| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 500.0, | |
| "global_step": 18000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 4.5903, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 0.000398, | |
| "loss": 2.9083, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.000598, | |
| "loss": 1.2823, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 0.0007980000000000001, | |
| "loss": 0.3962, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 0.000998, | |
| "loss": 0.3034, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "eval_loss": 0.38567885756492615, | |
| "eval_runtime": 133.1769, | |
| "eval_samples_per_second": 19.59, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.4354438434616608, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 0.0009999210371864342, | |
| "loss": 0.2639, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.44, | |
| "learning_rate": 0.0009996809758056298, | |
| "loss": 0.2354, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 0.000999279885163942, | |
| "loss": 0.2158, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 0.000998717894518502, | |
| "loss": 0.2063, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 0.000997995184978743, | |
| "loss": 0.1926, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "eval_loss": 0.4142022132873535, | |
| "eval_runtime": 135.7258, | |
| "eval_samples_per_second": 19.223, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 0.4174355711104041, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "learning_rate": 0.0009971119894480345, | |
| "loss": 0.1771, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 0.0009960685925486266, | |
| "loss": 0.1709, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 36.11, | |
| "learning_rate": 0.0009948653305299256, | |
| "loss": 0.1724, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 38.89, | |
| "learning_rate": 0.0009935025911601335, | |
| "loss": 0.1632, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "learning_rate": 0.0009919808136012837, | |
| "loss": 0.1609, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "eval_loss": 0.4701250195503235, | |
| "eval_runtime": 133.1213, | |
| "eval_samples_per_second": 19.599, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.42958956411072224, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "learning_rate": 0.0009903004882677155, | |
| "loss": 0.1521, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 47.22, | |
| "learning_rate": 0.0009884621566680308, | |
| "loss": 0.1531, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 0.0009864664112305846, | |
| "loss": 0.1514, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 52.78, | |
| "learning_rate": 0.0009843138951125682, | |
| "loss": 0.1468, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 55.56, | |
| "learning_rate": 0.000982005301992741, | |
| "loss": 0.1501, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 55.56, | |
| "eval_loss": 0.45478546619415283, | |
| "eval_runtime": 133.376, | |
| "eval_samples_per_second": 19.561, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 0.43805281578110083, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 58.33, | |
| "learning_rate": 0.0009795413758478834, | |
| "loss": 0.1427, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 61.11, | |
| "learning_rate": 0.0009769229107130409, | |
| "loss": 0.1441, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 63.89, | |
| "learning_rate": 0.0009741507504256327, | |
| "loss": 0.1399, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 0.0009712257883535152, | |
| "loss": 0.1379, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 69.44, | |
| "learning_rate": 0.0009681489671070803, | |
| "loss": 0.1382, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 69.44, | |
| "eval_loss": 0.523300290107727, | |
| "eval_runtime": 131.9566, | |
| "eval_samples_per_second": 19.772, | |
| "eval_steps_per_second": 0.25, | |
| "eval_wer": 0.4544066178810054, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 72.22, | |
| "learning_rate": 0.0009649212782354872, | |
| "loss": 0.1372, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 0.0009615437619071199, | |
| "loss": 0.1345, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 77.78, | |
| "learning_rate": 0.0009580175065743799, | |
| "loss": 0.1346, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 80.56, | |
| "learning_rate": 0.0009543436486229162, | |
| "loss": 0.1367, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "learning_rate": 0.0009505233720054086, | |
| "loss": 0.1287, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "eval_loss": 0.48360341787338257, | |
| "eval_runtime": 132.2498, | |
| "eval_samples_per_second": 19.728, | |
| "eval_steps_per_second": 0.25, | |
| "eval_wer": 0.43658924594336623, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 86.11, | |
| "learning_rate": 0.0009465579078600233, | |
| "loss": 0.1226, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 88.89, | |
| "learning_rate": 0.0009424485341136599, | |
| "loss": 0.1258, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 91.67, | |
| "learning_rate": 0.0009381965750701229, | |
| "loss": 0.1222, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 94.44, | |
| "learning_rate": 0.0009338034009833446, | |
| "loss": 0.1182, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 97.22, | |
| "learning_rate": 0.0009292704276158022, | |
| "loss": 0.1164, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 97.22, | |
| "eval_loss": 0.5202445983886719, | |
| "eval_runtime": 132.7793, | |
| "eval_samples_per_second": 19.649, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.43948456888323256, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 0.0009245991157822682, | |
| "loss": 0.1212, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 102.78, | |
| "learning_rate": 0.000919790970879042, | |
| "loss": 0.1139, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 105.56, | |
| "learning_rate": 0.000914847542398815, | |
| "loss": 0.115, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 108.33, | |
| "learning_rate": 0.0009097704234313245, | |
| "loss": 0.1123, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 111.11, | |
| "learning_rate": 0.0009045612501499573, | |
| "loss": 0.1101, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 111.11, | |
| "eval_loss": 0.5120447874069214, | |
| "eval_runtime": 131.3001, | |
| "eval_samples_per_second": 19.871, | |
| "eval_steps_per_second": 0.251, | |
| "eval_wer": 0.43827553293032134, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 113.89, | |
| "learning_rate": 0.0008992217012844713, | |
| "loss": 0.1092, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 116.67, | |
| "learning_rate": 0.0008937534975799996, | |
| "loss": 0.1104, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 119.44, | |
| "learning_rate": 0.0008881584012425165, | |
| "loss": 0.1021, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 122.22, | |
| "learning_rate": 0.0008824382153709421, | |
| "loss": 0.1063, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "learning_rate": 0.0008765947833760678, | |
| "loss": 0.1107, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "eval_loss": 0.5130612850189209, | |
| "eval_runtime": 133.0961, | |
| "eval_samples_per_second": 19.602, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.4446706967865097, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 127.78, | |
| "learning_rate": 0.000870629988386491, | |
| "loss": 0.1054, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 130.56, | |
| "learning_rate": 0.0008645457526417502, | |
| "loss": 0.1037, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 133.33, | |
| "learning_rate": 0.0008583440368728572, | |
| "loss": 0.1, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 136.11, | |
| "learning_rate": 0.0008520268396704223, | |
| "loss": 0.0967, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 138.89, | |
| "learning_rate": 0.0008455961968405809, | |
| "loss": 0.0942, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 138.89, | |
| "eval_loss": 0.5256752967834473, | |
| "eval_runtime": 133.1352, | |
| "eval_samples_per_second": 19.597, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.43748011454024816, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 141.67, | |
| "learning_rate": 0.0008390541807489265, | |
| "loss": 0.0987, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 144.44, | |
| "learning_rate": 0.0008324028996526611, | |
| "loss": 0.094, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 147.22, | |
| "learning_rate": 0.0008256444970211793, | |
| "loss": 0.0935, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 0.000818781150845306, | |
| "loss": 0.0906, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 152.78, | |
| "learning_rate": 0.0008118150729354068, | |
| "loss": 0.0909, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 152.78, | |
| "eval_loss": 0.5312919020652771, | |
| "eval_runtime": 133.5185, | |
| "eval_samples_per_second": 19.54, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 0.47133312122176263, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 155.56, | |
| "learning_rate": 0.0008047485082086017, | |
| "loss": 0.0881, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 158.33, | |
| "learning_rate": 0.0007975837339653084, | |
| "loss": 0.0921, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 161.11, | |
| "learning_rate": 0.0007903230591553503, | |
| "loss": 0.0838, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 163.89, | |
| "learning_rate": 0.0007829688236338643, | |
| "loss": 0.0775, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 166.67, | |
| "learning_rate": 0.0007755233974072504, | |
| "loss": 0.0807, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 166.67, | |
| "eval_loss": 0.5847254991531372, | |
| "eval_runtime": 133.1812, | |
| "eval_samples_per_second": 19.59, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.44708876869233216, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 169.44, | |
| "learning_rate": 0.0007679891798694021, | |
| "loss": 0.0801, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 172.22, | |
| "learning_rate": 0.000760368599028469, | |
| "loss": 0.0952, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 175.0, | |
| "learning_rate": 0.0007526641107243966, | |
| "loss": 0.0813, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 177.78, | |
| "learning_rate": 0.0007448781978374969, | |
| "loss": 0.0795, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 180.56, | |
| "learning_rate": 0.0007370133694883051, | |
| "loss": 0.0767, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 180.56, | |
| "eval_loss": 0.5318175554275513, | |
| "eval_runtime": 135.804, | |
| "eval_samples_per_second": 19.212, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 0.4534521158129176, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 183.33, | |
| "learning_rate": 0.0007290721602289801, | |
| "loss": 0.0762, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 186.11, | |
| "learning_rate": 0.0007210571292265096, | |
| "loss": 0.0729, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 188.89, | |
| "learning_rate": 0.0007129708594379816, | |
| "loss": 0.0751, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 191.67, | |
| "learning_rate": 0.0007048159567781905, | |
| "loss": 0.0743, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 194.44, | |
| "learning_rate": 0.0006965950492798443, | |
| "loss": 0.0729, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 194.44, | |
| "eval_loss": 0.5796982645988464, | |
| "eval_runtime": 132.2018, | |
| "eval_samples_per_second": 19.735, | |
| "eval_steps_per_second": 0.25, | |
| "eval_wer": 0.4584155265669742, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 197.22, | |
| "learning_rate": 0.0006883107862466433, | |
| "loss": 0.0791, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 0.0006799658373995054, | |
| "loss": 0.0737, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 202.78, | |
| "learning_rate": 0.0006715628920162093, | |
| "loss": 0.0694, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 205.56, | |
| "learning_rate": 0.0006631046580647384, | |
| "loss": 0.0687, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 208.33, | |
| "learning_rate": 0.0006545938613305993, | |
| "loss": 0.072, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 208.33, | |
| "eval_loss": 0.5266388654708862, | |
| "eval_runtime": 130.3709, | |
| "eval_samples_per_second": 20.012, | |
| "eval_steps_per_second": 0.253, | |
| "eval_wer": 0.44339802736239264, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 211.11, | |
| "learning_rate": 0.000646033244538398, | |
| "loss": 0.0647, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 213.89, | |
| "learning_rate": 0.000637425566467959, | |
| "loss": 0.0631, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 216.67, | |
| "learning_rate": 0.0006287736010652693, | |
| "loss": 0.0606, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 219.44, | |
| "learning_rate": 0.000620080136548535, | |
| "loss": 0.0645, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 222.22, | |
| "learning_rate": 0.0006113479745096386, | |
| "loss": 0.0593, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 222.22, | |
| "eval_loss": 0.579177737236023, | |
| "eval_runtime": 133.0716, | |
| "eval_samples_per_second": 19.606, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.43888005090677695, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 225.0, | |
| "learning_rate": 0.0006025799290112865, | |
| "loss": 0.0596, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 227.78, | |
| "learning_rate": 0.0005937788256801371, | |
| "loss": 0.0606, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 230.56, | |
| "learning_rate": 0.0005849475007962031, | |
| "loss": 0.0591, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 233.33, | |
| "learning_rate": 0.0005760888003788196, | |
| "loss": 0.0544, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 236.11, | |
| "learning_rate": 0.0005672055792694753, | |
| "loss": 0.0524, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 236.11, | |
| "eval_loss": 0.5740395784378052, | |
| "eval_runtime": 132.5642, | |
| "eval_samples_per_second": 19.681, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.43732103086223356, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 238.89, | |
| "learning_rate": 0.0005583007002117978, | |
| "loss": 0.0504, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 241.67, | |
| "learning_rate": 0.0005493770329289953, | |
| "loss": 0.0494, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 244.44, | |
| "learning_rate": 0.0005404374531990474, | |
| "loss": 0.0484, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 247.22, | |
| "learning_rate": 0.0005314848419279442, | |
| "loss": 0.0489, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "learning_rate": 0.0005225220842212748, | |
| "loss": 0.0494, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "eval_loss": 0.5520852208137512, | |
| "eval_runtime": 133.6258, | |
| "eval_samples_per_second": 19.525, | |
| "eval_steps_per_second": 0.247, | |
| "eval_wer": 0.44358892777601017, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 252.78, | |
| "learning_rate": 0.0005135520684544602, | |
| "loss": 0.0499, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 255.56, | |
| "learning_rate": 0.0005045776853419344, | |
| "loss": 0.0487, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 258.33, | |
| "learning_rate": 0.0004956018270055703, | |
| "loss": 0.0523, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 261.11, | |
| "learning_rate": 0.00048662738604265206, | |
| "loss": 0.0479, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 263.89, | |
| "learning_rate": 0.0004776572545936954, | |
| "loss": 0.0469, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 263.89, | |
| "eval_loss": 0.5630596876144409, | |
| "eval_runtime": 130.7272, | |
| "eval_samples_per_second": 19.958, | |
| "eval_steps_per_second": 0.252, | |
| "eval_wer": 0.4345847916003818, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 266.67, | |
| "learning_rate": 0.0004686943234104146, | |
| "loss": 0.0467, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 269.44, | |
| "learning_rate": 0.00045974148092413597, | |
| "loss": 0.0438, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 272.22, | |
| "learning_rate": 0.00045080161231496064, | |
| "loss": 0.0422, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 275.0, | |
| "learning_rate": 0.00044187759858197486, | |
| "loss": 0.042, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 277.78, | |
| "learning_rate": 0.0004329723156148063, | |
| "loss": 0.0417, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 277.78, | |
| "eval_loss": 0.5777372121810913, | |
| "eval_runtime": 130.7332, | |
| "eval_samples_per_second": 19.957, | |
| "eval_steps_per_second": 0.252, | |
| "eval_wer": 0.43057588291441296, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 280.56, | |
| "learning_rate": 0.0004240886332668285, | |
| "loss": 0.0406, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 283.33, | |
| "learning_rate": 0.00041522941443031025, | |
| "loss": 0.0385, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 286.11, | |
| "learning_rate": 0.0004063975141138074, | |
| "loss": 0.0384, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 288.89, | |
| "learning_rate": 0.00039759577852209704, | |
| "loss": 0.0369, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 291.67, | |
| "learning_rate": 0.0003888270441389465, | |
| "loss": 0.0379, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 291.67, | |
| "eval_loss": 0.5934527516365051, | |
| "eval_runtime": 131.6738, | |
| "eval_samples_per_second": 19.814, | |
| "eval_steps_per_second": 0.251, | |
| "eval_wer": 0.42828507795100224, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 294.44, | |
| "learning_rate": 0.0003800941368130171, | |
| "loss": 0.0399, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 297.22, | |
| "learning_rate": 0.0003713998708471946, | |
| "loss": 0.0359, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "learning_rate": 0.0003627470480916394, | |
| "loss": 0.0345, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 302.78, | |
| "learning_rate": 0.00035413845704085215, | |
| "loss": 0.0346, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 305.56, | |
| "learning_rate": 0.00034557687193504045, | |
| "loss": 0.0319, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 305.56, | |
| "eval_loss": 0.5892442464828491, | |
| "eval_runtime": 130.692, | |
| "eval_samples_per_second": 19.963, | |
| "eval_steps_per_second": 0.253, | |
| "eval_wer": 0.41880369074132995, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 308.33, | |
| "learning_rate": 0.00033706505186608136, | |
| "loss": 0.0318, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 311.11, | |
| "learning_rate": 0.00032860573988836573, | |
| "loss": 0.0301, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 313.89, | |
| "learning_rate": 0.0003202016621348088, | |
| "loss": 0.0322, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 316.67, | |
| "learning_rate": 0.00031185552693831594, | |
| "loss": 0.0331, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 319.44, | |
| "learning_rate": 0.0003035700239589846, | |
| "loss": 0.0311, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 319.44, | |
| "eval_loss": 0.6010639071464539, | |
| "eval_runtime": 132.2371, | |
| "eval_samples_per_second": 19.73, | |
| "eval_steps_per_second": 0.25, | |
| "eval_wer": 0.4210626789691378, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 322.22, | |
| "learning_rate": 0.00029534782331732204, | |
| "loss": 0.0291, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 325.0, | |
| "learning_rate": 0.00028719157473376303, | |
| "loss": 0.0287, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 327.78, | |
| "learning_rate": 0.00027918443530500935, | |
| "loss": 0.0279, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 330.56, | |
| "learning_rate": 0.0002711672294401907, | |
| "loss": 0.0294, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 333.33, | |
| "learning_rate": 0.00026322376817181157, | |
| "loss": 0.0269, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 333.33, | |
| "eval_loss": 0.5972930192947388, | |
| "eval_runtime": 133.0245, | |
| "eval_samples_per_second": 19.613, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.4232262169901368, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 336.11, | |
| "learning_rate": 0.0002553566113926006, | |
| "loss": 0.0251, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 338.89, | |
| "learning_rate": 0.00024756829440508705, | |
| "loss": 0.0261, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 341.67, | |
| "learning_rate": 0.0002398613271045611, | |
| "loss": 0.025, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 344.44, | |
| "learning_rate": 0.00023223819317022654, | |
| "loss": 0.0264, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 347.22, | |
| "learning_rate": 0.00022470134926479935, | |
| "loss": 0.0271, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 347.22, | |
| "eval_loss": 0.6094068288803101, | |
| "eval_runtime": 132.9769, | |
| "eval_samples_per_second": 19.62, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.4428253261215399, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 350.0, | |
| "learning_rate": 0.00021725322424281286, | |
| "loss": 0.0246, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 352.78, | |
| "learning_rate": 0.00020989621836788614, | |
| "loss": 0.0245, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 355.56, | |
| "learning_rate": 0.00020263270253920428, | |
| "loss": 0.0229, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 358.33, | |
| "learning_rate": 0.00019546501752746288, | |
| "loss": 0.0229, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 361.11, | |
| "learning_rate": 0.0001883954732205217, | |
| "loss": 0.0236, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 361.11, | |
| "eval_loss": 0.611167311668396, | |
| "eval_runtime": 132.3797, | |
| "eval_samples_per_second": 19.708, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.42373528475978367, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 363.89, | |
| "learning_rate": 0.00018142634787900996, | |
| "loss": 0.0229, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 366.67, | |
| "learning_rate": 0.00017455988740212576, | |
| "loss": 0.0221, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 369.44, | |
| "learning_rate": 0.0001677983046038624, | |
| "loss": 0.0208, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 372.22, | |
| "learning_rate": 0.00016114377849989813, | |
| "loss": 0.0208, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 375.0, | |
| "learning_rate": 0.00015459845360537756, | |
| "loss": 0.0202, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 375.0, | |
| "eval_loss": 0.605171799659729, | |
| "eval_runtime": 132.1498, | |
| "eval_samples_per_second": 19.743, | |
| "eval_steps_per_second": 0.25, | |
| "eval_wer": 0.4233534839325485, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 377.78, | |
| "learning_rate": 0.0001481644392438113, | |
| "loss": 0.0197, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 380.56, | |
| "learning_rate": 0.00014184380886731607, | |
| "loss": 0.0195, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 383.33, | |
| "learning_rate": 0.0001356385993884165, | |
| "loss": 0.0186, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 386.11, | |
| "learning_rate": 0.00012955081052362072, | |
| "loss": 0.0186, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 388.89, | |
| "learning_rate": 0.00012358240414898276, | |
| "loss": 0.0175, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 388.89, | |
| "eval_loss": 0.631072998046875, | |
| "eval_runtime": 131.5592, | |
| "eval_samples_per_second": 19.831, | |
| "eval_steps_per_second": 0.251, | |
| "eval_wer": 0.4188673242125358, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 391.67, | |
| "learning_rate": 0.00011773530366786184, | |
| "loss": 0.0164, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 394.44, | |
| "learning_rate": 0.0001120113933910767, | |
| "loss": 0.0166, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 397.22, | |
| "learning_rate": 0.000106412517929661, | |
| "loss": 0.0163, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "learning_rate": 0.00010094048160041091, | |
| "loss": 0.0159, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 402.78, | |
| "learning_rate": 9.559704784441814e-05, | |
| "loss": 0.0158, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 402.78, | |
| "eval_loss": 0.618556797504425, | |
| "eval_runtime": 132.4333, | |
| "eval_samples_per_second": 19.7, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.4137130130448616, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 405.56, | |
| "learning_rate": 9.038393865877726e-05, | |
| "loss": 0.015, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 408.33, | |
| "learning_rate": 8.530283404164612e-05, | |
| "loss": 0.0156, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 411.11, | |
| "learning_rate": 8.03553714508441e-05, | |
| "loss": 0.0146, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 413.89, | |
| "learning_rate": 7.554314527615763e-05, | |
| "loss": 0.0146, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 416.67, | |
| "learning_rate": 7.086770632552514e-05, | |
| "loss": 0.0141, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 416.67, | |
| "eval_loss": 0.6352781057357788, | |
| "eval_runtime": 132.6132, | |
| "eval_samples_per_second": 19.674, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.412249443207127, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 419.44, | |
| "learning_rate": 6.633056132526849e-05, | |
| "loss": 0.0138, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 422.22, | |
| "learning_rate": 6.193317243452734e-05, | |
| "loss": 0.0138, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 425.0, | |
| "learning_rate": 5.7676956774059205e-05, | |
| "loss": 0.0131, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 427.78, | |
| "learning_rate": 5.356328596955179e-05, | |
| "loss": 0.0131, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 430.56, | |
| "learning_rate": 4.963246731369253e-05, | |
| "loss": 0.013, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 430.56, | |
| "eval_loss": 0.6343812942504883, | |
| "eval_runtime": 131.5808, | |
| "eval_samples_per_second": 19.828, | |
| "eval_steps_per_second": 0.251, | |
| "eval_wer": 0.40687241489023224, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 433.33, | |
| "learning_rate": 4.580635924447718e-05, | |
| "loss": 0.0127, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 436.11, | |
| "learning_rate": 4.212662149914887e-05, | |
| "loss": 0.0126, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 438.89, | |
| "learning_rate": 3.859443992523443e-05, | |
| "loss": 0.0125, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 441.67, | |
| "learning_rate": 3.521095281819786e-05, | |
| "loss": 0.0127, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 444.44, | |
| "learning_rate": 3.1977250554608205e-05, | |
| "loss": 0.0117, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 444.44, | |
| "eval_loss": 0.6442536115646362, | |
| "eval_runtime": 132.782, | |
| "eval_samples_per_second": 19.649, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.40801781737193765, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 447.22, | |
| "learning_rate": 2.8894375240750613e-05, | |
| "loss": 0.0122, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 450.0, | |
| "learning_rate": 2.5963320376792666e-05, | |
| "loss": 0.0127, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 452.78, | |
| "learning_rate": 2.318503053661475e-05, | |
| "loss": 0.0125, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 455.56, | |
| "learning_rate": 2.056040106340823e-05, | |
| "loss": 0.0121, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 458.33, | |
| "learning_rate": 1.809027778113803e-05, | |
| "loss": 0.0122, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 458.33, | |
| "eval_loss": 0.6512282490730286, | |
| "eval_runtime": 133.1307, | |
| "eval_samples_per_second": 19.597, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 0.4069360483614381, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 461.11, | |
| "learning_rate": 1.5775456721964065e-05, | |
| "loss": 0.0113, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 463.89, | |
| "learning_rate": 1.3616683869708658e-05, | |
| "loss": 0.0113, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 466.67, | |
| "learning_rate": 1.1614654919452406e-05, | |
| "loss": 0.0117, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 469.44, | |
| "learning_rate": 9.770015053337078e-06, | |
| "loss": 0.0112, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 472.22, | |
| "learning_rate": 8.083358732646206e-06, | |
| "loss": 0.0115, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 472.22, | |
| "eval_loss": 0.6556624174118042, | |
| "eval_runtime": 132.4623, | |
| "eval_samples_per_second": 19.696, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.4069360483614381, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 475.0, | |
| "learning_rate": 6.555229506231608e-06, | |
| "loss": 0.0111, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 477.78, | |
| "learning_rate": 5.186119835347169e-06, | |
| "loss": 0.0118, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 480.56, | |
| "learning_rate": 3.976470934946075e-06, | |
| "loss": 0.0112, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 483.33, | |
| "learning_rate": 2.92667263149321e-06, | |
| "loss": 0.0117, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 486.11, | |
| "learning_rate": 2.037063237338044e-06, | |
| "loss": 0.0111, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 486.11, | |
| "eval_loss": 0.6513079404830933, | |
| "eval_runtime": 131.4726, | |
| "eval_samples_per_second": 19.844, | |
| "eval_steps_per_second": 0.251, | |
| "eval_wer": 0.4059497295577474, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 488.89, | |
| "learning_rate": 1.3144256109794418e-06, | |
| "loss": 0.0109, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 491.67, | |
| "learning_rate": 7.44394288673067e-07, | |
| "loss": 0.0112, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 494.44, | |
| "learning_rate": 3.352551457218578e-07, | |
| "loss": 0.0113, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 497.22, | |
| "learning_rate": 8.714003300058781e-08, | |
| "loss": 0.011, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "learning_rate": 1.2890911316176812e-10, | |
| "loss": 0.0113, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "eval_loss": 0.6521313190460205, | |
| "eval_runtime": 132.4384, | |
| "eval_samples_per_second": 19.7, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 0.40601336302895324, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "step": 18000, | |
| "total_flos": 5.3110650126518e+20, | |
| "train_loss": 0.11533763631184896, | |
| "train_runtime": 176050.2174, | |
| "train_samples_per_second": 16.203, | |
| "train_steps_per_second": 0.102 | |
| } | |
| ], | |
| "max_steps": 18000, | |
| "num_train_epochs": 500, | |
| "total_flos": 5.3110650126518e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |