| { | |
| "best_metric": 0.5734274387359619, | |
| "best_model_checkpoint": "./Hubert-common_voice_JSUT-ja-demo-kana/checkpoint-10200", | |
| "epoch": 19.962245885769605, | |
| "eval_steps": 100, | |
| "global_step": 10320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1936108422071636, | |
| "eval_cer": 6.16957909562534, | |
| "eval_loss": 41.956024169921875, | |
| "eval_runtime": 229.6586, | |
| "eval_samples_per_second": 23.779, | |
| "eval_steps_per_second": 2.974, | |
| "eval_wer": 1.5293902215711408, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3872216844143272, | |
| "eval_cer": 5.972248763190861, | |
| "eval_loss": 41.465721130371094, | |
| "eval_runtime": 227.3403, | |
| "eval_samples_per_second": 24.021, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 1.4235488005859733, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5808325266214908, | |
| "eval_cer": 3.732693519049338, | |
| "eval_loss": 40.27690505981445, | |
| "eval_runtime": 227.0532, | |
| "eval_samples_per_second": 24.052, | |
| "eval_steps_per_second": 3.008, | |
| "eval_wer": 1.1847646951107855, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7744433688286544, | |
| "eval_cer": 0.9962559540239647, | |
| "eval_loss": 36.3010139465332, | |
| "eval_runtime": 226.5945, | |
| "eval_samples_per_second": 24.1, | |
| "eval_steps_per_second": 3.014, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.968054211035818, | |
| "grad_norm": 145.3476104736328, | |
| "learning_rate": 1.188e-06, | |
| "loss": 31.4419, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.968054211035818, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 24.542619705200195, | |
| "eval_runtime": 252.7587, | |
| "eval_samples_per_second": 21.606, | |
| "eval_steps_per_second": 2.702, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1606969990319458, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 18.864185333251953, | |
| "eval_runtime": 245.7063, | |
| "eval_samples_per_second": 22.226, | |
| "eval_steps_per_second": 2.78, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3543078412391094, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 17.665119171142578, | |
| "eval_runtime": 248.5924, | |
| "eval_samples_per_second": 21.968, | |
| "eval_steps_per_second": 2.747, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.547918683446273, | |
| "eval_cer": 0.9991719217784573, | |
| "eval_loss": 17.200698852539062, | |
| "eval_runtime": 253.0263, | |
| "eval_samples_per_second": 21.583, | |
| "eval_steps_per_second": 2.699, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7415295256534367, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 16.761735916137695, | |
| "eval_runtime": 259.4018, | |
| "eval_samples_per_second": 21.052, | |
| "eval_steps_per_second": 2.633, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9351403678606003, | |
| "grad_norm": 84.73486328125, | |
| "learning_rate": 2.3880000000000003e-06, | |
| "loss": 14.8315, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9351403678606003, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 16.289474487304688, | |
| "eval_runtime": 251.3459, | |
| "eval_samples_per_second": 21.727, | |
| "eval_steps_per_second": 2.717, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.127783155856728, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 15.787718772888184, | |
| "eval_runtime": 256.1973, | |
| "eval_samples_per_second": 21.316, | |
| "eval_steps_per_second": 2.666, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.3213939980638916, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 15.248761177062988, | |
| "eval_runtime": 250.6433, | |
| "eval_samples_per_second": 21.788, | |
| "eval_steps_per_second": 2.725, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.515004840271055, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 14.66799259185791, | |
| "eval_runtime": 257.3763, | |
| "eval_samples_per_second": 21.218, | |
| "eval_steps_per_second": 2.654, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.708615682478219, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 14.063650131225586, | |
| "eval_runtime": 244.7195, | |
| "eval_samples_per_second": 22.315, | |
| "eval_steps_per_second": 2.791, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.9022265246853824, | |
| "grad_norm": 72.20648193359375, | |
| "learning_rate": 3.588e-06, | |
| "loss": 12.4363, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.9022265246853824, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 13.421667098999023, | |
| "eval_runtime": 256.8698, | |
| "eval_samples_per_second": 21.26, | |
| "eval_steps_per_second": 2.659, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.09486931268151, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 12.737384796142578, | |
| "eval_runtime": 223.924, | |
| "eval_samples_per_second": 24.388, | |
| "eval_steps_per_second": 3.05, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.2884801548886737, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 12.031902313232422, | |
| "eval_runtime": 222.2009, | |
| "eval_samples_per_second": 24.577, | |
| "eval_steps_per_second": 3.074, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.4820909970958374, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 11.298233032226562, | |
| "eval_runtime": 225.5639, | |
| "eval_samples_per_second": 24.21, | |
| "eval_steps_per_second": 3.028, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.675701839303001, | |
| "eval_cer": 0.9991719217784573, | |
| "eval_loss": 10.557957649230957, | |
| "eval_runtime": 227.931, | |
| "eval_samples_per_second": 23.959, | |
| "eval_steps_per_second": 2.997, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.8693126815101646, | |
| "grad_norm": 87.52015686035156, | |
| "learning_rate": 4.788e-06, | |
| "loss": 9.8267, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8693126815101646, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 9.812894821166992, | |
| "eval_runtime": 227.4635, | |
| "eval_samples_per_second": 24.008, | |
| "eval_steps_per_second": 3.003, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.061955469506293, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 9.064043998718262, | |
| "eval_runtime": 223.8694, | |
| "eval_samples_per_second": 24.394, | |
| "eval_steps_per_second": 3.051, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.255566311713456, | |
| "eval_cer": 0.9991719217784573, | |
| "eval_loss": 8.337604522705078, | |
| "eval_runtime": 224.9536, | |
| "eval_samples_per_second": 24.276, | |
| "eval_steps_per_second": 3.036, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.44917715392062, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 7.628673076629639, | |
| "eval_runtime": 228.2923, | |
| "eval_samples_per_second": 23.921, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.642787996127783, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 6.967807769775391, | |
| "eval_runtime": 228.5115, | |
| "eval_samples_per_second": 23.898, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.836398838334947, | |
| "grad_norm": 34.805992126464844, | |
| "learning_rate": 5.988e-06, | |
| "loss": 6.9778, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.836398838334947, | |
| "eval_cer": 0.9991719217784573, | |
| "eval_loss": 6.363549709320068, | |
| "eval_runtime": 227.6007, | |
| "eval_samples_per_second": 23.994, | |
| "eval_steps_per_second": 3.001, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.029041626331074, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 5.8258442878723145, | |
| "eval_runtime": 265.7388, | |
| "eval_samples_per_second": 20.55, | |
| "eval_steps_per_second": 2.57, | |
| "eval_wer": 1.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.2226524685382385, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 5.367654800415039, | |
| "eval_runtime": 226.9427, | |
| "eval_samples_per_second": 24.063, | |
| "eval_steps_per_second": 3.01, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.416263310745402, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.988757133483887, | |
| "eval_runtime": 227.6207, | |
| "eval_samples_per_second": 23.992, | |
| "eval_steps_per_second": 3.001, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.609874152952566, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.695638656616211, | |
| "eval_runtime": 230.3237, | |
| "eval_samples_per_second": 23.71, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 1.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.803484995159729, | |
| "grad_norm": 3.1527926921844482, | |
| "learning_rate": 7.1880000000000005e-06, | |
| "loss": 4.8731, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.803484995159729, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.478826999664307, | |
| "eval_runtime": 234.3555, | |
| "eval_samples_per_second": 23.302, | |
| "eval_steps_per_second": 2.914, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.997095837366892, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.328735828399658, | |
| "eval_runtime": 226.0428, | |
| "eval_samples_per_second": 24.159, | |
| "eval_steps_per_second": 3.022, | |
| "eval_wer": 1.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.18973862536302, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.20569372177124, | |
| "eval_runtime": 227.3968, | |
| "eval_samples_per_second": 24.015, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.383349467570184, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.1447553634643555, | |
| "eval_runtime": 226.8138, | |
| "eval_samples_per_second": 24.077, | |
| "eval_steps_per_second": 3.011, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.5769603097773475, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.109466552734375, | |
| "eval_runtime": 223.6971, | |
| "eval_samples_per_second": 24.412, | |
| "eval_steps_per_second": 3.053, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.770571151984511, | |
| "grad_norm": 1.822492241859436, | |
| "learning_rate": 8.388e-06, | |
| "loss": 4.1216, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.770571151984511, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.085766792297363, | |
| "eval_runtime": 228.1097, | |
| "eval_samples_per_second": 23.94, | |
| "eval_steps_per_second": 2.994, | |
| "eval_wer": 1.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.964181994191675, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.072465896606445, | |
| "eval_runtime": 276.0608, | |
| "eval_samples_per_second": 19.782, | |
| "eval_steps_per_second": 2.474, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 7.156824782187803, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.064761161804199, | |
| "eval_runtime": 223.6339, | |
| "eval_samples_per_second": 24.419, | |
| "eval_steps_per_second": 3.054, | |
| "eval_wer": 1.0, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.350435624394966, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.057798385620117, | |
| "eval_runtime": 223.5104, | |
| "eval_samples_per_second": 24.433, | |
| "eval_steps_per_second": 3.056, | |
| "eval_wer": 1.0, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.54404646660213, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.0493927001953125, | |
| "eval_runtime": 227.6334, | |
| "eval_samples_per_second": 23.99, | |
| "eval_steps_per_second": 3.0, | |
| "eval_wer": 1.0, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.737657308809293, | |
| "grad_norm": 0.8575032949447632, | |
| "learning_rate": 9.588e-06, | |
| "loss": 4.0264, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.737657308809293, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.036725044250488, | |
| "eval_runtime": 232.0149, | |
| "eval_samples_per_second": 23.537, | |
| "eval_steps_per_second": 2.944, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.931268151016457, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.027583122253418, | |
| "eval_runtime": 228.3311, | |
| "eval_samples_per_second": 23.917, | |
| "eval_steps_per_second": 2.991, | |
| "eval_wer": 1.0, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 8.123910939012585, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 4.012051582336426, | |
| "eval_runtime": 230.306, | |
| "eval_samples_per_second": 23.712, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 1.0, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.317521781219748, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 3.9720146656036377, | |
| "eval_runtime": 232.5487, | |
| "eval_samples_per_second": 23.483, | |
| "eval_steps_per_second": 2.937, | |
| "eval_wer": 1.0, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.511132623426912, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 3.903093099594116, | |
| "eval_runtime": 225.8081, | |
| "eval_samples_per_second": 24.184, | |
| "eval_steps_per_second": 3.025, | |
| "eval_wer": 1.0, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.704743465634076, | |
| "grad_norm": 1.4530484676361084, | |
| "learning_rate": 1.0787999999999999e-05, | |
| "loss": 3.937, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.704743465634076, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 3.8090686798095703, | |
| "eval_runtime": 230.0214, | |
| "eval_samples_per_second": 23.741, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 1.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.89835430784124, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 3.6690022945404053, | |
| "eval_runtime": 229.4694, | |
| "eval_samples_per_second": 23.798, | |
| "eval_steps_per_second": 2.976, | |
| "eval_wer": 1.0, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 9.090997095837366, | |
| "eval_cer": 0.9991365338202717, | |
| "eval_loss": 3.475857973098755, | |
| "eval_runtime": 229.5875, | |
| "eval_samples_per_second": 23.786, | |
| "eval_steps_per_second": 2.975, | |
| "eval_wer": 1.0, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.28460793804453, | |
| "eval_cer": 0.998676490363859, | |
| "eval_loss": 3.2108352184295654, | |
| "eval_runtime": 224.0108, | |
| "eval_samples_per_second": 24.378, | |
| "eval_steps_per_second": 3.049, | |
| "eval_wer": 1.0, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.478218780251694, | |
| "eval_cer": 0.6453277278807568, | |
| "eval_loss": 2.681295394897461, | |
| "eval_runtime": 226.867, | |
| "eval_samples_per_second": 24.071, | |
| "eval_steps_per_second": 3.011, | |
| "eval_wer": 1.0, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.671829622458858, | |
| "grad_norm": 13.226855278015137, | |
| "learning_rate": 1.1988000000000001e-05, | |
| "loss": 3.1866, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.671829622458858, | |
| "eval_cer": 0.5372387483986949, | |
| "eval_loss": 2.3876163959503174, | |
| "eval_runtime": 226.669, | |
| "eval_samples_per_second": 24.092, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 1.000183116645303, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.86544046466602, | |
| "eval_cer": 0.4901869191951363, | |
| "eval_loss": 2.16782808303833, | |
| "eval_runtime": 225.6765, | |
| "eval_samples_per_second": 24.198, | |
| "eval_steps_per_second": 3.026, | |
| "eval_wer": 1.0, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 10.058083252662149, | |
| "eval_cer": 0.4530012527337198, | |
| "eval_loss": 1.9945032596588135, | |
| "eval_runtime": 231.2758, | |
| "eval_samples_per_second": 23.612, | |
| "eval_steps_per_second": 2.953, | |
| "eval_wer": 1.000183116645303, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 10.251694094869313, | |
| "eval_cer": 0.4269698706924008, | |
| "eval_loss": 1.8575785160064697, | |
| "eval_runtime": 226.1334, | |
| "eval_samples_per_second": 24.149, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 1.0, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 10.445304937076477, | |
| "eval_cer": 0.43989355302177774, | |
| "eval_loss": 1.7787507772445679, | |
| "eval_runtime": 226.9349, | |
| "eval_samples_per_second": 24.064, | |
| "eval_steps_per_second": 3.01, | |
| "eval_wer": 1.0, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.63891577928364, | |
| "grad_norm": 8.360028266906738, | |
| "learning_rate": 1.3188e-05, | |
| "loss": 1.9458, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.63891577928364, | |
| "eval_cer": 0.40938205547416323, | |
| "eval_loss": 1.6519718170166016, | |
| "eval_runtime": 227.1281, | |
| "eval_samples_per_second": 24.044, | |
| "eval_steps_per_second": 3.007, | |
| "eval_wer": 1.0, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.832526621490803, | |
| "eval_cer": 0.38743444380746117, | |
| "eval_loss": 1.5545194149017334, | |
| "eval_runtime": 225.52, | |
| "eval_samples_per_second": 24.215, | |
| "eval_steps_per_second": 3.029, | |
| "eval_wer": 1.0, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 11.025169409486931, | |
| "eval_cer": 0.3800312829550361, | |
| "eval_loss": 1.4698398113250732, | |
| "eval_runtime": 230.154, | |
| "eval_samples_per_second": 23.728, | |
| "eval_steps_per_second": 2.968, | |
| "eval_wer": 1.0, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 11.218780251694096, | |
| "eval_cer": 0.3777310656729728, | |
| "eval_loss": 1.4052294492721558, | |
| "eval_runtime": 229.123, | |
| "eval_samples_per_second": 23.834, | |
| "eval_steps_per_second": 2.981, | |
| "eval_wer": 1.0, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 11.412391093901258, | |
| "eval_cer": 0.3658124013560666, | |
| "eval_loss": 1.3275731801986694, | |
| "eval_runtime": 230.2917, | |
| "eval_samples_per_second": 23.713, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 1.0, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 11.606001936108422, | |
| "grad_norm": 4.284916877746582, | |
| "learning_rate": 1.4388000000000002e-05, | |
| "loss": 1.4263, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.606001936108422, | |
| "eval_cer": 0.36683865214344863, | |
| "eval_loss": 1.2710145711898804, | |
| "eval_runtime": 237.3574, | |
| "eval_samples_per_second": 23.008, | |
| "eval_steps_per_second": 2.878, | |
| "eval_wer": 1.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.799612778315586, | |
| "eval_cer": 0.35358940059876426, | |
| "eval_loss": 1.2150152921676636, | |
| "eval_runtime": 229.8802, | |
| "eval_samples_per_second": 23.756, | |
| "eval_steps_per_second": 2.971, | |
| "eval_wer": 1.0, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 11.99322362052275, | |
| "eval_cer": 0.3531364347339887, | |
| "eval_loss": 1.1585793495178223, | |
| "eval_runtime": 229.1811, | |
| "eval_samples_per_second": 23.828, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 12.185866408518876, | |
| "eval_cer": 0.3518553906476704, | |
| "eval_loss": 1.1155860424041748, | |
| "eval_runtime": 229.1686, | |
| "eval_samples_per_second": 23.83, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 12.37947725072604, | |
| "eval_cer": 0.34836613797057137, | |
| "eval_loss": 1.0729304552078247, | |
| "eval_runtime": 235.2642, | |
| "eval_samples_per_second": 23.212, | |
| "eval_steps_per_second": 2.903, | |
| "eval_wer": 1.0, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 12.573088092933205, | |
| "grad_norm": 7.104965686798096, | |
| "learning_rate": 1.5588e-05, | |
| "loss": 1.1212, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.573088092933205, | |
| "eval_cer": 0.3466533607943889, | |
| "eval_loss": 1.0344808101654053, | |
| "eval_runtime": 229.0405, | |
| "eval_samples_per_second": 23.843, | |
| "eval_steps_per_second": 2.982, | |
| "eval_wer": 1.0, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.766698935140369, | |
| "eval_cer": 0.3428314613103453, | |
| "eval_loss": 0.988746166229248, | |
| "eval_runtime": 228.7064, | |
| "eval_samples_per_second": 23.878, | |
| "eval_steps_per_second": 2.986, | |
| "eval_wer": 1.0, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 12.96030977734753, | |
| "eval_cer": 0.3416636586902209, | |
| "eval_loss": 0.963031530380249, | |
| "eval_runtime": 229.2204, | |
| "eval_samples_per_second": 23.824, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 13.152952565343659, | |
| "eval_cer": 0.3380753197302022, | |
| "eval_loss": 0.9259727597236633, | |
| "eval_runtime": 230.8473, | |
| "eval_samples_per_second": 23.656, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 1.0, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 13.346563407550823, | |
| "eval_cer": 0.3397102433983764, | |
| "eval_loss": 0.9005178809165955, | |
| "eval_runtime": 229.79, | |
| "eval_samples_per_second": 23.765, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 1.0, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 13.540174249757987, | |
| "grad_norm": 8.212745666503906, | |
| "learning_rate": 1.6788e-05, | |
| "loss": 0.9141, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.540174249757987, | |
| "eval_cer": 0.33693582747662626, | |
| "eval_loss": 0.8763672709465027, | |
| "eval_runtime": 233.0771, | |
| "eval_samples_per_second": 23.43, | |
| "eval_steps_per_second": 2.93, | |
| "eval_wer": 1.0, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.73378509196515, | |
| "eval_cer": 0.3362776114543743, | |
| "eval_loss": 0.8511508703231812, | |
| "eval_runtime": 228.4896, | |
| "eval_samples_per_second": 23.9, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 1.0, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 13.927395934172313, | |
| "eval_cer": 0.3351098088342499, | |
| "eval_loss": 0.8273207545280457, | |
| "eval_runtime": 228.9007, | |
| "eval_samples_per_second": 23.858, | |
| "eval_steps_per_second": 2.984, | |
| "eval_wer": 1.0, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 14.120038722168442, | |
| "eval_cer": 0.3328520571020093, | |
| "eval_loss": 0.8083305358886719, | |
| "eval_runtime": 230.4678, | |
| "eval_samples_per_second": 23.695, | |
| "eval_steps_per_second": 2.964, | |
| "eval_wer": 1.0, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 14.313649564375606, | |
| "eval_cer": 0.33002102044716225, | |
| "eval_loss": 0.7850707769393921, | |
| "eval_runtime": 234.2407, | |
| "eval_samples_per_second": 23.314, | |
| "eval_steps_per_second": 2.916, | |
| "eval_wer": 0.999816883354697, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 14.507260406582768, | |
| "grad_norm": 5.25850772857666, | |
| "learning_rate": 1.7988e-05, | |
| "loss": 0.7811, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.507260406582768, | |
| "eval_cer": 0.3312312886171094, | |
| "eval_loss": 0.7742574214935303, | |
| "eval_runtime": 228.7569, | |
| "eval_samples_per_second": 23.873, | |
| "eval_steps_per_second": 2.986, | |
| "eval_wer": 1.0, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.700871248789932, | |
| "eval_cer": 0.3272112165672265, | |
| "eval_loss": 0.7509779334068298, | |
| "eval_runtime": 225.7839, | |
| "eval_samples_per_second": 24.187, | |
| "eval_steps_per_second": 3.025, | |
| "eval_wer": 0.999816883354697, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 14.894482090997096, | |
| "eval_cer": 0.3267299403359025, | |
| "eval_loss": 0.7366129159927368, | |
| "eval_runtime": 234.848, | |
| "eval_samples_per_second": 23.253, | |
| "eval_steps_per_second": 2.908, | |
| "eval_wer": 1.0, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 15.087124878993224, | |
| "eval_cer": 0.3253427323750274, | |
| "eval_loss": 0.7289799451828003, | |
| "eval_runtime": 233.7629, | |
| "eval_samples_per_second": 23.361, | |
| "eval_steps_per_second": 2.922, | |
| "eval_wer": 1.0, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 15.280735721200386, | |
| "eval_cer": 0.3247199043109611, | |
| "eval_loss": 0.7132401466369629, | |
| "eval_runtime": 240.827, | |
| "eval_samples_per_second": 22.676, | |
| "eval_steps_per_second": 2.836, | |
| "eval_wer": 1.0, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 15.47434656340755, | |
| "grad_norm": 13.825493812561035, | |
| "learning_rate": 1.9188e-05, | |
| "loss": 0.6725, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.47434656340755, | |
| "eval_cer": 0.3276924927985505, | |
| "eval_loss": 0.7190116047859192, | |
| "eval_runtime": 231.4321, | |
| "eval_samples_per_second": 23.597, | |
| "eval_steps_per_second": 2.951, | |
| "eval_wer": 1.0, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.667957405614715, | |
| "eval_cer": 0.3241253866134432, | |
| "eval_loss": 0.7005703449249268, | |
| "eval_runtime": 229.0636, | |
| "eval_samples_per_second": 23.841, | |
| "eval_steps_per_second": 2.982, | |
| "eval_wer": 1.0, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 15.861568247821879, | |
| "eval_cer": 0.322554161270003, | |
| "eval_loss": 0.683487594127655, | |
| "eval_runtime": 230.5943, | |
| "eval_samples_per_second": 23.682, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 1.0, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 16.054211035818007, | |
| "eval_cer": 0.32088384964364325, | |
| "eval_loss": 0.6697654724121094, | |
| "eval_runtime": 253.1459, | |
| "eval_samples_per_second": 21.573, | |
| "eval_steps_per_second": 2.698, | |
| "eval_wer": 0.999816883354697, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 16.24782187802517, | |
| "eval_cer": 0.3213722034666044, | |
| "eval_loss": 0.6627610921859741, | |
| "eval_runtime": 228.1606, | |
| "eval_samples_per_second": 23.935, | |
| "eval_steps_per_second": 2.994, | |
| "eval_wer": 0.999816883354697, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 16.44143272023233, | |
| "grad_norm": 4.491298675537109, | |
| "learning_rate": 2.0388e-05, | |
| "loss": 0.606, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.44143272023233, | |
| "eval_cer": 0.320508737286876, | |
| "eval_loss": 0.6538371443748474, | |
| "eval_runtime": 229.6742, | |
| "eval_samples_per_second": 23.777, | |
| "eval_steps_per_second": 2.974, | |
| "eval_wer": 1.0, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.635043562439495, | |
| "eval_cer": 0.3186119427281285, | |
| "eval_loss": 0.652263343334198, | |
| "eval_runtime": 229.4944, | |
| "eval_samples_per_second": 23.796, | |
| "eval_steps_per_second": 2.976, | |
| "eval_wer": 1.0, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 16.82865440464666, | |
| "eval_cer": 0.3183005286960953, | |
| "eval_loss": 0.6448878049850464, | |
| "eval_runtime": 253.2088, | |
| "eval_samples_per_second": 21.567, | |
| "eval_steps_per_second": 2.697, | |
| "eval_wer": 1.0, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 17.021297192642788, | |
| "eval_cer": 0.3179466491142394, | |
| "eval_loss": 0.6400793194770813, | |
| "eval_runtime": 227.3314, | |
| "eval_samples_per_second": 24.022, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 1.0, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 17.21490803484995, | |
| "eval_cer": 0.3199849955057293, | |
| "eval_loss": 0.6333277821540833, | |
| "eval_runtime": 227.9774, | |
| "eval_samples_per_second": 23.954, | |
| "eval_steps_per_second": 2.996, | |
| "eval_wer": 1.0, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 17.408518877057116, | |
| "grad_norm": 3.5857622623443604, | |
| "learning_rate": 2.1588e-05, | |
| "loss": 0.5492, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.408518877057116, | |
| "eval_cer": 0.320147780113383, | |
| "eval_loss": 0.6332593560218811, | |
| "eval_runtime": 229.5302, | |
| "eval_samples_per_second": 23.792, | |
| "eval_steps_per_second": 2.976, | |
| "eval_wer": 1.0, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.60212971926428, | |
| "eval_cer": 0.3179324939309652, | |
| "eval_loss": 0.6219143867492676, | |
| "eval_runtime": 227.7602, | |
| "eval_samples_per_second": 23.977, | |
| "eval_steps_per_second": 2.999, | |
| "eval_wer": 1.0, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 17.795740561471444, | |
| "eval_cer": 0.32011239215519743, | |
| "eval_loss": 0.6189157366752625, | |
| "eval_runtime": 234.0727, | |
| "eval_samples_per_second": 23.33, | |
| "eval_steps_per_second": 2.918, | |
| "eval_wer": 1.0, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 17.989351403678604, | |
| "eval_cer": 0.3165523635617272, | |
| "eval_loss": 0.6023225784301758, | |
| "eval_runtime": 252.0388, | |
| "eval_samples_per_second": 21.667, | |
| "eval_steps_per_second": 2.71, | |
| "eval_wer": 0.999816883354697, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 18.181994191674733, | |
| "eval_cer": 0.31535625057505434, | |
| "eval_loss": 0.6083632707595825, | |
| "eval_runtime": 237.956, | |
| "eval_samples_per_second": 22.95, | |
| "eval_steps_per_second": 2.87, | |
| "eval_wer": 1.0, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 18.375605033881897, | |
| "grad_norm": 4.794497489929199, | |
| "learning_rate": 2.2788000000000003e-05, | |
| "loss": 0.5057, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.375605033881897, | |
| "eval_cer": 0.31467680177789104, | |
| "eval_loss": 0.6001758575439453, | |
| "eval_runtime": 227.7296, | |
| "eval_samples_per_second": 23.98, | |
| "eval_steps_per_second": 2.999, | |
| "eval_wer": 0.999816883354697, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.56921587608906, | |
| "eval_cer": 0.31280123999405485, | |
| "eval_loss": 0.5875076055526733, | |
| "eval_runtime": 226.4211, | |
| "eval_samples_per_second": 24.119, | |
| "eval_steps_per_second": 3.017, | |
| "eval_wer": 1.0, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 18.762826718296225, | |
| "eval_cer": 0.31375671486506573, | |
| "eval_loss": 0.5903081297874451, | |
| "eval_runtime": 227.027, | |
| "eval_samples_per_second": 24.054, | |
| "eval_steps_per_second": 3.008, | |
| "eval_wer": 0.999816883354697, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 18.95643756050339, | |
| "eval_cer": 0.3126738433445867, | |
| "eval_loss": 0.5929713249206543, | |
| "eval_runtime": 243.7928, | |
| "eval_samples_per_second": 22.4, | |
| "eval_steps_per_second": 2.802, | |
| "eval_wer": 1.0, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 19.149080348499517, | |
| "eval_cer": 0.31413890481347007, | |
| "eval_loss": 0.5854855179786682, | |
| "eval_runtime": 230.7398, | |
| "eval_samples_per_second": 23.667, | |
| "eval_steps_per_second": 2.96, | |
| "eval_wer": 1.0, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 19.34269119070668, | |
| "grad_norm": 4.687788009643555, | |
| "learning_rate": 2.39856e-05, | |
| "loss": 0.4709, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.34269119070668, | |
| "eval_cer": 0.3119943945474234, | |
| "eval_loss": 0.5880363583564758, | |
| "eval_runtime": 227.5508, | |
| "eval_samples_per_second": 23.999, | |
| "eval_steps_per_second": 3.002, | |
| "eval_wer": 1.0, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.53630203291384, | |
| "eval_cer": 0.3131197316177251, | |
| "eval_loss": 0.5854519605636597, | |
| "eval_runtime": 235.4586, | |
| "eval_samples_per_second": 23.193, | |
| "eval_steps_per_second": 2.901, | |
| "eval_wer": 1.0, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 19.729912875121006, | |
| "eval_cer": 0.310649652136371, | |
| "eval_loss": 0.5734274387359619, | |
| "eval_runtime": 227.5849, | |
| "eval_samples_per_second": 23.995, | |
| "eval_steps_per_second": 3.001, | |
| "eval_wer": 1.0, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 19.92352371732817, | |
| "eval_cer": 0.31090444543530726, | |
| "eval_loss": 0.5776930451393127, | |
| "eval_runtime": 227.4224, | |
| "eval_samples_per_second": 24.013, | |
| "eval_steps_per_second": 3.003, | |
| "eval_wer": 1.0, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 19.962245885769605, | |
| "step": 10320, | |
| "total_flos": 1.4589649587726338e+19, | |
| "train_loss": 5.084269817855007, | |
| "train_runtime": 64283.8792, | |
| "train_samples_per_second": 5.142, | |
| "train_steps_per_second": 0.161 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10320, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4589649587726338e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |