| { | |
| "best_metric": 15.012180267965894, | |
| "best_model_checkpoint": "./checkpoint-3000", | |
| "epoch": 1.0, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.651, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6439, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6508, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6457, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1e-05, | |
| "loss": 0.633, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6258, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6325, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6192, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1e-05, | |
| "loss": 0.63, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1e-05, | |
| "loss": 0.623, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6099, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6128, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6155, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6169, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6073, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6076, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6165, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6096, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6059, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5975, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.33536991477012634, | |
| "eval_runtime": 97.6999, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.061, | |
| "eval_wer": 15.77344701583435, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6067, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6052, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6001, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5959, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5945, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6036, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6011, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5921, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6054, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5883, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6017, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5966, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1e-05, | |
| "loss": 0.587, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5909, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5973, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5885, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5824, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5919, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5857, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5783, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.33269065618515015, | |
| "eval_runtime": 99.8788, | |
| "eval_samples_per_second": 1.612, | |
| "eval_steps_per_second": 0.06, | |
| "eval_wer": 16.35200974421437, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5788, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5876, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5825, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5743, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5813, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5882, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5824, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5829, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5779, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5697, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5713, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5695, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5779, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5734, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5682, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5708, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5762, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5762, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1e-05, | |
| "loss": 0.572, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5788, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.3284420371055603, | |
| "eval_runtime": 99.6761, | |
| "eval_samples_per_second": 1.615, | |
| "eval_steps_per_second": 0.06, | |
| "eval_wer": 15.012180267965894, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 3000, | |
| "total_flos": 1.1081679372288e+20, | |
| "train_loss": 0.598131457010905, | |
| "train_runtime": 63177.9957, | |
| "train_samples_per_second": 6.078, | |
| "train_steps_per_second": 0.047 | |
| } | |
| ], | |
| "max_steps": 3000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.1081679372288e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |