| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 200.0, | |
| "global_step": 10400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 0.000148, | |
| "loss": 5.649, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "eval_loss": 3.0038444995880127, | |
| "eval_runtime": 14.1536, | |
| "eval_samples_per_second": 20.772, | |
| "eval_steps_per_second": 0.707, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 0.00029800000000000003, | |
| "loss": 2.9978, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 0.00039600000000000003, | |
| "loss": 1.6272, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_loss": 0.7362223267555237, | |
| "eval_runtime": 13.8179, | |
| "eval_samples_per_second": 21.277, | |
| "eval_steps_per_second": 0.724, | |
| "eval_wer": 0.781941309255079, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 0.0003835, | |
| "loss": 1.2822, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 28.85, | |
| "learning_rate": 0.000371, | |
| "loss": 1.1354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 28.85, | |
| "eval_loss": 0.6409761309623718, | |
| "eval_runtime": 13.8178, | |
| "eval_samples_per_second": 21.277, | |
| "eval_steps_per_second": 0.724, | |
| "eval_wer": 0.7110609480812641, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 34.62, | |
| "learning_rate": 0.00035850000000000004, | |
| "loss": 1.0424, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_loss": 0.6907294392585754, | |
| "eval_runtime": 13.9619, | |
| "eval_samples_per_second": 21.057, | |
| "eval_steps_per_second": 0.716, | |
| "eval_wer": 0.7431151241534989, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 40.38, | |
| "learning_rate": 0.000346, | |
| "loss": 0.9872, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 0.00033350000000000003, | |
| "loss": 0.9293, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 48.08, | |
| "eval_loss": 0.7248561978340149, | |
| "eval_runtime": 14.1849, | |
| "eval_samples_per_second": 20.726, | |
| "eval_steps_per_second": 0.705, | |
| "eval_wer": 0.7101580135440181, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 51.92, | |
| "learning_rate": 0.000321, | |
| "loss": 0.8747, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "learning_rate": 0.0003085, | |
| "loss": 0.8246, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "eval_loss": 0.7421836853027344, | |
| "eval_runtime": 14.4192, | |
| "eval_samples_per_second": 20.39, | |
| "eval_steps_per_second": 0.694, | |
| "eval_wer": 0.6966139954853273, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 63.46, | |
| "learning_rate": 0.000296, | |
| "loss": 0.7837, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 67.31, | |
| "eval_loss": 0.7412946820259094, | |
| "eval_runtime": 14.1823, | |
| "eval_samples_per_second": 20.73, | |
| "eval_steps_per_second": 0.705, | |
| "eval_wer": 0.6812641083521445, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 69.23, | |
| "learning_rate": 0.0002835, | |
| "loss": 0.7527, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 0.00027100000000000003, | |
| "loss": 0.7147, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "eval_loss": 0.7873469591140747, | |
| "eval_runtime": 13.9067, | |
| "eval_samples_per_second": 21.141, | |
| "eval_steps_per_second": 0.719, | |
| "eval_wer": 0.6930022573363431, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 80.77, | |
| "learning_rate": 0.0002585, | |
| "loss": 0.6779, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 86.54, | |
| "learning_rate": 0.000246, | |
| "loss": 0.6276, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 86.54, | |
| "eval_loss": 0.8037810921669006, | |
| "eval_runtime": 14.0837, | |
| "eval_samples_per_second": 20.875, | |
| "eval_steps_per_second": 0.71, | |
| "eval_wer": 0.6677200902934537, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 92.31, | |
| "learning_rate": 0.0002335, | |
| "loss": 0.6041, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "eval_loss": 0.8240488767623901, | |
| "eval_runtime": 13.9338, | |
| "eval_samples_per_second": 21.1, | |
| "eval_steps_per_second": 0.718, | |
| "eval_wer": 0.6830699774266366, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 98.08, | |
| "learning_rate": 0.000221, | |
| "loss": 0.5588, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 103.85, | |
| "learning_rate": 0.0002085, | |
| "loss": 0.5336, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 105.77, | |
| "eval_loss": 0.8747946619987488, | |
| "eval_runtime": 13.9807, | |
| "eval_samples_per_second": 21.029, | |
| "eval_steps_per_second": 0.715, | |
| "eval_wer": 0.6749435665914221, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 109.62, | |
| "learning_rate": 0.000196, | |
| "loss": 0.5, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "learning_rate": 0.00018350000000000002, | |
| "loss": 0.4705, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "eval_loss": 0.9005643129348755, | |
| "eval_runtime": 13.867, | |
| "eval_samples_per_second": 21.201, | |
| "eval_steps_per_second": 0.721, | |
| "eval_wer": 0.6496613995485327, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 121.15, | |
| "learning_rate": 0.00017104166666666667, | |
| "loss": 0.43, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "eval_loss": 0.8953593969345093, | |
| "eval_runtime": 13.9621, | |
| "eval_samples_per_second": 21.057, | |
| "eval_steps_per_second": 0.716, | |
| "eval_wer": 0.655079006772009, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 126.92, | |
| "learning_rate": 0.00015854166666666667, | |
| "loss": 0.4068, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 132.69, | |
| "learning_rate": 0.0001460416666666667, | |
| "loss": 0.3859, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 134.62, | |
| "eval_loss": 0.9073536396026611, | |
| "eval_runtime": 13.9518, | |
| "eval_samples_per_second": 21.072, | |
| "eval_steps_per_second": 0.717, | |
| "eval_wer": 0.6613995485327314, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 138.46, | |
| "learning_rate": 0.00013354166666666668, | |
| "loss": 0.3622, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 144.23, | |
| "learning_rate": 0.00012104166666666668, | |
| "loss": 0.3342, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 144.23, | |
| "eval_loss": 0.9693499803543091, | |
| "eval_runtime": 13.8467, | |
| "eval_samples_per_second": 21.233, | |
| "eval_steps_per_second": 0.722, | |
| "eval_wer": 0.6559819413092551, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 0.00010854166666666667, | |
| "loss": 0.3155, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "eval_loss": 1.0072590112686157, | |
| "eval_runtime": 13.7449, | |
| "eval_samples_per_second": 21.39, | |
| "eval_steps_per_second": 0.728, | |
| "eval_wer": 0.6690744920993228, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 155.77, | |
| "learning_rate": 9.604166666666668e-05, | |
| "loss": 0.2894, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 161.54, | |
| "learning_rate": 8.358333333333334e-05, | |
| "loss": 0.2673, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 163.46, | |
| "eval_loss": 1.0170269012451172, | |
| "eval_runtime": 14.0595, | |
| "eval_samples_per_second": 20.911, | |
| "eval_steps_per_second": 0.711, | |
| "eval_wer": 0.6632054176072235, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 167.31, | |
| "learning_rate": 7.108333333333333e-05, | |
| "loss": 0.2517, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 173.08, | |
| "learning_rate": 5.858333333333333e-05, | |
| "loss": 0.2409, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 173.08, | |
| "eval_loss": 1.0304286479949951, | |
| "eval_runtime": 13.8942, | |
| "eval_samples_per_second": 21.16, | |
| "eval_steps_per_second": 0.72, | |
| "eval_wer": 0.6708803611738149, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 178.85, | |
| "learning_rate": 4.608333333333333e-05, | |
| "loss": 0.2189, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 182.69, | |
| "eval_loss": 0.9965260624885559, | |
| "eval_runtime": 14.2442, | |
| "eval_samples_per_second": 20.64, | |
| "eval_steps_per_second": 0.702, | |
| "eval_wer": 0.654627539503386, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 184.62, | |
| "learning_rate": 3.3625000000000004e-05, | |
| "loss": 0.203, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 190.38, | |
| "learning_rate": 2.1125000000000002e-05, | |
| "loss": 0.1973, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "eval_loss": 1.0360474586486816, | |
| "eval_runtime": 14.1087, | |
| "eval_samples_per_second": 20.838, | |
| "eval_steps_per_second": 0.709, | |
| "eval_wer": 0.655079006772009, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 196.15, | |
| "learning_rate": 8.625e-06, | |
| "loss": 0.1881, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "step": 10400, | |
| "total_flos": 3.0179570579437056e+19, | |
| "train_loss": 0.8014375554598295, | |
| "train_runtime": 12807.5215, | |
| "train_samples_per_second": 12.93, | |
| "train_steps_per_second": 0.812 | |
| } | |
| ], | |
| "max_steps": 10400, | |
| "num_train_epochs": 200, | |
| "total_flos": 3.0179570579437056e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |