069, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.812643502219501, | |
| "grad_norm": 0.8575685620307922, | |
| "learning_rate": 0.0001902706766917293, | |
| "loss": 0.5806, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 2.8317771314863003, | |
| "grad_norm": 0.9127354025840759, | |
| "learning_rate": 0.0001895187969924812, | |
| "loss": 0.5755, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.8317771314863003, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.601, | |
| "eval_samples_per_second": 22.25, | |
| "eval_steps_per_second": 2.782, | |
| "eval_wer": 0.5319441506240745, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.8509107607530995, | |
| "grad_norm": 1.893068552017212, | |
| "learning_rate": 0.00018876691729323306, | |
| "loss": 0.574, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 2.870044390019899, | |
| "grad_norm": 0.7603012323379517, | |
| "learning_rate": 0.00018801654135338345, | |
| "loss": 0.5659, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.870044390019899, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7507, | |
| "eval_samples_per_second": 22.239, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.5286783372117622, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.8891780192866983, | |
| "grad_norm": 1.636072039604187, | |
| "learning_rate": 0.00018726466165413532, | |
| "loss": 0.5645, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 2.9083116485534974, | |
| "grad_norm": 0.8112033605575562, | |
| "learning_rate": 0.0001865127819548872, | |
| "loss": 0.5581, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.9083116485534974, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6028, | |
| "eval_samples_per_second": 22.249, | |
| "eval_steps_per_second": 2.782, | |
| "eval_wer": 0.5277924687962767, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.927445277820297, | |
| "grad_norm": 0.8813944458961487, | |
| "learning_rate": 0.00018576090225563908, | |
| "loss": 0.573, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 2.9465789070870962, | |
| "grad_norm": 0.7851129174232483, | |
| "learning_rate": 0.00018501052631578945, | |
| "loss": 0.5786, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.9465789070870962, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8095, | |
| "eval_samples_per_second": 22.165, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.5194626613073832, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.9657125363538954, | |
| "grad_norm": 0.9913876056671143, | |
| "learning_rate": 0.00018425864661654134, | |
| "loss": 0.5783, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 2.984846165620695, | |
| "grad_norm": 0.9374109506607056, | |
| "learning_rate": 0.0001835067669172932, | |
| "loss": 0.5485, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.984846165620695, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.6351, | |
| "eval_samples_per_second": 22.177, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.5255976306325365, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 3.0039797948874942, | |
| "grad_norm": 0.9412303566932678, | |
| "learning_rate": 0.0001827548872180451, | |
| "loss": 0.543, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 3.0231134241542934, | |
| "grad_norm": 0.9871559739112854, | |
| "learning_rate": 0.00018200300751879697, | |
| "loss": 0.5113, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 3.0231134241542934, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.7492, | |
| "eval_samples_per_second": 22.169, | |
| "eval_steps_per_second": 2.772, | |
| "eval_wer": 0.5220277131372963, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 3.042247053421093, | |
| "grad_norm": 1.3727389574050903, | |
| "learning_rate": 0.00018125112781954887, | |
| "loss": 0.5075, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 3.061380682687892, | |
| "grad_norm": 1.3193981647491455, | |
| "learning_rate": 0.00018049924812030073, | |
| "loss": 0.4973, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.061380682687892, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.9407, | |
| "eval_samples_per_second": 22.226, | |
| "eval_steps_per_second": 2.779, | |
| "eval_wer": 0.5146102178971864, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.0805143119546914, | |
| "grad_norm": 1.4254885911941528, | |
| "learning_rate": 0.00017974887218045113, | |
| "loss": 0.5063, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 3.099647941221491, | |
| "grad_norm": 1.1925376653671265, | |
| "learning_rate": 0.000178996992481203, | |
| "loss": 0.5085, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 3.099647941221491, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.3194, | |
| "eval_samples_per_second": 22.269, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.5240903321345463, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 3.11878157048829, | |
| "grad_norm": 0.8363128304481506, | |
| "learning_rate": 0.00017824661654135339, | |
| "loss": 0.5263, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 3.1379151997550894, | |
| "grad_norm": 0.8608238697052002, | |
| "learning_rate": 0.00017749473684210525, | |
| "loss": 0.5111, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 3.1379151997550894, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.6205, | |
| "eval_samples_per_second": 22.178, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.5104585360693886, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 3.157048829021889, | |
| "grad_norm": 0.7947099208831787, | |
| "learning_rate": 0.00017674285714285715, | |
| "loss": 0.496, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 3.176182458288688, | |
| "grad_norm": 0.629405677318573, | |
| "learning_rate": 0.00017599097744360902, | |
| "loss": 0.5047, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 3.176182458288688, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.567, | |
| "eval_samples_per_second": 22.252, | |
| "eval_steps_per_second": 2.782, | |
| "eval_wer": 0.5117675058176433, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 3.1953160875554873, | |
| "grad_norm": 0.6956018209457397, | |
| "learning_rate": 0.0001752390977443609, | |
| "loss": 0.519, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 3.214449716822287, | |
| "grad_norm": 0.4928194284439087, | |
| "learning_rate": 0.00017448721804511275, | |
| "loss": 0.4994, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 3.214449716822287, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.5623, | |
| "eval_samples_per_second": 22.182, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.49931246033425003, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 3.233583346089086, | |
| "grad_norm": 0.9308450222015381, | |
| "learning_rate": 0.00017373533834586465, | |
| "loss": 0.5108, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 3.2527169753558853, | |
| "grad_norm": 1.1794687509536743, | |
| "learning_rate": 0.00017298345864661652, | |
| "loss": 0.5077, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 3.2527169753558853, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1672, | |
| "eval_samples_per_second": 22.21, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.5099825470700232, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 3.271850604622685, | |
| "grad_norm": 0.6575067639350891, | |
| "learning_rate": 0.0001722315789473684, | |
| "loss": 0.5123, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 3.290984233889484, | |
| "grad_norm": 0.60300612449646, | |
| "learning_rate": 0.00017148120300751877, | |
| "loss": 0.5035, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 3.290984233889484, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8684, | |
| "eval_samples_per_second": 22.161, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.49292627459276495, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 3.3101178631562833, | |
| "grad_norm": 1.378197193145752, | |
| "learning_rate": 0.00017072932330827064, | |
| "loss": 0.5036, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 3.329251492423083, | |
| "grad_norm": 0.6758792996406555, | |
| "learning_rate": 0.00016997894736842103, | |
| "loss": 0.5045, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 3.329251492423083, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6861, | |
| "eval_samples_per_second": 22.244, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.5026311614131584, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 3.348385121689882, | |
| "grad_norm": 0.708972692489624, | |
| "learning_rate": 0.0001692270676691729, | |
| "loss": 0.5057, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 3.3675187509566813, | |
| "grad_norm": 1.0322130918502808, | |
| "learning_rate": 0.0001684751879699248, | |
| "loss": 0.4951, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 3.3675187509566813, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1978, | |
| "eval_samples_per_second": 22.208, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.49707795642056274, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 3.386652380223481, | |
| "grad_norm": 0.5131962299346924, | |
| "learning_rate": 0.00016772330827067667, | |
| "loss": 0.4925, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 3.40578600949028, | |
| "grad_norm": 0.7158399820327759, | |
| "learning_rate": 0.00016697142857142856, | |
| "loss": 0.4915, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 3.40578600949028, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 319.2292, | |
| "eval_samples_per_second": 21.997, | |
| "eval_steps_per_second": 2.75, | |
| "eval_wer": 0.4984133700021155, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 3.4249196387570793, | |
| "grad_norm": 1.1134260892868042, | |
| "learning_rate": 0.00016621954887218043, | |
| "loss": 0.5056, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 3.444053268023879, | |
| "grad_norm": 0.7996990084648132, | |
| "learning_rate": 0.00016546766917293232, | |
| "loss": 0.4875, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 3.444053268023879, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.288, | |
| "eval_samples_per_second": 22.201, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.49683996192088004, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 3.463186897290678, | |
| "grad_norm": 1.4016754627227783, | |
| "learning_rate": 0.0001647157894736842, | |
| "loss": 0.5019, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 3.4823205265574773, | |
| "grad_norm": 0.6514917016029358, | |
| "learning_rate": 0.00016396541353383458, | |
| "loss": 0.4964, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 3.4823205265574773, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.9586, | |
| "eval_samples_per_second": 22.154, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.49888935900148085, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 3.501454155824277, | |
| "grad_norm": 0.3953873813152313, | |
| "learning_rate": 0.00016321353383458645, | |
| "loss": 0.4878, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 3.520587785091076, | |
| "grad_norm": 0.6485087871551514, | |
| "learning_rate": 0.00016246165413533832, | |
| "loss": 0.4767, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 3.520587785091076, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.0491, | |
| "eval_samples_per_second": 22.148, | |
| "eval_steps_per_second": 2.769, | |
| "eval_wer": 0.4921594034271208, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 3.5397214143578752, | |
| "grad_norm": 0.649442732334137, | |
| "learning_rate": 0.00016170977443609021, | |
| "loss": 0.491, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 3.558855043624675, | |
| "grad_norm": 0.7617647051811218, | |
| "learning_rate": 0.00016095939849624058, | |
| "loss": 0.4765, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 3.558855043624675, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.7633, | |
| "eval_samples_per_second": 22.098, | |
| "eval_steps_per_second": 2.763, | |
| "eval_wer": 0.48691030251745293, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 3.577988672891474, | |
| "grad_norm": 0.5942517518997192, | |
| "learning_rate": 0.00016020751879699247, | |
| "loss": 0.4805, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 3.597122302158273, | |
| "grad_norm": 0.9230866432189941, | |
| "learning_rate": 0.00015945563909774434, | |
| "loss": 0.4967, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 3.597122302158273, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.3012, | |
| "eval_samples_per_second": 22.13, | |
| "eval_steps_per_second": 2.767, | |
| "eval_wer": 0.49814893166913476, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 3.616255931425073, | |
| "grad_norm": 0.601637065410614, | |
| "learning_rate": 0.00015870375939849624, | |
| "loss": 0.5055, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 3.635389560691872, | |
| "grad_norm": 0.6463965773582458, | |
| "learning_rate": 0.0001579518796992481, | |
| "loss": 0.4941, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 3.635389560691872, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.5729, | |
| "eval_samples_per_second": 22.042, | |
| "eval_steps_per_second": 2.756, | |
| "eval_wer": 0.49616564417177916, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 3.654523189958671, | |
| "grad_norm": 0.8088521957397461, | |
| "learning_rate": 0.0001572, | |
| "loss": 0.4851, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 3.673656819225471, | |
| "grad_norm": 1.0758212804794312, | |
| "learning_rate": 0.00015644812030075187, | |
| "loss": 0.4808, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 3.673656819225471, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.3887, | |
| "eval_samples_per_second": 22.124, | |
| "eval_steps_per_second": 2.766, | |
| "eval_wer": 0.4856277766024963, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 3.69279044849227, | |
| "grad_norm": 2.3342583179473877, | |
| "learning_rate": 0.00015569624060150376, | |
| "loss": 0.4682, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 3.711924077759069, | |
| "grad_norm": 0.9281033873558044, | |
| "learning_rate": 0.0001549443609022556, | |
| "loss": 0.4838, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 3.711924077759069, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.8326, | |
| "eval_samples_per_second": 22.093, | |
| "eval_steps_per_second": 2.762, | |
| "eval_wer": 0.47491802411677597, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 3.731057707025869, | |
| "grad_norm": 0.6656193733215332, | |
| "learning_rate": 0.00015419248120300753, | |
| "loss": 0.4893, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 3.750191336292668, | |
| "grad_norm": 0.8286083936691284, | |
| "learning_rate": 0.0001534421052631579, | |
| "loss": 0.4644, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 3.750191336292668, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.0552, | |
| "eval_samples_per_second": 22.148, | |
| "eval_steps_per_second": 2.769, | |
| "eval_wer": 0.4738073831182568, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 3.769324965559467, | |
| "grad_norm": 0.7517048716545105, | |
| "learning_rate": 0.00015269022556390978, | |
| "loss": 0.4828, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 3.788458594826267, | |
| "grad_norm": 1.126383662223816, | |
| "learning_rate": 0.00015193834586466163, | |
| "loss": 0.4818, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 3.788458594826267, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.5174, | |
| "eval_samples_per_second": 22.115, | |
| "eval_steps_per_second": 2.765, | |
| "eval_wer": 0.47370160778506454, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 3.807592224093066, | |
| "grad_norm": 1.257995843887329, | |
| "learning_rate": 0.00015118646616541352, | |
| "loss": 0.485, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 3.826725853359865, | |
| "grad_norm": 1.0233116149902344, | |
| "learning_rate": 0.0001504345864661654, | |
| "loss": 0.4741, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 3.826725853359865, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.1035, | |
| "eval_samples_per_second": 22.144, | |
| "eval_steps_per_second": 2.769, | |
| "eval_wer": 0.48123810027501585, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 3.8458594826266648, | |
| "grad_norm": 1.8550606966018677, | |
| "learning_rate": 0.00014968270676691728, | |
| "loss": 0.4763, | |
| "step": 100500 | |
| }, | |
| { | |
| "epoch": 3.864993111893464, | |
| "grad_norm": 0.8199677467346191, | |
| "learning_rate": 0.00014893082706766915, | |
| "loss": 0.4734, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 3.864993111893464, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.4217, | |
| "eval_samples_per_second": 22.122, | |
| "eval_steps_per_second": 2.766, | |
| "eval_wer": 0.4772186376137085, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 3.884126741160263, | |
| "grad_norm": 0.5380846261978149, | |
| "learning_rate": 0.00014817894736842105, | |
| "loss": 0.4602, | |
| "step": 101500 | |
| }, | |
| { | |
| "epoch": 3.9032603704270628, | |
| "grad_norm": 1.026496410369873, | |
| "learning_rate": 0.00014742706766917291, | |
| "loss": 0.4733, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 3.9032603704270628, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.6547, | |
| "eval_samples_per_second": 22.106, | |
| "eval_steps_per_second": 2.764, | |
| "eval_wer": 0.47355616670192513, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 3.922393999693862, | |
| "grad_norm": 1.4769624471664429, | |
| "learning_rate": 0.0001466766917293233, | |
| "loss": 0.477, | |
| "step": 102500 | |
| }, | |
| { | |
| "epoch": 3.941527628960661, | |
| "grad_norm": 1.2881931066513062, | |
| "learning_rate": 0.00014592481203007517, | |
| "loss": 0.4937, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 3.941527628960661, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.1763, | |
| "eval_samples_per_second": 22.139, | |
| "eval_steps_per_second": 2.768, | |
| "eval_wer": 0.4694573725407235, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 3.9606612582274607, | |
| "grad_norm": 0.6641072630882263, | |
| "learning_rate": 0.00014517443609022554, | |
| "loss": 0.4728, | |
| "step": 103500 | |
| }, | |
| { | |
| "epoch": 3.97979488749426, | |
| "grad_norm": 0.7772675156593323, | |
| "learning_rate": 0.00014442255639097743, | |
| "loss": 0.4864, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 3.97979488749426, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8284, | |
| "eval_samples_per_second": 22.163, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.47485191453353076, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 3.998928516761059, | |
| "grad_norm": 0.4977366626262665, | |
| "learning_rate": 0.0001436706766917293, | |
| "loss": 0.4671, | |
| "step": 104500 | |
| }, | |
| { | |
| "epoch": 4.018062146027859, | |
| "grad_norm": 1.0952422618865967, | |
| "learning_rate": 0.00014291879699248117, | |
| "loss": 0.4126, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 4.018062146027859, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.0988, | |
| "eval_samples_per_second": 22.075, | |
| "eval_steps_per_second": 2.76, | |
| "eval_wer": 0.4635207319653057, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 4.0371957752946575, | |
| "grad_norm": 0.694837212562561, | |
| "learning_rate": 0.00014216691729323306, | |
| "loss": 0.4202, | |
| "step": 105500 | |
| }, | |
| { | |
| "epoch": 4.056329404561457, | |
| "grad_norm": 0.7097035646438599, | |
| "learning_rate": 0.00014141503759398493, | |
| "loss": 0.4228, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 4.056329404561457, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7201, | |
| "eval_samples_per_second": 22.241, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.47007880262322826, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 4.075463033828257, | |
| "grad_norm": 1.1219637393951416, | |
| "learning_rate": 0.00014066315789473683, | |
| "loss": 0.4349, | |
| "step": 106500 | |
| }, | |
| { | |
| "epoch": 4.094596663095055, | |
| "grad_norm": 0.5968381762504578, | |
| "learning_rate": 0.0001399127819548872, | |
| "loss": 0.4098, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 4.094596663095055, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2686, | |
| "eval_samples_per_second": 22.203, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.45890628305479164, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 4.113730292361855, | |
| "grad_norm": 0.5920900106430054, | |
| "learning_rate": 0.0001391609022556391, | |
| "loss": 0.4182, | |
| "step": 107500 | |
| }, | |
| { | |
| "epoch": 4.132863921628655, | |
| "grad_norm": 0.45823031663894653, | |
| "learning_rate": 0.00013840902255639095, | |
| "loss": 0.4193, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 4.132863921628655, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.9696, | |
| "eval_samples_per_second": 22.154, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.46152422255130104, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 4.151997550895453, | |
| "grad_norm": 0.6653383374214172, | |
| "learning_rate": 0.00013765714285714285, | |
| "loss": 0.4087, | |
| "step": 108500 | |
| }, | |
| { | |
| "epoch": 4.171131180162253, | |
| "grad_norm": 0.5999200940132141, | |
| "learning_rate": 0.00013690526315789472, | |
| "loss": 0.4083, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 4.171131180162253, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.122, | |
| "eval_samples_per_second": 22.073, | |
| "eval_steps_per_second": 2.76, | |
| "eval_wer": 0.46397027713137295, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 4.190264809429053, | |
| "grad_norm": 0.5787246823310852, | |
| "learning_rate": 0.0001361533834586466, | |
| "loss": 0.4075, | |
| "step": 109500 | |
| }, | |
| { | |
| "epoch": 4.209398438695851, | |
| "grad_norm": 2.8409461975097656, | |
| "learning_rate": 0.00013540150375939848, | |
| "loss": 0.406, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 4.209398438695851, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.8914, | |
| "eval_samples_per_second": 22.159, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.46135233763486355, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 4.228532067962651, | |
| "grad_norm": 0.8015612959861755, | |
| "learning_rate": 0.00013464962406015038, | |
| "loss": 0.4167, | |
| "step": 110500 | |
| }, | |
| { | |
| "epoch": 4.247665697229451, | |
| "grad_norm": 3.532646417617798, | |
| "learning_rate": 0.00013389774436090224, | |
| "loss": 0.4125, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 4.247665697229451, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.906, | |
| "eval_samples_per_second": 22.158, | |
| "eval_steps_per_second": 2.771, | |
| "eval_wer": 0.4608366828855511, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 4.266799326496249, | |
| "grad_norm": 0.7524324059486389, | |
| "learning_rate": 0.00013314736842105264, | |
| "loss": 0.4205, | |
| "step": 111500 | |
| }, | |
| { | |
| "epoch": 4.285932955763049, | |
| "grad_norm": 2.5353856086730957, | |
| "learning_rate": 0.0001323954887218045, | |
| "loss": 0.4104, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 4.285932955763049, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.4796, | |
| "eval_samples_per_second": 22.049, | |
| "eval_steps_per_second": 2.757, | |
| "eval_wer": 0.44868574148508567, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 4.305066585029849, | |
| "grad_norm": 0.9442459940910339, | |
| "learning_rate": 0.00013164360902255637, | |
| "loss": 0.4114, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 4.324200214296647, | |
| "grad_norm": 1.0531048774719238, | |
| "learning_rate": 0.00013089172932330827, | |
| "loss": 0.3988, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 4.324200214296647, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.0914, | |
| "eval_samples_per_second": 22.145, | |
| "eval_steps_per_second": 2.769, | |
| "eval_wer": 0.45987148297017133, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 4.343333843563447, | |
| "grad_norm": 0.7277682423591614, | |
| "learning_rate": 0.00013013984962406013, | |
| "loss": 0.4164, | |
| "step": 113500 | |
| }, | |
| { | |
| "epoch": 4.362467472830247, | |
| "grad_norm": 0.48210740089416504, | |
| "learning_rate": 0.00012938796992481203, | |
| "loss": 0.4034, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 4.362467472830247, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.084, | |
| "eval_samples_per_second": 22.076, | |
| "eval_steps_per_second": 2.76, | |
| "eval_wer": 0.45389517664480644, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 4.381601102097045, | |
| "grad_norm": 1.0025782585144043, | |
| "learning_rate": 0.0001286375939849624, | |
| "loss": 0.423, | |
| "step": 114500 | |
| }, | |
| { | |
| "epoch": 4.400734731363845, | |
| "grad_norm": 0.5586313009262085, | |
| "learning_rate": 0.00012788571428571426, | |
| "loss": 0.4023, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 4.400734731363845, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.3491, | |
| "eval_samples_per_second": 22.058, | |
| "eval_steps_per_second": 2.758, | |
| "eval_wer": 0.4479585360693886, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 4.419868360630645, | |
| "grad_norm": 0.6510444283485413, | |
| "learning_rate": 0.00012713383458646616, | |
| "loss": 0.4041, | |
| "step": 115500 | |
| }, | |
| { | |
| "epoch": 4.439001989897443, | |
| "grad_norm": 0.6380518674850464, | |
| "learning_rate": 0.00012638195488721802, | |
| "loss": 0.4026, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 4.439001989897443, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.4508, | |
| "eval_samples_per_second": 22.051, | |
| "eval_steps_per_second": 2.757, | |
| "eval_wer": 0.45242754389676326, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 4.458135619164243, | |
| "grad_norm": 0.7297781109809875, | |
| "learning_rate": 0.00012563157894736842, | |
| "loss": 0.4124, | |
| "step": 116500 | |
| }, | |
| { | |
| "epoch": 4.477269248431043, | |
| "grad_norm": 0.9323301911354065, | |
| "learning_rate": 0.00012487969924812028, | |
| "loss": 0.4182, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 4.477269248431043, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.9248, | |
| "eval_samples_per_second": 22.157, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.44729744023693674, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 4.496402877697841, | |
| "grad_norm": 0.7702882289886475, | |
| "learning_rate": 0.00012412781954887218, | |
| "loss": 0.4137, | |
| "step": 117500 | |
| }, | |
| { | |
| "epoch": 4.515536506964641, | |
| "grad_norm": 0.45166295766830444, | |
| "learning_rate": 0.00012337593984962405, | |
| "loss": 0.4046, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 4.515536506964641, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.8786, | |
| "eval_samples_per_second": 22.372, | |
| "eval_steps_per_second": 2.797, | |
| "eval_wer": 0.44563147873915804, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 4.5346701362314406, | |
| "grad_norm": 0.8825483322143555, | |
| "learning_rate": 0.00012262406015037594, | |
| "loss": 0.4001, | |
| "step": 118500 | |
| }, | |
| { | |
| "epoch": 4.553803765498239, | |
| "grad_norm": 0.5982456207275391, | |
| "learning_rate": 0.00012187218045112781, | |
| "loss": 0.4126, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 4.553803765498239, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.2679, | |
| "eval_samples_per_second": 22.415, | |
| "eval_steps_per_second": 2.803, | |
| "eval_wer": 0.4406071504125238, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 4.572937394765039, | |
| "grad_norm": 0.49651646614074707, | |
| "learning_rate": 0.00012112030075187969, | |
| "loss": 0.3899, | |
| "step": 119500 | |
| }, | |
| { | |
| "epoch": 4.5920710240318385, | |
| "grad_norm": 0.6609870195388794, | |
| "learning_rate": 0.00012036842105263157, | |
| "loss": 0.4144, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 4.5920710240318385, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.9675, | |
| "eval_samples_per_second": 22.365, | |
| "eval_steps_per_second": 2.796, | |
| "eval_wer": 0.44486460757351387, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 4.611204653298637, | |
| "grad_norm": 0.8339817523956299, | |
| "learning_rate": 0.00011961804511278195, | |
| "loss": 0.4067, | |
| "step": 120500 | |
| }, | |
| { | |
| "epoch": 4.630338282565437, | |
| "grad_norm": 0.683476984500885, | |
| "learning_rate": 0.00011886766917293232, | |
| "loss": 0.4074, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 4.630338282565437, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.2537, | |
| "eval_samples_per_second": 22.345, | |
| "eval_steps_per_second": 2.794, | |
| "eval_wer": 0.44750899090332136, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 4.6494719118322365, | |
| "grad_norm": 0.6157557964324951, | |
| "learning_rate": 0.0001181172932330827, | |
| "loss": 0.3947, | |
| "step": 121500 | |
| }, | |
| { | |
| "epoch": 4.668605541099035, | |
| "grad_norm": 0.7728904485702515, | |
| "learning_rate": 0.00011736541353383457, | |
| "loss": 0.3922, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 4.668605541099035, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.5029, | |
| "eval_samples_per_second": 22.47, | |
| "eval_steps_per_second": 2.81, | |
| "eval_wer": 0.4387693039983076, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 4.687739170365835, | |
| "grad_norm": 0.6580181121826172, | |
| "learning_rate": 0.00011661353383458646, | |
| "loss": 0.3969, | |
| "step": 122500 | |
| }, | |
| { | |
| "epoch": 4.7068727996326345, | |
| "grad_norm": 3.9013619422912598, | |
| "learning_rate": 0.00011586165413533834, | |
| "loss": 0.3866, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 4.7068727996326345, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.7382, | |
| "eval_samples_per_second": 22.453, | |
| "eval_steps_per_second": 2.807, | |
| "eval_wer": 0.44741643748677806, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 4.726006428899433, | |
| "grad_norm": 0.7487606406211853, | |
| "learning_rate": 0.00011510977443609022, | |
| "loss": 0.4057, | |
| "step": 123500 | |
| }, | |
| { | |
| "epoch": 4.745140058166233, | |
| "grad_norm": 0.9207865595817566, | |
| "learning_rate": 0.0001143578947368421, | |
| "loss": 0.3873, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 4.745140058166233, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.9132, | |
| "eval_samples_per_second": 22.441, | |
| "eval_steps_per_second": 2.806, | |
| "eval_wer": 0.4344986249206685, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 4.7642736874330325, | |
| "grad_norm": 0.6548067331314087, | |
| "learning_rate": 0.00011360601503759398, | |
| "loss": 0.4058, | |
| "step": 124500 | |
| }, | |
| { | |
| "epoch": 4.783407316699831, | |
| "grad_norm": 0.97373366355896, | |
| "learning_rate": 0.00011285413533834586, | |
| "loss": 0.3917, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 4.783407316699831, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.7438, | |
| "eval_samples_per_second": 22.381, | |
| "eval_steps_per_second": 2.798, | |
| "eval_wer": 0.4338243071715676, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 4.802540945966631, | |
| "grad_norm": 0.7585910558700562, | |
| "learning_rate": 0.00011210375939849623, | |
| "loss": 0.4095, | |
| "step": 125500 | |
| }, | |
| { | |
| "epoch": 4.8216745752334305, | |
| "grad_norm": 1.3345550298690796, | |
| "learning_rate": 0.0001113533834586466, | |
| "loss": 0.3864, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 4.8216745752334305, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.643, | |
| "eval_samples_per_second": 22.46, | |
| "eval_steps_per_second": 2.808, | |
| "eval_wer": 0.4350803892532262, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 4.840808204500229, | |
| "grad_norm": 0.6124061346054077, | |
| "learning_rate": 0.00011060150375939849, | |
| "loss": 0.3909, | |
| "step": 126500 | |
| }, | |
| { | |
| "epoch": 4.859941833767029, | |
| "grad_norm": 0.7190678119659424, | |
| "learning_rate": 0.00010984962406015037, | |
| "loss": 0.3826, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 4.859941833767029, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.9649, | |
| "eval_samples_per_second": 22.437, | |
| "eval_steps_per_second": 2.805, | |
| "eval_wer": 0.430783266342289, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 4.879075463033828, | |
| "grad_norm": 0.9486848711967468, | |
| "learning_rate": 0.00010909774436090225, | |
| "loss": 0.3974, | |
| "step": 127500 | |
| }, | |
| { | |
| "epoch": 4.898209092300627, | |
| "grad_norm": 3.170478343963623, | |
| "learning_rate": 0.00010834736842105263, | |
| "loss": 0.391, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 4.898209092300627, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.5327, | |
| "eval_samples_per_second": 22.396, | |
| "eval_steps_per_second": 2.8, | |
| "eval_wer": 0.431497249841337, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 4.917342721567427, | |
| "grad_norm": 0.462166428565979, | |
| "learning_rate": 0.000107596992481203, | |
| "loss": 0.3817, | |
| "step": 128500 | |
| }, | |
| { | |
| "epoch": 4.936476350834226, | |
| "grad_norm": 1.1014796495437622, | |
| "learning_rate": 0.00010684511278195487, | |
| "loss": 0.394, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 4.936476350834226, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.0162, | |
| "eval_samples_per_second": 22.433, | |
| "eval_steps_per_second": 2.805, | |
| "eval_wer": 0.42787444467950075, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 4.955609980101025, | |
| "grad_norm": 0.624528706073761, | |
| "learning_rate": 0.00010609323308270676, | |
| "loss": 0.3821, | |
| "step": 129500 | |
| }, | |
| { | |
| "epoch": 4.974743609367825, | |
| "grad_norm": 0.5960122346878052, | |
| "learning_rate": 0.00010534135338345864, | |
| "loss": 0.3957, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 4.974743609367825, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.106, | |
| "eval_samples_per_second": 22.427, | |
| "eval_steps_per_second": 2.804, | |
| "eval_wer": 0.4235112121853184, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 4.993877238634624, | |
| "grad_norm": 0.7740840911865234, | |
| "learning_rate": 0.0001045894736842105, | |
| "loss": 0.4012, | |
| "step": 130500 | |
| }, | |
| { | |
| "epoch": 5.013010867901423, | |
| "grad_norm": 0.39604371786117554, | |
| "learning_rate": 0.00010383759398496239, | |
| "loss": 0.3515, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 5.013010867901423, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.7575, | |
| "eval_samples_per_second": 22.452, | |
| "eval_steps_per_second": 2.807, | |
| "eval_wer": 0.42159403427120795, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 5.032144497168223, | |
| "grad_norm": 0.582255482673645, | |
| "learning_rate": 0.00010308571428571427, | |
| "loss": 0.332, | |
| "step": 131500 | |
| }, | |
| { | |
| "epoch": 5.051278126435022, | |
| "grad_norm": 1.186954379081726, | |
| "learning_rate": 0.00010233383458646615, | |
| "loss": 0.3389, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 5.051278126435022, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 312.5025, | |
| "eval_samples_per_second": 22.47, | |
| "eval_steps_per_second": 2.81, | |
| "eval_wer": 0.42548127776602496, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 5.070411755701821, | |
| "grad_norm": 0.7782790064811707, | |
| "learning_rate": 0.00010158195488721803, | |
| "loss": 0.3352, | |
| "step": 132500 | |
| }, | |
| { | |
| "epoch": 5.089545384968621, | |
| "grad_norm": 0.5987495183944702, | |
| "learning_rate": 0.00010083157894736841, | |
| "loss": 0.333, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 5.089545384968621, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 514.1295, | |
| "eval_samples_per_second": 13.658, | |
| "eval_steps_per_second": 1.708, | |
| "eval_wer": 0.42526972709964034, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 5.10867901423542, | |
| "grad_norm": 0.714470624923706, | |
| "learning_rate": 0.0001000812030075188, | |
| "loss": 0.3334, | |
| "step": 133500 | |
| }, | |
| { | |
| "epoch": 5.127812643502219, | |
| "grad_norm": 0.554375171661377, | |
| "learning_rate": 9.932932330827067e-05, | |
| "loss": 0.3313, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 5.127812643502219, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 313.7907, | |
| "eval_samples_per_second": 22.378, | |
| "eval_steps_per_second": 2.798, | |
| "eval_wer": 0.4178390099428813, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 5.146946272769019, | |
| "grad_norm": 0.8462457060813904, | |
| "learning_rate": 9.857744360902255e-05, | |
| "loss": 0.3391, | |
| "step": 134500 | |
| }, | |
| { | |
| "epoch": 5.166079902035818, | |
| "grad_norm": 0.5063708424568176, | |
| "learning_rate": 9.782556390977443e-05, | |
| "loss": 0.3351, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 5.166079902035818, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.6864, | |
| "eval_samples_per_second": 22.314, | |
| "eval_steps_per_second": 2.79, | |
| "eval_wer": 0.422321239686905, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 5.185213531302617, | |
| "grad_norm": 0.5310961008071899, | |
| "learning_rate": 9.707368421052631e-05, | |
| "loss": 0.3435, | |
| "step": 135500 | |
| }, | |
| { | |
| "epoch": 5.204347160569417, | |
| "grad_norm": 0.7004749774932861, | |
| "learning_rate": 9.63218045112782e-05, | |
| "loss": 0.3262, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 5.204347160569417, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.217, | |
| "eval_samples_per_second": 22.348, | |
| "eval_steps_per_second": 2.794, | |
| "eval_wer": 0.416318489528242, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 5.223480789836216, | |
| "grad_norm": 0.5730725526809692, | |
| "learning_rate": 9.556992481203008e-05, | |
| "loss": 0.333, | |
| "step": 136500 | |
| }, | |
| { | |
| "epoch": 5.242614419103015, | |
| "grad_norm": 1.6563917398452759, | |
| "learning_rate": 9.481804511278196e-05, | |
| "loss": 0.3333, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 5.242614419103015, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.3953, | |
| "eval_samples_per_second": 22.264, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.42156759043790987, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 5.261748048369815, | |
| "grad_norm": 0.5844420790672302, | |
| "learning_rate": 9.406616541353384e-05, | |
| "loss": 0.3369, | |
| "step": 137500 | |
| }, | |
| { | |
| "epoch": 5.280881677636614, | |
| "grad_norm": 0.6321229934692383, | |
| "learning_rate": 9.331428571428571e-05, | |
| "loss": 0.3229, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 5.280881677636614, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.3276, | |
| "eval_samples_per_second": 22.269, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.41326422678231434, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 5.300015306903413, | |
| "grad_norm": 0.6538165807723999, | |
| "learning_rate": 9.256240601503759e-05, | |
| "loss": 0.3382, | |
| "step": 138500 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "grad_norm": 0.7837240099906921, | |
| "learning_rate": 9.181052631578947e-05, | |
| "loss": 0.3345, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2249, | |
| "eval_samples_per_second": 22.206, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.4136079966151893, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 5.338282565437012, | |
| "grad_norm": 0.8858345150947571, | |
| "learning_rate": 9.106015037593984e-05, | |
| "loss": 0.3293, | |
| "step": 139500 | |
| }, | |
| { | |
| "epoch": 5.357416194703811, | |
| "grad_norm": 1.3112056255340576, | |
| "learning_rate": 9.030827067669172e-05, | |
| "loss": 0.3365, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 5.357416194703811, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1322, | |
| "eval_samples_per_second": 22.212, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.4193330865242226, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 5.376549823970611, | |
| "grad_norm": 0.720756471157074, | |
| "learning_rate": 8.95563909774436e-05, | |
| "loss": 0.3318, | |
| "step": 140500 | |
| }, | |
| { | |
| "epoch": 5.39568345323741, | |
| "grad_norm": 0.46977701783180237, | |
| "learning_rate": 8.880451127819548e-05, | |
| "loss": 0.3165, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 5.39568345323741, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.295, | |
| "eval_samples_per_second": 22.342, | |
| "eval_steps_per_second": 2.794, | |
| "eval_wer": 0.4112148297017136, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 5.414817082504209, | |
| "grad_norm": 0.7260543704032898, | |
| "learning_rate": 8.805263157894736e-05, | |
| "loss": 0.337, | |
| "step": 141500 | |
| }, | |
| { | |
| "epoch": 5.433950711771009, | |
| "grad_norm": 1.3891347646713257, | |
| "learning_rate": 8.730075187969924e-05, | |
| "loss": 0.3224, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 5.433950711771009, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.0255, | |
| "eval_samples_per_second": 22.29, | |
| "eval_steps_per_second": 2.787, | |
| "eval_wer": 0.40745980537338694, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 5.453084341037808, | |
| "grad_norm": 0.4979989528656006, | |
| "learning_rate": 8.654887218045112e-05, | |
| "loss": 0.3341, | |
| "step": 142500 | |
| }, | |
| { | |
| "epoch": 5.472217970304607, | |
| "grad_norm": 0.9820772409439087, | |
| "learning_rate": 8.579699248120299e-05, | |
| "loss": 0.335, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 5.472217970304607, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.1397, | |
| "eval_samples_per_second": 22.353, | |
| "eval_steps_per_second": 2.795, | |
| "eval_wer": 0.4112941612016078, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 5.491351599571407, | |
| "grad_norm": 1.6596304178237915, | |
| "learning_rate": 8.504511278195487e-05, | |
| "loss": 0.3243, | |
| "step": 143500 | |
| }, | |
| { | |
| "epoch": 5.510485228838206, | |
| "grad_norm": 0.687848687171936, | |
| "learning_rate": 8.429323308270675e-05, | |
| "loss": 0.3377, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 5.510485228838206, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.8491, | |
| "eval_samples_per_second": 22.232, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 0.41762745927649675, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 5.529618858105005, | |
| "grad_norm": 1.0538053512573242, | |
| "learning_rate": 8.354135338345864e-05, | |
| "loss": 0.3195, | |
| "step": 144500 | |
| }, | |
| { | |
| "epoch": 5.548752487371805, | |
| "grad_norm": 0.6369953751564026, | |
| "learning_rate": 8.278947368421052e-05, | |
| "loss": 0.3411, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 5.548752487371805, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0558, | |
| "eval_samples_per_second": 22.218, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.4091125449545166, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 5.567886116638604, | |
| "grad_norm": 1.193766474723816, | |
| "learning_rate": 8.20390977443609e-05, | |
| "loss": 0.3385, | |
| "step": 145500 | |
| }, | |
| { | |
| "epoch": 5.587019745905403, | |
| "grad_norm": 0.5476765632629395, | |
| "learning_rate": 8.128721804511278e-05, | |
| "loss": 0.3247, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 5.587019745905403, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7439, | |
| "eval_samples_per_second": 22.24, | |
| "eval_steps_per_second": 2.781, | |
| "eval_wer": 0.4096414216204781, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 5.606153375172203, | |
| "grad_norm": 0.9929884076118469, | |
| "learning_rate": 8.053533834586466e-05, | |
| "loss": 0.3305, | |
| "step": 146500 | |
| }, | |
| { | |
| "epoch": 5.625287004439002, | |
| "grad_norm": 0.3942908048629761, | |
| "learning_rate": 7.978345864661654e-05, | |
| "loss": 0.3304, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 5.625287004439002, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.4131, | |
| "eval_samples_per_second": 22.334, | |
| "eval_steps_per_second": 2.793, | |
| "eval_wer": 0.40843822720541567, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 5.644420633705801, | |
| "grad_norm": 0.7728341221809387, | |
| "learning_rate": 7.903157894736842e-05, | |
| "loss": 0.3301, | |
| "step": 147500 | |
| }, | |
| { | |
| "epoch": 5.663554262972601, | |
| "grad_norm": 1.4606784582138062, | |
| "learning_rate": 7.82796992481203e-05, | |
| "loss": 0.3267, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 5.663554262972601, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.4256, | |
| "eval_samples_per_second": 22.333, | |
| "eval_steps_per_second": 2.792, | |
| "eval_wer": 0.40423365771102177, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 5.6826878922394, | |
| "grad_norm": 0.6086077690124512, | |
| "learning_rate": 7.752781954887217e-05, | |
| "loss": 0.3195, | |
| "step": 148500 | |
| }, | |
| { | |
| "epoch": 5.701821521506199, | |
| "grad_norm": 0.5943909883499146, | |
| "learning_rate": 7.677593984962405e-05, | |
| "loss": 0.3193, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 5.701821521506199, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.9231, | |
| "eval_samples_per_second": 22.298, | |
| "eval_steps_per_second": 2.788, | |
| "eval_wer": 0.4026470277131373, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 5.7209551507729985, | |
| "grad_norm": 0.5374177098274231, | |
| "learning_rate": 7.602556390977442e-05, | |
| "loss": 0.3374, | |
| "step": 149500 | |
| }, | |
| { | |
| "epoch": 5.740088780039798, | |
| "grad_norm": 0.675542950630188, | |
| "learning_rate": 7.52736842105263e-05, | |
| "loss": 0.3155, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 5.740088780039798, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.593, | |
| "eval_samples_per_second": 22.18, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.40481542204357945, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 5.759222409306597, | |
| "grad_norm": 1.1648385524749756, | |
| "learning_rate": 7.45218045112782e-05, | |
| "loss": 0.3247, | |
| "step": 150500 | |
| }, | |
| { | |
| "epoch": 5.7783560385733965, | |
| "grad_norm": 0.6115811467170715, | |
| "learning_rate": 7.376992481203008e-05, | |
| "loss": 0.3238, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 5.7783560385733965, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.3837, | |
| "eval_samples_per_second": 22.055, | |
| "eval_steps_per_second": 2.758, | |
| "eval_wer": 0.40333456737888723, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 5.797489667840196, | |
| "grad_norm": 1.0145585536956787, | |
| "learning_rate": 7.301804511278196e-05, | |
| "loss": 0.3175, | |
| "step": 151500 | |
| }, | |
| { | |
| "epoch": 5.816623297106995, | |
| "grad_norm": 0.7065938115119934, | |
| "learning_rate": 7.226766917293232e-05, | |
| "loss": 0.3172, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 5.816623297106995, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.6878, | |
| "eval_samples_per_second": 22.034, | |
| "eval_steps_per_second": 2.755, | |
| "eval_wer": 0.40486830971017557, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 5.8357569263737945, | |
| "grad_norm": 0.7309425473213196, | |
| "learning_rate": 7.15172932330827e-05, | |
| "loss": 0.3229, | |
| "step": 152500 | |
| }, | |
| { | |
| "epoch": 5.854890555640594, | |
| "grad_norm": 1.0197374820709229, | |
| "learning_rate": 7.076541353383458e-05, | |
| "loss": 0.3148, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 5.854890555640594, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 323.1191, | |
| "eval_samples_per_second": 21.732, | |
| "eval_steps_per_second": 2.717, | |
| "eval_wer": 0.3989052253014597, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 5.874024184907393, | |
| "grad_norm": 0.7352388501167297, | |
| "learning_rate": 7.001503759398496e-05, | |
| "loss": 0.313, | |
| "step": 153500 | |
| }, | |
| { | |
| "epoch": 5.8931578141741925, | |
| "grad_norm": 0.6405870914459229, | |
| "learning_rate": 6.926315789473684e-05, | |
| "loss": 0.3217, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 5.8931578141741925, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 319.2909, | |
| "eval_samples_per_second": 21.992, | |
| "eval_steps_per_second": 2.75, | |
| "eval_wer": 0.39782102813623865, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 5.912291443440992, | |
| "grad_norm": 0.711800754070282, | |
| "learning_rate": 6.851127819548872e-05, | |
| "loss": 0.3206, | |
| "step": 154500 | |
| }, | |
| { | |
| "epoch": 5.931425072707791, | |
| "grad_norm": 0.41337111592292786, | |
| "learning_rate": 6.77593984962406e-05, | |
| "loss": 0.3145, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 5.931425072707791, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.549, | |
| "eval_samples_per_second": 22.044, | |
| "eval_steps_per_second": 2.756, | |
| "eval_wer": 0.39298180664269095, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 5.9505587019745905, | |
| "grad_norm": 0.3790297210216522, | |
| "learning_rate": 6.700751879699248e-05, | |
| "loss": 0.3185, | |
| "step": 155500 | |
| }, | |
| { | |
| "epoch": 5.96969233124139, | |
| "grad_norm": 0.7092337012290955, | |
| "learning_rate": 6.625563909774435e-05, | |
| "loss": 0.3178, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 5.96969233124139, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.815, | |
| "eval_samples_per_second": 22.025, | |
| "eval_steps_per_second": 2.754, | |
| "eval_wer": 0.39946054580071927, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 5.988825960508189, | |
| "grad_norm": 1.5943549871444702, | |
| "learning_rate": 6.550375939849623e-05, | |
| "loss": 0.3154, | |
| "step": 156500 | |
| }, | |
| { | |
| "epoch": 6.0079595897749885, | |
| "grad_norm": 1.1709485054016113, | |
| "learning_rate": 6.475187969924812e-05, | |
| "loss": 0.2895, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 6.0079595897749885, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.8329, | |
| "eval_samples_per_second": 22.024, | |
| "eval_steps_per_second": 2.754, | |
| "eval_wer": 0.39980431563359425, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 6.027093219041788, | |
| "grad_norm": 0.6642709374427795, | |
| "learning_rate": 6.4e-05, | |
| "loss": 0.2714, | |
| "step": 157500 | |
| }, | |
| { | |
| "epoch": 6.046226848308587, | |
| "grad_norm": 0.7695789337158203, | |
| "learning_rate": 6.324812030075188e-05, | |
| "loss": 0.269, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 6.046226848308587, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.0108, | |
| "eval_samples_per_second": 22.081, | |
| "eval_steps_per_second": 2.761, | |
| "eval_wer": 0.3926248148931669, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 6.065360477575386, | |
| "grad_norm": 0.6437819600105286, | |
| "learning_rate": 6.249624060150375e-05, | |
| "loss": 0.2651, | |
| "step": 158500 | |
| }, | |
| { | |
| "epoch": 6.084494106842186, | |
| "grad_norm": 0.9133914113044739, | |
| "learning_rate": 6.174436090225563e-05, | |
| "loss": 0.2757, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 6.084494106842186, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.9915, | |
| "eval_samples_per_second": 22.082, | |
| "eval_steps_per_second": 2.761, | |
| "eval_wer": 0.39225460122699385, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 6.103627736108985, | |
| "grad_norm": 0.8651337623596191, | |
| "learning_rate": 6.099398496240601e-05, | |
| "loss": 0.2763, | |
| "step": 159500 | |
| }, | |
| { | |
| "epoch": 6.122761365375784, | |
| "grad_norm": 0.6800199151039124, | |
| "learning_rate": 6.0243609022556384e-05, | |
| "loss": 0.2573, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 6.122761365375784, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.7524, | |
| "eval_samples_per_second": 22.03, | |
| "eval_steps_per_second": 2.754, | |
| "eval_wer": 0.3906283054791623, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 6.141894994642584, | |
| "grad_norm": 0.7264246940612793, | |
| "learning_rate": 5.9491729323308265e-05, | |
| "loss": 0.2632, | |
| "step": 160500 | |
| }, | |
| { | |
| "epoch": 6.161028623909383, | |
| "grad_norm": 0.2874845862388611, | |
| "learning_rate": 5.873984962406015e-05, | |
| "loss": 0.2666, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 6.161028623909383, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.0588, | |
| "eval_samples_per_second": 22.078, | |
| "eval_steps_per_second": 2.76, | |
| "eval_wer": 0.38831447006558073, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 6.180162253176182, | |
| "grad_norm": 0.36712953448295593, | |
| "learning_rate": 5.798796992481202e-05, | |
| "loss": 0.2719, | |
| "step": 161500 | |
| }, | |
| { | |
| "epoch": 6.199295882442982, | |
| "grad_norm": 1.4700485467910767, | |
| "learning_rate": 5.72360902255639e-05, | |
| "loss": 0.2691, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 6.199295882442982, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.6549, | |
| "eval_samples_per_second": 22.036, | |
| "eval_steps_per_second": 2.755, | |
| "eval_wer": 0.391950497144066, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 6.218429511709781, | |
| "grad_norm": 0.664314866065979, | |
| "learning_rate": 5.6484210526315785e-05, | |
| "loss": 0.2734, | |
| "step": 162500 | |
| }, | |
| { | |
| "epoch": 6.23756314097658, | |
| "grad_norm": 0.7786546945571899, | |
| "learning_rate": 5.5732330827067666e-05, | |
| "loss": 0.2699, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 6.23756314097658, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 323.3374, | |
| "eval_samples_per_second": 21.717, | |
| "eval_steps_per_second": 2.715, | |
| "eval_wer": 0.39616828855510894, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 6.25669677024338, | |
| "grad_norm": 1.2886419296264648, | |
| "learning_rate": 5.498045112781954e-05, | |
| "loss": 0.2702, | |
| "step": 163500 | |
| }, | |
| { | |
| "epoch": 6.275830399510179, | |
| "grad_norm": 0.6407492160797119, | |
| "learning_rate": 5.422857142857142e-05, | |
| "loss": 0.259, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 6.275830399510179, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 318.7872, | |
| "eval_samples_per_second": 22.027, | |
| "eval_steps_per_second": 2.754, | |
| "eval_wer": 0.39016553839644597, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 6.294964028776978, | |
| "grad_norm": 0.36012986302375793, | |
| "learning_rate": 5.3476691729323304e-05, | |
| "loss": 0.2591, | |
| "step": 164500 | |
| }, | |
| { | |
| "epoch": 6.314097658043778, | |
| "grad_norm": 0.5582063794136047, | |
| "learning_rate": 5.2724812030075185e-05, | |
| "loss": 0.2707, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 6.314097658043778, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.8204, | |
| "eval_samples_per_second": 22.094, | |
| "eval_steps_per_second": 2.763, | |
| "eval_wer": 0.38777237148297017, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 6.333231287310577, | |
| "grad_norm": 0.5500897765159607, | |
| "learning_rate": 5.197293233082706e-05, | |
| "loss": 0.2574, | |
| "step": 165500 | |
| }, | |
| { | |
| "epoch": 6.352364916577376, | |
| "grad_norm": 0.5922083854675293, | |
| "learning_rate": 5.122105263157894e-05, | |
| "loss": 0.265, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 6.352364916577376, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.5005, | |
| "eval_samples_per_second": 22.186, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.38557753331922995, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 6.371498545844176, | |
| "grad_norm": 0.5268240571022034, | |
| "learning_rate": 5.046917293233082e-05, | |
| "loss": 0.2693, | |
| "step": 166500 | |
| }, | |
| { | |
| "epoch": 6.390632175110975, | |
| "grad_norm": 2.8765857219696045, | |
| "learning_rate": 4.9717293233082705e-05, | |
| "loss": 0.2657, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 6.390632175110975, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.5854, | |
| "eval_samples_per_second": 22.111, | |
| "eval_steps_per_second": 2.765, | |
| "eval_wer": 0.38506187856991747, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 6.409765804377774, | |
| "grad_norm": 0.6294525265693665, | |
| "learning_rate": 4.896541353383458e-05, | |
| "loss": 0.2634, | |
| "step": 167500 | |
| }, | |
| { | |
| "epoch": 6.428899433644574, | |
| "grad_norm": 1.270578384399414, | |
| "learning_rate": 4.821503759398496e-05, | |
| "loss": 0.2625, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 6.428899433644574, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.2046, | |
| "eval_samples_per_second": 22.137, | |
| "eval_steps_per_second": 2.768, | |
| "eval_wer": 0.3841495663211339, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 6.448033062911373, | |
| "grad_norm": 1.4967974424362183, | |
| "learning_rate": 4.746315789473684e-05, | |
| "loss": 0.2679, | |
| "step": 168500 | |
| }, | |
| { | |
| "epoch": 6.467166692178172, | |
| "grad_norm": 0.8087161779403687, | |
| "learning_rate": 4.671278195488721e-05, | |
| "loss": 0.2615, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 6.467166692178172, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.9216, | |
| "eval_samples_per_second": 22.157, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.3831975883224032, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 6.486300321444972, | |
| "grad_norm": 0.5716475248336792, | |
| "learning_rate": 4.596090225563909e-05, | |
| "loss": 0.2507, | |
| "step": 169500 | |
| }, | |
| { | |
| "epoch": 6.505433950711771, | |
| "grad_norm": 0.7699230909347534, | |
| "learning_rate": 4.520902255639097e-05, | |
| "loss": 0.2629, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 6.505433950711771, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.644, | |
| "eval_samples_per_second": 22.107, | |
| "eval_steps_per_second": 2.764, | |
| "eval_wer": 0.38342236090543685, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 6.52456757997857, | |
| "grad_norm": 0.8144583702087402, | |
| "learning_rate": 4.445714285714285e-05, | |
| "loss": 0.2646, | |
| "step": 170500 | |
| }, | |
| { | |
| "epoch": 6.54370120924537, | |
| "grad_norm": 0.9461275339126587, | |
| "learning_rate": 4.370526315789473e-05, | |
| "loss": 0.276, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 6.54370120924537, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.3907, | |
| "eval_samples_per_second": 22.194, | |
| "eval_steps_per_second": 2.775, | |
| "eval_wer": 0.38307859107256187, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 6.562834838512169, | |
| "grad_norm": 0.5300208926200867, | |
| "learning_rate": 4.295338345864661e-05, | |
| "loss": 0.2657, | |
| "step": 171500 | |
| }, | |
| { | |
| "epoch": 6.581968467778968, | |
| "grad_norm": 0.611358106136322, | |
| "learning_rate": 4.220150375939849e-05, | |
| "loss": 0.2623, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 6.581968467778968, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1573, | |
| "eval_samples_per_second": 22.21, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.38129363232494184, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 6.601102097045768, | |
| "grad_norm": 0.8705514073371887, | |
| "learning_rate": 4.144962406015037e-05, | |
| "loss": 0.2614, | |
| "step": 172500 | |
| }, | |
| { | |
| "epoch": 6.620235726312567, | |
| "grad_norm": 0.4813309609889984, | |
| "learning_rate": 4.069924812030075e-05, | |
| "loss": 0.27, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 6.620235726312567, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2461, | |
| "eval_samples_per_second": 22.204, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.3814787391580283, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 6.639369355579366, | |
| "grad_norm": 2.4199442863464355, | |
| "learning_rate": 3.994736842105263e-05, | |
| "loss": 0.261, | |
| "step": 173500 | |
| }, | |
| { | |
| "epoch": 6.658502984846166, | |
| "grad_norm": 0.6200481653213501, | |
| "learning_rate": 3.919548872180451e-05, | |
| "loss": 0.2712, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 6.658502984846166, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.1791, | |
| "eval_samples_per_second": 22.139, | |
| "eval_steps_per_second": 2.768, | |
| "eval_wer": 0.3812143008250476, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 6.677636614112965, | |
| "grad_norm": 0.2569734454154968, | |
| "learning_rate": 3.844360902255639e-05, | |
| "loss": 0.2614, | |
| "step": 174500 | |
| }, | |
| { | |
| "epoch": 6.696770243379764, | |
| "grad_norm": 0.44579431414604187, | |
| "learning_rate": 3.769172932330827e-05, | |
| "loss": 0.263, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 6.696770243379764, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.5662, | |
| "eval_samples_per_second": 22.182, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.38161095832451875, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 6.715903872646564, | |
| "grad_norm": 0.38563570380210876, | |
| "learning_rate": 3.6939849624060146e-05, | |
| "loss": 0.2488, | |
| "step": 175500 | |
| }, | |
| { | |
| "epoch": 6.735037501913363, | |
| "grad_norm": 0.5862724781036377, | |
| "learning_rate": 3.6189473684210524e-05, | |
| "loss": 0.2616, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 6.735037501913363, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2674, | |
| "eval_samples_per_second": 22.203, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.379574783160567, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 6.754171131180162, | |
| "grad_norm": 0.5157662034034729, | |
| "learning_rate": 3.5437593984962405e-05, | |
| "loss": 0.2525, | |
| "step": 176500 | |
| }, | |
| { | |
| "epoch": 6.773304760446962, | |
| "grad_norm": 0.807600200176239, | |
| "learning_rate": 3.4687218045112776e-05, | |
| "loss": 0.253, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 6.773304760446962, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2322, | |
| "eval_samples_per_second": 22.205, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.3794161201607785, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 6.792438389713761, | |
| "grad_norm": 0.4601055383682251, | |
| "learning_rate": 3.393533834586466e-05, | |
| "loss": 0.2513, | |
| "step": 177500 | |
| }, | |
| { | |
| "epoch": 6.81157201898056, | |
| "grad_norm": 0.4807584881782532, | |
| "learning_rate": 3.3184962406015036e-05, | |
| "loss": 0.2572, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 6.81157201898056, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.7771, | |
| "eval_samples_per_second": 22.308, | |
| "eval_steps_per_second": 2.789, | |
| "eval_wer": 0.38285381848952826, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 6.83070564824736, | |
| "grad_norm": 2.668820381164551, | |
| "learning_rate": 3.243308270676692e-05, | |
| "loss": 0.2522, | |
| "step": 178500 | |
| }, | |
| { | |
| "epoch": 6.8498392775141586, | |
| "grad_norm": 0.5270944833755493, | |
| "learning_rate": 3.168270676691729e-05, | |
| "loss": 0.2517, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 6.8498392775141586, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 314.726, | |
| "eval_samples_per_second": 22.311, | |
| "eval_steps_per_second": 2.79, | |
| "eval_wer": 0.37728739158028346, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 6.868972906780958, | |
| "grad_norm": 0.5133803486824036, | |
| "learning_rate": 3.093082706766917e-05, | |
| "loss": 0.2479, | |
| "step": 179500 | |
| }, | |
| { | |
| "epoch": 6.888106536047758, | |
| "grad_norm": 0.6608215570449829, | |
| "learning_rate": 3.0178947368421048e-05, | |
| "loss": 0.2471, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 6.888106536047758, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.1185, | |
| "eval_samples_per_second": 22.284, | |
| "eval_steps_per_second": 2.786, | |
| "eval_wer": 0.37833192299555746, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 6.9072401653145565, | |
| "grad_norm": 0.9923522472381592, | |
| "learning_rate": 2.942706766917293e-05, | |
| "loss": 0.2533, | |
| "step": 180500 | |
| }, | |
| { | |
| "epoch": 6.926373794581356, | |
| "grad_norm": 0.6495700478553772, | |
| "learning_rate": 2.8675187969924808e-05, | |
| "loss": 0.2441, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 6.926373794581356, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.6279, | |
| "eval_samples_per_second": 22.248, | |
| "eval_steps_per_second": 2.782, | |
| "eval_wer": 0.37630896974825473, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 6.945507423848156, | |
| "grad_norm": 0.5995193123817444, | |
| "learning_rate": 2.792330827067669e-05, | |
| "loss": 0.2509, | |
| "step": 181500 | |
| }, | |
| { | |
| "epoch": 6.9646410531149545, | |
| "grad_norm": 0.6942078471183777, | |
| "learning_rate": 2.7172932330827067e-05, | |
| "loss": 0.245, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 6.9646410531149545, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.6284, | |
| "eval_samples_per_second": 22.177, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.3749338904167548, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 6.983774682381754, | |
| "grad_norm": 0.7296892404556274, | |
| "learning_rate": 2.6421052631578945e-05, | |
| "loss": 0.2513, | |
| "step": 182500 | |
| }, | |
| { | |
| "epoch": 7.002908311648554, | |
| "grad_norm": 0.9255119562149048, | |
| "learning_rate": 2.5669172932330827e-05, | |
| "loss": 0.235, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 7.002908311648554, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.5684, | |
| "eval_samples_per_second": 22.252, | |
| "eval_steps_per_second": 2.782, | |
| "eval_wer": 0.3724085043367887, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 7.0220419409153525, | |
| "grad_norm": 0.6719674468040466, | |
| "learning_rate": 2.4917293233082705e-05, | |
| "loss": 0.2154, | |
| "step": 183500 | |
| }, | |
| { | |
| "epoch": 7.041175570182152, | |
| "grad_norm": 0.5619477033615112, | |
| "learning_rate": 2.4165413533834586e-05, | |
| "loss": 0.2281, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 7.041175570182152, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.4942, | |
| "eval_samples_per_second": 22.187, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.37427279458430296, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 7.060309199448952, | |
| "grad_norm": 0.7847068309783936, | |
| "learning_rate": 2.3413533834586465e-05, | |
| "loss": 0.2184, | |
| "step": 184500 | |
| }, | |
| { | |
| "epoch": 7.0794428287157505, | |
| "grad_norm": 0.7864698171615601, | |
| "learning_rate": 2.2661654135338346e-05, | |
| "loss": 0.2155, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 7.0794428287157505, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.7964, | |
