| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.2075471698113207, | |
| "eval_steps": 500, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07547169811320754, | |
| "grad_norm": 9.590713500976562, | |
| "learning_rate": 0.00018449999999999999, | |
| "loss": 5.1054, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07547169811320754, | |
| "eval_loss": 2.816082239151001, | |
| "eval_runtime": 195.1014, | |
| "eval_samples_per_second": 35.992, | |
| "eval_steps_per_second": 0.564, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "grad_norm": 6.886263847351074, | |
| "learning_rate": 0.00029204166666666663, | |
| "loss": 1.4196, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "eval_loss": 1.168688178062439, | |
| "eval_runtime": 193.3054, | |
| "eval_samples_per_second": 36.326, | |
| "eval_steps_per_second": 0.569, | |
| "eval_wer": 0.7941110033351493, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22641509433962265, | |
| "grad_norm": 3.8018798828125, | |
| "learning_rate": 0.00027124999999999995, | |
| "loss": 1.0028, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.22641509433962265, | |
| "eval_loss": 0.9889835119247437, | |
| "eval_runtime": 194.0625, | |
| "eval_samples_per_second": 36.184, | |
| "eval_steps_per_second": 0.567, | |
| "eval_wer": 0.6952523950623846, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3018867924528302, | |
| "grad_norm": 9.146143913269043, | |
| "learning_rate": 0.00025041666666666664, | |
| "loss": 0.8942, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3018867924528302, | |
| "eval_loss": 0.8935320973396301, | |
| "eval_runtime": 191.4777, | |
| "eval_samples_per_second": 36.673, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.6194873702812953, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.37735849056603776, | |
| "grad_norm": 5.679136276245117, | |
| "learning_rate": 0.0002295833333333333, | |
| "loss": 0.8285, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.37735849056603776, | |
| "eval_loss": 0.8221403360366821, | |
| "eval_runtime": 191.0145, | |
| "eval_samples_per_second": 36.762, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.6075419551150029, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.4528301886792453, | |
| "grad_norm": 7.3102006912231445, | |
| "learning_rate": 0.00020879166666666664, | |
| "loss": 0.763, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4528301886792453, | |
| "eval_loss": 0.7164860963821411, | |
| "eval_runtime": 191.5603, | |
| "eval_samples_per_second": 36.657, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.5306873596513374, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5283018867924528, | |
| "grad_norm": 4.605156898498535, | |
| "learning_rate": 0.00018795833333333333, | |
| "loss": 0.7203, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5283018867924528, | |
| "eval_loss": 0.6891956925392151, | |
| "eval_runtime": 191.0308, | |
| "eval_samples_per_second": 36.758, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.5053747724524642, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6037735849056604, | |
| "grad_norm": 4.3389763832092285, | |
| "learning_rate": 0.000167125, | |
| "loss": 0.7051, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6037735849056604, | |
| "eval_loss": 0.6848454475402832, | |
| "eval_runtime": 192.1753, | |
| "eval_samples_per_second": 36.54, | |
| "eval_steps_per_second": 0.572, | |
| "eval_wer": 0.5069825535816315, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6792452830188679, | |
| "grad_norm": 5.629897117614746, | |
| "learning_rate": 0.00014629166666666667, | |
| "loss": 0.6568, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.6792452830188679, | |
| "eval_loss": 0.6341728568077087, | |
| "eval_runtime": 190.6714, | |
| "eval_samples_per_second": 36.828, | |
| "eval_steps_per_second": 0.577, | |
| "eval_wer": 0.4925789606558684, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "grad_norm": 5.2013421058654785, | |
| "learning_rate": 0.0001255, | |
| "loss": 0.6315, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "eval_loss": 0.59560626745224, | |
| "eval_runtime": 192.113, | |
| "eval_samples_per_second": 36.551, | |
| "eval_steps_per_second": 0.573, | |
| "eval_wer": 0.4493947567732763, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.8301886792452831, | |
| "grad_norm": 6.034324645996094, | |
| "learning_rate": 0.00010466666666666667, | |
| "loss": 0.6171, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.8301886792452831, | |
| "eval_loss": 0.5532522201538086, | |
| "eval_runtime": 191.0439, | |
| "eval_samples_per_second": 36.756, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.4304601443016782, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.9056603773584906, | |
| "grad_norm": 11.971189498901367, | |
| "learning_rate": 8.383333333333333e-05, | |
| "loss": 0.5717, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9056603773584906, | |
| "eval_loss": 0.5359871983528137, | |
| "eval_runtime": 188.0002, | |
| "eval_samples_per_second": 37.351, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.42121208094712925, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9811320754716981, | |
| "grad_norm": 13.140325546264648, | |
| "learning_rate": 6.304166666666666e-05, | |
| "loss": 0.5699, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.9811320754716981, | |
| "eval_loss": 0.518449068069458, | |
| "eval_runtime": 186.6376, | |
| "eval_samples_per_second": 37.624, | |
| "eval_steps_per_second": 0.589, | |
| "eval_wer": 0.4039516868414409, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.0566037735849056, | |
| "grad_norm": 1.923995018005371, | |
| "learning_rate": 4.2208333333333334e-05, | |
| "loss": 0.4905, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.0566037735849056, | |
| "eval_loss": 0.5081383585929871, | |
| "eval_runtime": 187.1769, | |
| "eval_samples_per_second": 37.515, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.39667016569446845, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.1320754716981132, | |
| "grad_norm": 4.82823371887207, | |
| "learning_rate": 2.1374999999999998e-05, | |
| "loss": 0.4706, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.1320754716981132, | |
| "eval_loss": 0.49912941455841064, | |
| "eval_runtime": 187.5036, | |
| "eval_samples_per_second": 37.45, | |
| "eval_steps_per_second": 0.587, | |
| "eval_wer": 0.3825323217156752, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.2075471698113207, | |
| "grad_norm": 2.875157594680786, | |
| "learning_rate": 6.249999999999999e-07, | |
| "loss": 0.4614, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.2075471698113207, | |
| "eval_loss": 0.49337518215179443, | |
| "eval_runtime": 189.5607, | |
| "eval_samples_per_second": 37.044, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.37760267874938547, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.2075471698113207, | |
| "step": 8000, | |
| "total_flos": 9.085819982279884e+18, | |
| "train_loss": 0.9942695980072022, | |
| "train_runtime": 6320.527, | |
| "train_samples_per_second": 10.126, | |
| "train_steps_per_second": 1.266 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.085819982279884e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |