| { | |
| "best_metric": 0.441910982131958, | |
| "best_model_checkpoint": "./Hubert-common_voice-ja-demo-roma-debug-40epochs-cosine/checkpoint-9700", | |
| "epoch": 40.0, | |
| "eval_steps": 100, | |
| "global_step": 15040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26595744680851063, | |
| "eval_cer": 3.477914534858724, | |
| "eval_loss": 16.817241668701172, | |
| "eval_runtime": 221.1037, | |
| "eval_samples_per_second": 22.428, | |
| "eval_steps_per_second": 2.804, | |
| "eval_wer": 2.9493849566444847, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "eval_cer": 2.9561261222527526, | |
| "eval_loss": 16.550180435180664, | |
| "eval_runtime": 213.5908, | |
| "eval_samples_per_second": 23.217, | |
| "eval_steps_per_second": 2.903, | |
| "eval_wer": 2.7606372252470255, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "eval_cer": 1.6838657478883865, | |
| "eval_loss": 15.8340425491333, | |
| "eval_runtime": 213.9597, | |
| "eval_samples_per_second": 23.177, | |
| "eval_steps_per_second": 2.898, | |
| "eval_wer": 1.9100625126033475, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "eval_cer": 0.9275934445195761, | |
| "eval_loss": 13.291926383972168, | |
| "eval_runtime": 211.7589, | |
| "eval_samples_per_second": 23.418, | |
| "eval_steps_per_second": 2.928, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 41.734230041503906, | |
| "learning_rate": 1.1904e-06, | |
| "loss": 12.6588, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 7.879184246063232, | |
| "eval_runtime": 213.1723, | |
| "eval_samples_per_second": 23.263, | |
| "eval_steps_per_second": 2.908, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 6.154232501983643, | |
| "eval_runtime": 212.0332, | |
| "eval_samples_per_second": 23.388, | |
| "eval_steps_per_second": 2.924, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.775670051574707, | |
| "eval_runtime": 210.8145, | |
| "eval_samples_per_second": 23.523, | |
| "eval_steps_per_second": 2.941, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.6188063621521, | |
| "eval_runtime": 211.2888, | |
| "eval_samples_per_second": 23.47, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.475332260131836, | |
| "eval_runtime": 208.2147, | |
| "eval_samples_per_second": 23.817, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 41.6483154296875, | |
| "learning_rate": 2.3904e-06, | |
| "loss": 5.2353, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.3238911628723145, | |
| "eval_runtime": 208.8415, | |
| "eval_samples_per_second": 23.745, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.167562961578369, | |
| "eval_runtime": 207.9905, | |
| "eval_samples_per_second": 23.842, | |
| "eval_steps_per_second": 2.981, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.008359432220459, | |
| "eval_runtime": 208.6683, | |
| "eval_samples_per_second": 23.765, | |
| "eval_steps_per_second": 2.971, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.840206146240234, | |
| "eval_runtime": 207.1918, | |
| "eval_samples_per_second": 23.934, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.670183181762695, | |
| "eval_runtime": 206.5028, | |
| "eval_samples_per_second": 24.014, | |
| "eval_steps_per_second": 3.002, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 25.022253036499023, | |
| "learning_rate": 3.5904e-06, | |
| "loss": 4.4502, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.495745658874512, | |
| "eval_runtime": 206.573, | |
| "eval_samples_per_second": 24.006, | |
| "eval_steps_per_second": 3.001, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.321905136108398, | |
| "eval_runtime": 208.0558, | |
| "eval_samples_per_second": 23.835, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.5212765957446805, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.150215148925781, | |
| "eval_runtime": 213.3593, | |
| "eval_samples_per_second": 23.242, | |
| "eval_steps_per_second": 2.906, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.787234042553192, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.9856085777282715, | |
| "eval_runtime": 207.2181, | |
| "eval_samples_per_second": 23.931, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.053191489361702, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.834258794784546, | |
| "eval_runtime": 206.1057, | |
| "eval_samples_per_second": 24.06, | |
| "eval_steps_per_second": 3.008, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "grad_norm": 12.537736892700195, | |
| "learning_rate": 4.7904e-06, | |
| "loss": 3.7863, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.690730333328247, | |
| "eval_runtime": 207.5053, | |
| "eval_samples_per_second": 23.898, | |
| "eval_steps_per_second": 2.988, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.585106382978723, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.5544023513793945, | |
| "eval_runtime": 207.2908, | |
| "eval_samples_per_second": 23.923, | |
| "eval_steps_per_second": 2.991, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.851063829787234, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.4332330226898193, | |
| "eval_runtime": 208.1267, | |
| "eval_samples_per_second": 23.827, | |
| "eval_steps_per_second": 2.979, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.117021276595745, | |
| "eval_cer": 0.9275817187616037, | |
| "eval_loss": 3.3063294887542725, | |
| "eval_runtime": 207.0236, | |
| "eval_samples_per_second": 23.954, | |
| "eval_steps_per_second": 2.995, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.382978723404255, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.2074904441833496, | |
| "eval_runtime": 207.8656, | |
| "eval_samples_per_second": 23.857, | |
| "eval_steps_per_second": 2.983, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "grad_norm": 6.3363356590271, | |
| "learning_rate": 5.9904e-06, | |
| "loss": 3.2473, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.1271653175354004, | |
| "eval_runtime": 209.3454, | |
| "eval_samples_per_second": 23.688, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.914893617021277, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.0657362937927246, | |
| "eval_runtime": 209.9086, | |
| "eval_samples_per_second": 23.625, | |
| "eval_steps_per_second": 2.954, | |
| "eval_wer": 1.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.180851063829787, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.0163941383361816, | |
| "eval_runtime": 209.8668, | |
| "eval_samples_per_second": 23.629, | |
| "eval_steps_per_second": 2.954, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.446808510638298, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.9747605323791504, | |
| "eval_runtime": 209.2203, | |
| "eval_samples_per_second": 23.702, | |
| "eval_steps_per_second": 2.963, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.712765957446808, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.9446794986724854, | |
| "eval_runtime": 208.1917, | |
| "eval_samples_per_second": 23.819, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 1.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "grad_norm": 1.2298308610916138, | |
| "learning_rate": 7.190400000000001e-06, | |
| "loss": 2.9649, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.918757915496826, | |
| "eval_runtime": 208.2245, | |
| "eval_samples_per_second": 23.816, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.24468085106383, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.9005508422851562, | |
| "eval_runtime": 205.5088, | |
| "eval_samples_per_second": 24.13, | |
| "eval_steps_per_second": 3.017, | |
| "eval_wer": 1.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.51063829787234, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8877005577087402, | |
| "eval_runtime": 205.8033, | |
| "eval_samples_per_second": 24.096, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.77659574468085, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8712358474731445, | |
| "eval_runtime": 206.9211, | |
| "eval_samples_per_second": 23.966, | |
| "eval_steps_per_second": 2.996, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.042553191489361, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8528554439544678, | |
| "eval_runtime": 205.6952, | |
| "eval_samples_per_second": 24.108, | |
| "eval_steps_per_second": 3.014, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "grad_norm": 1.270645022392273, | |
| "learning_rate": 8.3904e-06, | |
| "loss": 2.8673, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8438620567321777, | |
| "eval_runtime": 206.4519, | |
| "eval_samples_per_second": 24.02, | |
| "eval_steps_per_second": 3.003, | |
| "eval_wer": 1.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.574468085106384, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8312649726867676, | |
| "eval_runtime": 206.5101, | |
| "eval_samples_per_second": 24.013, | |
| "eval_steps_per_second": 3.002, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.840425531914894, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8182146549224854, | |
| "eval_runtime": 206.3851, | |
| "eval_samples_per_second": 24.028, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 1.0, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.106382978723405, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.7310709953308105, | |
| "eval_runtime": 207.1152, | |
| "eval_samples_per_second": 23.943, | |
| "eval_steps_per_second": 2.994, | |
| "eval_wer": 1.0, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 10.372340425531915, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.4996562004089355, | |
| "eval_runtime": 207.3947, | |
| "eval_samples_per_second": 23.911, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 1.0, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "grad_norm": 2.881579875946045, | |
| "learning_rate": 9.5904e-06, | |
| "loss": 2.6801, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "eval_cer": 0.8951130949356452, | |
| "eval_loss": 2.2397685050964355, | |
| "eval_runtime": 207.9025, | |
| "eval_samples_per_second": 23.853, | |
| "eval_steps_per_second": 2.982, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.904255319148936, | |
| "eval_cer": 0.6153560526408361, | |
| "eval_loss": 1.9111369848251343, | |
| "eval_runtime": 209.5566, | |
| "eval_samples_per_second": 23.664, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 1.0, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.170212765957446, | |
| "eval_cer": 0.4341266460032754, | |
| "eval_loss": 1.5446799993515015, | |
| "eval_runtime": 210.5656, | |
| "eval_samples_per_second": 23.551, | |
| "eval_steps_per_second": 2.944, | |
| "eval_wer": 1.0, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 11.436170212765958, | |
| "eval_cer": 0.3958537719809105, | |
| "eval_loss": 1.3182002305984497, | |
| "eval_runtime": 210.4243, | |
| "eval_samples_per_second": 23.567, | |
| "eval_steps_per_second": 2.946, | |
| "eval_wer": 1.0, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.702127659574469, | |
| "eval_cer": 0.3706043064800447, | |
| "eval_loss": 1.1702227592468262, | |
| "eval_runtime": 210.384, | |
| "eval_samples_per_second": 23.571, | |
| "eval_steps_per_second": 2.947, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "grad_norm": 3.1133100986480713, | |
| "learning_rate": 1.0790400000000001e-05, | |
| "loss": 1.5214, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "eval_cer": 0.32139911744128324, | |
| "eval_loss": 1.0558491945266724, | |
| "eval_runtime": 211.4587, | |
| "eval_samples_per_second": 23.451, | |
| "eval_steps_per_second": 2.932, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.23404255319149, | |
| "eval_cer": 0.3023643036658628, | |
| "eval_loss": 0.9717462658882141, | |
| "eval_runtime": 211.083, | |
| "eval_samples_per_second": 23.493, | |
| "eval_steps_per_second": 2.937, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_cer": 0.2873631506329955, | |
| "eval_loss": 0.8958828449249268, | |
| "eval_runtime": 215.6014, | |
| "eval_samples_per_second": 23.001, | |
| "eval_steps_per_second": 2.876, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.76595744680851, | |
| "eval_cer": 0.2747306007105809, | |
| "eval_loss": 0.8398593068122864, | |
| "eval_runtime": 211.3172, | |
| "eval_samples_per_second": 23.467, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 13.03191489361702, | |
| "eval_cer": 0.2656900413137539, | |
| "eval_loss": 0.7890844345092773, | |
| "eval_runtime": 211.3236, | |
| "eval_samples_per_second": 23.466, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "grad_norm": 2.623598098754883, | |
| "learning_rate": 1.19904e-05, | |
| "loss": 0.8719, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "eval_cer": 0.25801748701372307, | |
| "eval_loss": 0.7483934760093689, | |
| "eval_runtime": 210.0979, | |
| "eval_samples_per_second": 23.603, | |
| "eval_steps_per_second": 2.951, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.563829787234042, | |
| "eval_cer": 0.2522914085371335, | |
| "eval_loss": 0.7144805788993835, | |
| "eval_runtime": 209.8962, | |
| "eval_samples_per_second": 23.626, | |
| "eval_steps_per_second": 2.954, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.829787234042554, | |
| "eval_cer": 0.24809358718296481, | |
| "eval_loss": 0.6851741075515747, | |
| "eval_runtime": 209.9795, | |
| "eval_samples_per_second": 23.617, | |
| "eval_steps_per_second": 2.953, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 14.095744680851064, | |
| "eval_cer": 0.24873068669947274, | |
| "eval_loss": 0.6617541313171387, | |
| "eval_runtime": 215.3335, | |
| "eval_samples_per_second": 23.029, | |
| "eval_steps_per_second": 2.879, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 14.361702127659575, | |
| "eval_cer": 0.2476675513099626, | |
| "eval_loss": 0.6399756073951721, | |
| "eval_runtime": 210.9211, | |
| "eval_samples_per_second": 23.511, | |
| "eval_steps_per_second": 2.939, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "grad_norm": 3.613203763961792, | |
| "learning_rate": 1.31904e-05, | |
| "loss": 0.6568, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "eval_cer": 0.2448690037405168, | |
| "eval_loss": 0.6200099587440491, | |
| "eval_runtime": 208.8016, | |
| "eval_samples_per_second": 23.75, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.893617021276595, | |
| "eval_cer": 0.2420548218271076, | |
| "eval_loss": 0.6031706929206848, | |
| "eval_runtime": 208.8818, | |
| "eval_samples_per_second": 23.741, | |
| "eval_steps_per_second": 2.968, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 15.159574468085106, | |
| "eval_cer": 0.2395025151750851, | |
| "eval_loss": 0.5874983668327332, | |
| "eval_runtime": 215.2428, | |
| "eval_samples_per_second": 23.039, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 15.425531914893616, | |
| "eval_cer": 0.24089788037381715, | |
| "eval_loss": 0.5776127576828003, | |
| "eval_runtime": 208.795, | |
| "eval_samples_per_second": 23.751, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 15.691489361702128, | |
| "eval_cer": 0.23602778222922294, | |
| "eval_loss": 0.5617058873176575, | |
| "eval_runtime": 209.2365, | |
| "eval_samples_per_second": 23.7, | |
| "eval_steps_per_second": 2.963, | |
| "eval_wer": 0.999395039322444, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 15.957446808510639, | |
| "grad_norm": 2.9839932918548584, | |
| "learning_rate": 1.43904e-05, | |
| "loss": 0.548, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.957446808510639, | |
| "eval_cer": 0.23474576602422542, | |
| "eval_loss": 0.5485217571258545, | |
| "eval_runtime": 212.04, | |
| "eval_samples_per_second": 23.387, | |
| "eval_steps_per_second": 2.924, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.22340425531915, | |
| "eval_cer": 0.23343248113130113, | |
| "eval_loss": 0.5394352078437805, | |
| "eval_runtime": 220.2211, | |
| "eval_samples_per_second": 22.518, | |
| "eval_steps_per_second": 2.815, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 16.48936170212766, | |
| "eval_cer": 0.23171270329532884, | |
| "eval_loss": 0.5321902632713318, | |
| "eval_runtime": 221.3122, | |
| "eval_samples_per_second": 22.407, | |
| "eval_steps_per_second": 2.801, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 16.75531914893617, | |
| "eval_cer": 0.23198630431468809, | |
| "eval_loss": 0.5243075489997864, | |
| "eval_runtime": 210.0529, | |
| "eval_samples_per_second": 23.608, | |
| "eval_steps_per_second": 2.952, | |
| "eval_wer": 0.9969751966122202, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 17.02127659574468, | |
| "eval_cer": 0.22724518950779177, | |
| "eval_loss": 0.5120757222175598, | |
| "eval_runtime": 207.2228, | |
| "eval_samples_per_second": 23.931, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 17.28723404255319, | |
| "grad_norm": 2.14619779586792, | |
| "learning_rate": 1.5590400000000002e-05, | |
| "loss": 0.4681, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.28723404255319, | |
| "eval_cer": 0.22661981574925638, | |
| "eval_loss": 0.5070440173149109, | |
| "eval_runtime": 221.9297, | |
| "eval_samples_per_second": 22.345, | |
| "eval_steps_per_second": 2.794, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.5531914893617, | |
| "eval_cer": 0.22627976876805278, | |
| "eval_loss": 0.5013606548309326, | |
| "eval_runtime": 207.1035, | |
| "eval_samples_per_second": 23.945, | |
| "eval_steps_per_second": 2.994, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 17.819148936170212, | |
| "eval_cer": 0.2241613151610142, | |
| "eval_loss": 0.4942595660686493, | |
| "eval_runtime": 207.4255, | |
| "eval_samples_per_second": 23.907, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 18.085106382978722, | |
| "eval_cer": 0.2228128529941723, | |
| "eval_loss": 0.4929651618003845, | |
| "eval_runtime": 209.5569, | |
| "eval_samples_per_second": 23.664, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 18.351063829787233, | |
| "eval_cer": 0.2245013621422178, | |
| "eval_loss": 0.4968652129173279, | |
| "eval_runtime": 229.2892, | |
| "eval_samples_per_second": 21.628, | |
| "eval_steps_per_second": 2.704, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 18.617021276595743, | |
| "grad_norm": 3.0399532318115234, | |
| "learning_rate": 1.6790399999999998e-05, | |
| "loss": 0.4198, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.617021276595743, | |
| "eval_cer": 0.22252752621684052, | |
| "eval_loss": 0.48828795552253723, | |
| "eval_runtime": 211.562, | |
| "eval_samples_per_second": 23.44, | |
| "eval_steps_per_second": 2.931, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.882978723404257, | |
| "eval_cer": 0.22148784234327548, | |
| "eval_loss": 0.4804530441761017, | |
| "eval_runtime": 212.7369, | |
| "eval_samples_per_second": 23.31, | |
| "eval_steps_per_second": 2.914, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 19.148936170212767, | |
| "eval_cer": 0.22081556555284995, | |
| "eval_loss": 0.47766780853271484, | |
| "eval_runtime": 210.7245, | |
| "eval_samples_per_second": 23.533, | |
| "eval_steps_per_second": 2.942, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 19.414893617021278, | |
| "eval_cer": 0.22092500596059364, | |
| "eval_loss": 0.47176772356033325, | |
| "eval_runtime": 209.4286, | |
| "eval_samples_per_second": 23.679, | |
| "eval_steps_per_second": 2.96, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 19.680851063829788, | |
| "eval_cer": 0.21988532208702857, | |
| "eval_loss": 0.47212347388267517, | |
| "eval_runtime": 211.3146, | |
| "eval_samples_per_second": 23.467, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 19.9468085106383, | |
| "grad_norm": 2.2681851387023926, | |
| "learning_rate": 1.79904e-05, | |
| "loss": 0.3795, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.9468085106383, | |
| "eval_cer": 0.22045206705570125, | |
| "eval_loss": 0.4674960672855377, | |
| "eval_runtime": 211.1631, | |
| "eval_samples_per_second": 23.484, | |
| "eval_steps_per_second": 2.936, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.21276595744681, | |
| "eval_cer": 0.21619561691166986, | |
| "eval_loss": 0.46922406554222107, | |
| "eval_runtime": 210.3069, | |
| "eval_samples_per_second": 23.58, | |
| "eval_steps_per_second": 2.948, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 20.47872340425532, | |
| "eval_cer": 0.21727829523113423, | |
| "eval_loss": 0.47320201992988586, | |
| "eval_runtime": 208.0889, | |
| "eval_samples_per_second": 23.831, | |
| "eval_steps_per_second": 2.979, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 20.74468085106383, | |
| "eval_cer": 0.217309563919061, | |
| "eval_loss": 0.46535801887512207, | |
| "eval_runtime": 208.908, | |
| "eval_samples_per_second": 23.738, | |
| "eval_steps_per_second": 2.968, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 21.01063829787234, | |
| "eval_cer": 0.21583211841452118, | |
| "eval_loss": 0.4556586742401123, | |
| "eval_runtime": 207.993, | |
| "eval_samples_per_second": 23.842, | |
| "eval_steps_per_second": 2.981, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 21.27659574468085, | |
| "grad_norm": 1.6410722732543945, | |
| "learning_rate": 1.9190400000000002e-05, | |
| "loss": 0.3504, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.27659574468085, | |
| "eval_cer": 0.21436249008196304, | |
| "eval_loss": 0.4562208950519562, | |
| "eval_runtime": 209.2953, | |
| "eval_samples_per_second": 23.694, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.54255319148936, | |
| "eval_cer": 0.2143742158399356, | |
| "eval_loss": 0.4679461419582367, | |
| "eval_runtime": 207.9541, | |
| "eval_samples_per_second": 23.847, | |
| "eval_steps_per_second": 2.981, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 21.80851063829787, | |
| "eval_cer": 0.21689525380403132, | |
| "eval_loss": 0.45842525362968445, | |
| "eval_runtime": 208.594, | |
| "eval_samples_per_second": 23.773, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 22.074468085106382, | |
| "eval_cer": 0.2133853435842515, | |
| "eval_loss": 0.45605719089508057, | |
| "eval_runtime": 209.0293, | |
| "eval_samples_per_second": 23.724, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 22.340425531914892, | |
| "eval_cer": 0.21432731280804543, | |
| "eval_loss": 0.45950454473495483, | |
| "eval_runtime": 208.8441, | |
| "eval_samples_per_second": 23.745, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 22.606382978723403, | |
| "grad_norm": 2.5605006217956543, | |
| "learning_rate": 2.03904e-05, | |
| "loss": 0.3134, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.606382978723403, | |
| "eval_cer": 0.21549988860529926, | |
| "eval_loss": 0.4543595314025879, | |
| "eval_runtime": 207.7389, | |
| "eval_samples_per_second": 23.871, | |
| "eval_steps_per_second": 2.985, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.872340425531917, | |
| "eval_cer": 0.21343615520213252, | |
| "eval_loss": 0.4543818533420563, | |
| "eval_runtime": 210.3188, | |
| "eval_samples_per_second": 23.578, | |
| "eval_steps_per_second": 2.948, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 23.138297872340427, | |
| "eval_cer": 0.21289286174940492, | |
| "eval_loss": 0.45519956946372986, | |
| "eval_runtime": 210.0976, | |
| "eval_samples_per_second": 23.603, | |
| "eval_steps_per_second": 2.951, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 23.404255319148938, | |
| "eval_cer": 0.21207987586330893, | |
| "eval_loss": 0.45236727595329285, | |
| "eval_runtime": 209.5064, | |
| "eval_samples_per_second": 23.67, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 23.670212765957448, | |
| "eval_cer": 0.21132551876707564, | |
| "eval_loss": 0.45537006855010986, | |
| "eval_runtime": 209.1429, | |
| "eval_samples_per_second": 23.711, | |
| "eval_steps_per_second": 2.964, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 23.93617021276596, | |
| "grad_norm": 1.3679530620574951, | |
| "learning_rate": 2.15904e-05, | |
| "loss": 0.3014, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 23.93617021276596, | |
| "eval_cer": 0.21025847479157464, | |
| "eval_loss": 0.46165505051612854, | |
| "eval_runtime": 209.5827, | |
| "eval_samples_per_second": 23.661, | |
| "eval_steps_per_second": 2.958, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.20212765957447, | |
| "eval_cer": 0.21297103346922183, | |
| "eval_loss": 0.4606345295906067, | |
| "eval_runtime": 209.6362, | |
| "eval_samples_per_second": 23.655, | |
| "eval_steps_per_second": 2.958, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 24.46808510638298, | |
| "eval_cer": 0.21053598439692472, | |
| "eval_loss": 0.4560868740081787, | |
| "eval_runtime": 209.0063, | |
| "eval_samples_per_second": 23.727, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 24.73404255319149, | |
| "eval_cer": 0.20888656110878767, | |
| "eval_loss": 0.45663759112358093, | |
| "eval_runtime": 211.6105, | |
| "eval_samples_per_second": 23.435, | |
| "eval_steps_per_second": 2.93, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.21188444656376662, | |
| "eval_loss": 0.4485580623149872, | |
| "eval_runtime": 212.8182, | |
| "eval_samples_per_second": 23.302, | |
| "eval_steps_per_second": 2.913, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 25.26595744680851, | |
| "grad_norm": 9.48050594329834, | |
| "learning_rate": 2.27904e-05, | |
| "loss": 0.2791, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.26595744680851, | |
| "eval_cer": 0.21173982888210532, | |
| "eval_loss": 0.4541599452495575, | |
| "eval_runtime": 212.1403, | |
| "eval_samples_per_second": 23.376, | |
| "eval_steps_per_second": 2.923, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.53191489361702, | |
| "eval_cer": 0.20951193486732306, | |
| "eval_loss": 0.4539582133293152, | |
| "eval_runtime": 211.4226, | |
| "eval_samples_per_second": 23.455, | |
| "eval_steps_per_second": 2.933, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 25.79787234042553, | |
| "eval_cer": 0.20911325909625675, | |
| "eval_loss": 0.441910982131958, | |
| "eval_runtime": 213.904, | |
| "eval_samples_per_second": 23.183, | |
| "eval_steps_per_second": 2.898, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 26.06382978723404, | |
| "eval_cer": 0.20742474994821122, | |
| "eval_loss": 0.45685461163520813, | |
| "eval_runtime": 212.9009, | |
| "eval_samples_per_second": 23.293, | |
| "eval_steps_per_second": 2.912, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 26.329787234042552, | |
| "eval_cer": 0.20901163586049476, | |
| "eval_loss": 0.45425865054130554, | |
| "eval_runtime": 210.5454, | |
| "eval_samples_per_second": 23.553, | |
| "eval_steps_per_second": 2.945, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 26.595744680851062, | |
| "grad_norm": 2.0923683643341064, | |
| "learning_rate": 2.39904e-05, | |
| "loss": 0.2564, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.595744680851062, | |
| "eval_cer": 0.20882402373293413, | |
| "eval_loss": 0.4689125120639801, | |
| "eval_runtime": 211.8844, | |
| "eval_samples_per_second": 23.404, | |
| "eval_steps_per_second": 2.926, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.861702127659573, | |
| "eval_cer": 0.20887483535081514, | |
| "eval_loss": 0.45904749631881714, | |
| "eval_runtime": 213.2034, | |
| "eval_samples_per_second": 23.259, | |
| "eval_steps_per_second": 2.908, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 27.127659574468087, | |
| "eval_cer": 0.2093282313257533, | |
| "eval_loss": 0.4985686242580414, | |
| "eval_runtime": 211.4066, | |
| "eval_samples_per_second": 23.457, | |
| "eval_steps_per_second": 2.933, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 27.393617021276597, | |
| "eval_cer": 0.20998096518622458, | |
| "eval_loss": 0.46933168172836304, | |
| "eval_runtime": 210.8303, | |
| "eval_samples_per_second": 23.521, | |
| "eval_steps_per_second": 2.941, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 27.659574468085108, | |
| "eval_cer": 0.208472250993758, | |
| "eval_loss": 0.5127565860748291, | |
| "eval_runtime": 210.3509, | |
| "eval_samples_per_second": 23.575, | |
| "eval_steps_per_second": 2.947, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 27.925531914893618, | |
| "grad_norm": 4.9774322509765625, | |
| "learning_rate": 2.51904e-05, | |
| "loss": 0.2449, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 27.925531914893618, | |
| "eval_cer": 0.20994578791230697, | |
| "eval_loss": 0.4511733651161194, | |
| "eval_runtime": 213.3431, | |
| "eval_samples_per_second": 23.244, | |
| "eval_steps_per_second": 2.906, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 28.19148936170213, | |
| "eval_cer": 0.20911325909625675, | |
| "eval_loss": 0.4650718867778778, | |
| "eval_runtime": 212.7967, | |
| "eval_samples_per_second": 23.304, | |
| "eval_steps_per_second": 2.914, | |
| "eval_wer": 0.999395039322444, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 28.45744680851064, | |
| "eval_cer": 0.206795467603685, | |
| "eval_loss": 0.46036314964294434, | |
| "eval_runtime": 211.0991, | |
| "eval_samples_per_second": 23.491, | |
| "eval_steps_per_second": 2.937, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 28.72340425531915, | |
| "eval_cer": 0.20801885501881984, | |
| "eval_loss": 0.46872854232788086, | |
| "eval_runtime": 211.1242, | |
| "eval_samples_per_second": 23.489, | |
| "eval_steps_per_second": 2.937, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 28.98936170212766, | |
| "eval_cer": 0.20636943173068278, | |
| "eval_loss": 0.468840092420578, | |
| "eval_runtime": 212.878, | |
| "eval_samples_per_second": 23.295, | |
| "eval_steps_per_second": 2.912, | |
| "eval_wer": 0.999395039322444, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 29.25531914893617, | |
| "grad_norm": 4.162810802459717, | |
| "learning_rate": 2.63904e-05, | |
| "loss": 0.2258, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.25531914893617, | |
| "eval_cer": 0.20923051667598214, | |
| "eval_loss": 0.4758950173854828, | |
| "eval_runtime": 212.8392, | |
| "eval_samples_per_second": 23.299, | |
| "eval_steps_per_second": 2.913, | |
| "eval_wer": 0.999395039322444, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.52127659574468, | |
| "eval_cer": 0.20683064487760264, | |
| "eval_loss": 0.4816320836544037, | |
| "eval_runtime": 213.093, | |
| "eval_samples_per_second": 23.272, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 29.78723404255319, | |
| "eval_cer": 0.20534147361509028, | |
| "eval_loss": 0.4749976098537445, | |
| "eval_runtime": 212.0731, | |
| "eval_samples_per_second": 23.383, | |
| "eval_steps_per_second": 2.924, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 30.0531914893617, | |
| "eval_cer": 0.2047864544043901, | |
| "eval_loss": 0.47532856464385986, | |
| "eval_runtime": 213.0921, | |
| "eval_samples_per_second": 23.272, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 30.319148936170212, | |
| "eval_cer": 0.2059551216156531, | |
| "eval_loss": 0.48292842507362366, | |
| "eval_runtime": 212.9065, | |
| "eval_samples_per_second": 23.292, | |
| "eval_steps_per_second": 2.912, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 30.585106382978722, | |
| "grad_norm": 2.1135425567626953, | |
| "learning_rate": 2.7590400000000003e-05, | |
| "loss": 0.2124, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 30.585106382978722, | |
| "eval_cer": 0.20811266108260015, | |
| "eval_loss": 0.4800090193748474, | |
| "eval_runtime": 215.1825, | |
| "eval_samples_per_second": 23.046, | |
| "eval_steps_per_second": 2.881, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 30.851063829787233, | |
| "eval_cer": 0.20605283626542426, | |
| "eval_loss": 0.5290209054946899, | |
| "eval_runtime": 213.4725, | |
| "eval_samples_per_second": 23.23, | |
| "eval_steps_per_second": 2.904, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 31.117021276595743, | |
| "eval_cer": 0.20547045695278818, | |
| "eval_loss": 0.5368531942367554, | |
| "eval_runtime": 214.9789, | |
| "eval_samples_per_second": 23.067, | |
| "eval_steps_per_second": 2.884, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 31.382978723404257, | |
| "eval_cer": 0.2040555488241019, | |
| "eval_loss": 0.5169620513916016, | |
| "eval_runtime": 214.7111, | |
| "eval_samples_per_second": 23.096, | |
| "eval_steps_per_second": 2.888, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 31.648936170212767, | |
| "eval_cer": 0.20704170852110831, | |
| "eval_loss": 0.5228501558303833, | |
| "eval_runtime": 216.5107, | |
| "eval_samples_per_second": 22.904, | |
| "eval_steps_per_second": 2.864, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 31.914893617021278, | |
| "grad_norm": 4.3140645027160645, | |
| "learning_rate": 2.87904e-05, | |
| "loss": 0.2007, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 31.914893617021278, | |
| "eval_cer": 0.20597857313159817, | |
| "eval_loss": 0.5034534931182861, | |
| "eval_runtime": 215.0115, | |
| "eval_samples_per_second": 23.064, | |
| "eval_steps_per_second": 2.884, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 32.180851063829785, | |
| "eval_cer": 0.20488807764015213, | |
| "eval_loss": 0.5103185772895813, | |
| "eval_runtime": 214.9574, | |
| "eval_samples_per_second": 23.07, | |
| "eval_steps_per_second": 2.884, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 32.4468085106383, | |
| "eval_cer": 0.2032073856640883, | |
| "eval_loss": 0.48684972524642944, | |
| "eval_runtime": 215.7924, | |
| "eval_samples_per_second": 22.98, | |
| "eval_steps_per_second": 2.873, | |
| "eval_wer": 0.9971768501714056, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 32.712765957446805, | |
| "eval_cer": 0.2042666124676076, | |
| "eval_loss": 0.4866673946380615, | |
| "eval_runtime": 215.5038, | |
| "eval_samples_per_second": 23.011, | |
| "eval_steps_per_second": 2.877, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 32.97872340425532, | |
| "eval_cer": 0.20395001700234905, | |
| "eval_loss": 0.5048738718032837, | |
| "eval_runtime": 220.2273, | |
| "eval_samples_per_second": 22.518, | |
| "eval_steps_per_second": 2.815, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 33.244680851063826, | |
| "grad_norm": 3.023064613342285, | |
| "learning_rate": 2.99904e-05, | |
| "loss": 0.1867, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 33.244680851063826, | |
| "eval_cer": 0.20403600589414767, | |
| "eval_loss": 0.5125642418861389, | |
| "eval_runtime": 217.1156, | |
| "eval_samples_per_second": 22.84, | |
| "eval_steps_per_second": 2.856, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 33.51063829787234, | |
| "eval_cer": 0.20373895335884337, | |
| "eval_loss": 0.5321324467658997, | |
| "eval_runtime": 218.7742, | |
| "eval_samples_per_second": 22.667, | |
| "eval_steps_per_second": 2.834, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 33.776595744680854, | |
| "eval_cer": 0.20402428013617513, | |
| "eval_loss": 0.5186927914619446, | |
| "eval_runtime": 216.188, | |
| "eval_samples_per_second": 22.938, | |
| "eval_steps_per_second": 2.868, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 34.04255319148936, | |
| "eval_cer": 0.20636943173068278, | |
| "eval_loss": 0.5318589806556702, | |
| "eval_runtime": 217.8562, | |
| "eval_samples_per_second": 22.763, | |
| "eval_steps_per_second": 2.846, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 34.308510638297875, | |
| "eval_cer": 0.2040711831680653, | |
| "eval_loss": 0.5274991989135742, | |
| "eval_runtime": 218.7876, | |
| "eval_samples_per_second": 22.666, | |
| "eval_steps_per_second": 2.834, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 34.57446808510638, | |
| "grad_norm": 4.690342426300049, | |
| "learning_rate": 2.7264773627685888e-05, | |
| "loss": 0.1749, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 34.57446808510638, | |
| "eval_cer": 0.20429788115553435, | |
| "eval_loss": 0.543329656124115, | |
| "eval_runtime": 215.6846, | |
| "eval_samples_per_second": 22.992, | |
| "eval_steps_per_second": 2.875, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 34.840425531914896, | |
| "eval_cer": 0.2023240452301571, | |
| "eval_loss": 0.5093534588813782, | |
| "eval_runtime": 216.6521, | |
| "eval_samples_per_second": 22.889, | |
| "eval_steps_per_second": 2.862, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 35.1063829787234, | |
| "eval_cer": 0.20042838102459673, | |
| "eval_loss": 0.5363013744354248, | |
| "eval_runtime": 216.3501, | |
| "eval_samples_per_second": 22.921, | |
| "eval_steps_per_second": 2.866, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 35.37234042553192, | |
| "eval_cer": 0.20222633058038594, | |
| "eval_loss": 0.5330685377120972, | |
| "eval_runtime": 215.7086, | |
| "eval_samples_per_second": 22.989, | |
| "eval_steps_per_second": 2.874, | |
| "eval_wer": 0.999395039322444, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 35.638297872340424, | |
| "eval_cer": 0.20093258861741586, | |
| "eval_loss": 0.5052955746650696, | |
| "eval_runtime": 213.6416, | |
| "eval_samples_per_second": 23.212, | |
| "eval_steps_per_second": 2.902, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 35.90425531914894, | |
| "grad_norm": 3.094069242477417, | |
| "learning_rate": 1.9986694366624682e-05, | |
| "loss": 0.1604, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 35.90425531914894, | |
| "eval_cer": 0.2026210977654614, | |
| "eval_loss": 0.5156988501548767, | |
| "eval_runtime": 221.1798, | |
| "eval_samples_per_second": 22.421, | |
| "eval_steps_per_second": 2.803, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 36.170212765957444, | |
| "eval_cer": 0.20183938056729217, | |
| "eval_loss": 0.5299074649810791, | |
| "eval_runtime": 215.0871, | |
| "eval_samples_per_second": 23.056, | |
| "eval_steps_per_second": 2.883, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 36.43617021276596, | |
| "eval_cer": 0.20501706097785005, | |
| "eval_loss": 0.5117177963256836, | |
| "eval_runtime": 214.4648, | |
| "eval_samples_per_second": 23.123, | |
| "eval_steps_per_second": 2.891, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 36.702127659574465, | |
| "eval_cer": 0.2038327594226237, | |
| "eval_loss": 0.5067043900489807, | |
| "eval_runtime": 213.8511, | |
| "eval_samples_per_second": 23.189, | |
| "eval_steps_per_second": 2.899, | |
| "eval_wer": 0.999395039322444, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 36.96808510638298, | |
| "eval_cer": 0.20278525837707692, | |
| "eval_loss": 0.4994255602359772, | |
| "eval_runtime": 216.1066, | |
| "eval_samples_per_second": 22.947, | |
| "eval_steps_per_second": 2.869, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 37.234042553191486, | |
| "grad_norm": 2.801684617996216, | |
| "learning_rate": 1.089715454232877e-05, | |
| "loss": 0.1412, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 37.234042553191486, | |
| "eval_cer": 0.20241394270794655, | |
| "eval_loss": 0.5346146821975708, | |
| "eval_runtime": 215.4888, | |
| "eval_samples_per_second": 23.013, | |
| "eval_steps_per_second": 2.877, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "eval_cer": 0.20152669368802448, | |
| "eval_loss": 0.5349961519241333, | |
| "eval_runtime": 213.0582, | |
| "eval_samples_per_second": 23.275, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 0.999395039322444, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 37.765957446808514, | |
| "eval_cer": 0.20097949164930604, | |
| "eval_loss": 0.5236727595329285, | |
| "eval_runtime": 211.9375, | |
| "eval_samples_per_second": 23.398, | |
| "eval_steps_per_second": 2.925, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 38.03191489361702, | |
| "eval_cer": 0.19928316532927884, | |
| "eval_loss": 0.5304979085922241, | |
| "eval_runtime": 212.7349, | |
| "eval_samples_per_second": 23.311, | |
| "eval_steps_per_second": 2.914, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 38.297872340425535, | |
| "eval_cer": 0.19734059809182833, | |
| "eval_loss": 0.5309327244758606, | |
| "eval_runtime": 214.8976, | |
| "eval_samples_per_second": 23.076, | |
| "eval_steps_per_second": 2.885, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 38.56382978723404, | |
| "grad_norm": 1.7099072933197021, | |
| "learning_rate": 3.292383984744315e-06, | |
| "loss": 0.1286, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 38.56382978723404, | |
| "eval_cer": 0.19924407946937037, | |
| "eval_loss": 0.5269559621810913, | |
| "eval_runtime": 214.5389, | |
| "eval_samples_per_second": 23.115, | |
| "eval_steps_per_second": 2.89, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 38.829787234042556, | |
| "eval_cer": 0.19992026484578673, | |
| "eval_loss": 0.5362734198570251, | |
| "eval_runtime": 221.7091, | |
| "eval_samples_per_second": 22.367, | |
| "eval_steps_per_second": 2.796, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 39.09574468085106, | |
| "eval_cer": 0.19989681332984166, | |
| "eval_loss": 0.5347363352775574, | |
| "eval_runtime": 213.4871, | |
| "eval_samples_per_second": 23.229, | |
| "eval_steps_per_second": 2.904, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 39.361702127659576, | |
| "eval_cer": 0.19991244767380506, | |
| "eval_loss": 0.5318508744239807, | |
| "eval_runtime": 214.7174, | |
| "eval_samples_per_second": 23.095, | |
| "eval_steps_per_second": 2.888, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 39.62765957446808, | |
| "eval_cer": 0.1994590516988669, | |
| "eval_loss": 0.5322226881980896, | |
| "eval_runtime": 213.8602, | |
| "eval_samples_per_second": 23.188, | |
| "eval_steps_per_second": 2.899, | |
| "eval_wer": 0.999395039322444, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 39.8936170212766, | |
| "grad_norm": 2.6511354446411133, | |
| "learning_rate": 2.4271246633469913e-08, | |
| "loss": 0.1217, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 39.8936170212766, | |
| "eval_cer": 0.19924017088337953, | |
| "eval_loss": 0.5321789383888245, | |
| "eval_runtime": 215.7329, | |
| "eval_samples_per_second": 22.987, | |
| "eval_steps_per_second": 2.874, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 15040, | |
| "total_flos": 2.1018408189762245e+19, | |
| "train_loss": 1.5314149432993949, | |
| "train_runtime": 82631.5291, | |
| "train_samples_per_second": 5.824, | |
| "train_steps_per_second": 0.182 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 15040, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "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": 2.1018408189762245e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |