| { | |
| "best_metric": 0.3024933338165283, | |
| "best_model_checkpoint": "./Hubert-common_voice-ja-demo-roma-cosine-3e-4/checkpoint-4400", | |
| "epoch": 50.0, | |
| "eval_steps": 100, | |
| "global_step": 18800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26595744680851063, | |
| "eval_cer": 1.6163371077245385, | |
| "eval_loss": 15.823559761047363, | |
| "eval_runtime": 210.3767, | |
| "eval_samples_per_second": 23.572, | |
| "eval_steps_per_second": 2.947, | |
| "eval_wer": 1.792498487598306, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 6.4206719398498535, | |
| "eval_runtime": 221.1891, | |
| "eval_samples_per_second": 22.42, | |
| "eval_steps_per_second": 2.803, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 5.419256210327148, | |
| "eval_runtime": 241.9799, | |
| "eval_samples_per_second": 20.493, | |
| "eval_steps_per_second": 2.562, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.907326698303223, | |
| "eval_runtime": 261.1648, | |
| "eval_samples_per_second": 18.988, | |
| "eval_steps_per_second": 2.374, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 20.339691162109375, | |
| "learning_rate": 1.1903999999999998e-05, | |
| "loss": 7.456, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 4.358646869659424, | |
| "eval_runtime": 256.9257, | |
| "eval_samples_per_second": 19.301, | |
| "eval_steps_per_second": 2.413, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.8341214656829834, | |
| "eval_runtime": 284.7473, | |
| "eval_samples_per_second": 17.415, | |
| "eval_steps_per_second": 2.177, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.4148387908935547, | |
| "eval_runtime": 301.2939, | |
| "eval_samples_per_second": 16.459, | |
| "eval_steps_per_second": 2.058, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 3.104691982269287, | |
| "eval_runtime": 236.1726, | |
| "eval_samples_per_second": 20.997, | |
| "eval_steps_per_second": 2.625, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.961205244064331, | |
| "eval_runtime": 231.5216, | |
| "eval_samples_per_second": 21.419, | |
| "eval_steps_per_second": 2.678, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 1.9079935550689697, | |
| "learning_rate": 2.3903999999999997e-05, | |
| "loss": 3.2563, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.902658700942993, | |
| "eval_runtime": 265.314, | |
| "eval_samples_per_second": 18.691, | |
| "eval_steps_per_second": 2.337, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.876744031906128, | |
| "eval_runtime": 268.5469, | |
| "eval_samples_per_second": 18.466, | |
| "eval_steps_per_second": 2.309, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.8588435649871826, | |
| "eval_runtime": 250.7138, | |
| "eval_samples_per_second": 19.78, | |
| "eval_steps_per_second": 2.473, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "eval_cer": 0.9276012616915579, | |
| "eval_loss": 2.803792953491211, | |
| "eval_runtime": 253.4886, | |
| "eval_samples_per_second": 19.563, | |
| "eval_steps_per_second": 2.446, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "eval_cer": 0.8359292858622536, | |
| "eval_loss": 1.9932550191879272, | |
| "eval_runtime": 266.0365, | |
| "eval_samples_per_second": 18.64, | |
| "eval_steps_per_second": 2.331, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 4.3663716316223145, | |
| "learning_rate": 3.5903999999999994e-05, | |
| "loss": 2.5032, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_cer": 0.38112231138141156, | |
| "eval_loss": 1.1380858421325684, | |
| "eval_runtime": 256.0779, | |
| "eval_samples_per_second": 19.365, | |
| "eval_steps_per_second": 2.421, | |
| "eval_wer": 0.9997983464408147, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "eval_cer": 0.2746446118187823, | |
| "eval_loss": 0.8252492547035217, | |
| "eval_runtime": 240.6394, | |
| "eval_samples_per_second": 20.608, | |
| "eval_steps_per_second": 2.576, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.5212765957446805, | |
| "eval_cer": 0.2507318827267859, | |
| "eval_loss": 0.6762787103652954, | |
| "eval_runtime": 212.2043, | |
| "eval_samples_per_second": 23.369, | |
| "eval_steps_per_second": 2.922, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.787234042553192, | |
| "eval_cer": 0.24174213494783992, | |
| "eval_loss": 0.6010992527008057, | |
| "eval_runtime": 212.3652, | |
| "eval_samples_per_second": 23.351, | |
| "eval_steps_per_second": 2.919, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.053191489361702, | |
| "eval_cer": 0.23079418558748002, | |
| "eval_loss": 0.5514463782310486, | |
| "eval_runtime": 209.5514, | |
| "eval_samples_per_second": 23.665, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "grad_norm": 2.237888813018799, | |
| "learning_rate": 4.7903999999999994e-05, | |
| "loss": 0.678, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "eval_cer": 0.22485313488139397, | |
| "eval_loss": 0.5128617882728577, | |
| "eval_runtime": 210.7223, | |
| "eval_samples_per_second": 23.533, | |
| "eval_steps_per_second": 2.942, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.585106382978723, | |
| "eval_cer": 0.23253350635340653, | |
| "eval_loss": 0.4957906901836395, | |
| "eval_runtime": 209.1903, | |
| "eval_samples_per_second": 23.706, | |
| "eval_steps_per_second": 2.964, | |
| "eval_wer": 0.994353700342811, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.851063829787234, | |
| "eval_cer": 0.21945146904204466, | |
| "eval_loss": 0.47169229388237, | |
| "eval_runtime": 208.7444, | |
| "eval_samples_per_second": 23.756, | |
| "eval_steps_per_second": 2.97, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.117021276595745, | |
| "eval_cer": 0.217309563919061, | |
| "eval_loss": 0.4651758670806885, | |
| "eval_runtime": 239.8374, | |
| "eval_samples_per_second": 20.677, | |
| "eval_steps_per_second": 2.585, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.382978723404255, | |
| "eval_cer": 0.21218149909907094, | |
| "eval_loss": 0.45553550124168396, | |
| "eval_runtime": 209.0918, | |
| "eval_samples_per_second": 23.717, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "grad_norm": 3.0910284519195557, | |
| "learning_rate": 5.9903999999999994e-05, | |
| "loss": 0.4238, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "eval_cer": 0.21164211423233417, | |
| "eval_loss": 0.4360628128051758, | |
| "eval_runtime": 211.1218, | |
| "eval_samples_per_second": 23.489, | |
| "eval_steps_per_second": 2.937, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.914893617021277, | |
| "eval_cer": 0.20628344283888417, | |
| "eval_loss": 0.4180741608142853, | |
| "eval_runtime": 210.3304, | |
| "eval_samples_per_second": 23.577, | |
| "eval_steps_per_second": 2.948, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.180851063829787, | |
| "eval_cer": 0.2011983724647934, | |
| "eval_loss": 0.4185584485530853, | |
| "eval_runtime": 209.9592, | |
| "eval_samples_per_second": 23.619, | |
| "eval_steps_per_second": 2.953, | |
| "eval_wer": 0.9957652752571083, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.446808510638298, | |
| "eval_cer": 0.20350052961340176, | |
| "eval_loss": 0.4253988265991211, | |
| "eval_runtime": 212.3471, | |
| "eval_samples_per_second": 23.353, | |
| "eval_steps_per_second": 2.92, | |
| "eval_wer": 0.999395039322444, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.712765957446808, | |
| "eval_cer": 0.1980519607421623, | |
| "eval_loss": 0.4013788104057312, | |
| "eval_runtime": 213.4065, | |
| "eval_samples_per_second": 23.237, | |
| "eval_steps_per_second": 2.905, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "grad_norm": 2.3959779739379883, | |
| "learning_rate": 7.1904e-05, | |
| "loss": 0.3375, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "eval_cer": 0.19799333195229962, | |
| "eval_loss": 0.38772091269493103, | |
| "eval_runtime": 212.9922, | |
| "eval_samples_per_second": 23.283, | |
| "eval_steps_per_second": 2.911, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.24468085106383, | |
| "eval_cer": 0.19256821459700524, | |
| "eval_loss": 0.3868231475353241, | |
| "eval_runtime": 211.4945, | |
| "eval_samples_per_second": 23.447, | |
| "eval_steps_per_second": 2.932, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.51063829787234, | |
| "eval_cer": 0.19027778320636943, | |
| "eval_loss": 0.374016672372818, | |
| "eval_runtime": 207.4949, | |
| "eval_samples_per_second": 23.899, | |
| "eval_steps_per_second": 2.988, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.77659574468085, | |
| "eval_cer": 0.18545067950767452, | |
| "eval_loss": 0.36447063088417053, | |
| "eval_runtime": 208.0584, | |
| "eval_samples_per_second": 23.835, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.042553191489361, | |
| "eval_cer": 0.18235117081693356, | |
| "eval_loss": 0.35864943265914917, | |
| "eval_runtime": 233.9479, | |
| "eval_samples_per_second": 21.197, | |
| "eval_steps_per_second": 2.65, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "grad_norm": 1.6243752241134644, | |
| "learning_rate": 8.390399999999999e-05, | |
| "loss": 0.2553, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "eval_cer": 0.1774537125704034, | |
| "eval_loss": 0.3328181505203247, | |
| "eval_runtime": 208.1328, | |
| "eval_samples_per_second": 23.826, | |
| "eval_steps_per_second": 2.979, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.574468085106384, | |
| "eval_cer": 0.174365929637635, | |
| "eval_loss": 0.34012970328330994, | |
| "eval_runtime": 208.9838, | |
| "eval_samples_per_second": 23.729, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.840425531914894, | |
| "eval_cer": 0.1727282321074705, | |
| "eval_loss": 0.31242451071739197, | |
| "eval_runtime": 209.0076, | |
| "eval_samples_per_second": 23.726, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.106382978723405, | |
| "eval_cer": 0.1708716537618186, | |
| "eval_loss": 0.322468101978302, | |
| "eval_runtime": 207.4311, | |
| "eval_samples_per_second": 23.907, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 10.372340425531915, | |
| "eval_cer": 0.17599971858180866, | |
| "eval_loss": 0.3310896158218384, | |
| "eval_runtime": 206.3784, | |
| "eval_samples_per_second": 24.029, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "grad_norm": 2.6917850971221924, | |
| "learning_rate": 9.5904e-05, | |
| "loss": 0.2035, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "eval_cer": 0.17048861233471568, | |
| "eval_loss": 0.3098428249359131, | |
| "eval_runtime": 207.2374, | |
| "eval_samples_per_second": 23.929, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.904255319148936, | |
| "eval_cer": 0.1713641355966652, | |
| "eval_loss": 0.32444971799850464, | |
| "eval_runtime": 207.2878, | |
| "eval_samples_per_second": 23.923, | |
| "eval_steps_per_second": 2.991, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.170212765957446, | |
| "eval_cer": 0.16859294812915532, | |
| "eval_loss": 0.3280084431171417, | |
| "eval_runtime": 206.6779, | |
| "eval_samples_per_second": 23.994, | |
| "eval_steps_per_second": 3.0, | |
| "eval_wer": 0.9925388183101431, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 11.436170212765958, | |
| "eval_cer": 0.17051206385066076, | |
| "eval_loss": 0.31342190504074097, | |
| "eval_runtime": 214.4373, | |
| "eval_samples_per_second": 23.126, | |
| "eval_steps_per_second": 2.891, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.702127659574469, | |
| "eval_cer": 0.16671291826755835, | |
| "eval_loss": 0.3024933338165283, | |
| "eval_runtime": 208.7095, | |
| "eval_samples_per_second": 23.76, | |
| "eval_steps_per_second": 2.971, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "grad_norm": 1.652050256729126, | |
| "learning_rate": 0.00010787999999999998, | |
| "loss": 0.1772, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "eval_cer": 0.16904634410409347, | |
| "eval_loss": 0.31564900279045105, | |
| "eval_runtime": 207.6579, | |
| "eval_samples_per_second": 23.881, | |
| "eval_steps_per_second": 2.986, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.23404255319149, | |
| "eval_cer": 0.16571232025390176, | |
| "eval_loss": 0.3212853670120239, | |
| "eval_runtime": 208.1891, | |
| "eval_samples_per_second": 23.82, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 0.9967735430530349, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_cer": 0.17018374262742966, | |
| "eval_loss": 0.31842583417892456, | |
| "eval_runtime": 207.4188, | |
| "eval_samples_per_second": 23.908, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.76595744680851, | |
| "eval_cer": 0.16590774955344406, | |
| "eval_loss": 0.3348098397254944, | |
| "eval_runtime": 206.5664, | |
| "eval_samples_per_second": 24.007, | |
| "eval_steps_per_second": 3.001, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 13.03191489361702, | |
| "eval_cer": 0.16545044499251504, | |
| "eval_loss": 0.31754830479621887, | |
| "eval_runtime": 204.6212, | |
| "eval_samples_per_second": 24.235, | |
| "eval_steps_per_second": 3.03, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "grad_norm": 1.4975948333740234, | |
| "learning_rate": 0.00011988, | |
| "loss": 0.1542, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "eval_cer": 0.16795975719863826, | |
| "eval_loss": 0.3414091467857361, | |
| "eval_runtime": 212.8965, | |
| "eval_samples_per_second": 23.293, | |
| "eval_steps_per_second": 2.912, | |
| "eval_wer": 0.9997983464408147, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.563829787234042, | |
| "eval_cer": 0.17008993656364937, | |
| "eval_loss": 0.3142850399017334, | |
| "eval_runtime": 206.8293, | |
| "eval_samples_per_second": 23.976, | |
| "eval_steps_per_second": 2.998, | |
| "eval_wer": 0.999395039322444, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.829787234042554, | |
| "eval_cer": 0.16879228601468846, | |
| "eval_loss": 0.3204251229763031, | |
| "eval_runtime": 205.7182, | |
| "eval_samples_per_second": 24.106, | |
| "eval_steps_per_second": 3.014, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 14.095744680851064, | |
| "eval_cer": 0.16618525915879412, | |
| "eval_loss": 0.35486334562301636, | |
| "eval_runtime": 206.3465, | |
| "eval_samples_per_second": 24.032, | |
| "eval_steps_per_second": 3.005, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 14.361702127659575, | |
| "eval_cer": 0.1666073864458055, | |
| "eval_loss": 0.40914931893348694, | |
| "eval_runtime": 206.0512, | |
| "eval_samples_per_second": 24.067, | |
| "eval_steps_per_second": 3.009, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "grad_norm": 2.959421157836914, | |
| "learning_rate": 0.000131856, | |
| "loss": 0.1449, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "eval_cer": 0.16760407587347126, | |
| "eval_loss": 0.3907996714115143, | |
| "eval_runtime": 205.2139, | |
| "eval_samples_per_second": 24.165, | |
| "eval_steps_per_second": 3.021, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.893617021276595, | |
| "eval_cer": 0.1662243450187026, | |
| "eval_loss": 0.3705514073371887, | |
| "eval_runtime": 205.3677, | |
| "eval_samples_per_second": 24.147, | |
| "eval_steps_per_second": 3.019, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 15.159574468085106, | |
| "eval_cer": 0.164105891411664, | |
| "eval_loss": 0.3971535861492157, | |
| "eval_runtime": 205.6896, | |
| "eval_samples_per_second": 24.109, | |
| "eval_steps_per_second": 3.014, | |
| "eval_wer": 0.9971768501714056, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 15.425531914893616, | |
| "eval_cer": 0.1653097358968446, | |
| "eval_loss": 0.3461838662624359, | |
| "eval_runtime": 206.3509, | |
| "eval_samples_per_second": 24.032, | |
| "eval_steps_per_second": 3.005, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 15.691489361702128, | |
| "eval_cer": 0.16992186736604298, | |
| "eval_loss": 0.35435542464256287, | |
| "eval_runtime": 205.3227, | |
| "eval_samples_per_second": 24.152, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 15.957446808510639, | |
| "grad_norm": 1.915339469909668, | |
| "learning_rate": 0.00014385599999999999, | |
| "loss": 0.1396, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.957446808510639, | |
| "eval_cer": 0.1681590950841714, | |
| "eval_loss": 0.33965829014778137, | |
| "eval_runtime": 205.3964, | |
| "eval_samples_per_second": 24.144, | |
| "eval_steps_per_second": 3.019, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.22340425531915, | |
| "eval_cer": 0.16804574609043688, | |
| "eval_loss": 0.3451753556728363, | |
| "eval_runtime": 205.8971, | |
| "eval_samples_per_second": 24.085, | |
| "eval_steps_per_second": 3.011, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 16.48936170212766, | |
| "eval_cer": 0.16652921472598858, | |
| "eval_loss": 0.3533567488193512, | |
| "eval_runtime": 205.4803, | |
| "eval_samples_per_second": 24.134, | |
| "eval_steps_per_second": 3.017, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 16.75531914893617, | |
| "eval_cer": 0.17028145727720082, | |
| "eval_loss": 0.3502165377140045, | |
| "eval_runtime": 205.7767, | |
| "eval_samples_per_second": 24.099, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 17.02127659574468, | |
| "eval_cer": 0.17007430221968597, | |
| "eval_loss": 0.34754782915115356, | |
| "eval_runtime": 205.7613, | |
| "eval_samples_per_second": 24.101, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 17.28723404255319, | |
| "grad_norm": 1.2938112020492554, | |
| "learning_rate": 0.00015585599999999998, | |
| "loss": 0.1293, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.28723404255319, | |
| "eval_cer": 0.1681239178102538, | |
| "eval_loss": 0.3350105881690979, | |
| "eval_runtime": 206.0245, | |
| "eval_samples_per_second": 24.07, | |
| "eval_steps_per_second": 3.009, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.5531914893617, | |
| "eval_cer": 0.16944501987515975, | |
| "eval_loss": 0.3442031443119049, | |
| "eval_runtime": 205.3927, | |
| "eval_samples_per_second": 24.144, | |
| "eval_steps_per_second": 3.019, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 17.819148936170212, | |
| "eval_cer": 0.16870629712288984, | |
| "eval_loss": 0.33415013551712036, | |
| "eval_runtime": 205.9912, | |
| "eval_samples_per_second": 24.074, | |
| "eval_steps_per_second": 3.01, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 18.085106382978722, | |
| "eval_cer": 0.1696013633147936, | |
| "eval_loss": 0.3668520450592041, | |
| "eval_runtime": 205.5898, | |
| "eval_samples_per_second": 24.121, | |
| "eval_steps_per_second": 3.016, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 18.351063829787233, | |
| "eval_cer": 0.16906197844805684, | |
| "eval_loss": 0.34041330218315125, | |
| "eval_runtime": 205.1961, | |
| "eval_samples_per_second": 24.167, | |
| "eval_steps_per_second": 3.022, | |
| "eval_wer": 0.9969751966122202, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 18.617021276595743, | |
| "grad_norm": 2.553234815597534, | |
| "learning_rate": 0.000167856, | |
| "loss": 0.1276, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.617021276595743, | |
| "eval_cer": 0.16789331123679385, | |
| "eval_loss": 0.3463904857635498, | |
| "eval_runtime": 204.7479, | |
| "eval_samples_per_second": 24.22, | |
| "eval_steps_per_second": 3.028, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.882978723404257, | |
| "eval_cer": 0.16948801432105906, | |
| "eval_loss": 0.3496261537075043, | |
| "eval_runtime": 205.3143, | |
| "eval_samples_per_second": 24.153, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 19.148936170212767, | |
| "eval_cer": 0.16978897544235422, | |
| "eval_loss": 0.3436301350593567, | |
| "eval_runtime": 205.1734, | |
| "eval_samples_per_second": 24.17, | |
| "eval_steps_per_second": 3.022, | |
| "eval_wer": 0.9967735430530349, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 19.414893617021278, | |
| "eval_cer": 0.16896035521229485, | |
| "eval_loss": 0.36047497391700745, | |
| "eval_runtime": 205.4789, | |
| "eval_samples_per_second": 24.134, | |
| "eval_steps_per_second": 3.017, | |
| "eval_wer": 0.9953619681387377, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 19.680851063829788, | |
| "eval_cer": 0.16866330267699053, | |
| "eval_loss": 0.3582172989845276, | |
| "eval_runtime": 205.1102, | |
| "eval_samples_per_second": 24.177, | |
| "eval_steps_per_second": 3.023, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 19.9468085106383, | |
| "grad_norm": 1.7617504596710205, | |
| "learning_rate": 0.000179856, | |
| "loss": 0.1264, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.9468085106383, | |
| "eval_cer": 0.17402197407044054, | |
| "eval_loss": 0.3575660288333893, | |
| "eval_runtime": 204.7059, | |
| "eval_samples_per_second": 24.225, | |
| "eval_steps_per_second": 3.029, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.21276595744681, | |
| "eval_cer": 0.17258752301180003, | |
| "eval_loss": 0.3668830990791321, | |
| "eval_runtime": 205.0195, | |
| "eval_samples_per_second": 24.188, | |
| "eval_steps_per_second": 3.024, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 20.47872340425532, | |
| "eval_cer": 0.17064495577434952, | |
| "eval_loss": 0.361765593290329, | |
| "eval_runtime": 204.8268, | |
| "eval_samples_per_second": 24.211, | |
| "eval_steps_per_second": 3.027, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 20.74468085106383, | |
| "eval_cer": 0.17463562207100336, | |
| "eval_loss": 0.34747639298439026, | |
| "eval_runtime": 204.8046, | |
| "eval_samples_per_second": 24.213, | |
| "eval_steps_per_second": 3.027, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 21.01063829787234, | |
| "eval_cer": 0.1715165704503082, | |
| "eval_loss": 0.34254932403564453, | |
| "eval_runtime": 204.6469, | |
| "eval_samples_per_second": 24.232, | |
| "eval_steps_per_second": 3.03, | |
| "eval_wer": 0.9971768501714056, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 21.27659574468085, | |
| "grad_norm": 1.5520191192626953, | |
| "learning_rate": 0.000191856, | |
| "loss": 0.1219, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.27659574468085, | |
| "eval_cer": 0.1716377366160244, | |
| "eval_loss": 0.3685350716114044, | |
| "eval_runtime": 204.5671, | |
| "eval_samples_per_second": 24.241, | |
| "eval_steps_per_second": 3.031, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.54255319148936, | |
| "eval_cer": 0.1755893170527698, | |
| "eval_loss": 0.38032132387161255, | |
| "eval_runtime": 204.7288, | |
| "eval_samples_per_second": 24.222, | |
| "eval_steps_per_second": 3.028, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 21.80851063829787, | |
| "eval_cer": 0.17974805254703005, | |
| "eval_loss": 0.36628487706184387, | |
| "eval_runtime": 205.2564, | |
| "eval_samples_per_second": 24.16, | |
| "eval_steps_per_second": 3.021, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 22.074468085106382, | |
| "eval_cer": 0.1709576426536172, | |
| "eval_loss": 0.3641684055328369, | |
| "eval_runtime": 205.4639, | |
| "eval_samples_per_second": 24.136, | |
| "eval_steps_per_second": 3.018, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 22.340425531914892, | |
| "eval_cer": 0.17367801850324607, | |
| "eval_loss": 0.38052114844322205, | |
| "eval_runtime": 205.4218, | |
| "eval_samples_per_second": 24.141, | |
| "eval_steps_per_second": 3.018, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 22.606382978723403, | |
| "grad_norm": 2.203418493270874, | |
| "learning_rate": 0.00020385599999999998, | |
| "loss": 0.1177, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.606382978723403, | |
| "eval_cer": 0.17468252510289353, | |
| "eval_loss": 0.36295458674430847, | |
| "eval_runtime": 205.462, | |
| "eval_samples_per_second": 24.136, | |
| "eval_steps_per_second": 3.018, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.872340425531917, | |
| "eval_cer": 0.1753078988614289, | |
| "eval_loss": 0.4001345634460449, | |
| "eval_runtime": 205.3238, | |
| "eval_samples_per_second": 24.152, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 23.138297872340427, | |
| "eval_cer": 0.17578865493830298, | |
| "eval_loss": 0.375848650932312, | |
| "eval_runtime": 205.2532, | |
| "eval_samples_per_second": 24.16, | |
| "eval_steps_per_second": 3.021, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 23.404255319148938, | |
| "eval_cer": 0.17470988520482944, | |
| "eval_loss": 0.37709999084472656, | |
| "eval_runtime": 205.004, | |
| "eval_samples_per_second": 24.19, | |
| "eval_steps_per_second": 3.024, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 23.670212765957448, | |
| "eval_cer": 0.17936501111992714, | |
| "eval_loss": 0.4001403748989105, | |
| "eval_runtime": 204.8778, | |
| "eval_samples_per_second": 24.205, | |
| "eval_steps_per_second": 3.026, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 23.93617021276596, | |
| "grad_norm": 1.3760440349578857, | |
| "learning_rate": 0.000215856, | |
| "loss": 0.1241, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 23.93617021276596, | |
| "eval_cer": 0.17689087618772156, | |
| "eval_loss": 0.3929387629032135, | |
| "eval_runtime": 205.6986, | |
| "eval_samples_per_second": 24.108, | |
| "eval_steps_per_second": 3.014, | |
| "eval_wer": 0.9997983464408147, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.20212765957447, | |
| "eval_cer": 0.17517500693774013, | |
| "eval_loss": 0.3732432425022125, | |
| "eval_runtime": 205.6021, | |
| "eval_samples_per_second": 24.119, | |
| "eval_steps_per_second": 3.016, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 24.46808510638298, | |
| "eval_cer": 0.17383045335688907, | |
| "eval_loss": 0.3813112676143646, | |
| "eval_runtime": 205.3982, | |
| "eval_samples_per_second": 24.143, | |
| "eval_steps_per_second": 3.019, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 24.73404255319149, | |
| "eval_cer": 0.17939627980785391, | |
| "eval_loss": 0.41280829906463623, | |
| "eval_runtime": 219.7977, | |
| "eval_samples_per_second": 22.562, | |
| "eval_steps_per_second": 2.821, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.17513592107783169, | |
| "eval_loss": 0.3756047785282135, | |
| "eval_runtime": 219.9345, | |
| "eval_samples_per_second": 22.548, | |
| "eval_steps_per_second": 2.819, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 25.26595744680851, | |
| "grad_norm": 9.107454299926758, | |
| "learning_rate": 0.00022783199999999997, | |
| "loss": 0.121, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.26595744680851, | |
| "eval_cer": 0.17971678385910328, | |
| "eval_loss": 0.39162904024124146, | |
| "eval_runtime": 219.1957, | |
| "eval_samples_per_second": 22.624, | |
| "eval_steps_per_second": 2.829, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.53191489361702, | |
| "eval_cer": 0.18236680516089696, | |
| "eval_loss": 0.38820743560791016, | |
| "eval_runtime": 219.4833, | |
| "eval_samples_per_second": 22.594, | |
| "eval_steps_per_second": 2.825, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 25.79787234042553, | |
| "eval_cer": 0.18378562187557407, | |
| "eval_loss": 0.39170223474502563, | |
| "eval_runtime": 219.5415, | |
| "eval_samples_per_second": 22.588, | |
| "eval_steps_per_second": 2.824, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 26.06382978723404, | |
| "eval_cer": 0.17702376811141032, | |
| "eval_loss": 0.3927652835845947, | |
| "eval_runtime": 219.475, | |
| "eval_samples_per_second": 22.595, | |
| "eval_steps_per_second": 2.825, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 26.329787234042552, | |
| "eval_cer": 0.18011155104417875, | |
| "eval_loss": 0.3928898870944977, | |
| "eval_runtime": 219.0682, | |
| "eval_samples_per_second": 22.637, | |
| "eval_steps_per_second": 2.83, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 26.595744680851062, | |
| "grad_norm": 1.6058417558670044, | |
| "learning_rate": 0.000239832, | |
| "loss": 0.1206, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.595744680851062, | |
| "eval_cer": 0.17812598936082893, | |
| "eval_loss": 0.39845964312553406, | |
| "eval_runtime": 219.0245, | |
| "eval_samples_per_second": 22.641, | |
| "eval_steps_per_second": 2.831, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.861702127659573, | |
| "eval_cer": 0.1770628539713188, | |
| "eval_loss": 0.3799264132976532, | |
| "eval_runtime": 219.8342, | |
| "eval_samples_per_second": 22.558, | |
| "eval_steps_per_second": 2.82, | |
| "eval_wer": 0.999395039322444, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 27.127659574468087, | |
| "eval_cer": 0.1785715681637854, | |
| "eval_loss": 0.4022609293460846, | |
| "eval_runtime": 221.0183, | |
| "eval_samples_per_second": 22.437, | |
| "eval_steps_per_second": 2.805, | |
| "eval_wer": 0.999395039322444, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 27.393617021276597, | |
| "eval_cer": 0.17838395603622478, | |
| "eval_loss": 0.4000260531902313, | |
| "eval_runtime": 218.5093, | |
| "eval_samples_per_second": 22.695, | |
| "eval_steps_per_second": 2.837, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 27.659574468085108, | |
| "eval_cer": 0.18251142284255825, | |
| "eval_loss": 0.37564513087272644, | |
| "eval_runtime": 218.7597, | |
| "eval_samples_per_second": 22.669, | |
| "eval_steps_per_second": 2.834, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 27.925531914893618, | |
| "grad_norm": 5.426018238067627, | |
| "learning_rate": 0.00025183199999999996, | |
| "loss": 0.124, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 27.925531914893618, | |
| "eval_cer": 0.17786411409944225, | |
| "eval_loss": 0.39711853861808777, | |
| "eval_runtime": 219.1126, | |
| "eval_samples_per_second": 22.632, | |
| "eval_steps_per_second": 2.83, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 28.19148936170213, | |
| "eval_cer": 0.178911615144989, | |
| "eval_loss": 0.42398887872695923, | |
| "eval_runtime": 216.8565, | |
| "eval_samples_per_second": 22.868, | |
| "eval_steps_per_second": 2.859, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 28.45744680851064, | |
| "eval_cer": 0.17917349040637567, | |
| "eval_loss": 0.37179791927337646, | |
| "eval_runtime": 219.2208, | |
| "eval_samples_per_second": 22.621, | |
| "eval_steps_per_second": 2.828, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 28.72340425531915, | |
| "eval_cer": 0.17998256770648083, | |
| "eval_loss": 0.41135987639427185, | |
| "eval_runtime": 220.5483, | |
| "eval_samples_per_second": 22.485, | |
| "eval_steps_per_second": 2.811, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 28.98936170212766, | |
| "eval_cer": 0.18003337932436184, | |
| "eval_loss": 0.41742414236068726, | |
| "eval_runtime": 218.3591, | |
| "eval_samples_per_second": 22.71, | |
| "eval_steps_per_second": 2.839, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 29.25531914893617, | |
| "grad_norm": 2.8614120483398438, | |
| "learning_rate": 0.000263832, | |
| "loss": 0.122, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.25531914893617, | |
| "eval_cer": 0.18530997041200406, | |
| "eval_loss": 0.4062168598175049, | |
| "eval_runtime": 220.2511, | |
| "eval_samples_per_second": 22.515, | |
| "eval_steps_per_second": 2.815, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.52127659574468, | |
| "eval_cer": 0.1861112305401275, | |
| "eval_loss": 0.4202740788459778, | |
| "eval_runtime": 217.0728, | |
| "eval_samples_per_second": 22.845, | |
| "eval_steps_per_second": 2.856, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 29.78723404255319, | |
| "eval_cer": 0.18606432750823734, | |
| "eval_loss": 0.43755000829696655, | |
| "eval_runtime": 219.2588, | |
| "eval_samples_per_second": 22.617, | |
| "eval_steps_per_second": 2.828, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 30.0531914893617, | |
| "eval_cer": 0.18122940663756074, | |
| "eval_loss": 0.409412682056427, | |
| "eval_runtime": 223.5339, | |
| "eval_samples_per_second": 22.185, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 30.319148936170212, | |
| "eval_cer": 0.18185868898208696, | |
| "eval_loss": 0.40995094180107117, | |
| "eval_runtime": 225.4254, | |
| "eval_samples_per_second": 21.998, | |
| "eval_steps_per_second": 2.75, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 30.585106382978722, | |
| "grad_norm": 1.4252536296844482, | |
| "learning_rate": 0.000275832, | |
| "loss": 0.125, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 30.585106382978722, | |
| "eval_cer": 0.18692812501221434, | |
| "eval_loss": 0.39966416358947754, | |
| "eval_runtime": 219.2997, | |
| "eval_samples_per_second": 22.613, | |
| "eval_steps_per_second": 2.827, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 30.851063829787233, | |
| "eval_cer": 0.18203848393766586, | |
| "eval_loss": 0.4437381625175476, | |
| "eval_runtime": 219.4853, | |
| "eval_samples_per_second": 22.594, | |
| "eval_steps_per_second": 2.825, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 31.117021276595743, | |
| "eval_cer": 0.18580245224685066, | |
| "eval_loss": 0.44233861565589905, | |
| "eval_runtime": 216.9437, | |
| "eval_samples_per_second": 22.858, | |
| "eval_steps_per_second": 2.858, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 31.382978723404257, | |
| "eval_cer": 0.18947652307824597, | |
| "eval_loss": 0.4216567575931549, | |
| "eval_runtime": 217.1103, | |
| "eval_samples_per_second": 22.841, | |
| "eval_steps_per_second": 2.856, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 31.648936170212767, | |
| "eval_cer": 0.1966175096835218, | |
| "eval_loss": 0.4611862301826477, | |
| "eval_runtime": 226.5232, | |
| "eval_samples_per_second": 21.892, | |
| "eval_steps_per_second": 2.737, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 31.914893617021278, | |
| "grad_norm": 5.578627586364746, | |
| "learning_rate": 0.00028783199999999996, | |
| "loss": 0.1294, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 31.914893617021278, | |
| "eval_cer": 0.18622067094787118, | |
| "eval_loss": 0.4385882318019867, | |
| "eval_runtime": 216.3818, | |
| "eval_samples_per_second": 22.918, | |
| "eval_steps_per_second": 2.865, | |
| "eval_wer": 0.9973785037305909, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 32.180851063829785, | |
| "eval_cer": 0.1892146478168593, | |
| "eval_loss": 0.42784062027931213, | |
| "eval_runtime": 236.4807, | |
| "eval_samples_per_second": 20.97, | |
| "eval_steps_per_second": 2.622, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 32.4468085106383, | |
| "eval_cer": 0.1856109315332992, | |
| "eval_loss": 0.4187084436416626, | |
| "eval_runtime": 222.7058, | |
| "eval_samples_per_second": 22.267, | |
| "eval_steps_per_second": 2.784, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 32.712765957446805, | |
| "eval_cer": 0.18286710416772525, | |
| "eval_loss": 0.4047478437423706, | |
| "eval_runtime": 222.5261, | |
| "eval_samples_per_second": 22.285, | |
| "eval_steps_per_second": 2.786, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 32.97872340425532, | |
| "eval_cer": 0.18515753555836106, | |
| "eval_loss": 0.4230673015117645, | |
| "eval_runtime": 219.6428, | |
| "eval_samples_per_second": 22.578, | |
| "eval_steps_per_second": 2.823, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 33.244680851063826, | |
| "grad_norm": 3.530928134918213, | |
| "learning_rate": 0.000299832, | |
| "loss": 0.1275, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 33.244680851063826, | |
| "eval_cer": 0.1842781037104207, | |
| "eval_loss": 0.41239944100379944, | |
| "eval_runtime": 212.9629, | |
| "eval_samples_per_second": 23.286, | |
| "eval_steps_per_second": 2.911, | |
| "eval_wer": 0.999395039322444, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 33.51063829787234, | |
| "eval_cer": 0.18698284521608619, | |
| "eval_loss": 0.41910141706466675, | |
| "eval_runtime": 209.3165, | |
| "eval_samples_per_second": 23.691, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 0.999395039322444, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 33.776595744680854, | |
| "eval_cer": 0.19266983783276723, | |
| "eval_loss": 0.4845859110355377, | |
| "eval_runtime": 209.6184, | |
| "eval_samples_per_second": 23.657, | |
| "eval_steps_per_second": 2.958, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 34.04255319148936, | |
| "eval_cer": 0.18445399008000876, | |
| "eval_loss": 0.421185702085495, | |
| "eval_runtime": 209.5402, | |
| "eval_samples_per_second": 23.666, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 34.308510638297875, | |
| "eval_cer": 0.183660547123867, | |
| "eval_loss": 0.43256160616874695, | |
| "eval_runtime": 219.5902, | |
| "eval_samples_per_second": 22.583, | |
| "eval_steps_per_second": 2.823, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 34.57446808510638, | |
| "grad_norm": 1.3231768608093262, | |
| "learning_rate": 0.00029548990895071305, | |
| "loss": 0.134, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 34.57446808510638, | |
| "eval_cer": 0.18797171747177024, | |
| "eval_loss": 0.4103536009788513, | |
| "eval_runtime": 213.3906, | |
| "eval_samples_per_second": 23.239, | |
| "eval_steps_per_second": 2.905, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 34.840425531914896, | |
| "eval_cer": 0.18773329372632863, | |
| "eval_loss": 0.3964749276638031, | |
| "eval_runtime": 219.4036, | |
| "eval_samples_per_second": 22.602, | |
| "eval_steps_per_second": 2.826, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 35.1063829787234, | |
| "eval_cer": 0.18439145270415522, | |
| "eval_loss": 0.4146926701068878, | |
| "eval_runtime": 211.7608, | |
| "eval_samples_per_second": 23.418, | |
| "eval_steps_per_second": 2.928, | |
| "eval_wer": 0.999395039322444, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 35.37234042553192, | |
| "eval_cer": 0.18589625831063097, | |
| "eval_loss": 0.4251463711261749, | |
| "eval_runtime": 208.6917, | |
| "eval_samples_per_second": 23.762, | |
| "eval_steps_per_second": 2.971, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 35.638297872340424, | |
| "eval_cer": 0.19017225138461658, | |
| "eval_loss": 0.4457683861255646, | |
| "eval_runtime": 211.8043, | |
| "eval_samples_per_second": 23.413, | |
| "eval_steps_per_second": 2.927, | |
| "eval_wer": 0.999395039322444, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 35.90425531914894, | |
| "grad_norm": 3.084670066833496, | |
| "learning_rate": 0.00028198284613067677, | |
| "loss": 0.1293, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 35.90425531914894, | |
| "eval_cer": 0.19420591212716976, | |
| "eval_loss": 0.43544086813926697, | |
| "eval_runtime": 209.3943, | |
| "eval_samples_per_second": 23.683, | |
| "eval_steps_per_second": 2.961, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 36.170212765957444, | |
| "eval_cer": 0.1863222941836332, | |
| "eval_loss": 0.4197835326194763, | |
| "eval_runtime": 210.9951, | |
| "eval_samples_per_second": 23.503, | |
| "eval_steps_per_second": 2.938, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 36.43617021276596, | |
| "eval_cer": 0.18905830437722546, | |
| "eval_loss": 0.4278818666934967, | |
| "eval_runtime": 214.736, | |
| "eval_samples_per_second": 23.093, | |
| "eval_steps_per_second": 2.887, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 36.702127659574465, | |
| "eval_cer": 0.18459079058968836, | |
| "eval_loss": 0.4114536941051483, | |
| "eval_runtime": 212.7503, | |
| "eval_samples_per_second": 23.309, | |
| "eval_steps_per_second": 2.914, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 36.96808510638298, | |
| "eval_cer": 0.18683822753442486, | |
| "eval_loss": 0.43593207001686096, | |
| "eval_runtime": 213.278, | |
| "eval_samples_per_second": 23.251, | |
| "eval_steps_per_second": 2.907, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 37.234042553191486, | |
| "grad_norm": 2.2776243686676025, | |
| "learning_rate": 0.00026031324776349936, | |
| "loss": 0.1193, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 37.234042553191486, | |
| "eval_cer": 0.19045757816194836, | |
| "eval_loss": 0.43160808086395264, | |
| "eval_runtime": 209.1934, | |
| "eval_samples_per_second": 23.705, | |
| "eval_steps_per_second": 2.964, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "eval_cer": 0.18992991905318413, | |
| "eval_loss": 0.4389212727546692, | |
| "eval_runtime": 215.9667, | |
| "eval_samples_per_second": 22.962, | |
| "eval_steps_per_second": 2.871, | |
| "eval_wer": 0.999395039322444, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 37.765957446808514, | |
| "eval_cer": 0.1825231486005308, | |
| "eval_loss": 0.42151185870170593, | |
| "eval_runtime": 208.7788, | |
| "eval_samples_per_second": 23.752, | |
| "eval_steps_per_second": 2.97, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 38.03191489361702, | |
| "eval_cer": 0.18895668114146344, | |
| "eval_loss": 0.47925981879234314, | |
| "eval_runtime": 211.6717, | |
| "eval_samples_per_second": 23.428, | |
| "eval_steps_per_second": 2.929, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 38.297872340425535, | |
| "eval_cer": 0.1853920507178118, | |
| "eval_loss": 0.4381818473339081, | |
| "eval_runtime": 209.7616, | |
| "eval_samples_per_second": 23.641, | |
| "eval_steps_per_second": 2.956, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 38.56382978723404, | |
| "grad_norm": 1.5227301120758057, | |
| "learning_rate": 0.00023182127942485896, | |
| "loss": 0.1132, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 38.56382978723404, | |
| "eval_cer": 0.182441068294723, | |
| "eval_loss": 0.40106824040412903, | |
| "eval_runtime": 209.0948, | |
| "eval_samples_per_second": 23.717, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.9975801572897761, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 38.829787234042556, | |
| "eval_cer": 0.1797675954769843, | |
| "eval_loss": 0.4283459186553955, | |
| "eval_runtime": 209.0149, | |
| "eval_samples_per_second": 23.726, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 0.994958661020367, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 39.09574468085106, | |
| "eval_cer": 0.18034997478962037, | |
| "eval_loss": 0.4303878843784332, | |
| "eval_runtime": 211.9384, | |
| "eval_samples_per_second": 23.398, | |
| "eval_steps_per_second": 2.925, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 39.361702127659576, | |
| "eval_cer": 0.18114732633175296, | |
| "eval_loss": 0.40491461753845215, | |
| "eval_runtime": 210.1586, | |
| "eval_samples_per_second": 23.596, | |
| "eval_steps_per_second": 2.95, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 39.62765957446808, | |
| "eval_cer": 0.1785129393739227, | |
| "eval_loss": 0.41461843252182007, | |
| "eval_runtime": 209.4294, | |
| "eval_samples_per_second": 23.679, | |
| "eval_steps_per_second": 2.96, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 39.8936170212766, | |
| "grad_norm": 3.955990791320801, | |
| "learning_rate": 0.00019826903907816743, | |
| "loss": 0.0949, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 39.8936170212766, | |
| "eval_cer": 0.18166325968254465, | |
| "eval_loss": 0.44991689920425415, | |
| "eval_runtime": 208.9423, | |
| "eval_samples_per_second": 23.734, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 0.9997983464408147, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 40.159574468085104, | |
| "eval_cer": 0.17907968434259539, | |
| "eval_loss": 0.420516699552536, | |
| "eval_runtime": 210.185, | |
| "eval_samples_per_second": 23.593, | |
| "eval_steps_per_second": 2.95, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 40.42553191489362, | |
| "eval_cer": 0.1807017475287965, | |
| "eval_loss": 0.4418542683124542, | |
| "eval_runtime": 210.2936, | |
| "eval_samples_per_second": 23.581, | |
| "eval_steps_per_second": 2.948, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 40.691489361702125, | |
| "eval_cer": 0.18009591670021535, | |
| "eval_loss": 0.4282812476158142, | |
| "eval_runtime": 209.7755, | |
| "eval_samples_per_second": 23.64, | |
| "eval_steps_per_second": 2.956, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 40.95744680851064, | |
| "eval_cer": 0.17690651053168496, | |
| "eval_loss": 0.4327284097671509, | |
| "eval_runtime": 210.8473, | |
| "eval_samples_per_second": 23.519, | |
| "eval_steps_per_second": 2.941, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 41.223404255319146, | |
| "grad_norm": 1.6714894771575928, | |
| "learning_rate": 0.0001617315793726526, | |
| "loss": 0.0876, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 41.223404255319146, | |
| "eval_cer": 0.17806736057096625, | |
| "eval_loss": 0.44882500171661377, | |
| "eval_runtime": 210.3046, | |
| "eval_samples_per_second": 23.58, | |
| "eval_steps_per_second": 2.948, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 41.48936170212766, | |
| "eval_cer": 0.17360766395541086, | |
| "eval_loss": 0.41944339871406555, | |
| "eval_runtime": 210.2221, | |
| "eval_samples_per_second": 23.589, | |
| "eval_steps_per_second": 2.949, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 41.755319148936174, | |
| "eval_cer": 0.17544079078511768, | |
| "eval_loss": 0.4319663345813751, | |
| "eval_runtime": 209.0871, | |
| "eval_samples_per_second": 23.717, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 42.02127659574468, | |
| "eval_cer": 0.1729080270630494, | |
| "eval_loss": 0.4347435235977173, | |
| "eval_runtime": 209.3479, | |
| "eval_samples_per_second": 23.688, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 42.287234042553195, | |
| "eval_cer": 0.17443628418547022, | |
| "eval_loss": 0.48187553882598877, | |
| "eval_runtime": 208.6066, | |
| "eval_samples_per_second": 23.772, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 0.999395039322444, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 42.5531914893617, | |
| "grad_norm": 0.7495588660240173, | |
| "learning_rate": 0.00012446857512509515, | |
| "loss": 0.0725, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 42.5531914893617, | |
| "eval_cer": 0.17520236703967607, | |
| "eval_loss": 0.44911447167396545, | |
| "eval_runtime": 208.5357, | |
| "eval_samples_per_second": 23.78, | |
| "eval_steps_per_second": 2.973, | |
| "eval_wer": 0.9989917322040734, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 42.819148936170215, | |
| "eval_cer": 0.17413923165016593, | |
| "eval_loss": 0.45369845628738403, | |
| "eval_runtime": 210.1991, | |
| "eval_samples_per_second": 23.592, | |
| "eval_steps_per_second": 2.95, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 43.08510638297872, | |
| "eval_cer": 0.17170809116385966, | |
| "eval_loss": 0.4587579071521759, | |
| "eval_runtime": 209.6677, | |
| "eval_samples_per_second": 23.652, | |
| "eval_steps_per_second": 2.957, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 43.351063829787236, | |
| "eval_cer": 0.17147357600440888, | |
| "eval_loss": 0.4416778087615967, | |
| "eval_runtime": 209.0934, | |
| "eval_samples_per_second": 23.717, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 43.61702127659574, | |
| "eval_cer": 0.17290411847705855, | |
| "eval_loss": 0.45538103580474854, | |
| "eval_runtime": 210.428, | |
| "eval_samples_per_second": 23.566, | |
| "eval_steps_per_second": 2.946, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 43.88297872340426, | |
| "grad_norm": 1.7286139726638794, | |
| "learning_rate": 8.878457276467852e-05, | |
| "loss": 0.0615, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 43.88297872340426, | |
| "eval_cer": 0.17189570329142026, | |
| "eval_loss": 0.44644975662231445, | |
| "eval_runtime": 208.6857, | |
| "eval_samples_per_second": 23.763, | |
| "eval_steps_per_second": 2.971, | |
| "eval_wer": 0.9995966928816293, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 44.148936170212764, | |
| "eval_cer": 0.17184489167353925, | |
| "eval_loss": 0.47256070375442505, | |
| "eval_runtime": 214.9431, | |
| "eval_samples_per_second": 23.071, | |
| "eval_steps_per_second": 2.884, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 44.41489361702128, | |
| "eval_cer": 0.1704495264748072, | |
| "eval_loss": 0.4455854296684265, | |
| "eval_runtime": 212.2795, | |
| "eval_samples_per_second": 23.361, | |
| "eval_steps_per_second": 2.921, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 44.680851063829785, | |
| "eval_cer": 0.16927304209156255, | |
| "eval_loss": 0.42471006512641907, | |
| "eval_runtime": 208.6276, | |
| "eval_samples_per_second": 23.77, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 44.9468085106383, | |
| "eval_cer": 0.1683897016576313, | |
| "eval_loss": 0.4499261975288391, | |
| "eval_runtime": 208.8575, | |
| "eval_samples_per_second": 23.743, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 45.212765957446805, | |
| "grad_norm": 1.9694031476974487, | |
| "learning_rate": 5.6886464673787096e-05, | |
| "loss": 0.0524, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 45.212765957446805, | |
| "eval_cer": 0.16670119250958582, | |
| "eval_loss": 0.46098417043685913, | |
| "eval_runtime": 209.0801, | |
| "eval_samples_per_second": 23.718, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.9987900786448881, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 45.47872340425532, | |
| "eval_cer": 0.16754153849761771, | |
| "eval_loss": 0.42516496777534485, | |
| "eval_runtime": 215.2938, | |
| "eval_samples_per_second": 23.034, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 45.744680851063826, | |
| "eval_cer": 0.16701778797484435, | |
| "eval_loss": 0.41850271821022034, | |
| "eval_runtime": 208.7292, | |
| "eval_samples_per_second": 23.758, | |
| "eval_steps_per_second": 2.97, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 46.01063829787234, | |
| "eval_cer": 0.16651748896801605, | |
| "eval_loss": 0.4377266764640808, | |
| "eval_runtime": 208.7268, | |
| "eval_samples_per_second": 23.758, | |
| "eval_steps_per_second": 2.97, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 46.276595744680854, | |
| "eval_cer": 0.16658784351585126, | |
| "eval_loss": 0.4386991262435913, | |
| "eval_runtime": 208.9381, | |
| "eval_samples_per_second": 23.734, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 0.9991933857632587, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 46.54255319148936, | |
| "grad_norm": 0.5704839825630188, | |
| "learning_rate": 3.074700298507786e-05, | |
| "loss": 0.0466, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 46.54255319148936, | |
| "eval_cer": 0.16593510965537997, | |
| "eval_loss": 0.4388461112976074, | |
| "eval_runtime": 208.6064, | |
| "eval_samples_per_second": 23.772, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 46.808510638297875, | |
| "eval_cer": 0.16539963337463406, | |
| "eval_loss": 0.4407631754875183, | |
| "eval_runtime": 209.5029, | |
| "eval_samples_per_second": 23.67, | |
| "eval_steps_per_second": 2.959, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 47.07446808510638, | |
| "eval_cer": 0.16512603235527484, | |
| "eval_loss": 0.4277360141277313, | |
| "eval_runtime": 208.9388, | |
| "eval_samples_per_second": 23.734, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 47.340425531914896, | |
| "eval_cer": 0.16503222629149453, | |
| "eval_loss": 0.4244243800640106, | |
| "eval_runtime": 208.9653, | |
| "eval_samples_per_second": 23.731, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 0.9985884250857028, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 47.6063829787234, | |
| "eval_cer": 0.16441857829093168, | |
| "eval_loss": 0.42959603667259216, | |
| "eval_runtime": 209.1668, | |
| "eval_samples_per_second": 23.708, | |
| "eval_steps_per_second": 2.964, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 47.87234042553192, | |
| "grad_norm": 1.8080158233642578, | |
| "learning_rate": 1.1982793882402725e-05, | |
| "loss": 0.0393, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 47.87234042553192, | |
| "eval_cer": 0.1648250712339797, | |
| "eval_loss": 0.43411195278167725, | |
| "eval_runtime": 215.2833, | |
| "eval_samples_per_second": 23.035, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.9983867715265174, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 48.138297872340424, | |
| "eval_cer": 0.16460228183250145, | |
| "eval_loss": 0.43372273445129395, | |
| "eval_runtime": 259.5957, | |
| "eval_samples_per_second": 19.103, | |
| "eval_steps_per_second": 2.388, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 48.40425531914894, | |
| "eval_cer": 0.16421142323341684, | |
| "eval_loss": 0.43311166763305664, | |
| "eval_runtime": 209.0523, | |
| "eval_samples_per_second": 23.721, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 48.670212765957444, | |
| "eval_cer": 0.16413716009959078, | |
| "eval_loss": 0.4280834197998047, | |
| "eval_runtime": 210.4305, | |
| "eval_samples_per_second": 23.566, | |
| "eval_steps_per_second": 2.946, | |
| "eval_wer": 0.9979834644081468, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 48.93617021276596, | |
| "eval_cer": 0.1640824398957189, | |
| "eval_loss": 0.42681965231895447, | |
| "eval_runtime": 261.2993, | |
| "eval_samples_per_second": 18.978, | |
| "eval_steps_per_second": 2.373, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 49.202127659574465, | |
| "grad_norm": 0.8930312395095825, | |
| "learning_rate": 1.7543179002012464e-06, | |
| "loss": 0.0373, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 49.202127659574465, | |
| "eval_cer": 0.16407071413774638, | |
| "eval_loss": 0.42752090096473694, | |
| "eval_runtime": 209.2156, | |
| "eval_samples_per_second": 23.703, | |
| "eval_steps_per_second": 2.963, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 49.46808510638298, | |
| "eval_cer": 0.16406680555175554, | |
| "eval_loss": 0.4268626868724823, | |
| "eval_runtime": 212.2555, | |
| "eval_samples_per_second": 23.363, | |
| "eval_steps_per_second": 2.921, | |
| "eval_wer": 0.9981851179673321, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 49.734042553191486, | |
| "eval_cer": 0.16399254241792946, | |
| "eval_loss": 0.4268469214439392, | |
| "eval_runtime": 208.7487, | |
| "eval_samples_per_second": 23.756, | |
| "eval_steps_per_second": 2.97, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_cer": 0.16415279444355416, | |
| "eval_loss": 0.42705675959587097, | |
| "eval_runtime": 208.3387, | |
| "eval_samples_per_second": 23.803, | |
| "eval_steps_per_second": 2.976, | |
| "eval_wer": 0.9977818108489614, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 18800, | |
| "total_flos": 2.6273093616415412e+19, | |
| "train_loss": 0.4882018888757584, | |
| "train_runtime": 107059.6152, | |
| "train_samples_per_second": 5.619, | |
| "train_steps_per_second": 0.176 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 18800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "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": 2.6273093616415412e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |