|
{ |
|
"best_metric": 0.0491877980530262, |
|
"best_model_checkpoint": "./vit-base-beans-demo-v7/checkpoint-100", |
|
"epoch": 4.0, |
|
"eval_steps": 100, |
|
"global_step": 260, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.15384615384615385, |
|
"grad_norm": 1.6902730464935303, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.7748, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 1.2790460586547852, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.2739, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 0.7741395235061646, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.1927, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 5.114573001861572, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.2443, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 15.121728897094727, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.2038, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 3.094726324081421, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.2427, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 0.9401968717575073, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.179, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 0.16439810395240784, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.1334, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.1418611705303192, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.1166, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 1.5957568883895874, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.0862, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9924812030075187, |
|
"eval_loss": 0.0491877980530262, |
|
"eval_runtime": 1.3534, |
|
"eval_samples_per_second": 98.272, |
|
"eval_steps_per_second": 12.561, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 1.3301724195480347, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.1008, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 0.32133230566978455, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.0803, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 0.4987194836139679, |
|
"learning_rate": 0.0001, |
|
"loss": 0.0777, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 3.434934377670288, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0583, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 0.08973214775323868, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.0796, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 0.0839068666100502, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0468, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.265321284532547, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.0204, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 0.07917013019323349, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.0646, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 0.07372036576271057, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0452, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.0797925740480423, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.031, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 0.9849624060150376, |
|
"eval_loss": 0.0571388304233551, |
|
"eval_runtime": 1.3229, |
|
"eval_samples_per_second": 100.536, |
|
"eval_steps_per_second": 12.85, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.08364936709403992, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.0151, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.0659966766834259, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.013, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.06812237203121185, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0129, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.06594033539295197, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0123, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.06312059611082077, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0123, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.0624660886824131, |
|
"learning_rate": 0.0, |
|
"loss": 0.0121, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.12038601376116276, |
|
"train_runtime": 77.0575, |
|
"train_samples_per_second": 53.674, |
|
"train_steps_per_second": 3.374 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 260, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 100, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|