| { | |
| "best_metric": 0.38558459281921387, | |
| "best_model_checkpoint": "./Hubert-common_voice_JSUT-ja-demo-roma/checkpoint-10300", | |
| "epoch": 20.0, | |
| "eval_steps": 100, | |
| "global_step": 10340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19342359767891681, | |
| "eval_cer": 3.3906697851852106, | |
| "eval_loss": 16.271892547607422, | |
| "eval_runtime": 235.1592, | |
| "eval_samples_per_second": 23.214, | |
| "eval_steps_per_second": 2.904, | |
| "eval_wer": 3.0917750503755266, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "eval_cer": 2.893664453340365, | |
| "eval_loss": 16.005735397338867, | |
| "eval_runtime": 236.1908, | |
| "eval_samples_per_second": 23.113, | |
| "eval_steps_per_second": 2.892, | |
| "eval_wer": 2.8734200403004215, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "eval_cer": 1.690616021560241, | |
| "eval_loss": 15.374765396118164, | |
| "eval_runtime": 231.2146, | |
| "eval_samples_per_second": 23.61, | |
| "eval_steps_per_second": 2.954, | |
| "eval_wer": 1.9646455394760944, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "eval_cer": 0.9291963879062664, | |
| "eval_loss": 13.043855667114258, | |
| "eval_runtime": 226.4398, | |
| "eval_samples_per_second": 24.108, | |
| "eval_steps_per_second": 3.016, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "grad_norm": 47.46343994140625, | |
| "learning_rate": 1.1904e-06, | |
| "loss": 11.8785, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 7.6484761238098145, | |
| "eval_runtime": 226.1704, | |
| "eval_samples_per_second": 24.137, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 6.007350444793701, | |
| "eval_runtime": 228.9926, | |
| "eval_samples_per_second": 23.839, | |
| "eval_steps_per_second": 2.983, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 5.646536827087402, | |
| "eval_runtime": 323.7978, | |
| "eval_samples_per_second": 16.859, | |
| "eval_steps_per_second": 2.109, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 5.502617359161377, | |
| "eval_runtime": 258.5966, | |
| "eval_samples_per_second": 21.11, | |
| "eval_steps_per_second": 2.641, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 5.361776351928711, | |
| "eval_runtime": 256.6054, | |
| "eval_samples_per_second": 21.274, | |
| "eval_steps_per_second": 2.662, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "grad_norm": 29.7659969329834, | |
| "learning_rate": 2.3904e-06, | |
| "loss": 4.9912, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 5.219295024871826, | |
| "eval_runtime": 262.3733, | |
| "eval_samples_per_second": 20.806, | |
| "eval_steps_per_second": 2.603, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 5.06879997253418, | |
| "eval_runtime": 246.2278, | |
| "eval_samples_per_second": 22.171, | |
| "eval_steps_per_second": 2.774, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 4.91414737701416, | |
| "eval_runtime": 265.8995, | |
| "eval_samples_per_second": 20.53, | |
| "eval_steps_per_second": 2.569, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 4.753695964813232, | |
| "eval_runtime": 261.3741, | |
| "eval_samples_per_second": 20.886, | |
| "eval_steps_per_second": 2.613, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.7079303675048356, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 4.589428901672363, | |
| "eval_runtime": 262.0559, | |
| "eval_samples_per_second": 20.831, | |
| "eval_steps_per_second": 2.606, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "grad_norm": 19.784141540527344, | |
| "learning_rate": 3.5904e-06, | |
| "loss": 4.3024, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.9013539651837523, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 4.4225029945373535, | |
| "eval_runtime": 247.2412, | |
| "eval_samples_per_second": 22.08, | |
| "eval_steps_per_second": 2.762, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.094777562862669, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 4.2562456130981445, | |
| "eval_runtime": 265.0317, | |
| "eval_samples_per_second": 20.598, | |
| "eval_steps_per_second": 2.577, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.288201160541586, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 4.094423770904541, | |
| "eval_runtime": 262.7201, | |
| "eval_samples_per_second": 20.779, | |
| "eval_steps_per_second": 2.6, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.481624758220503, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.9344263076782227, | |
| "eval_runtime": 233.1037, | |
| "eval_samples_per_second": 23.419, | |
| "eval_steps_per_second": 2.93, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.6750483558994196, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.7835137844085693, | |
| "eval_runtime": 231.2092, | |
| "eval_samples_per_second": 23.611, | |
| "eval_steps_per_second": 2.954, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.8684719535783367, | |
| "grad_norm": 19.985584259033203, | |
| "learning_rate": 4.7904e-06, | |
| "loss": 3.6966, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8684719535783367, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.641073703765869, | |
| "eval_runtime": 230.0297, | |
| "eval_samples_per_second": 23.732, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.061895551257253, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.5155856609344482, | |
| "eval_runtime": 229.2143, | |
| "eval_samples_per_second": 23.816, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.397294044494629, | |
| "eval_runtime": 229.4391, | |
| "eval_samples_per_second": 23.793, | |
| "eval_steps_per_second": 2.977, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.448742746615087, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.290855646133423, | |
| "eval_runtime": 231.2121, | |
| "eval_samples_per_second": 23.61, | |
| "eval_steps_per_second": 2.954, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.6421663442940035, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.1956613063812256, | |
| "eval_runtime": 230.2419, | |
| "eval_samples_per_second": 23.71, | |
| "eval_steps_per_second": 2.966, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.835589941972921, | |
| "grad_norm": 3.1049797534942627, | |
| "learning_rate": 5.9904e-06, | |
| "loss": 3.2011, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.835589941972921, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.115924596786499, | |
| "eval_runtime": 231.315, | |
| "eval_samples_per_second": 23.6, | |
| "eval_steps_per_second": 2.953, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.029013539651838, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.0544307231903076, | |
| "eval_runtime": 231.5463, | |
| "eval_samples_per_second": 23.576, | |
| "eval_steps_per_second": 2.95, | |
| "eval_wer": 1.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.222437137330754, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 3.0038533210754395, | |
| "eval_runtime": 237.992, | |
| "eval_samples_per_second": 22.938, | |
| "eval_steps_per_second": 2.87, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.415860735009671, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.9653820991516113, | |
| "eval_runtime": 231.2651, | |
| "eval_samples_per_second": 23.605, | |
| "eval_steps_per_second": 2.953, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.609284332688588, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.9386982917785645, | |
| "eval_runtime": 235.1119, | |
| "eval_samples_per_second": 23.219, | |
| "eval_steps_per_second": 2.905, | |
| "eval_wer": 1.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.802707930367505, | |
| "grad_norm": 0.9443885684013367, | |
| "learning_rate": 7.190400000000001e-06, | |
| "loss": 2.9439, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.802707930367505, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.9090564250946045, | |
| "eval_runtime": 234.7068, | |
| "eval_samples_per_second": 23.259, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.996131528046422, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.8868346214294434, | |
| "eval_runtime": 233.0298, | |
| "eval_samples_per_second": 23.426, | |
| "eval_steps_per_second": 2.931, | |
| "eval_wer": 1.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.189555125725338, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.8660459518432617, | |
| "eval_runtime": 237.6658, | |
| "eval_samples_per_second": 22.969, | |
| "eval_steps_per_second": 2.874, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.382978723404255, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.853316068649292, | |
| "eval_runtime": 238.1324, | |
| "eval_samples_per_second": 22.924, | |
| "eval_steps_per_second": 2.868, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.576402321083172, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.7336666584014893, | |
| "eval_runtime": 248.0434, | |
| "eval_samples_per_second": 22.008, | |
| "eval_steps_per_second": 2.754, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.769825918762089, | |
| "grad_norm": 5.135930061340332, | |
| "learning_rate": 8.3904e-06, | |
| "loss": 2.7884, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.769825918762089, | |
| "eval_cer": 0.929223783383958, | |
| "eval_loss": 2.523024559020996, | |
| "eval_runtime": 234.9913, | |
| "eval_samples_per_second": 23.231, | |
| "eval_steps_per_second": 2.906, | |
| "eval_wer": 1.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.963249516441006, | |
| "eval_cer": 0.9181765570048525, | |
| "eval_loss": 2.272446870803833, | |
| "eval_runtime": 233.6368, | |
| "eval_samples_per_second": 23.365, | |
| "eval_steps_per_second": 2.923, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 7.156673114119923, | |
| "eval_cer": 0.6315890404391495, | |
| "eval_loss": 1.9632701873779297, | |
| "eval_runtime": 237.5728, | |
| "eval_samples_per_second": 22.978, | |
| "eval_steps_per_second": 2.875, | |
| "eval_wer": 1.0, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.350096711798839, | |
| "eval_cer": 0.4242155476184769, | |
| "eval_loss": 1.5858280658721924, | |
| "eval_runtime": 277.2902, | |
| "eval_samples_per_second": 19.687, | |
| "eval_steps_per_second": 2.463, | |
| "eval_wer": 1.0, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.543520309477756, | |
| "eval_cer": 0.3861015892801496, | |
| "eval_loss": 1.351006031036377, | |
| "eval_runtime": 236.5551, | |
| "eval_samples_per_second": 23.077, | |
| "eval_steps_per_second": 2.887, | |
| "eval_wer": 0.9998168162667155, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.7369439071566735, | |
| "grad_norm": 3.2181670665740967, | |
| "learning_rate": 9.5904e-06, | |
| "loss": 1.7651, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.7369439071566735, | |
| "eval_cer": 0.33344405672233657, | |
| "eval_loss": 1.1917191743850708, | |
| "eval_runtime": 237.0965, | |
| "eval_samples_per_second": 23.024, | |
| "eval_steps_per_second": 2.881, | |
| "eval_wer": 0.999267265066862, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.93036750483559, | |
| "eval_cer": 0.29824771675815615, | |
| "eval_loss": 1.0715813636779785, | |
| "eval_runtime": 240.0983, | |
| "eval_samples_per_second": 22.737, | |
| "eval_steps_per_second": 2.845, | |
| "eval_wer": 0.9979849789338707, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 8.123791102514506, | |
| "eval_cer": 0.27819422708796343, | |
| "eval_loss": 0.9761540293693542, | |
| "eval_runtime": 238.5129, | |
| "eval_samples_per_second": 22.888, | |
| "eval_steps_per_second": 2.864, | |
| "eval_wer": 0.9976186114673017, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.317214700193423, | |
| "eval_cer": 0.2595721511271527, | |
| "eval_loss": 0.904407799243927, | |
| "eval_runtime": 234.8451, | |
| "eval_samples_per_second": 23.245, | |
| "eval_steps_per_second": 2.908, | |
| "eval_wer": 0.9965195090675948, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.51063829787234, | |
| "eval_cer": 0.2565654974505084, | |
| "eval_loss": 0.8529276251792908, | |
| "eval_runtime": 237.5826, | |
| "eval_samples_per_second": 22.977, | |
| "eval_steps_per_second": 2.875, | |
| "eval_wer": 0.9963363253343103, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.704061895551257, | |
| "grad_norm": 2.711160898208618, | |
| "learning_rate": 1.0790400000000001e-05, | |
| "loss": 0.9278, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.704061895551257, | |
| "eval_cer": 0.2466140901790637, | |
| "eval_loss": 0.7957596182823181, | |
| "eval_runtime": 234.6716, | |
| "eval_samples_per_second": 23.262, | |
| "eval_steps_per_second": 2.91, | |
| "eval_wer": 0.9970690602674482, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.897485493230175, | |
| "eval_cer": 0.24351497676521047, | |
| "eval_loss": 0.7534636855125427, | |
| "eval_runtime": 235.3455, | |
| "eval_samples_per_second": 23.196, | |
| "eval_steps_per_second": 2.902, | |
| "eval_wer": 0.9965195090675948, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "eval_cer": 0.24028573483232255, | |
| "eval_loss": 0.7190229296684265, | |
| "eval_runtime": 235.8068, | |
| "eval_samples_per_second": 23.15, | |
| "eval_steps_per_second": 2.896, | |
| "eval_wer": 0.9974354277340172, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.284332688588007, | |
| "eval_cer": 0.2355908348429383, | |
| "eval_loss": 0.6800413727760315, | |
| "eval_runtime": 236.222, | |
| "eval_samples_per_second": 23.11, | |
| "eval_steps_per_second": 2.891, | |
| "eval_wer": 0.9974354277340172, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.477756286266924, | |
| "eval_cer": 0.23300196220108965, | |
| "eval_loss": 0.6568382978439331, | |
| "eval_runtime": 236.9709, | |
| "eval_samples_per_second": 23.037, | |
| "eval_steps_per_second": 2.882, | |
| "eval_wer": 0.9963363253343103, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.671179883945841, | |
| "grad_norm": 5.1745381355285645, | |
| "learning_rate": 1.19904e-05, | |
| "loss": 0.6673, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.671179883945841, | |
| "eval_cer": 0.23294032237628373, | |
| "eval_loss": 0.6317699551582336, | |
| "eval_runtime": 235.798, | |
| "eval_samples_per_second": 23.151, | |
| "eval_steps_per_second": 2.897, | |
| "eval_wer": 0.9959699578677413, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.864603481624759, | |
| "eval_cer": 0.22926932836562006, | |
| "eval_loss": 0.613182783126831, | |
| "eval_runtime": 236.7731, | |
| "eval_samples_per_second": 23.056, | |
| "eval_steps_per_second": 2.885, | |
| "eval_wer": 0.9972522440007328, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 10.058027079303676, | |
| "eval_cer": 0.2261496683434982, | |
| "eval_loss": 0.5896427631378174, | |
| "eval_runtime": 236.3817, | |
| "eval_samples_per_second": 23.094, | |
| "eval_steps_per_second": 2.889, | |
| "eval_wer": 0.9970690602674482, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 10.251450676982591, | |
| "eval_cer": 0.22313274136271954, | |
| "eval_loss": 0.5742546916007996, | |
| "eval_runtime": 234.8696, | |
| "eval_samples_per_second": 23.243, | |
| "eval_steps_per_second": 2.908, | |
| "eval_wer": 0.9961531416010259, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 10.444874274661508, | |
| "eval_cer": 0.2214684660929597, | |
| "eval_loss": 0.5562453866004944, | |
| "eval_runtime": 234.6327, | |
| "eval_samples_per_second": 23.266, | |
| "eval_steps_per_second": 2.911, | |
| "eval_wer": 0.9959699578677413, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "grad_norm": 3.1148133277893066, | |
| "learning_rate": 1.31904e-05, | |
| "loss": 0.5392, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "eval_cer": 0.2236943486553957, | |
| "eval_loss": 0.5472539067268372, | |
| "eval_runtime": 233.6555, | |
| "eval_samples_per_second": 23.363, | |
| "eval_steps_per_second": 2.923, | |
| "eval_wer": 0.9972522440007328, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.831721470019342, | |
| "eval_cer": 0.21852002780640986, | |
| "eval_loss": 0.5307034850120544, | |
| "eval_runtime": 233.484, | |
| "eval_samples_per_second": 23.381, | |
| "eval_steps_per_second": 2.925, | |
| "eval_wer": 0.9963363253343103, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 11.02514506769826, | |
| "eval_cer": 0.21733859783096304, | |
| "eval_loss": 0.5194967985153198, | |
| "eval_runtime": 235.0242, | |
| "eval_samples_per_second": 23.227, | |
| "eval_steps_per_second": 2.906, | |
| "eval_wer": 0.9976186114673017, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 11.218568665377177, | |
| "eval_cer": 0.21635578506877978, | |
| "eval_loss": 0.5090118050575256, | |
| "eval_runtime": 234.9867, | |
| "eval_samples_per_second": 23.231, | |
| "eval_steps_per_second": 2.907, | |
| "eval_wer": 0.9978017952005862, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 11.411992263056092, | |
| "eval_cer": 0.21347583547645874, | |
| "eval_loss": 0.4978716969490051, | |
| "eval_runtime": 235.5271, | |
| "eval_samples_per_second": 23.178, | |
| "eval_steps_per_second": 2.9, | |
| "eval_wer": 0.9974354277340172, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 11.60541586073501, | |
| "grad_norm": 2.333935499191284, | |
| "learning_rate": 1.43904e-05, | |
| "loss": 0.4572, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.60541586073501, | |
| "eval_cer": 0.21267451775398177, | |
| "eval_loss": 0.49008145928382874, | |
| "eval_runtime": 235.4425, | |
| "eval_samples_per_second": 23.186, | |
| "eval_steps_per_second": 2.901, | |
| "eval_wer": 0.9974354277340172, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.798839458413926, | |
| "eval_cer": 0.21371212147154808, | |
| "eval_loss": 0.487173467874527, | |
| "eval_runtime": 235.8902, | |
| "eval_samples_per_second": 23.142, | |
| "eval_steps_per_second": 2.895, | |
| "eval_wer": 0.999267265066862, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 11.992263056092844, | |
| "eval_cer": 0.211890322205062, | |
| "eval_loss": 0.4753509759902954, | |
| "eval_runtime": 236.7331, | |
| "eval_samples_per_second": 23.06, | |
| "eval_steps_per_second": 2.885, | |
| "eval_wer": 0.9972522440007328, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 12.18568665377176, | |
| "eval_cer": 0.21195538646457937, | |
| "eval_loss": 0.47239425778388977, | |
| "eval_runtime": 236.344, | |
| "eval_samples_per_second": 23.098, | |
| "eval_steps_per_second": 2.89, | |
| "eval_wer": 0.9968858765341637, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 12.379110251450676, | |
| "eval_cer": 0.20878093548707446, | |
| "eval_loss": 0.4649977684020996, | |
| "eval_runtime": 242.8375, | |
| "eval_samples_per_second": 22.48, | |
| "eval_steps_per_second": 2.813, | |
| "eval_wer": 0.9987177138670086, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 12.572533849129593, | |
| "grad_norm": 3.1772515773773193, | |
| "learning_rate": 1.5590400000000002e-05, | |
| "loss": 0.41, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.572533849129593, | |
| "eval_cer": 0.20764744759758783, | |
| "eval_loss": 0.4591744542121887, | |
| "eval_runtime": 237.1668, | |
| "eval_samples_per_second": 23.018, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.9976186114673017, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.76595744680851, | |
| "eval_cer": 0.20641122666675799, | |
| "eval_loss": 0.4502531886100769, | |
| "eval_runtime": 237.1617, | |
| "eval_samples_per_second": 23.018, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.9981681626671551, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 12.959381044487428, | |
| "eval_cer": 0.20994524328896408, | |
| "eval_loss": 0.44779568910598755, | |
| "eval_runtime": 236.6236, | |
| "eval_samples_per_second": 23.07, | |
| "eval_steps_per_second": 2.886, | |
| "eval_wer": 0.9963363253343103, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 13.152804642166345, | |
| "eval_cer": 0.2061133008468627, | |
| "eval_loss": 0.4495759606361389, | |
| "eval_runtime": 237.3684, | |
| "eval_samples_per_second": 22.998, | |
| "eval_steps_per_second": 2.877, | |
| "eval_wer": 0.9981681626671551, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 13.346228239845262, | |
| "eval_cer": 0.20519897677890822, | |
| "eval_loss": 0.443766713142395, | |
| "eval_runtime": 251.4879, | |
| "eval_samples_per_second": 21.707, | |
| "eval_steps_per_second": 2.716, | |
| "eval_wer": 0.9981681626671551, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 13.539651837524177, | |
| "grad_norm": 2.66337513923645, | |
| "learning_rate": 1.6790399999999998e-05, | |
| "loss": 0.3688, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.539651837524177, | |
| "eval_cer": 0.20395933141336695, | |
| "eval_loss": 0.4365153908729553, | |
| "eval_runtime": 238.8337, | |
| "eval_samples_per_second": 22.857, | |
| "eval_steps_per_second": 2.86, | |
| "eval_wer": 0.9990840813335776, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.733075435203094, | |
| "eval_cer": 0.20464421835565494, | |
| "eval_loss": 0.4288468062877655, | |
| "eval_runtime": 234.5955, | |
| "eval_samples_per_second": 23.27, | |
| "eval_steps_per_second": 2.911, | |
| "eval_wer": 0.9979849789338707, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 13.926499032882012, | |
| "eval_cer": 0.20250394666100494, | |
| "eval_loss": 0.4299309551715851, | |
| "eval_runtime": 237.0206, | |
| "eval_samples_per_second": 23.032, | |
| "eval_steps_per_second": 2.882, | |
| "eval_wer": 0.9981681626671551, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 14.119922630560929, | |
| "eval_cer": 0.20263407518003965, | |
| "eval_loss": 0.4274175465106964, | |
| "eval_runtime": 237.1451, | |
| "eval_samples_per_second": 23.02, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.9985345301337242, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 14.313346228239846, | |
| "eval_cer": 0.20056229217961846, | |
| "eval_loss": 0.42421066761016846, | |
| "eval_runtime": 244.6511, | |
| "eval_samples_per_second": 22.313, | |
| "eval_steps_per_second": 2.792, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 14.506769825918763, | |
| "grad_norm": 1.9870613813400269, | |
| "learning_rate": 1.79904e-05, | |
| "loss": 0.3394, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.506769825918763, | |
| "eval_cer": 0.20014108671011133, | |
| "eval_loss": 0.4253482520580292, | |
| "eval_runtime": 244.8721, | |
| "eval_samples_per_second": 22.293, | |
| "eval_steps_per_second": 2.789, | |
| "eval_wer": 0.9970690602674482, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.700193423597678, | |
| "eval_cer": 0.19957263054801228, | |
| "eval_loss": 0.4177948236465454, | |
| "eval_runtime": 234.464, | |
| "eval_samples_per_second": 23.283, | |
| "eval_steps_per_second": 2.913, | |
| "eval_wer": 0.9974354277340172, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 14.893617021276595, | |
| "eval_cer": 0.20044928583414093, | |
| "eval_loss": 0.41819530725479126, | |
| "eval_runtime": 240.7737, | |
| "eval_samples_per_second": 22.673, | |
| "eval_steps_per_second": 2.837, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 15.087040618955513, | |
| "eval_cer": 0.1979117797129639, | |
| "eval_loss": 0.41940802335739136, | |
| "eval_runtime": 239.6423, | |
| "eval_samples_per_second": 22.78, | |
| "eval_steps_per_second": 2.85, | |
| "eval_wer": 0.9970690602674482, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 15.28046421663443, | |
| "eval_cer": 0.19965481698108684, | |
| "eval_loss": 0.416002482175827, | |
| "eval_runtime": 233.6247, | |
| "eval_samples_per_second": 23.367, | |
| "eval_steps_per_second": 2.923, | |
| "eval_wer": 0.9978017952005862, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 15.473887814313347, | |
| "grad_norm": 3.6397812366485596, | |
| "learning_rate": 1.9190400000000002e-05, | |
| "loss": 0.3157, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.473887814313347, | |
| "eval_cer": 0.20096637547556837, | |
| "eval_loss": 0.40957844257354736, | |
| "eval_runtime": 233.3898, | |
| "eval_samples_per_second": 23.39, | |
| "eval_steps_per_second": 2.926, | |
| "eval_wer": 0.9974354277340172, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.667311411992262, | |
| "eval_cer": 0.19801108831959563, | |
| "eval_loss": 0.40878182649612427, | |
| "eval_runtime": 233.2723, | |
| "eval_samples_per_second": 23.402, | |
| "eval_steps_per_second": 2.928, | |
| "eval_wer": 0.9978017952005862, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 15.86073500967118, | |
| "eval_cer": 0.19737414346326781, | |
| "eval_loss": 0.4118936359882355, | |
| "eval_runtime": 235.4949, | |
| "eval_samples_per_second": 23.181, | |
| "eval_steps_per_second": 2.9, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 16.054158607350097, | |
| "eval_cer": 0.1965180347854078, | |
| "eval_loss": 0.40991291403770447, | |
| "eval_runtime": 243.05, | |
| "eval_samples_per_second": 22.46, | |
| "eval_steps_per_second": 2.81, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 16.247582205029012, | |
| "eval_cer": 0.19766864484845165, | |
| "eval_loss": 0.4085560739040375, | |
| "eval_runtime": 232.1832, | |
| "eval_samples_per_second": 23.512, | |
| "eval_steps_per_second": 2.942, | |
| "eval_wer": 0.9985345301337242, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 16.44100580270793, | |
| "grad_norm": 2.502861976623535, | |
| "learning_rate": 2.03904e-05, | |
| "loss": 0.2917, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.44100580270793, | |
| "eval_cer": 0.19676116964992005, | |
| "eval_loss": 0.40965768694877625, | |
| "eval_runtime": 233.1901, | |
| "eval_samples_per_second": 23.41, | |
| "eval_steps_per_second": 2.929, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.634429400386846, | |
| "eval_cer": 0.19489485273218524, | |
| "eval_loss": 0.41127797961235046, | |
| "eval_runtime": 235.9189, | |
| "eval_samples_per_second": 23.139, | |
| "eval_steps_per_second": 2.895, | |
| "eval_wer": 0.9979849789338707, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 16.827852998065765, | |
| "eval_cer": 0.19560713515216477, | |
| "eval_loss": 0.40175729990005493, | |
| "eval_runtime": 236.2968, | |
| "eval_samples_per_second": 23.102, | |
| "eval_steps_per_second": 2.89, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 17.02127659574468, | |
| "eval_cer": 0.19336413041617154, | |
| "eval_loss": 0.4042558968067169, | |
| "eval_runtime": 232.2491, | |
| "eval_samples_per_second": 23.505, | |
| "eval_steps_per_second": 2.941, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 17.214700193423596, | |
| "eval_cer": 0.19456953143459843, | |
| "eval_loss": 0.4046263098716736, | |
| "eval_runtime": 233.5316, | |
| "eval_samples_per_second": 23.376, | |
| "eval_steps_per_second": 2.925, | |
| "eval_wer": 0.9979849789338707, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 17.408123791102515, | |
| "grad_norm": 1.7895680665969849, | |
| "learning_rate": 2.15904e-05, | |
| "loss": 0.2785, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.408123791102515, | |
| "eval_cer": 0.19269636564744075, | |
| "eval_loss": 0.4045611321926117, | |
| "eval_runtime": 235.9629, | |
| "eval_samples_per_second": 23.135, | |
| "eval_steps_per_second": 2.895, | |
| "eval_wer": 0.9981681626671551, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.60154738878143, | |
| "eval_cer": 0.19476814864786196, | |
| "eval_loss": 0.40159401297569275, | |
| "eval_runtime": 235.3013, | |
| "eval_samples_per_second": 23.2, | |
| "eval_steps_per_second": 2.903, | |
| "eval_wer": 0.9989008976002931, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 17.79497098646035, | |
| "eval_cer": 0.19216557826716754, | |
| "eval_loss": 0.40133848786354065, | |
| "eval_runtime": 232.8231, | |
| "eval_samples_per_second": 23.447, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 17.988394584139265, | |
| "eval_cer": 0.19296004712022163, | |
| "eval_loss": 0.3879222273826599, | |
| "eval_runtime": 258.2005, | |
| "eval_samples_per_second": 21.142, | |
| "eval_steps_per_second": 2.645, | |
| "eval_wer": 0.9989008976002931, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 18.181818181818183, | |
| "eval_cer": 0.19279224981936108, | |
| "eval_loss": 0.40087568759918213, | |
| "eval_runtime": 233.7931, | |
| "eval_samples_per_second": 23.35, | |
| "eval_steps_per_second": 2.921, | |
| "eval_wer": 0.9979849789338707, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 18.3752417794971, | |
| "grad_norm": 2.837678909301758, | |
| "learning_rate": 2.27904e-05, | |
| "loss": 0.2647, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.3752417794971, | |
| "eval_cer": 0.19262787695321196, | |
| "eval_loss": 0.39038729667663574, | |
| "eval_runtime": 234.0768, | |
| "eval_samples_per_second": 23.321, | |
| "eval_steps_per_second": 2.918, | |
| "eval_wer": 0.9985345301337242, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.568665377176014, | |
| "eval_cer": 0.19593930531917444, | |
| "eval_loss": 0.3944104015827179, | |
| "eval_runtime": 237.4314, | |
| "eval_samples_per_second": 22.992, | |
| "eval_steps_per_second": 2.877, | |
| "eval_wer": 0.9983513464004397, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 18.762088974854933, | |
| "eval_cer": 0.19592218314561724, | |
| "eval_loss": 0.39569512009620667, | |
| "eval_runtime": 246.3149, | |
| "eval_samples_per_second": 22.163, | |
| "eval_steps_per_second": 2.773, | |
| "eval_wer": 0.9989008976002931, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 18.95551257253385, | |
| "eval_cer": 0.1937613648426986, | |
| "eval_loss": 0.39492446184158325, | |
| "eval_runtime": 243.7942, | |
| "eval_samples_per_second": 22.392, | |
| "eval_steps_per_second": 2.802, | |
| "eval_wer": 0.9981681626671551, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 19.148936170212767, | |
| "eval_cer": 0.1932579729401169, | |
| "eval_loss": 0.40386101603507996, | |
| "eval_runtime": 249.0686, | |
| "eval_samples_per_second": 21.918, | |
| "eval_steps_per_second": 2.742, | |
| "eval_wer": 0.9972522440007328, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 19.342359767891683, | |
| "grad_norm": 3.1278374195098877, | |
| "learning_rate": 2.39904e-05, | |
| "loss": 0.248, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.342359767891683, | |
| "eval_cer": 0.19337097928559444, | |
| "eval_loss": 0.40820688009262085, | |
| "eval_runtime": 231.3525, | |
| "eval_samples_per_second": 23.596, | |
| "eval_steps_per_second": 2.952, | |
| "eval_wer": 0.9990840813335776, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.535783365570598, | |
| "eval_cer": 0.19222379365726203, | |
| "eval_loss": 0.4074006974697113, | |
| "eval_runtime": 253.2468, | |
| "eval_samples_per_second": 21.556, | |
| "eval_steps_per_second": 2.697, | |
| "eval_wer": 0.999267265066862, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 19.729206963249517, | |
| "eval_cer": 0.19064170482057674, | |
| "eval_loss": 0.39546000957489014, | |
| "eval_runtime": 235.3118, | |
| "eval_samples_per_second": 23.199, | |
| "eval_steps_per_second": 2.903, | |
| "eval_wer": 0.9989008976002931, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 19.922630560928432, | |
| "eval_cer": 0.19093620620576057, | |
| "eval_loss": 0.38558459281921387, | |
| "eval_runtime": 246.5841, | |
| "eval_samples_per_second": 22.138, | |
| "eval_steps_per_second": 2.77, | |
| "eval_wer": 0.9979849789338707, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 10340, | |
| "total_flos": 1.4614133359859188e+19, | |
| "train_loss": 1.9748668191050192, | |
| "train_runtime": 66473.8913, | |
| "train_samples_per_second": 4.974, | |
| "train_steps_per_second": 0.156 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4614133359859188e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |