|
{ |
|
"best_metric": 0.017244208604097366, |
|
"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": 1.3774149417877197, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.8222, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 2.181103229522705, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.3165, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 0.6552544236183167, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.2321, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 1.0163898468017578, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.2035, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 3.2018964290618896, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.2585, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 3.0455973148345947, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.1975, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 1.6109272241592407, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.2491, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 0.19894865155220032, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.1941, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.4901449978351593, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.0616, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 1.5833218097686768, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.0758, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9849624060150376, |
|
"eval_loss": 0.06387122720479965, |
|
"eval_runtime": 3.6073, |
|
"eval_samples_per_second": 36.869, |
|
"eval_steps_per_second": 4.713, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 0.11933541297912598, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.065, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 4.058487892150879, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.1206, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 0.1607927531003952, |
|
"learning_rate": 0.0001, |
|
"loss": 0.0331, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 0.0955452099442482, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0605, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 0.09269090741872787, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.0546, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 0.08798301964998245, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0577, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.1001884862780571, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.0527, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 4.74984073638916, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.0853, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 0.1210957020521164, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0212, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.1006399542093277, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.0186, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 1.0, |
|
"eval_loss": 0.017244208604097366, |
|
"eval_runtime": 1.2663, |
|
"eval_samples_per_second": 105.029, |
|
"eval_steps_per_second": 13.425, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.0646267831325531, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.0391, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.07148883491754532, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.0137, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.06836055964231491, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0161, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.06472332030534744, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0124, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.06633427739143372, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0132, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.06199250370264053, |
|
"learning_rate": 0.0, |
|
"loss": 0.0122, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.1264148608996318, |
|
"train_runtime": 94.7614, |
|
"train_samples_per_second": 43.646, |
|
"train_steps_per_second": 2.744 |
|
} |
|
], |
|
"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 |
|
} |
|
|