| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 499.97196261682245, | |
| "global_step": 13000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 25.9068, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 7.4625e-06, | |
| "loss": 12.5481, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 7.2704, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.37, | |
| "learning_rate": 1.49625e-05, | |
| "loss": 5.7673, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.22, | |
| "learning_rate": 1.8712499999999997e-05, | |
| "loss": 4.6694, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 19.22, | |
| "eval_loss": 4.045496463775635, | |
| "eval_runtime": 13.5539, | |
| "eval_samples_per_second": 26.561, | |
| "eval_steps_per_second": 3.32, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 23.07, | |
| "learning_rate": 2.2462499999999997e-05, | |
| "loss": 3.9455, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 26.9, | |
| "learning_rate": 2.6212499999999997e-05, | |
| "loss": 3.583, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 30.75, | |
| "learning_rate": 2.99625e-05, | |
| "loss": 3.5138, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 34.6, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 3.4479, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 38.45, | |
| "learning_rate": 3.7462499999999996e-05, | |
| "loss": 3.3907, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 38.45, | |
| "eval_loss": 3.283632516860962, | |
| "eval_runtime": 13.4556, | |
| "eval_samples_per_second": 26.755, | |
| "eval_steps_per_second": 3.344, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 42.3, | |
| "learning_rate": 4.12125e-05, | |
| "loss": 3.3556, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 3.3117, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 49.97, | |
| "learning_rate": 4.871249999999999e-05, | |
| "loss": 3.0309, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 53.82, | |
| "learning_rate": 5.2462499999999994e-05, | |
| "loss": 2.5273, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 57.67, | |
| "learning_rate": 5.62125e-05, | |
| "loss": 2.0866, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 57.67, | |
| "eval_loss": 1.2787948846817017, | |
| "eval_runtime": 13.4904, | |
| "eval_samples_per_second": 26.686, | |
| "eval_steps_per_second": 3.336, | |
| "eval_wer": 0.7714621716973736, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 61.52, | |
| "learning_rate": 5.9962499999999994e-05, | |
| "loss": 1.8478, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 65.37, | |
| "learning_rate": 6.37125e-05, | |
| "loss": 1.7048, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 69.22, | |
| "learning_rate": 6.7425e-05, | |
| "loss": 1.6017, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 73.07, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 1.4931, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 76.9, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.4106, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 76.9, | |
| "eval_loss": 0.7866009473800659, | |
| "eval_runtime": 13.5793, | |
| "eval_samples_per_second": 26.511, | |
| "eval_steps_per_second": 3.314, | |
| "eval_wer": 0.689141513132105, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 80.75, | |
| "learning_rate": 7.433863636363635e-05, | |
| "loss": 1.3808, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 84.6, | |
| "learning_rate": 7.365681818181818e-05, | |
| "loss": 1.3114, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 88.45, | |
| "learning_rate": 7.2975e-05, | |
| "loss": 1.2661, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 92.3, | |
| "learning_rate": 7.229318181818181e-05, | |
| "loss": 1.2103, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "learning_rate": 7.161136363636363e-05, | |
| "loss": 1.1711, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "eval_loss": 0.655582070350647, | |
| "eval_runtime": 13.3729, | |
| "eval_samples_per_second": 26.92, | |
| "eval_steps_per_second": 3.365, | |
| "eval_wer": 0.6272050176401411, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 99.97, | |
| "learning_rate": 7.092954545454545e-05, | |
| "loss": 1.1192, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 103.82, | |
| "learning_rate": 7.024772727272726e-05, | |
| "loss": 1.0931, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 107.67, | |
| "learning_rate": 6.956590909090908e-05, | |
| "loss": 1.0688, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 111.52, | |
| "learning_rate": 6.88840909090909e-05, | |
| "loss": 1.0559, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 115.37, | |
| "learning_rate": 6.82090909090909e-05, | |
| "loss": 1.038, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 115.37, | |
| "eval_loss": 0.6194556951522827, | |
| "eval_runtime": 15.6261, | |
| "eval_samples_per_second": 23.038, | |
| "eval_steps_per_second": 2.88, | |
| "eval_wer": 0.5680125441003528, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 119.22, | |
| "learning_rate": 6.752727272727272e-05, | |
| "loss": 1.0038, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 123.07, | |
| "learning_rate": 6.684545454545454e-05, | |
| "loss": 0.9754, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 126.9, | |
| "learning_rate": 6.616363636363636e-05, | |
| "loss": 0.9372, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 130.75, | |
| "learning_rate": 6.548181818181817e-05, | |
| "loss": 0.9232, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 134.6, | |
| "learning_rate": 6.479999999999999e-05, | |
| "loss": 0.8989, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 134.6, | |
| "eval_loss": 0.6562529802322388, | |
| "eval_runtime": 13.2747, | |
| "eval_samples_per_second": 27.119, | |
| "eval_steps_per_second": 3.39, | |
| "eval_wer": 0.560172481379851, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 138.45, | |
| "learning_rate": 6.411818181818182e-05, | |
| "loss": 0.882, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 142.3, | |
| "learning_rate": 6.343636363636364e-05, | |
| "loss": 0.868, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 146.15, | |
| "learning_rate": 6.275454545454545e-05, | |
| "loss": 0.8438, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 149.97, | |
| "learning_rate": 6.207272727272727e-05, | |
| "loss": 0.8209, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 153.82, | |
| "learning_rate": 6.139090909090908e-05, | |
| "loss": 0.8021, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 153.82, | |
| "eval_loss": 0.6643799543380737, | |
| "eval_runtime": 13.2721, | |
| "eval_samples_per_second": 27.125, | |
| "eval_steps_per_second": 3.391, | |
| "eval_wer": 0.5327322618580949, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 157.67, | |
| "learning_rate": 6.07090909090909e-05, | |
| "loss": 0.7682, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 161.52, | |
| "learning_rate": 6.0027272727272725e-05, | |
| "loss": 0.7496, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 165.37, | |
| "learning_rate": 5.934545454545454e-05, | |
| "loss": 0.7427, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 169.22, | |
| "learning_rate": 5.866363636363636e-05, | |
| "loss": 0.7336, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 173.07, | |
| "learning_rate": 5.7981818181818174e-05, | |
| "loss": 0.7161, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 173.07, | |
| "eval_loss": 0.6844300031661987, | |
| "eval_runtime": 13.407, | |
| "eval_samples_per_second": 26.852, | |
| "eval_steps_per_second": 3.356, | |
| "eval_wer": 0.5252842022736182, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 176.9, | |
| "learning_rate": 5.73e-05, | |
| "loss": 0.6994, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 180.75, | |
| "learning_rate": 5.6618181818181806e-05, | |
| "loss": 0.6778, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 184.6, | |
| "learning_rate": 5.593636363636363e-05, | |
| "loss": 0.6709, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 188.45, | |
| "learning_rate": 5.5254545454545446e-05, | |
| "loss": 0.6551, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 192.3, | |
| "learning_rate": 5.457272727272727e-05, | |
| "loss": 0.6449, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 192.3, | |
| "eval_loss": 0.7017893195152283, | |
| "eval_runtime": 13.3622, | |
| "eval_samples_per_second": 26.942, | |
| "eval_steps_per_second": 3.368, | |
| "eval_wer": 0.5331242649941199, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 196.15, | |
| "learning_rate": 5.389090909090909e-05, | |
| "loss": 0.6229, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 199.97, | |
| "learning_rate": 5.32090909090909e-05, | |
| "loss": 0.5977, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 203.82, | |
| "learning_rate": 5.2527272727272725e-05, | |
| "loss": 0.5985, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 207.67, | |
| "learning_rate": 5.185227272727272e-05, | |
| "loss": 0.5819, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 211.52, | |
| "learning_rate": 5.117045454545454e-05, | |
| "loss": 0.5659, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 211.52, | |
| "eval_loss": 0.7450936436653137, | |
| "eval_runtime": 13.5305, | |
| "eval_samples_per_second": 26.606, | |
| "eval_steps_per_second": 3.326, | |
| "eval_wer": 0.546452371618973, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 215.37, | |
| "learning_rate": 5.048863636363636e-05, | |
| "loss": 0.5615, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 219.22, | |
| "learning_rate": 4.980681818181818e-05, | |
| "loss": 0.5475, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 223.07, | |
| "learning_rate": 4.9124999999999995e-05, | |
| "loss": 0.529, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 226.9, | |
| "learning_rate": 4.844318181818181e-05, | |
| "loss": 0.5217, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 230.75, | |
| "learning_rate": 4.7761363636363634e-05, | |
| "loss": 0.5118, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 230.75, | |
| "eval_loss": 0.7857084274291992, | |
| "eval_runtime": 13.2806, | |
| "eval_samples_per_second": 27.107, | |
| "eval_steps_per_second": 3.388, | |
| "eval_wer": 0.5386123088984712, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 234.6, | |
| "learning_rate": 4.707954545454545e-05, | |
| "loss": 0.4996, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 238.45, | |
| "learning_rate": 4.639772727272727e-05, | |
| "loss": 0.4808, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 242.3, | |
| "learning_rate": 4.5715909090909083e-05, | |
| "loss": 0.4632, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 246.15, | |
| "learning_rate": 4.5034090909090907e-05, | |
| "loss": 0.4643, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 249.97, | |
| "learning_rate": 4.435227272727272e-05, | |
| "loss": 0.4385, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 249.97, | |
| "eval_loss": 0.8062427043914795, | |
| "eval_runtime": 13.503, | |
| "eval_samples_per_second": 26.661, | |
| "eval_steps_per_second": 3.333, | |
| "eval_wer": 0.5382203057624461, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 253.82, | |
| "learning_rate": 4.367045454545454e-05, | |
| "loss": 0.439, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 257.67, | |
| "learning_rate": 4.298863636363636e-05, | |
| "loss": 0.4274, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 261.52, | |
| "learning_rate": 4.230681818181818e-05, | |
| "loss": 0.4177, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 265.37, | |
| "learning_rate": 4.1625e-05, | |
| "loss": 0.411, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 269.22, | |
| "learning_rate": 4.094318181818181e-05, | |
| "loss": 0.3984, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 269.22, | |
| "eval_loss": 0.8315810561180115, | |
| "eval_runtime": 13.3889, | |
| "eval_samples_per_second": 26.888, | |
| "eval_steps_per_second": 3.361, | |
| "eval_wer": 0.5621324970599765, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 273.07, | |
| "learning_rate": 4.0261363636363635e-05, | |
| "loss": 0.3935, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 276.9, | |
| "learning_rate": 3.957954545454545e-05, | |
| "loss": 0.3842, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 280.75, | |
| "learning_rate": 3.8897727272727274e-05, | |
| "loss": 0.3756, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 284.6, | |
| "learning_rate": 3.8215909090909084e-05, | |
| "loss": 0.3671, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 288.45, | |
| "learning_rate": 3.754090909090909e-05, | |
| "loss": 0.3666, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 288.45, | |
| "eval_loss": 0.8735576868057251, | |
| "eval_runtime": 13.3799, | |
| "eval_samples_per_second": 26.906, | |
| "eval_steps_per_second": 3.363, | |
| "eval_wer": 0.5503724029792239, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 292.3, | |
| "learning_rate": 3.6859090909090904e-05, | |
| "loss": 0.3454, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 296.15, | |
| "learning_rate": 3.617727272727273e-05, | |
| "loss": 0.3486, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 299.97, | |
| "learning_rate": 3.5495454545454544e-05, | |
| "loss": 0.3347, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 303.82, | |
| "learning_rate": 3.481363636363636e-05, | |
| "loss": 0.3262, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 307.67, | |
| "learning_rate": 3.413181818181818e-05, | |
| "loss": 0.3256, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 307.67, | |
| "eval_loss": 0.9133402705192566, | |
| "eval_runtime": 13.8839, | |
| "eval_samples_per_second": 25.929, | |
| "eval_steps_per_second": 3.241, | |
| "eval_wer": 0.568796550372403, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 311.52, | |
| "learning_rate": 3.345e-05, | |
| "loss": 0.3062, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 315.37, | |
| "learning_rate": 3.2768181818181816e-05, | |
| "loss": 0.3053, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 319.22, | |
| "learning_rate": 3.208636363636363e-05, | |
| "loss": 0.3003, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 323.07, | |
| "learning_rate": 3.140454545454545e-05, | |
| "loss": 0.2975, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 326.9, | |
| "learning_rate": 3.072272727272727e-05, | |
| "loss": 0.289, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 326.9, | |
| "eval_loss": 0.9556481242179871, | |
| "eval_runtime": 13.3982, | |
| "eval_samples_per_second": 26.869, | |
| "eval_steps_per_second": 3.359, | |
| "eval_wer": 0.5684045472363779, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 330.75, | |
| "learning_rate": 3.004090909090909e-05, | |
| "loss": 0.2802, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 334.6, | |
| "learning_rate": 2.9359090909090905e-05, | |
| "loss": 0.2736, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 338.45, | |
| "learning_rate": 2.8677272727272725e-05, | |
| "loss": 0.2683, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 342.3, | |
| "learning_rate": 2.799545454545454e-05, | |
| "loss": 0.2681, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 346.15, | |
| "learning_rate": 2.731363636363636e-05, | |
| "loss": 0.2663, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 346.15, | |
| "eval_loss": 0.9343604445457458, | |
| "eval_runtime": 13.4662, | |
| "eval_samples_per_second": 26.734, | |
| "eval_steps_per_second": 3.342, | |
| "eval_wer": 0.5707565660525284, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 349.97, | |
| "learning_rate": 2.6631818181818177e-05, | |
| "loss": 0.2597, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 353.82, | |
| "learning_rate": 2.5949999999999997e-05, | |
| "loss": 0.2636, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 357.67, | |
| "learning_rate": 2.5268181818181817e-05, | |
| "loss": 0.2538, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 361.52, | |
| "learning_rate": 2.4586363636363637e-05, | |
| "loss": 0.2505, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 365.37, | |
| "learning_rate": 2.3904545454545453e-05, | |
| "loss": 0.2445, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 365.37, | |
| "eval_loss": 0.9471534490585327, | |
| "eval_runtime": 13.371, | |
| "eval_samples_per_second": 26.924, | |
| "eval_steps_per_second": 3.365, | |
| "eval_wer": 0.5589964719717758, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 369.22, | |
| "learning_rate": 2.3222727272727273e-05, | |
| "loss": 0.2397, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 373.07, | |
| "learning_rate": 2.254090909090909e-05, | |
| "loss": 0.2395, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 376.9, | |
| "learning_rate": 2.185909090909091e-05, | |
| "loss": 0.2277, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 380.75, | |
| "learning_rate": 2.1184090909090906e-05, | |
| "loss": 0.2305, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 384.6, | |
| "learning_rate": 2.0509090909090907e-05, | |
| "loss": 0.2289, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 384.6, | |
| "eval_loss": 0.9713197350502014, | |
| "eval_runtime": 13.5048, | |
| "eval_samples_per_second": 26.657, | |
| "eval_steps_per_second": 3.332, | |
| "eval_wer": 0.5672285378283026, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 388.45, | |
| "learning_rate": 1.9827272727272727e-05, | |
| "loss": 0.2139, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 392.3, | |
| "learning_rate": 1.9145454545454543e-05, | |
| "loss": 0.2214, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 396.15, | |
| "learning_rate": 1.846363636363636e-05, | |
| "loss": 0.2053, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 399.97, | |
| "learning_rate": 1.7781818181818183e-05, | |
| "loss": 0.2007, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 403.82, | |
| "learning_rate": 1.71e-05, | |
| "loss": 0.2048, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 403.82, | |
| "eval_loss": 0.9977540969848633, | |
| "eval_runtime": 15.7748, | |
| "eval_samples_per_second": 22.821, | |
| "eval_steps_per_second": 2.853, | |
| "eval_wer": 0.5762446099568796, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 407.67, | |
| "learning_rate": 1.641818181818182e-05, | |
| "loss": 0.1995, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 411.52, | |
| "learning_rate": 1.5736363636363635e-05, | |
| "loss": 0.1994, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 415.37, | |
| "learning_rate": 1.5054545454545453e-05, | |
| "loss": 0.2005, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 419.22, | |
| "learning_rate": 1.4372727272727271e-05, | |
| "loss": 0.1988, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 423.07, | |
| "learning_rate": 1.369090909090909e-05, | |
| "loss": 0.1857, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 423.07, | |
| "eval_loss": 1.0229874849319458, | |
| "eval_runtime": 13.4188, | |
| "eval_samples_per_second": 26.828, | |
| "eval_steps_per_second": 3.354, | |
| "eval_wer": 0.5797726381811055, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 426.9, | |
| "learning_rate": 1.3009090909090907e-05, | |
| "loss": 0.1793, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 430.75, | |
| "learning_rate": 1.2327272727272726e-05, | |
| "loss": 0.1835, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 434.6, | |
| "learning_rate": 1.1645454545454545e-05, | |
| "loss": 0.1729, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 438.45, | |
| "learning_rate": 1.0963636363636363e-05, | |
| "loss": 0.1751, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 442.3, | |
| "learning_rate": 1.0281818181818181e-05, | |
| "loss": 0.1751, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 442.3, | |
| "eval_loss": 1.0409098863601685, | |
| "eval_runtime": 13.4411, | |
| "eval_samples_per_second": 26.784, | |
| "eval_steps_per_second": 3.348, | |
| "eval_wer": 0.5754606036848294, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 446.15, | |
| "learning_rate": 9.6e-06, | |
| "loss": 0.1743, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 449.97, | |
| "learning_rate": 8.918181818181818e-06, | |
| "loss": 0.1704, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 453.82, | |
| "learning_rate": 8.236363636363636e-06, | |
| "loss": 0.1691, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 457.67, | |
| "learning_rate": 7.554545454545453e-06, | |
| "loss": 0.1635, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 461.52, | |
| "learning_rate": 6.872727272727273e-06, | |
| "loss": 0.1688, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 461.52, | |
| "eval_loss": 1.0444588661193848, | |
| "eval_runtime": 13.5066, | |
| "eval_samples_per_second": 26.654, | |
| "eval_steps_per_second": 3.332, | |
| "eval_wer": 0.5727165817326538, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 465.37, | |
| "learning_rate": 6.190909090909091e-06, | |
| "loss": 0.1666, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 469.22, | |
| "learning_rate": 5.509090909090908e-06, | |
| "loss": 0.1619, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 473.07, | |
| "learning_rate": 4.827272727272726e-06, | |
| "loss": 0.161, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 476.9, | |
| "learning_rate": 4.145454545454545e-06, | |
| "loss": 0.1527, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 480.75, | |
| "learning_rate": 3.4636363636363636e-06, | |
| "loss": 0.1633, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 480.75, | |
| "eval_loss": 1.048424482345581, | |
| "eval_runtime": 13.4794, | |
| "eval_samples_per_second": 26.707, | |
| "eval_steps_per_second": 3.338, | |
| "eval_wer": 0.5738925911407291, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 484.6, | |
| "learning_rate": 2.7818181818181817e-06, | |
| "loss": 0.1604, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 488.45, | |
| "learning_rate": 2.1e-06, | |
| "loss": 0.156, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 492.3, | |
| "learning_rate": 1.418181818181818e-06, | |
| "loss": 0.1579, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 496.15, | |
| "learning_rate": 7.363636363636363e-07, | |
| "loss": 0.1557, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 499.97, | |
| "learning_rate": 5.454545454545454e-08, | |
| "loss": 0.1488, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 499.97, | |
| "eval_loss": 1.0443373918533325, | |
| "eval_runtime": 13.3989, | |
| "eval_samples_per_second": 26.868, | |
| "eval_steps_per_second": 3.358, | |
| "eval_wer": 0.5715405723245786, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 499.97, | |
| "step": 13000, | |
| "total_flos": 6.567667376811244e+19, | |
| "train_loss": 1.1417412889187153, | |
| "train_runtime": 29000.6737, | |
| "train_samples_per_second": 14.758, | |
| "train_steps_per_second": 0.448 | |
| } | |
| ], | |
| "max_steps": 13000, | |
| "num_train_epochs": 500, | |
| "total_flos": 6.567667376811244e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |