| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.98804780876495, | |
| "global_step": 3100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 17.9778, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 7.4625e-06, | |
| "loss": 10.1788, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 7.1097, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "learning_rate": 1.49625e-05, | |
| "loss": 5.9162, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 1.8712499999999997e-05, | |
| "loss": 4.917, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_loss": 4.89633846282959, | |
| "eval_runtime": 284.283, | |
| "eval_samples_per_second": 9.473, | |
| "eval_steps_per_second": 0.594, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.2462499999999997e-05, | |
| "loss": 4.1291, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 22.57, | |
| "learning_rate": 2.6212499999999997e-05, | |
| "loss": 3.6844, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 25.8, | |
| "learning_rate": 2.99625e-05, | |
| "loss": 3.5213, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 3.4601, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 32.25, | |
| "learning_rate": 3.7462499999999996e-05, | |
| "loss": 3.3585, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 32.25, | |
| "eval_loss": 3.3068907260894775, | |
| "eval_runtime": 283.3986, | |
| "eval_samples_per_second": 9.503, | |
| "eval_steps_per_second": 0.596, | |
| "eval_wer": 1.0000410576449335, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "learning_rate": 4.12125e-05, | |
| "loss": 3.2956, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 38.7, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 2.8135, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 41.92, | |
| "learning_rate": 4.871249999999999e-05, | |
| "loss": 2.0147, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 5.2424999999999994e-05, | |
| "loss": 1.7332, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 48.38, | |
| "learning_rate": 5.6175e-05, | |
| "loss": 1.5873, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 48.38, | |
| "eval_loss": 0.8273974061012268, | |
| "eval_runtime": 283.1526, | |
| "eval_samples_per_second": 9.511, | |
| "eval_steps_per_second": 0.597, | |
| "eval_wer": 1.006076531450156, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "learning_rate": 5.9925e-05, | |
| "loss": 1.476, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 54.83, | |
| "learning_rate": 6.367499999999999e-05, | |
| "loss": 1.4204, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "learning_rate": 6.7425e-05, | |
| "loss": 1.357, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 61.29, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 1.3037, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 64.51, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.2654, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 64.51, | |
| "eval_loss": 0.6249768137931824, | |
| "eval_runtime": 282.7219, | |
| "eval_samples_per_second": 9.525, | |
| "eval_steps_per_second": 0.598, | |
| "eval_wer": 1.0075546066677616, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 67.73, | |
| "learning_rate": 6.838636363636363e-05, | |
| "loss": 1.2231, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 70.96, | |
| "learning_rate": 6.156818181818181e-05, | |
| "loss": 1.1802, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 74.19, | |
| "learning_rate": 5.481818181818182e-05, | |
| "loss": 1.1484, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 77.41, | |
| "learning_rate": 4.7999999999999994e-05, | |
| "loss": 1.1185, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 80.64, | |
| "learning_rate": 4.118181818181818e-05, | |
| "loss": 1.0917, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 80.64, | |
| "eval_loss": 0.5459514856338501, | |
| "eval_runtime": 282.9764, | |
| "eval_samples_per_second": 9.517, | |
| "eval_steps_per_second": 0.597, | |
| "eval_wer": 1.0056248973558877, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 83.86, | |
| "learning_rate": 3.436363636363636e-05, | |
| "loss": 1.0663, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 87.1, | |
| "learning_rate": 2.754545454545454e-05, | |
| "loss": 1.0487, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 90.32, | |
| "learning_rate": 2.0727272727272725e-05, | |
| "loss": 1.0214, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 93.54, | |
| "learning_rate": 1.3909090909090907e-05, | |
| "loss": 1.016, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 96.76, | |
| "learning_rate": 7.09090909090909e-06, | |
| "loss": 1.0001, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 96.76, | |
| "eval_loss": 0.5304259061813354, | |
| "eval_runtime": 282.7392, | |
| "eval_samples_per_second": 9.525, | |
| "eval_steps_per_second": 0.598, | |
| "eval_wer": 1.0082525866316308, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 99.99, | |
| "learning_rate": 2.727272727272727e-07, | |
| "loss": 0.9947, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 99.99, | |
| "step": 3100, | |
| "total_flos": 5.0567037565307716e+19, | |
| "train_loss": 3.0460862658100742, | |
| "train_runtime": 81000.8052, | |
| "train_samples_per_second": 4.939, | |
| "train_steps_per_second": 0.038 | |
| } | |
| ], | |
| "max_steps": 3100, | |
| "num_train_epochs": 100, | |
| "total_flos": 5.0567037565307716e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |