|
{ |
|
"best_metric": 0.014574897475540638, |
|
"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": 1.2454960346221924, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.819, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 0.8512015342712402, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.299, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 0.6652663350105286, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.2006, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 3.5982391834259033, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.3113, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 2.6809399127960205, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.2997, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 4.859236717224121, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.2482, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 0.2121334671974182, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.1262, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 0.16424687206745148, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.1094, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.1939377784729004, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.0624, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 1.6206365823745728, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.1528, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9924812030075187, |
|
"eval_loss": 0.04189983382821083, |
|
"eval_runtime": 1.8015, |
|
"eval_samples_per_second": 73.827, |
|
"eval_steps_per_second": 9.437, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 1.522310495376587, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.0933, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 0.1764295995235443, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.0998, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 0.17484644055366516, |
|
"learning_rate": 0.0001, |
|
"loss": 0.0914, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 0.134028360247612, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0949, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 0.17579883337020874, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.065, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 3.210505485534668, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0431, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.31449979543685913, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.0367, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 0.8201821446418762, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.065, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 0.19602012634277344, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0275, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.07666312903165817, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.0316, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 1.0, |
|
"eval_loss": 0.014574897475540638, |
|
"eval_runtime": 1.5803, |
|
"eval_samples_per_second": 84.16, |
|
"eval_steps_per_second": 10.757, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.06805969029664993, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.0145, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.07413840293884277, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.0139, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.0722680538892746, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0129, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.07481131702661514, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0172, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.07286258786916733, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0125, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.06795915216207504, |
|
"learning_rate": 0.0, |
|
"loss": 0.0133, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.12927936476010543, |
|
"train_runtime": 94.1288, |
|
"train_samples_per_second": 43.94, |
|
"train_steps_per_second": 2.762 |
|
} |
|
], |
|
"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 |
|
} |
|
|