|
{ |
|
"best_metric": 0.5595238095238095, |
|
"best_model_checkpoint": "videomae-base-finetuned-ElderReact-anger-balanced/checkpoint-25", |
|
"epoch": 9.0625, |
|
"eval_steps": 500, |
|
"global_step": 240, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 4.018902778625488, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 0.7172, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"grad_norm": 4.251977920532227, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 0.7241, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_accuracy": 0.5595238095238095, |
|
"eval_loss": 0.6945744752883911, |
|
"eval_runtime": 171.7134, |
|
"eval_samples_per_second": 2.446, |
|
"eval_steps_per_second": 0.157, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"grad_norm": 4.4508538246154785, |
|
"learning_rate": 4.8611111111111115e-05, |
|
"loss": 0.7158, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"grad_norm": 4.41381311416626, |
|
"learning_rate": 4.62962962962963e-05, |
|
"loss": 0.7154, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"grad_norm": 13.169350624084473, |
|
"learning_rate": 4.3981481481481486e-05, |
|
"loss": 0.7205, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"eval_accuracy": 0.5095238095238095, |
|
"eval_loss": 0.7030633687973022, |
|
"eval_runtime": 176.8132, |
|
"eval_samples_per_second": 2.375, |
|
"eval_steps_per_second": 0.153, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"grad_norm": 2.2161309719085693, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 0.6693, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"grad_norm": 3.9168288707733154, |
|
"learning_rate": 3.935185185185186e-05, |
|
"loss": 0.6897, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_accuracy": 0.5119047619047619, |
|
"eval_loss": 0.6988625526428223, |
|
"eval_runtime": 174.7717, |
|
"eval_samples_per_second": 2.403, |
|
"eval_steps_per_second": 0.154, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"grad_norm": 5.149790287017822, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.7026, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"grad_norm": 3.6450681686401367, |
|
"learning_rate": 3.472222222222222e-05, |
|
"loss": 0.6506, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"grad_norm": 14.79007625579834, |
|
"learning_rate": 3.240740740740741e-05, |
|
"loss": 0.6771, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"eval_accuracy": 0.5190476190476191, |
|
"eval_loss": 0.7333685159683228, |
|
"eval_runtime": 166.0989, |
|
"eval_samples_per_second": 2.529, |
|
"eval_steps_per_second": 0.163, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 4.04, |
|
"grad_norm": 3.0915603637695312, |
|
"learning_rate": 3.0092592592592593e-05, |
|
"loss": 0.648, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"grad_norm": 5.750015735626221, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.6668, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 4.1, |
|
"eval_accuracy": 0.5238095238095238, |
|
"eval_loss": 0.7564190030097961, |
|
"eval_runtime": 180.3108, |
|
"eval_samples_per_second": 2.329, |
|
"eval_steps_per_second": 0.15, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 5.02, |
|
"grad_norm": 7.389021396636963, |
|
"learning_rate": 2.5462962962962965e-05, |
|
"loss": 0.7349, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"grad_norm": 2.2353854179382324, |
|
"learning_rate": 2.314814814814815e-05, |
|
"loss": 0.5991, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"grad_norm": 16.68378448486328, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 0.5922, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"eval_accuracy": 0.5166666666666667, |
|
"eval_loss": 0.7602722644805908, |
|
"eval_runtime": 176.2053, |
|
"eval_samples_per_second": 2.384, |
|
"eval_steps_per_second": 0.153, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 6.04, |
|
"grad_norm": 5.568227767944336, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.6442, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"grad_norm": 3.2906060218811035, |
|
"learning_rate": 1.6203703703703704e-05, |
|
"loss": 0.6092, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 6.1, |
|
"eval_accuracy": 0.5285714285714286, |
|
"eval_loss": 0.749624490737915, |
|
"eval_runtime": 158.2618, |
|
"eval_samples_per_second": 2.654, |
|
"eval_steps_per_second": 0.171, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 7.02, |
|
"grad_norm": 3.01987361907959, |
|
"learning_rate": 1.388888888888889e-05, |
|
"loss": 0.5057, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 7.06, |
|
"grad_norm": 4.032674789428711, |
|
"learning_rate": 1.1574074074074075e-05, |
|
"loss": 0.5524, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 7.1, |
|
"grad_norm": 4.746392250061035, |
|
"learning_rate": 9.259259259259259e-06, |
|
"loss": 0.5803, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 7.1, |
|
"eval_accuracy": 0.5404761904761904, |
|
"eval_loss": 0.8512407541275024, |
|
"eval_runtime": 165.3346, |
|
"eval_samples_per_second": 2.54, |
|
"eval_steps_per_second": 0.163, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 8.04, |
|
"grad_norm": 3.170111894607544, |
|
"learning_rate": 6.944444444444445e-06, |
|
"loss": 0.4992, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 8.08, |
|
"grad_norm": 3.854588270187378, |
|
"learning_rate": 4.6296296296296296e-06, |
|
"loss": 0.6102, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 8.1, |
|
"eval_accuracy": 0.5238095238095238, |
|
"eval_loss": 0.8049675226211548, |
|
"eval_runtime": 172.2303, |
|
"eval_samples_per_second": 2.439, |
|
"eval_steps_per_second": 0.157, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 9.02, |
|
"grad_norm": 3.0995934009552, |
|
"learning_rate": 2.3148148148148148e-06, |
|
"loss": 0.5465, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 9.06, |
|
"grad_norm": 2.566746234893799, |
|
"learning_rate": 0.0, |
|
"loss": 0.4901, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 9.06, |
|
"eval_accuracy": 0.5238095238095238, |
|
"eval_loss": 0.7804582715034485, |
|
"eval_runtime": 169.5522, |
|
"eval_samples_per_second": 2.477, |
|
"eval_steps_per_second": 0.159, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 9.06, |
|
"step": 240, |
|
"total_flos": 4.616669303992812e+18, |
|
"train_loss": 0.6358872016270956, |
|
"train_runtime": 3365.0495, |
|
"train_samples_per_second": 1.141, |
|
"train_steps_per_second": 0.071 |
|
}, |
|
{ |
|
"epoch": 9.06, |
|
"eval_accuracy": 0.4304461942257218, |
|
"eval_loss": 0.7508144974708557, |
|
"eval_runtime": 149.084, |
|
"eval_samples_per_second": 2.556, |
|
"eval_steps_per_second": 0.161, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 9.06, |
|
"eval_accuracy": 0.4304461942257218, |
|
"eval_loss": 0.7506275773048401, |
|
"eval_runtime": 147.9621, |
|
"eval_samples_per_second": 2.575, |
|
"eval_steps_per_second": 0.162, |
|
"step": 240 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 240, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 9223372036854775807, |
|
"save_steps": 500, |
|
"total_flos": 4.616669303992812e+18, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|