|
{ |
|
"best_metric": 0.024031903594732285, |
|
"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": 1.234106421470642, |
|
"learning_rate": 0.00019230769230769233, |
|
"loss": 0.8148, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3076923076923077, |
|
"grad_norm": 0.5447805523872375, |
|
"learning_rate": 0.00018461538461538463, |
|
"loss": 0.3096, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.46153846153846156, |
|
"grad_norm": 3.6802539825439453, |
|
"learning_rate": 0.00017692307692307693, |
|
"loss": 0.2258, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6153846153846154, |
|
"grad_norm": 1.9799599647521973, |
|
"learning_rate": 0.00016923076923076923, |
|
"loss": 0.2187, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7692307692307693, |
|
"grad_norm": 11.105630874633789, |
|
"learning_rate": 0.00016153846153846155, |
|
"loss": 0.1941, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9230769230769231, |
|
"grad_norm": 3.717527389526367, |
|
"learning_rate": 0.00015384615384615385, |
|
"loss": 0.2707, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0769230769230769, |
|
"grad_norm": 0.22495949268341064, |
|
"learning_rate": 0.00014615384615384615, |
|
"loss": 0.185, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2307692307692308, |
|
"grad_norm": 0.3134849965572357, |
|
"learning_rate": 0.00013846153846153847, |
|
"loss": 0.1708, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3846153846153846, |
|
"grad_norm": 0.7548186182975769, |
|
"learning_rate": 0.00013076923076923077, |
|
"loss": 0.0594, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"grad_norm": 1.528877854347229, |
|
"learning_rate": 0.0001230769230769231, |
|
"loss": 0.1062, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.5384615384615383, |
|
"eval_accuracy": 0.9849624060150376, |
|
"eval_loss": 0.05330296605825424, |
|
"eval_runtime": 1.2608, |
|
"eval_samples_per_second": 105.489, |
|
"eval_steps_per_second": 13.484, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6923076923076923, |
|
"grad_norm": 0.1376911699771881, |
|
"learning_rate": 0.00011538461538461538, |
|
"loss": 0.0669, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8461538461538463, |
|
"grad_norm": 0.17515841126441956, |
|
"learning_rate": 0.0001076923076923077, |
|
"loss": 0.0728, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 0.09629915654659271, |
|
"learning_rate": 0.0001, |
|
"loss": 0.0599, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1538461538461537, |
|
"grad_norm": 0.2766275703907013, |
|
"learning_rate": 9.230769230769232e-05, |
|
"loss": 0.0285, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.3076923076923075, |
|
"grad_norm": 0.0857153981924057, |
|
"learning_rate": 8.461538461538461e-05, |
|
"loss": 0.0523, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4615384615384617, |
|
"grad_norm": 0.0917874425649643, |
|
"learning_rate": 7.692307692307693e-05, |
|
"loss": 0.0171, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.6153846153846154, |
|
"grad_norm": 0.06923723220825195, |
|
"learning_rate": 6.923076923076924e-05, |
|
"loss": 0.015, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.769230769230769, |
|
"grad_norm": 3.2087008953094482, |
|
"learning_rate": 6.153846153846155e-05, |
|
"loss": 0.0323, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.9230769230769234, |
|
"grad_norm": 0.061998046934604645, |
|
"learning_rate": 5.384615384615385e-05, |
|
"loss": 0.0123, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"grad_norm": 0.05889332294464111, |
|
"learning_rate": 4.615384615384616e-05, |
|
"loss": 0.0118, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.076923076923077, |
|
"eval_accuracy": 0.9924812030075187, |
|
"eval_loss": 0.024031903594732285, |
|
"eval_runtime": 1.7763, |
|
"eval_samples_per_second": 74.877, |
|
"eval_steps_per_second": 9.571, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.230769230769231, |
|
"grad_norm": 0.05593299865722656, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.0112, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.3846153846153846, |
|
"grad_norm": 0.054984573274850845, |
|
"learning_rate": 3.0769230769230774e-05, |
|
"loss": 0.0108, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.5384615384615383, |
|
"grad_norm": 0.054389726370573044, |
|
"learning_rate": 2.307692307692308e-05, |
|
"loss": 0.0103, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.6923076923076925, |
|
"grad_norm": 0.05344008281826973, |
|
"learning_rate": 1.5384615384615387e-05, |
|
"loss": 0.0113, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.8461538461538463, |
|
"grad_norm": 0.05246325582265854, |
|
"learning_rate": 7.692307692307694e-06, |
|
"loss": 0.0102, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.0501759834587574, |
|
"learning_rate": 0.0, |
|
"loss": 0.0103, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 260, |
|
"total_flos": 3.205097416476426e+17, |
|
"train_loss": 0.11492267976013514, |
|
"train_runtime": 92.1679, |
|
"train_samples_per_second": 44.875, |
|
"train_steps_per_second": 2.821 |
|
} |
|
], |
|
"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 |
|
} |
|
|