|
{ |
|
"best_metric": 0.05263166502118111, |
|
"best_model_checkpoint": "./vit-base-beans-demo-v5/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": 0.05400194600224495, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.0122, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 6.026679992675781, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.0244, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 0.04543726146221161, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.0087, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 0.04408282786607742, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.0312, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 0.036833759397268295, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.0366, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 0.09263049811124802, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.028, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 0.03878260403871536, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.0395, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 0.03130094334483147, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.0061, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.028703533113002777, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.0491, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 9.305721282958984, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.078, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9849624060150376, |
|
"eval_loss": 0.05263166502118111, |
|
"eval_runtime": 1.6957, |
|
"eval_samples_per_second": 78.435, |
|
"eval_steps_per_second": 10.026, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 0.029910510405898094, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.0247, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 0.029740193858742714, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.1057, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 0.028273504227399826, |
|
"learning_rate": 0.0001, |
|
"loss": 0.016, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 0.03569089621305466, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0392, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 5.499910354614258, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.0212, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 0.027333833277225494, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0493, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.022924846038222313, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.0055, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 0.40536436438560486, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.0046, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 0.022467907518148422, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0066, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.020369328558444977, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.0038, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 0.9548872180451128, |
|
"eval_loss": 0.22457469999790192, |
|
"eval_runtime": 1.3037, |
|
"eval_samples_per_second": 102.017, |
|
"eval_steps_per_second": 13.04, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.02012537233531475, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.0223, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.019498102366924286, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.0038, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.02147608995437622, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0226, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.021528450772166252, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0036, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.02092928998172283, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0038, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.018939368426799774, |
|
"learning_rate": 0.0, |
|
"loss": 0.0036, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.025007283272078403, |
|
"train_runtime": 87.6347, |
|
"train_samples_per_second": 47.196, |
|
"train_steps_per_second": 2.967 |
|
} |
|
], |
|
"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 |
|
} |
|
|