|
{ |
|
"best_metric": 0.03179202973842621, |
|
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200", |
|
"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": 1.3854262828826904, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.8006, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 4.267582893371582, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.3319, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 1.7108104228973389, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.2789, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 1.280118465423584, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.2837, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 8.134176254272461, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.2461, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 2.2806813716888428, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.2131, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 0.2824852168560028, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.1639, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 0.39076101779937744, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.0845, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.1591070592403412, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.0721, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 2.908008575439453, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.0725, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9473684210526315, |
|
"eval_loss": 0.1633957028388977, |
|
"eval_runtime": 1.2488, |
|
"eval_samples_per_second": 106.503, |
|
"eval_steps_per_second": 13.613, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 3.4353740215301514, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.1748, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 3.038853883743286, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.1436, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 2.874983310699463, |
|
"learning_rate": 0.0001, |
|
"loss": 0.0415, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 0.20417748391628265, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0394, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 9.560956954956055, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.0438, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 0.3993445634841919, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0366, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.07924646139144897, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.0157, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 0.07146325707435608, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.0336, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 1.8434661626815796, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0176, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.07253334671258926, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.0145, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 0.9924812030075187, |
|
"eval_loss": 0.03179202973842621, |
|
"eval_runtime": 1.4947, |
|
"eval_samples_per_second": 88.979, |
|
"eval_steps_per_second": 11.373, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.0676412433385849, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.013, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.06383011490106583, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.0123, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.06610537320375443, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0119, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.06182776391506195, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0116, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.05951670929789543, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0115, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.11315698176622391, |
|
"learning_rate": 0.0, |
|
"loss": 0.0117, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.12231931924246825, |
|
"train_runtime": 89.2176, |
|
"train_samples_per_second": 46.359, |
|
"train_steps_per_second": 2.914 |
|
} |
|
], |
|
"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 |
|
} |
|
|