|
{ |
|
"best_metric": 0.031315796077251434, |
|
"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": 1.4313935041427612, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.8173, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 2.4101505279541016, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.3143, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 2.5627517700195312, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.2993, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 1.614366054534912, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.2584, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 0.40835028886795044, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.1474, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 3.403233051300049, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.2012, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 0.21098481118679047, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.1404, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 3.3641014099121094, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.1411, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.141947403550148, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.0396, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 1.2932701110839844, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.0664, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9924812030075187, |
|
"eval_loss": 0.03421594202518463, |
|
"eval_runtime": 1.2565, |
|
"eval_samples_per_second": 105.85, |
|
"eval_steps_per_second": 13.53, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 1.4846528768539429, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.0618, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 0.5059632658958435, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.1365, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 4.463663101196289, |
|
"learning_rate": 0.0001, |
|
"loss": 0.0462, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 0.5683491230010986, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0472, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 2.8736608028411865, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.0391, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 0.09224144369363785, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0203, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.6061734557151794, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.0184, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 10.278433799743652, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.0649, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 2.6682260036468506, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0632, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.05894143134355545, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.0131, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 0.9924812030075187, |
|
"eval_loss": 0.031315796077251434, |
|
"eval_runtime": 1.7955, |
|
"eval_samples_per_second": 74.074, |
|
"eval_steps_per_second": 9.468, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.0573604553937912, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.0113, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.06024244800209999, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.0109, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.05909183248877525, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0108, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.05563810095191002, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0103, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.07520989328622818, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0105, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.05178371071815491, |
|
"learning_rate": 0.0, |
|
"loss": 0.0103, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.11539229828004653, |
|
"train_runtime": 104.1168, |
|
"train_samples_per_second": 39.725, |
|
"train_steps_per_second": 2.497 |
|
} |
|
], |
|
"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 |
|
} |
|
|