| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 100, | |
| "global_step": 2370, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6329113924050633, | |
| "eval_loss": 3.940936803817749, | |
| "eval_runtime": 314.2197, | |
| "eval_samples_per_second": 8.045, | |
| "eval_steps_per_second": 1.006, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2658227848101267, | |
| "eval_loss": 3.044066905975342, | |
| "eval_runtime": 313.2217, | |
| "eval_samples_per_second": 8.071, | |
| "eval_steps_per_second": 1.009, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8987341772151898, | |
| "eval_loss": 2.9164648056030273, | |
| "eval_runtime": 317.0806, | |
| "eval_samples_per_second": 7.973, | |
| "eval_steps_per_second": 0.997, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.5316455696202533, | |
| "eval_loss": 1.4924770593643188, | |
| "eval_runtime": 316.4362, | |
| "eval_samples_per_second": 7.989, | |
| "eval_steps_per_second": 0.999, | |
| "eval_wer": 1.9968354430379747, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1645569620253164, | |
| "grad_norm": 0.8539223670959473, | |
| "learning_rate": 0.0002982, | |
| "loss": 3.7012, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.1645569620253164, | |
| "eval_loss": 0.30101653933525085, | |
| "eval_runtime": 312.8326, | |
| "eval_samples_per_second": 8.081, | |
| "eval_steps_per_second": 1.01, | |
| "eval_wer": 1.9446202531645569, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.7974683544303796, | |
| "eval_loss": 0.17126257717609406, | |
| "eval_runtime": 314.312, | |
| "eval_samples_per_second": 8.043, | |
| "eval_steps_per_second": 1.005, | |
| "eval_wer": 1.8259493670886076, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.430379746835443, | |
| "eval_loss": 0.09897234290838242, | |
| "eval_runtime": 312.6958, | |
| "eval_samples_per_second": 8.085, | |
| "eval_steps_per_second": 1.011, | |
| "eval_wer": 1.6162974683544304, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.063291139240507, | |
| "eval_loss": 0.06915320456027985, | |
| "eval_runtime": 317.6492, | |
| "eval_samples_per_second": 7.958, | |
| "eval_steps_per_second": 0.995, | |
| "eval_wer": 1.5439082278481013, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.69620253164557, | |
| "eval_loss": 0.046260952949523926, | |
| "eval_runtime": 321.8259, | |
| "eval_samples_per_second": 7.855, | |
| "eval_steps_per_second": 0.982, | |
| "eval_wer": 1.423259493670886, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.329113924050633, | |
| "grad_norm": 0.26903435587882996, | |
| "learning_rate": 0.00022026737967914436, | |
| "loss": 0.1686, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.329113924050633, | |
| "eval_loss": 0.038907092064619064, | |
| "eval_runtime": 320.4174, | |
| "eval_samples_per_second": 7.89, | |
| "eval_steps_per_second": 0.986, | |
| "eval_wer": 1.3469145569620253, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.962025316455696, | |
| "eval_loss": 0.029044821858406067, | |
| "eval_runtime": 320.649, | |
| "eval_samples_per_second": 7.884, | |
| "eval_steps_per_second": 0.986, | |
| "eval_wer": 1.3101265822784811, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.594936708860759, | |
| "eval_loss": 0.020351797342300415, | |
| "eval_runtime": 316.4006, | |
| "eval_samples_per_second": 7.99, | |
| "eval_steps_per_second": 0.999, | |
| "eval_wer": 1.1993670886075949, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.227848101265822, | |
| "eval_loss": 0.016085166484117508, | |
| "eval_runtime": 321.3591, | |
| "eval_samples_per_second": 7.867, | |
| "eval_steps_per_second": 0.983, | |
| "eval_wer": 1.1839398734177216, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.860759493670885, | |
| "eval_loss": 0.014270616695284843, | |
| "eval_runtime": 314.8714, | |
| "eval_samples_per_second": 8.029, | |
| "eval_steps_per_second": 1.004, | |
| "eval_wer": 1.1499208860759493, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.49367088607595, | |
| "grad_norm": 1.2659544944763184, | |
| "learning_rate": 0.00014005347593582887, | |
| "loss": 0.0553, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.49367088607595, | |
| "eval_loss": 0.011028471402823925, | |
| "eval_runtime": 316.4518, | |
| "eval_samples_per_second": 7.989, | |
| "eval_steps_per_second": 0.999, | |
| "eval_wer": 1.1459651898734178, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.126582278481013, | |
| "eval_loss": 0.008157163858413696, | |
| "eval_runtime": 315.7243, | |
| "eval_samples_per_second": 8.007, | |
| "eval_steps_per_second": 1.001, | |
| "eval_wer": 1.0953322784810127, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.759493670886076, | |
| "eval_loss": 0.008831300772726536, | |
| "eval_runtime": 315.0431, | |
| "eval_samples_per_second": 8.024, | |
| "eval_steps_per_second": 1.003, | |
| "eval_wer": 1.1119462025316456, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.39240506329114, | |
| "eval_loss": 0.005905392114073038, | |
| "eval_runtime": 319.3821, | |
| "eval_samples_per_second": 7.915, | |
| "eval_steps_per_second": 0.989, | |
| "eval_wer": 1.0573575949367089, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.025316455696203, | |
| "eval_loss": 0.005364276003092527, | |
| "eval_runtime": 316.4189, | |
| "eval_samples_per_second": 7.989, | |
| "eval_steps_per_second": 0.999, | |
| "eval_wer": 1.0510284810126582, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 12.658227848101266, | |
| "grad_norm": 0.34013208746910095, | |
| "learning_rate": 5.983957219251336e-05, | |
| "loss": 0.0295, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.658227848101266, | |
| "eval_loss": 0.004200654104351997, | |
| "eval_runtime": 316.6068, | |
| "eval_samples_per_second": 7.985, | |
| "eval_steps_per_second": 0.998, | |
| "eval_wer": 1.0356012658227849, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.291139240506329, | |
| "eval_loss": 0.003917561378329992, | |
| "eval_runtime": 317.6825, | |
| "eval_samples_per_second": 7.958, | |
| "eval_steps_per_second": 0.995, | |
| "eval_wer": 1.035996835443038, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 13.924050632911392, | |
| "eval_loss": 0.0033297832123935223, | |
| "eval_runtime": 317.8207, | |
| "eval_samples_per_second": 7.954, | |
| "eval_steps_per_second": 0.994, | |
| "eval_wer": 1.0268987341772151, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 14.556962025316455, | |
| "eval_loss": 0.0030845776200294495, | |
| "eval_runtime": 316.4074, | |
| "eval_samples_per_second": 7.99, | |
| "eval_steps_per_second": 0.999, | |
| "eval_wer": 1.0237341772151898, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 2370, | |
| "total_flos": 7.984243073297488e+18, | |
| "train_loss": 0.8373936769831533, | |
| "train_runtime": 15099.3203, | |
| "train_samples_per_second": 2.511, | |
| "train_steps_per_second": 0.157 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2370, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 7.984243073297488e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |