| { | |
| "best_metric": 16.017052375152254, | |
| "best_model_checkpoint": "./checkpoint-5000", | |
| "epoch": 1.0, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.800000000000001e-07, | |
| "loss": 1.7359, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.98e-06, | |
| "loss": 1.4111, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.9800000000000003e-06, | |
| "loss": 1.2405, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 1.1693, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.980000000000001e-06, | |
| "loss": 1.1121, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.98e-06, | |
| "loss": 1.0805, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.98e-06, | |
| "loss": 1.0618, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.980000000000002e-06, | |
| "loss": 1.0203, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.98e-06, | |
| "loss": 1.0208, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.980000000000001e-06, | |
| "loss": 0.9901, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9601, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9468, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9364, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9238, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9055, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8913, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8945, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1e-05, | |
| "loss": 0.881, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8714, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8528, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.44020676612854004, | |
| "eval_runtime": 98.0744, | |
| "eval_samples_per_second": 1.642, | |
| "eval_steps_per_second": 0.061, | |
| "eval_wer": 22.56394640682095, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8567, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8457, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8393, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8277, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8219, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8254, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8209, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8069, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8161, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7951, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8073, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7973, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7803, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7833, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7874, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1e-05, | |
| "loss": 0.776, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7668, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7736, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7652, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7558, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.38613995909690857, | |
| "eval_runtime": 95.704, | |
| "eval_samples_per_second": 1.682, | |
| "eval_steps_per_second": 0.063, | |
| "eval_wer": 19.701583434835566, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7529, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7618, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7524, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7414, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7448, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1e-05, | |
| "loss": 0.755, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7445, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7432, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7345, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1e-05, | |
| "loss": 0.725, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1e-05, | |
| "loss": 0.727, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7219, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7286, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7246, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7168, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7184, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7212, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7221, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7143, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1e-05, | |
| "loss": 0.721, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.3545149862766266, | |
| "eval_runtime": 97.6917, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.061, | |
| "eval_wer": 17.84409257003654, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7108, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7189, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7052, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7095, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6974, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6952, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6991, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6952, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7045, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6994, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1e-05, | |
| "loss": 0.695, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6858, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6873, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6955, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6851, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1e-05, | |
| "loss": 0.675, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6909, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6894, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6814, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6803, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.33644479513168335, | |
| "eval_runtime": 96.6414, | |
| "eval_samples_per_second": 1.666, | |
| "eval_steps_per_second": 0.062, | |
| "eval_wer": 16.56516443361754, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6872, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6905, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6769, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6774, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6674, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6762, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6702, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6751, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1e-05, | |
| "loss": 0.675, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1e-05, | |
| "loss": 0.662, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6689, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6557, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6627, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6796, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6731, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6562, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6627, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6521, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6599, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6557, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.32896530628204346, | |
| "eval_runtime": 102.0673, | |
| "eval_samples_per_second": 1.577, | |
| "eval_steps_per_second": 0.059, | |
| "eval_wer": 16.017052375152254, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 5000, | |
| "total_flos": 1.846946562048e+20, | |
| "train_loss": 0.7891104797363281, | |
| "train_runtime": 103940.9978, | |
| "train_samples_per_second": 6.157, | |
| "train_steps_per_second": 0.048 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.846946562048e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |