|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 20.0, |
|
"global_step": 280, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 0.00019357142857142858, |
|
"loss": 1.9919, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 0.00018642857142857144, |
|
"loss": 0.8872, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"learning_rate": 0.0001792857142857143, |
|
"loss": 0.1348, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"learning_rate": 0.00017214285714285717, |
|
"loss": 0.0191, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 3.57, |
|
"learning_rate": 0.000165, |
|
"loss": 0.004, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 3.57, |
|
"eval_accuracy": 0.7575757575757576, |
|
"eval_loss": 1.07438063621521, |
|
"eval_runtime": 1.2746, |
|
"eval_samples_per_second": 25.891, |
|
"eval_steps_per_second": 3.923, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 4.29, |
|
"learning_rate": 0.00015785714285714285, |
|
"loss": 0.0034, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0001507142857142857, |
|
"loss": 0.0012, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 5.71, |
|
"learning_rate": 0.0001435714285714286, |
|
"loss": 0.0008, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 6.43, |
|
"learning_rate": 0.00013642857142857144, |
|
"loss": 0.0008, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 7.14, |
|
"learning_rate": 0.0001292857142857143, |
|
"loss": 0.0005, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 7.14, |
|
"eval_accuracy": 0.7878787878787878, |
|
"eval_loss": 0.8181638121604919, |
|
"eval_runtime": 1.1472, |
|
"eval_samples_per_second": 28.766, |
|
"eval_steps_per_second": 4.358, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 7.86, |
|
"learning_rate": 0.00012214285714285715, |
|
"loss": 0.0005, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 8.57, |
|
"learning_rate": 0.00011499999999999999, |
|
"loss": 0.0004, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 9.29, |
|
"learning_rate": 0.00010785714285714286, |
|
"loss": 0.0004, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 0.00010071428571428571, |
|
"loss": 0.0004, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 10.71, |
|
"learning_rate": 9.357142857142858e-05, |
|
"loss": 0.0004, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 10.71, |
|
"eval_accuracy": 0.7878787878787878, |
|
"eval_loss": 0.7808332443237305, |
|
"eval_runtime": 1.1203, |
|
"eval_samples_per_second": 29.456, |
|
"eval_steps_per_second": 4.463, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 11.43, |
|
"learning_rate": 8.642857142857143e-05, |
|
"loss": 0.0003, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 12.14, |
|
"learning_rate": 7.928571428571429e-05, |
|
"loss": 0.0003, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 12.86, |
|
"learning_rate": 7.214285714285714e-05, |
|
"loss": 0.0003, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 13.57, |
|
"learning_rate": 6.500000000000001e-05, |
|
"loss": 0.0003, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 14.29, |
|
"learning_rate": 5.785714285714287e-05, |
|
"loss": 0.0003, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 14.29, |
|
"eval_accuracy": 0.8181818181818182, |
|
"eval_loss": 0.7710055112838745, |
|
"eval_runtime": 1.1654, |
|
"eval_samples_per_second": 28.316, |
|
"eval_steps_per_second": 4.29, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"learning_rate": 5.0714285714285716e-05, |
|
"loss": 0.0003, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 15.71, |
|
"learning_rate": 4.3571428571428576e-05, |
|
"loss": 0.0003, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 16.43, |
|
"learning_rate": 3.642857142857143e-05, |
|
"loss": 0.0003, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 17.14, |
|
"learning_rate": 2.9285714285714288e-05, |
|
"loss": 0.0003, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 17.86, |
|
"learning_rate": 2.214285714285714e-05, |
|
"loss": 0.0003, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 17.86, |
|
"eval_accuracy": 0.8181818181818182, |
|
"eval_loss": 0.7675150632858276, |
|
"eval_runtime": 1.1463, |
|
"eval_samples_per_second": 28.788, |
|
"eval_steps_per_second": 4.362, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 18.57, |
|
"learning_rate": 1.5e-05, |
|
"loss": 0.0003, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 19.29, |
|
"learning_rate": 7.857142857142858e-06, |
|
"loss": 0.0003, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"learning_rate": 7.142857142857143e-07, |
|
"loss": 0.0003, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"step": 280, |
|
"total_flos": 8.5248064484352e+16, |
|
"train_loss": 0.10890771941714254, |
|
"train_runtime": 100.0908, |
|
"train_samples_per_second": 10.99, |
|
"train_steps_per_second": 2.797 |
|
} |
|
], |
|
"max_steps": 280, |
|
"num_train_epochs": 20, |
|
"total_flos": 8.5248064484352e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|