|
{ |
|
"best_metric": 32.594792142530835, |
|
"best_model_checkpoint": "./checkpoint-1000", |
|
"epoch": 6.112, |
|
"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": 2.75e-07, |
|
"loss": 1.1711, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 5.875e-07, |
|
"loss": 0.9928, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 9.000000000000001e-07, |
|
"loss": 0.6807, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 1.2125e-06, |
|
"loss": 0.4894, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 1.525e-06, |
|
"loss": 0.4589, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 1.8375000000000002e-06, |
|
"loss": 0.4241, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 2.15e-06, |
|
"loss": 0.3693, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 2.4625e-06, |
|
"loss": 0.3465, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"eval_loss": 0.3499109745025635, |
|
"eval_runtime": 1155.2135, |
|
"eval_samples_per_second": 4.076, |
|
"eval_steps_per_second": 0.128, |
|
"eval_wer": 41.93238921882138, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 2.7750000000000005e-06, |
|
"loss": 0.3232, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 3.0875000000000005e-06, |
|
"loss": 0.2946, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 3.4000000000000005e-06, |
|
"loss": 0.3059, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 3.7125000000000005e-06, |
|
"loss": 0.2842, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.0250000000000004e-06, |
|
"loss": 0.2499, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"learning_rate": 4.3375e-06, |
|
"loss": 0.2456, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 4.65e-06, |
|
"loss": 0.2252, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 4.9625e-06, |
|
"loss": 0.2137, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_loss": 0.2953178584575653, |
|
"eval_runtime": 1162.2242, |
|
"eval_samples_per_second": 4.052, |
|
"eval_steps_per_second": 0.127, |
|
"eval_wer": 36.29511192325263, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 2.13, |
|
"learning_rate": 5.275e-06, |
|
"loss": 0.2182, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 5.5875e-06, |
|
"loss": 0.1958, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 5.9e-06, |
|
"loss": 0.1758, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"learning_rate": 6.2125e-06, |
|
"loss": 0.164, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 3.08, |
|
"learning_rate": 6.525e-06, |
|
"loss": 0.154, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 3.11, |
|
"learning_rate": 6.8375e-06, |
|
"loss": 0.1397, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 3.13, |
|
"learning_rate": 7.15e-06, |
|
"loss": 0.1419, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 7.4625e-06, |
|
"loss": 0.1255, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"eval_loss": 0.29274308681488037, |
|
"eval_runtime": 1130.2455, |
|
"eval_samples_per_second": 4.166, |
|
"eval_steps_per_second": 0.131, |
|
"eval_wer": 33.72316126084971, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 7.775000000000001e-06, |
|
"loss": 0.1077, |
|
"step": 625 |
|
}, |
|
{ |
|
"epoch": 4.06, |
|
"learning_rate": 8.0875e-06, |
|
"loss": 0.0996, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"learning_rate": 8.400000000000001e-06, |
|
"loss": 0.091, |
|
"step": 675 |
|
}, |
|
{ |
|
"epoch": 4.11, |
|
"learning_rate": 8.7125e-06, |
|
"loss": 0.0804, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 4.13, |
|
"learning_rate": 9.025e-06, |
|
"loss": 0.0781, |
|
"step": 725 |
|
}, |
|
{ |
|
"epoch": 5.01, |
|
"learning_rate": 9.3375e-06, |
|
"loss": 0.0648, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 5.04, |
|
"learning_rate": 9.65e-06, |
|
"loss": 0.0549, |
|
"step": 775 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"learning_rate": 9.9625e-06, |
|
"loss": 0.0509, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"eval_loss": 0.3148699104785919, |
|
"eval_runtime": 1070.2676, |
|
"eval_samples_per_second": 4.4, |
|
"eval_steps_per_second": 0.138, |
|
"eval_wer": 34.05664687071722, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 5.08, |
|
"learning_rate": 8.900000000000001e-06, |
|
"loss": 0.0518, |
|
"step": 825 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"learning_rate": 7.650000000000001e-06, |
|
"loss": 0.0406, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 5.13, |
|
"learning_rate": 6.4000000000000006e-06, |
|
"loss": 0.0361, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 6.01, |
|
"learning_rate": 5.150000000000001e-06, |
|
"loss": 0.0299, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 6.04, |
|
"learning_rate": 3.900000000000001e-06, |
|
"loss": 0.0246, |
|
"step": 925 |
|
}, |
|
{ |
|
"epoch": 6.06, |
|
"learning_rate": 2.6500000000000005e-06, |
|
"loss": 0.0218, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 6.09, |
|
"learning_rate": 1.4000000000000001e-06, |
|
"loss": 0.0202, |
|
"step": 975 |
|
}, |
|
{ |
|
"epoch": 6.11, |
|
"learning_rate": 1.5000000000000002e-07, |
|
"loss": 0.0164, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 6.11, |
|
"eval_loss": 0.32497844099998474, |
|
"eval_runtime": 1129.333, |
|
"eval_samples_per_second": 4.17, |
|
"eval_steps_per_second": 0.131, |
|
"eval_wer": 32.594792142530835, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 6.11, |
|
"step": 1000, |
|
"total_flos": 1.837596395077632e+19, |
|
"train_loss": 0.23146729975938796, |
|
"train_runtime": 12715.3657, |
|
"train_samples_per_second": 5.033, |
|
"train_steps_per_second": 0.079 |
|
} |
|
], |
|
"max_steps": 1000, |
|
"num_train_epochs": 9223372036854775807, |
|
"total_flos": 1.837596395077632e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|