| { |
| "best_global_step": 10500, |
| "best_metric": 40.10271240526526, |
| "best_model_checkpoint": "whisper-small-br/checkpoint-10500", |
| "epoch": 1.4972097454743434, |
| "eval_steps": 500, |
| "global_step": 11000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.013610997686130393, |
| "grad_norm": 22.095027923583984, |
| "learning_rate": 9.9e-06, |
| "loss": 4.058, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.027221995372260787, |
| "grad_norm": 18.20777702331543, |
| "learning_rate": 1.9900000000000003e-05, |
| "loss": 2.0289, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.04083299305839118, |
| "grad_norm": 17.343502044677734, |
| "learning_rate": 1.9909344810219313e-05, |
| "loss": 1.5023, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.054443990744521574, |
| "grad_norm": 12.90123462677002, |
| "learning_rate": 1.9817773911450943e-05, |
| "loss": 1.3212, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.06805498843065197, |
| "grad_norm": 12.617008209228516, |
| "learning_rate": 1.9726203012682573e-05, |
| "loss": 1.2295, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.06805498843065197, |
| "eval_cer": 44.18141757235164, |
| "eval_loss": 1.1764613389968872, |
| "eval_runtime": 1019.3249, |
| "eval_samples_per_second": 3.035, |
| "eval_steps_per_second": 0.38, |
| "eval_wer": 82.73583964898285, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.08166598611678236, |
| "grad_norm": 11.933979034423828, |
| "learning_rate": 1.96346321139142e-05, |
| "loss": 1.1777, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.09527698380291276, |
| "grad_norm": 12.346055030822754, |
| "learning_rate": 1.954306121514583e-05, |
| "loss": 1.1195, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.10888798148904315, |
| "grad_norm": 11.624801635742188, |
| "learning_rate": 1.9451490316377456e-05, |
| "loss": 1.0431, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.12249897917517354, |
| "grad_norm": 11.907363891601562, |
| "learning_rate": 1.9359919417609086e-05, |
| "loss": 0.9826, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.13610997686130394, |
| "grad_norm": 11.434345245361328, |
| "learning_rate": 1.9268348518840712e-05, |
| "loss": 1.0302, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.13610997686130394, |
| "eval_cer": 33.54789196147848, |
| "eval_loss": 0.9722283482551575, |
| "eval_runtime": 946.305, |
| "eval_samples_per_second": 3.27, |
| "eval_steps_per_second": 0.409, |
| "eval_wer": 66.64339848424412, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.14972097454743433, |
| "grad_norm": 9.866198539733887, |
| "learning_rate": 1.9176777620072342e-05, |
| "loss": 0.9639, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.16333197223356472, |
| "grad_norm": 12.207975387573242, |
| "learning_rate": 1.9085206721303972e-05, |
| "loss": 0.9652, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.1769429699196951, |
| "grad_norm": 13.966451644897461, |
| "learning_rate": 1.89936358225356e-05, |
| "loss": 0.9003, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.19055396760582552, |
| "grad_norm": 11.561500549316406, |
| "learning_rate": 1.890206492376723e-05, |
| "loss": 0.8899, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.2041649652919559, |
| "grad_norm": 10.504300117492676, |
| "learning_rate": 1.881049402499886e-05, |
| "loss": 0.8969, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.2041649652919559, |
| "eval_cer": 27.465649988859457, |
| "eval_loss": 0.860107421875, |
| "eval_runtime": 893.0785, |
| "eval_samples_per_second": 3.464, |
| "eval_steps_per_second": 0.433, |
| "eval_wer": 56.55165536497806, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.2177759629780863, |
| "grad_norm": 14.605524063110352, |
| "learning_rate": 1.8718923126230485e-05, |
| "loss": 0.8828, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.23138696066421668, |
| "grad_norm": 10.524106979370117, |
| "learning_rate": 1.8627352227462115e-05, |
| "loss": 0.8459, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.24499795835034707, |
| "grad_norm": 11.29123306274414, |
| "learning_rate": 1.853578132869374e-05, |
| "loss": 0.8757, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.2586089560364775, |
| "grad_norm": 9.073974609375, |
| "learning_rate": 1.844421042992537e-05, |
| "loss": 0.8445, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.2722199537226079, |
| "grad_norm": 11.95883846282959, |
| "learning_rate": 1.8352639531156998e-05, |
| "loss": 0.817, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.2722199537226079, |
| "eval_cer": 27.33939048845097, |
| "eval_loss": 0.7963160872459412, |
| "eval_runtime": 929.7295, |
| "eval_samples_per_second": 3.328, |
| "eval_steps_per_second": 0.416, |
| "eval_wer": 55.73145193458316, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.28583095140873827, |
| "grad_norm": 13.961465835571289, |
| "learning_rate": 1.8261068632388628e-05, |
| "loss": 0.8128, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.29944194909486865, |
| "grad_norm": 11.061758041381836, |
| "learning_rate": 1.8169497733620255e-05, |
| "loss": 0.7818, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.31305294678099904, |
| "grad_norm": 8.439574241638184, |
| "learning_rate": 1.8077926834851888e-05, |
| "loss": 0.7794, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.32666394446712943, |
| "grad_norm": 8.946911811828613, |
| "learning_rate": 1.7986355936083515e-05, |
| "loss": 0.8061, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.3402749421532598, |
| "grad_norm": 13.243382453918457, |
| "learning_rate": 1.7894785037315145e-05, |
| "loss": 0.7967, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.3402749421532598, |
| "eval_cer": 27.54091055380883, |
| "eval_loss": 0.7550674676895142, |
| "eval_runtime": 929.7973, |
| "eval_samples_per_second": 3.328, |
| "eval_steps_per_second": 0.416, |
| "eval_wer": 56.20761866773035, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.3538859398393902, |
| "grad_norm": 9.587485313415527, |
| "learning_rate": 1.780321413854677e-05, |
| "loss": 0.7408, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.3674969375255206, |
| "grad_norm": 10.000493049621582, |
| "learning_rate": 1.77116432397784e-05, |
| "loss": 0.7568, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.38110793521165104, |
| "grad_norm": 8.778080940246582, |
| "learning_rate": 1.7620072341010028e-05, |
| "loss": 0.7346, |
| "step": 2800 |
| }, |
| { |
| "epoch": 0.3947189328977814, |
| "grad_norm": 11.506646156311035, |
| "learning_rate": 1.7528501442241658e-05, |
| "loss": 0.7681, |
| "step": 2900 |
| }, |
| { |
| "epoch": 0.4083299305839118, |
| "grad_norm": 10.7967529296875, |
| "learning_rate": 1.7436930543473284e-05, |
| "loss": 0.7216, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.4083299305839118, |
| "eval_cer": 25.7019780655064, |
| "eval_loss": 0.721060574054718, |
| "eval_runtime": 897.2936, |
| "eval_samples_per_second": 3.448, |
| "eval_steps_per_second": 0.431, |
| "eval_wer": 52.54786597526925, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.4219409282700422, |
| "grad_norm": 10.669580459594727, |
| "learning_rate": 1.7345359644704914e-05, |
| "loss": 0.7571, |
| "step": 3100 |
| }, |
| { |
| "epoch": 0.4355519259561726, |
| "grad_norm": 11.027612686157227, |
| "learning_rate": 1.7253788745936544e-05, |
| "loss": 0.7199, |
| "step": 3200 |
| }, |
| { |
| "epoch": 0.449162923642303, |
| "grad_norm": 10.802726745605469, |
| "learning_rate": 1.716221784716817e-05, |
| "loss": 0.7132, |
| "step": 3300 |
| }, |
| { |
| "epoch": 0.46277392132843337, |
| "grad_norm": 8.402053833007812, |
| "learning_rate": 1.70706469483998e-05, |
| "loss": 0.7173, |
| "step": 3400 |
| }, |
| { |
| "epoch": 0.47638491901456376, |
| "grad_norm": 9.13488483428955, |
| "learning_rate": 1.697907604963143e-05, |
| "loss": 0.7426, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.47638491901456376, |
| "eval_cer": 27.62013220112396, |
| "eval_loss": 0.6920907497406006, |
| "eval_runtime": 936.9738, |
| "eval_samples_per_second": 3.302, |
| "eval_steps_per_second": 0.413, |
| "eval_wer": 54.05364978061428, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.48999591670069415, |
| "grad_norm": 10.873909950256348, |
| "learning_rate": 1.6887505150863057e-05, |
| "loss": 0.6771, |
| "step": 3600 |
| }, |
| { |
| "epoch": 0.5036069143868246, |
| "grad_norm": 12.089336395263672, |
| "learning_rate": 1.6795934252094687e-05, |
| "loss": 0.7418, |
| "step": 3700 |
| }, |
| { |
| "epoch": 0.517217912072955, |
| "grad_norm": 7.360589027404785, |
| "learning_rate": 1.6704363353326314e-05, |
| "loss": 0.6805, |
| "step": 3800 |
| }, |
| { |
| "epoch": 0.5308289097590854, |
| "grad_norm": 10.794859886169434, |
| "learning_rate": 1.6612792454557944e-05, |
| "loss": 0.7006, |
| "step": 3900 |
| }, |
| { |
| "epoch": 0.5444399074452158, |
| "grad_norm": 8.446488380432129, |
| "learning_rate": 1.652122155578957e-05, |
| "loss": 0.6652, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.5444399074452158, |
| "eval_cer": 31.48466318421509, |
| "eval_loss": 0.6702451705932617, |
| "eval_runtime": 981.908, |
| "eval_samples_per_second": 3.151, |
| "eval_steps_per_second": 0.394, |
| "eval_wer": 57.81810929397686, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.5580509051313461, |
| "grad_norm": 11.135995864868164, |
| "learning_rate": 1.64296506570212e-05, |
| "loss": 0.6538, |
| "step": 4100 |
| }, |
| { |
| "epoch": 0.5716619028174765, |
| "grad_norm": 13.993082046508789, |
| "learning_rate": 1.633807975825283e-05, |
| "loss": 0.657, |
| "step": 4200 |
| }, |
| { |
| "epoch": 0.5852729005036069, |
| "grad_norm": 10.156628608703613, |
| "learning_rate": 1.6246508859484457e-05, |
| "loss": 0.6605, |
| "step": 4300 |
| }, |
| { |
| "epoch": 0.5988838981897373, |
| "grad_norm": 10.285860061645508, |
| "learning_rate": 1.6154937960716086e-05, |
| "loss": 0.6817, |
| "step": 4400 |
| }, |
| { |
| "epoch": 0.6124948958758677, |
| "grad_norm": 6.604043960571289, |
| "learning_rate": 1.6063367061947713e-05, |
| "loss": 0.666, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.6124948958758677, |
| "eval_cer": 26.41299235015968, |
| "eval_loss": 0.6442458033561707, |
| "eval_runtime": 951.5023, |
| "eval_samples_per_second": 3.252, |
| "eval_steps_per_second": 0.407, |
| "eval_wer": 52.4406661348225, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.6261058935619981, |
| "grad_norm": 10.028115272521973, |
| "learning_rate": 1.5971796163179343e-05, |
| "loss": 0.6739, |
| "step": 4600 |
| }, |
| { |
| "epoch": 0.6397168912481285, |
| "grad_norm": 12.291666984558105, |
| "learning_rate": 1.588022526441097e-05, |
| "loss": 0.6252, |
| "step": 4700 |
| }, |
| { |
| "epoch": 0.6533278889342589, |
| "grad_norm": 8.521187782287598, |
| "learning_rate": 1.57886543656426e-05, |
| "loss": 0.6351, |
| "step": 4800 |
| }, |
| { |
| "epoch": 0.6669388866203892, |
| "grad_norm": 7.403680324554443, |
| "learning_rate": 1.569708346687423e-05, |
| "loss": 0.6406, |
| "step": 4900 |
| }, |
| { |
| "epoch": 0.6805498843065196, |
| "grad_norm": 7.44627046585083, |
| "learning_rate": 1.5605512568105856e-05, |
| "loss": 0.6574, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.6805498843065196, |
| "eval_cer": 26.51697076226079, |
| "eval_loss": 0.6287782192230225, |
| "eval_runtime": 964.4487, |
| "eval_samples_per_second": 3.208, |
| "eval_steps_per_second": 0.401, |
| "eval_wer": 52.03928998803351, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.69416088199265, |
| "grad_norm": 12.659337997436523, |
| "learning_rate": 1.5513941669337486e-05, |
| "loss": 0.6263, |
| "step": 5100 |
| }, |
| { |
| "epoch": 0.7077718796787804, |
| "grad_norm": 10.218366622924805, |
| "learning_rate": 1.5422370770569116e-05, |
| "loss": 0.6441, |
| "step": 5200 |
| }, |
| { |
| "epoch": 0.7213828773649108, |
| "grad_norm": 8.07555866241455, |
| "learning_rate": 1.5330799871800742e-05, |
| "loss": 0.6436, |
| "step": 5300 |
| }, |
| { |
| "epoch": 0.7349938750510412, |
| "grad_norm": 8.196616172790527, |
| "learning_rate": 1.5239228973032372e-05, |
| "loss": 0.6254, |
| "step": 5400 |
| }, |
| { |
| "epoch": 0.7486048727371717, |
| "grad_norm": 7.4557085037231445, |
| "learning_rate": 1.5147658074264e-05, |
| "loss": 0.6161, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.7486048727371717, |
| "eval_cer": 22.99655880969475, |
| "eval_loss": 0.6194547414779663, |
| "eval_runtime": 930.6325, |
| "eval_samples_per_second": 3.325, |
| "eval_steps_per_second": 0.416, |
| "eval_wer": 47.23274830474671, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.7622158704233021, |
| "grad_norm": 8.781782150268555, |
| "learning_rate": 1.5056087175495629e-05, |
| "loss": 0.6144, |
| "step": 5600 |
| }, |
| { |
| "epoch": 0.7758268681094325, |
| "grad_norm": 11.802079200744629, |
| "learning_rate": 1.4964516276727257e-05, |
| "loss": 0.6012, |
| "step": 5700 |
| }, |
| { |
| "epoch": 0.7894378657955629, |
| "grad_norm": 9.713030815124512, |
| "learning_rate": 1.4872945377958885e-05, |
| "loss": 0.6244, |
| "step": 5800 |
| }, |
| { |
| "epoch": 0.8030488634816932, |
| "grad_norm": 11.544571876525879, |
| "learning_rate": 1.4781374479190514e-05, |
| "loss": 0.6248, |
| "step": 5900 |
| }, |
| { |
| "epoch": 0.8166598611678236, |
| "grad_norm": 9.622050285339355, |
| "learning_rate": 1.4689803580422144e-05, |
| "loss": 0.6335, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.8166598611678236, |
| "eval_cer": 27.483474859505357, |
| "eval_loss": 0.6030585169792175, |
| "eval_runtime": 979.4668, |
| "eval_samples_per_second": 3.159, |
| "eval_steps_per_second": 0.395, |
| "eval_wer": 52.93927004387714, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.830270858853954, |
| "grad_norm": 5.885132312774658, |
| "learning_rate": 1.4598232681653772e-05, |
| "loss": 0.5996, |
| "step": 6100 |
| }, |
| { |
| "epoch": 0.8438818565400844, |
| "grad_norm": 10.961427688598633, |
| "learning_rate": 1.45066617828854e-05, |
| "loss": 0.5984, |
| "step": 6200 |
| }, |
| { |
| "epoch": 0.8574928542262148, |
| "grad_norm": 8.962092399597168, |
| "learning_rate": 1.4415090884117028e-05, |
| "loss": 0.5831, |
| "step": 6300 |
| }, |
| { |
| "epoch": 0.8711038519123452, |
| "grad_norm": 8.857563018798828, |
| "learning_rate": 1.4323519985348657e-05, |
| "loss": 0.6214, |
| "step": 6400 |
| }, |
| { |
| "epoch": 0.8847148495984756, |
| "grad_norm": 8.55826187133789, |
| "learning_rate": 1.4231949086580285e-05, |
| "loss": 0.6063, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.8847148495984756, |
| "eval_cer": 26.41299235015968, |
| "eval_loss": 0.5897173285484314, |
| "eval_runtime": 961.8494, |
| "eval_samples_per_second": 3.217, |
| "eval_steps_per_second": 0.402, |
| "eval_wer": 51.50827682489031, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.898325847284606, |
| "grad_norm": 9.004158020019531, |
| "learning_rate": 1.4140378187811913e-05, |
| "loss": 0.5816, |
| "step": 6600 |
| }, |
| { |
| "epoch": 0.9119368449707363, |
| "grad_norm": 10.888954162597656, |
| "learning_rate": 1.4048807289043541e-05, |
| "loss": 0.5822, |
| "step": 6700 |
| }, |
| { |
| "epoch": 0.9255478426568667, |
| "grad_norm": 7.482538223266602, |
| "learning_rate": 1.3957236390275171e-05, |
| "loss": 0.596, |
| "step": 6800 |
| }, |
| { |
| "epoch": 0.9391588403429971, |
| "grad_norm": 9.417503356933594, |
| "learning_rate": 1.3865665491506801e-05, |
| "loss": 0.5702, |
| "step": 6900 |
| }, |
| { |
| "epoch": 0.9527698380291275, |
| "grad_norm": 6.990769863128662, |
| "learning_rate": 1.377409459273843e-05, |
| "loss": 0.5917, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.9527698380291275, |
| "eval_cer": 23.621914688188546, |
| "eval_loss": 0.5820683240890503, |
| "eval_runtime": 940.7788, |
| "eval_samples_per_second": 3.289, |
| "eval_steps_per_second": 0.411, |
| "eval_wer": 46.91114878340647, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.9663808357152579, |
| "grad_norm": 7.560977935791016, |
| "learning_rate": 1.3682523693970058e-05, |
| "loss": 0.583, |
| "step": 7100 |
| }, |
| { |
| "epoch": 0.9799918334013883, |
| "grad_norm": 10.802068710327148, |
| "learning_rate": 1.3590952795201686e-05, |
| "loss": 0.6026, |
| "step": 7200 |
| }, |
| { |
| "epoch": 0.9936028310875187, |
| "grad_norm": 11.21953010559082, |
| "learning_rate": 1.3499381896433314e-05, |
| "loss": 0.5949, |
| "step": 7300 |
| }, |
| { |
| "epoch": 1.0072138287736492, |
| "grad_norm": 7.203009128570557, |
| "learning_rate": 1.3407810997664943e-05, |
| "loss": 0.4813, |
| "step": 7400 |
| }, |
| { |
| "epoch": 1.0208248264597795, |
| "grad_norm": 6.168055057525635, |
| "learning_rate": 1.3316240098896571e-05, |
| "loss": 0.3952, |
| "step": 7500 |
| }, |
| { |
| "epoch": 1.0208248264597795, |
| "eval_cer": 21.952813606317925, |
| "eval_loss": 0.5695406794548035, |
| "eval_runtime": 912.4196, |
| "eval_samples_per_second": 3.391, |
| "eval_steps_per_second": 0.424, |
| "eval_wer": 45.213402473075384, |
| "step": 7500 |
| }, |
| { |
| "epoch": 1.03443582414591, |
| "grad_norm": 8.780722618103027, |
| "learning_rate": 1.3224669200128199e-05, |
| "loss": 0.3778, |
| "step": 7600 |
| }, |
| { |
| "epoch": 1.0480468218320402, |
| "grad_norm": 6.931602954864502, |
| "learning_rate": 1.3133098301359827e-05, |
| "loss": 0.3839, |
| "step": 7700 |
| }, |
| { |
| "epoch": 1.0616578195181707, |
| "grad_norm": 7.048837661743164, |
| "learning_rate": 1.3041527402591459e-05, |
| "loss": 0.3773, |
| "step": 7800 |
| }, |
| { |
| "epoch": 1.075268817204301, |
| "grad_norm": 7.898579120635986, |
| "learning_rate": 1.2949956503823087e-05, |
| "loss": 0.3894, |
| "step": 7900 |
| }, |
| { |
| "epoch": 1.0888798148904315, |
| "grad_norm": 7.919825553894043, |
| "learning_rate": 1.2858385605054716e-05, |
| "loss": 0.3882, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.0888798148904315, |
| "eval_cer": 21.941425494516377, |
| "eval_loss": 0.5697298049926758, |
| "eval_runtime": 926.3774, |
| "eval_samples_per_second": 3.34, |
| "eval_steps_per_second": 0.418, |
| "eval_wer": 44.32090147586757, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.1024908125765618, |
| "grad_norm": 8.552074432373047, |
| "learning_rate": 1.2766814706286344e-05, |
| "loss": 0.3829, |
| "step": 8100 |
| }, |
| { |
| "epoch": 1.1161018102626923, |
| "grad_norm": 7.923656463623047, |
| "learning_rate": 1.2675243807517972e-05, |
| "loss": 0.3868, |
| "step": 8200 |
| }, |
| { |
| "epoch": 1.1297128079488226, |
| "grad_norm": 7.2799530029296875, |
| "learning_rate": 1.25836729087496e-05, |
| "loss": 0.3878, |
| "step": 8300 |
| }, |
| { |
| "epoch": 1.143323805634953, |
| "grad_norm": 6.550467491149902, |
| "learning_rate": 1.2492102009981229e-05, |
| "loss": 0.3888, |
| "step": 8400 |
| }, |
| { |
| "epoch": 1.1569348033210836, |
| "grad_norm": 7.603492259979248, |
| "learning_rate": 1.2400531111212857e-05, |
| "loss": 0.3817, |
| "step": 8500 |
| }, |
| { |
| "epoch": 1.1569348033210836, |
| "eval_cer": 20.560988290050254, |
| "eval_loss": 0.5587778091430664, |
| "eval_runtime": 913.5504, |
| "eval_samples_per_second": 3.387, |
| "eval_steps_per_second": 0.424, |
| "eval_wer": 43.44834463502194, |
| "step": 8500 |
| }, |
| { |
| "epoch": 1.1705458010072138, |
| "grad_norm": 8.294718742370605, |
| "learning_rate": 1.2308960212444485e-05, |
| "loss": 0.3732, |
| "step": 8600 |
| }, |
| { |
| "epoch": 1.1841567986933441, |
| "grad_norm": 9.819673538208008, |
| "learning_rate": 1.2217389313676115e-05, |
| "loss": 0.3886, |
| "step": 8700 |
| }, |
| { |
| "epoch": 1.1977677963794746, |
| "grad_norm": 7.964117527008057, |
| "learning_rate": 1.2125818414907743e-05, |
| "loss": 0.3816, |
| "step": 8800 |
| }, |
| { |
| "epoch": 1.2113787940656051, |
| "grad_norm": 5.106940269470215, |
| "learning_rate": 1.2034247516139372e-05, |
| "loss": 0.3689, |
| "step": 8900 |
| }, |
| { |
| "epoch": 1.2249897917517354, |
| "grad_norm": 6.9468231201171875, |
| "learning_rate": 1.1942676617371001e-05, |
| "loss": 0.3791, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.2249897917517354, |
| "eval_cer": 21.8027876117149, |
| "eval_loss": 0.5529155135154724, |
| "eval_runtime": 914.3402, |
| "eval_samples_per_second": 3.384, |
| "eval_steps_per_second": 0.423, |
| "eval_wer": 44.63751495811727, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.238600789437866, |
| "grad_norm": 6.808917999267578, |
| "learning_rate": 1.185110571860263e-05, |
| "loss": 0.3768, |
| "step": 9100 |
| }, |
| { |
| "epoch": 1.2522117871239962, |
| "grad_norm": 9.738892555236816, |
| "learning_rate": 1.1759534819834258e-05, |
| "loss": 0.376, |
| "step": 9200 |
| }, |
| { |
| "epoch": 1.2658227848101267, |
| "grad_norm": 8.172088623046875, |
| "learning_rate": 1.1667963921065886e-05, |
| "loss": 0.3928, |
| "step": 9300 |
| }, |
| { |
| "epoch": 1.279433782496257, |
| "grad_norm": 6.819339275360107, |
| "learning_rate": 1.1576393022297515e-05, |
| "loss": 0.3902, |
| "step": 9400 |
| }, |
| { |
| "epoch": 1.2930447801823874, |
| "grad_norm": 7.440240859985352, |
| "learning_rate": 1.1484822123529143e-05, |
| "loss": 0.3849, |
| "step": 9500 |
| }, |
| { |
| "epoch": 1.2930447801823874, |
| "eval_cer": 21.664644864209144, |
| "eval_loss": 0.5443253517150879, |
| "eval_runtime": 938.2102, |
| "eval_samples_per_second": 3.298, |
| "eval_steps_per_second": 0.412, |
| "eval_wer": 43.96190666134822, |
| "step": 9500 |
| }, |
| { |
| "epoch": 1.3066557778685177, |
| "grad_norm": 7.373666286468506, |
| "learning_rate": 1.1393251224760773e-05, |
| "loss": 0.3755, |
| "step": 9600 |
| }, |
| { |
| "epoch": 1.3202667755546482, |
| "grad_norm": 8.251206398010254, |
| "learning_rate": 1.1301680325992401e-05, |
| "loss": 0.3785, |
| "step": 9700 |
| }, |
| { |
| "epoch": 1.3338777732407785, |
| "grad_norm": 5.46920919418335, |
| "learning_rate": 1.121010942722403e-05, |
| "loss": 0.3764, |
| "step": 9800 |
| }, |
| { |
| "epoch": 1.347488770926909, |
| "grad_norm": 8.345499992370605, |
| "learning_rate": 1.1118538528455657e-05, |
| "loss": 0.3849, |
| "step": 9900 |
| }, |
| { |
| "epoch": 1.3610997686130393, |
| "grad_norm": 5.544594764709473, |
| "learning_rate": 1.1026967629687286e-05, |
| "loss": 0.3903, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.3610997686130393, |
| "eval_cer": 21.089792785878743, |
| "eval_loss": 0.5430769324302673, |
| "eval_runtime": 921.5241, |
| "eval_samples_per_second": 3.357, |
| "eval_steps_per_second": 0.42, |
| "eval_wer": 43.48075388911049, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.3747107662991698, |
| "grad_norm": 6.338021755218506, |
| "learning_rate": 1.0935396730918914e-05, |
| "loss": 0.3883, |
| "step": 10100 |
| }, |
| { |
| "epoch": 1.3883217639853, |
| "grad_norm": 8.008199691772461, |
| "learning_rate": 1.0843825832150542e-05, |
| "loss": 0.3963, |
| "step": 10200 |
| }, |
| { |
| "epoch": 1.4019327616714305, |
| "grad_norm": 6.975709438323975, |
| "learning_rate": 1.075225493338217e-05, |
| "loss": 0.3934, |
| "step": 10300 |
| }, |
| { |
| "epoch": 1.4155437593575608, |
| "grad_norm": 8.565428733825684, |
| "learning_rate": 1.06606840346138e-05, |
| "loss": 0.3879, |
| "step": 10400 |
| }, |
| { |
| "epoch": 1.4291547570436913, |
| "grad_norm": 6.089544296264648, |
| "learning_rate": 1.056911313584543e-05, |
| "loss": 0.3708, |
| "step": 10500 |
| }, |
| { |
| "epoch": 1.4291547570436913, |
| "eval_cer": 18.837422325650486, |
| "eval_loss": 0.5384772419929504, |
| "eval_runtime": 891.4907, |
| "eval_samples_per_second": 3.471, |
| "eval_steps_per_second": 0.434, |
| "eval_wer": 40.10271240526526, |
| "step": 10500 |
| }, |
| { |
| "epoch": 1.4427657547298218, |
| "grad_norm": 7.494001388549805, |
| "learning_rate": 1.0477542237077059e-05, |
| "loss": 0.3734, |
| "step": 10600 |
| }, |
| { |
| "epoch": 1.456376752415952, |
| "grad_norm": 4.784611701965332, |
| "learning_rate": 1.0385971338308687e-05, |
| "loss": 0.3804, |
| "step": 10700 |
| }, |
| { |
| "epoch": 1.4699877501020824, |
| "grad_norm": 8.518943786621094, |
| "learning_rate": 1.0294400439540315e-05, |
| "loss": 0.372, |
| "step": 10800 |
| }, |
| { |
| "epoch": 1.4835987477882129, |
| "grad_norm": 7.112024784088135, |
| "learning_rate": 1.0202829540771943e-05, |
| "loss": 0.3837, |
| "step": 10900 |
| }, |
| { |
| "epoch": 1.4972097454743434, |
| "grad_norm": 7.3742218017578125, |
| "learning_rate": 1.0111258642003572e-05, |
| "loss": 0.3664, |
| "step": 11000 |
| }, |
| { |
| "epoch": 1.4972097454743434, |
| "eval_cer": 19.681628004852325, |
| "eval_loss": 0.5333254337310791, |
| "eval_runtime": 915.2281, |
| "eval_samples_per_second": 3.381, |
| "eval_steps_per_second": 0.423, |
| "eval_wer": 41.19465496609494, |
| "step": 11000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 22041, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.539464947195904e+19, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