| "eval_samples_per_second": 22.236, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 0.3742463507510049, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 7.09857645798255, | |
| "grad_norm": 0.6666255593299866, | |
| "learning_rate": 2.1909774436090224e-05, | |
| "loss": 0.2106, | |
| "step": 185500 | |
| }, | |
| { | |
| "epoch": 7.11771008724935, | |
| "grad_norm": 0.5075043439865112, | |
| "learning_rate": 2.1157894736842106e-05, | |
| "loss": 0.2177, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 7.11771008724935, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0968, | |
| "eval_samples_per_second": 22.215, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.37367780833509623, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 7.1368437165161485, | |
| "grad_norm": 0.7393398284912109, | |
| "learning_rate": 2.0406015037593984e-05, | |
| "loss": 0.2208, | |
| "step": 186500 | |
| }, | |
| { | |
| "epoch": 7.155977345782948, | |
| "grad_norm": 0.3300219476222992, | |
| "learning_rate": 1.9654135338345865e-05, | |
| "loss": 0.2107, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 7.155977345782948, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.1907, | |
| "eval_samples_per_second": 22.279, | |
| "eval_steps_per_second": 2.786, | |
| "eval_wer": 0.37078220858895705, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 7.175110975049748, | |
| "grad_norm": 3.2175910472869873, | |
| "learning_rate": 1.8902255639097743e-05, | |
| "loss": 0.2032, | |
| "step": 187500 | |
| }, | |
| { | |
| "epoch": 7.194244604316546, | |
| "grad_norm": 0.28260278701782227, | |
| "learning_rate": 1.8151879699248118e-05, | |
| "loss": 0.2129, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 7.194244604316546, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.4927, | |
| "eval_samples_per_second": 22.257, | |
| "eval_steps_per_second": 2.783, | |
| "eval_wer": 0.37161518933784643, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 7.213378233583346, | |
| "grad_norm": 1.4389430284500122, | |
| "learning_rate": 1.74e-05, | |
| "loss": 0.221, | |
| "step": 188500 | |
| }, | |
| { | |
| "epoch": 7.232511862850146, | |
| "grad_norm": 0.5784205198287964, | |
| "learning_rate": 1.6648120300751878e-05, | |
| "loss": 0.2173, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 7.232511862850146, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.5523, | |
| "eval_samples_per_second": 22.113, | |
| "eval_steps_per_second": 2.765, | |
| "eval_wer": 0.36953934842394753, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 7.251645492116944, | |
| "grad_norm": 0.5264465808868408, | |
| "learning_rate": 1.589624060150376e-05, | |
| "loss": 0.2102, | |
| "step": 189500 | |
| }, | |
| { | |
| "epoch": 7.270779121383744, | |
| "grad_norm": 1.0469930171966553, | |
| "learning_rate": 1.5144360902255639e-05, | |
| "loss": 0.2145, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 7.270779121383744, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.5149, | |
| "eval_samples_per_second": 22.185, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.37215728792045694, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 7.289912750650544, | |
| "grad_norm": 0.43167009949684143, | |
| "learning_rate": 1.4392481203007517e-05, | |
| "loss": 0.2217, | |
| "step": 190500 | |
| }, | |
| { | |
| "epoch": 7.309046379917342, | |
| "grad_norm": 1.083001732826233, | |
| "learning_rate": 1.3640601503759397e-05, | |
| "loss": 0.2116, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 7.309046379917342, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.4082, | |
| "eval_samples_per_second": 22.263, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.37024011000634655, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 7.328180009184142, | |
| "grad_norm": 0.5683468580245972, | |
| "learning_rate": 1.2888721804511277e-05, | |
| "loss": 0.2085, | |
| "step": 191500 | |
| }, | |
| { | |
| "epoch": 7.347313638450942, | |
| "grad_norm": 0.45209017395973206, | |
| "learning_rate": 1.2136842105263156e-05, | |
| "loss": 0.212, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 7.347313638450942, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.0151, | |
| "eval_samples_per_second": 22.22, | |
| "eval_steps_per_second": 2.778, | |
| "eval_wer": 0.37038555108948595, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 7.36644726771774, | |
| "grad_norm": 1.5557799339294434, | |
| "learning_rate": 1.1384962406015036e-05, | |
| "loss": 0.2091, | |
| "step": 192500 | |
| }, | |
| { | |
| "epoch": 7.38558089698454, | |
| "grad_norm": 0.456394761800766, | |
| "learning_rate": 1.0633082706766916e-05, | |
| "loss": 0.2116, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 7.38558089698454, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.9212, | |
| "eval_samples_per_second": 22.157, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.37012111275650517, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 7.40471452625134, | |
| "grad_norm": 0.34570273756980896, | |
| "learning_rate": 9.881203007518796e-06, | |
| "loss": 0.2034, | |
| "step": 193500 | |
| }, | |
| { | |
| "epoch": 7.423848155518138, | |
| "grad_norm": 0.6514278054237366, | |
| "learning_rate": 9.129323308270676e-06, | |
| "loss": 0.2124, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 7.423848155518138, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.4723, | |
| "eval_samples_per_second": 22.259, | |
| "eval_steps_per_second": 2.783, | |
| "eval_wer": 0.36865348000846204, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 7.442981784784938, | |
| "grad_norm": 0.3534170091152191, | |
| "learning_rate": 8.378947368421052e-06, | |
| "loss": 0.2146, | |
| "step": 194500 | |
| }, | |
| { | |
| "epoch": 7.462115414051738, | |
| "grad_norm": 0.6505366563796997, | |
| "learning_rate": 7.627067669172932e-06, | |
| "loss": 0.2078, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 7.462115414051738, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.1812, | |
| "eval_samples_per_second": 22.209, | |
| "eval_steps_per_second": 2.777, | |
| "eval_wer": 0.3681113814258515, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 7.481249043318536, | |
| "grad_norm": 0.5068254470825195, | |
| "learning_rate": 6.8751879699248115e-06, | |
| "loss": 0.2097, | |
| "step": 195500 | |
| }, | |
| { | |
| "epoch": 7.500382672585336, | |
| "grad_norm": 0.32878100872039795, | |
| "learning_rate": 6.124812030075188e-06, | |
| "loss": 0.2158, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 7.500382672585336, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.2612, | |
| "eval_samples_per_second": 22.203, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 0.3682171567590438, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 7.519516301852136, | |
| "grad_norm": 0.9998613595962524, | |
| "learning_rate": 5.3729323308270675e-06, | |
| "loss": 0.2031, | |
| "step": 196500 | |
| }, | |
| { | |
| "epoch": 7.538649931118934, | |
| "grad_norm": 0.6963976621627808, | |
| "learning_rate": 4.622556390977443e-06, | |
| "loss": 0.2157, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 7.538649931118934, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.927, | |
| "eval_samples_per_second": 22.157, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.36727840067696216, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 7.557783560385734, | |
| "grad_norm": 0.6300442218780518, | |
| "learning_rate": 3.870676691729323e-06, | |
| "loss": 0.2082, | |
| "step": 197500 | |
| }, | |
| { | |
| "epoch": 7.576917189652534, | |
| "grad_norm": 0.3542906939983368, | |
| "learning_rate": 3.118796992481203e-06, | |
| "loss": 0.2045, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 7.576917189652534, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 317.109, | |
| "eval_samples_per_second": 22.144, | |
| "eval_steps_per_second": 2.769, | |
| "eval_wer": 0.3666834144277554, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 7.596050818919332, | |
| "grad_norm": 0.9721285700798035, | |
| "learning_rate": 2.366917293233083e-06, | |
| "loss": 0.209, | |
| "step": 198500 | |
| }, | |
| { | |
| "epoch": 7.615184448186132, | |
| "grad_norm": 0.8516126275062561, | |
| "learning_rate": 1.6165413533834587e-06, | |
| "loss": 0.2188, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 7.615184448186132, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 316.4647, | |
| "eval_samples_per_second": 22.189, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.36752961709329385, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 7.634318077452932, | |
| "grad_norm": 0.7036492824554443, | |
| "learning_rate": 8.646616541353383e-07, | |
| "loss": 0.2072, | |
| "step": 199500 | |
| }, | |
| { | |
| "epoch": 7.65345170671973, | |
| "grad_norm": 0.5840544104576111, | |
| "learning_rate": 1.1278195488721805e-07, | |
| "loss": 0.2041, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 7.65345170671973, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 315.8635, | |
| "eval_samples_per_second": 22.231, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 0.3670800719272266, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 7.65345170671973, | |
| "step": 200000, | |
| "total_flos": 1.6865103601997185e+21, | |
| "train_loss": 0.5320780529403687, | |
| "train_runtime": 333857.2134, | |
| "train_samples_per_second": 4.792, | |
| "train_steps_per_second": 0.599 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 200000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 4000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.6865103601997185e+21, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |