| { | |
| "best_metric": 13.768684731417652, | |
| "best_model_checkpoint": "./checkpoint-600", | |
| "epoch": 7.013, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 0.8523, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 0.5983, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.46e-06, | |
| "loss": 0.4674, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 0.5018, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.46e-06, | |
| "loss": 0.4059, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.96e-06, | |
| "loss": 0.3495, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3.46e-06, | |
| "loss": 0.2695, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3.96e-06, | |
| "loss": 0.2515, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.28808125853538513, | |
| "eval_runtime": 1946.1012, | |
| "eval_samples_per_second": 2.366, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 16.944235888335268, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 0.3219, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.3448, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.2704, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.2282, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.1848, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.2006, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.2534, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.2212, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 0.2615717649459839, | |
| "eval_runtime": 1940.3687, | |
| "eval_samples_per_second": 2.373, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 14.688417172889224, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.46e-06, | |
| "loss": 0.1822, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.1378, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 0.1197, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.1561, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 9.54e-06, | |
| "loss": 0.1683, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 9.040000000000002e-06, | |
| "loss": 0.1323, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 8.540000000000001e-06, | |
| "loss": 0.1061, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 8.040000000000001e-06, | |
| "loss": 0.0774, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_loss": 0.25427135825157166, | |
| "eval_runtime": 1983.6254, | |
| "eval_samples_per_second": 2.321, | |
| "eval_steps_per_second": 0.073, | |
| "eval_wer": 13.768684731417652, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.540000000000001e-06, | |
| "loss": 0.0755, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 7.04e-06, | |
| "loss": 0.0925, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 6.540000000000001e-06, | |
| "loss": 0.0908, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 6.040000000000001e-06, | |
| "loss": 0.0684, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 5.540000000000001e-06, | |
| "loss": 0.0529, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 5.04e-06, | |
| "loss": 0.0401, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 4.540000000000001e-06, | |
| "loss": 0.0469, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 4.04e-06, | |
| "loss": 0.0564, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "eval_loss": 0.27310237288475037, | |
| "eval_runtime": 1976.319, | |
| "eval_samples_per_second": 2.33, | |
| "eval_steps_per_second": 0.073, | |
| "eval_wer": 13.9768616476691, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 3.54e-06, | |
| "loss": 0.0452, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.04e-06, | |
| "loss": 0.0372, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 2.5400000000000002e-06, | |
| "loss": 0.026, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 2.04e-06, | |
| "loss": 0.0216, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 1.54e-06, | |
| "loss": 0.031, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 1.04e-06, | |
| "loss": 0.0347, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 5.4e-07, | |
| "loss": 0.0264, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 4e-08, | |
| "loss": 0.0221, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_loss": 0.28137078881263733, | |
| "eval_runtime": 2000.1135, | |
| "eval_samples_per_second": 2.302, | |
| "eval_steps_per_second": 0.072, | |
| "eval_wer": 13.970036175005118, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "step": 1000, | |
| "total_flos": 1.836038033915904e+19, | |
| "train_loss": 0.18922350925207138, | |
| "train_runtime": 17278.1753, | |
| "train_samples_per_second": 3.704, | |
| "train_steps_per_second": 0.058 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.836038033915904e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |