| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.19999594985925762, | |
| "global_step": 2469, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 19.7984, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 7.425e-06, | |
| "loss": 13.3485, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.1174999999999999e-05, | |
| "loss": 10.0074, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.4925e-05, | |
| "loss": 8.4765, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.8675e-05, | |
| "loss": 6.9106, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 6.717121124267578, | |
| "eval_runtime": 908.8081, | |
| "eval_samples_per_second": 17.541, | |
| "eval_steps_per_second": 2.193, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.2424999999999996e-05, | |
| "loss": 5.4175, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.6174999999999996e-05, | |
| "loss": 4.1471, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9925e-05, | |
| "loss": 3.3536, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.3675e-05, | |
| "loss": 3.0742, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3.7424999999999995e-05, | |
| "loss": 3.0034, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 3.01257061958313, | |
| "eval_runtime": 815.2192, | |
| "eval_samples_per_second": 19.554, | |
| "eval_steps_per_second": 2.445, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.1175e-05, | |
| "loss": 2.9695, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.4924999999999994e-05, | |
| "loss": 2.9321, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.8675e-05, | |
| "loss": 2.916, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5.2424999999999994e-05, | |
| "loss": 2.8952, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 5.6175e-05, | |
| "loss": 2.8699, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 2.850850820541382, | |
| "eval_runtime": 814.3164, | |
| "eval_samples_per_second": 19.576, | |
| "eval_steps_per_second": 2.447, | |
| "eval_wer": 0.9817464201091995, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5.9925e-05, | |
| "loss": 2.7752, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 6.367499999999999e-05, | |
| "loss": 2.4403, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 6.7425e-05, | |
| "loss": 1.9873, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 7.1175e-05, | |
| "loss": 1.7549, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.629, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.7786943316459656, | |
| "eval_runtime": 820.0379, | |
| "eval_samples_per_second": 19.439, | |
| "eval_steps_per_second": 2.43, | |
| "eval_wer": 0.586136293396518, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 5.948827292110874e-05, | |
| "loss": 1.5154, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.349680170575692e-05, | |
| "loss": 1.4719, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.7505330490405118e-05, | |
| "loss": 1.4058, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.1513859275053305e-05, | |
| "loss": 1.3579, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "step": 2469, | |
| "total_flos": 1.1608261301380424e+19, | |
| "train_loss": 4.42930416077822, | |
| "train_runtime": 8599.8864, | |
| "train_samples_per_second": 9.187, | |
| "train_steps_per_second": 0.287 | |
| } | |
| ], | |
| "max_steps": 2469, | |
| "num_train_epochs": 1, | |
| "total_flos": 1.1608261301380424e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |