|
{ |
|
"best_metric": 0.7025748491287231, |
|
"best_model_checkpoint": "bert_tiny_lda_100_v1_wnli/checkpoint-6", |
|
"epoch": 7.0, |
|
"eval_steps": 500, |
|
"global_step": 21, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 1.800396203994751, |
|
"learning_rate": 4.9e-05, |
|
"loss": 0.7082, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.352112676056338, |
|
"eval_loss": 0.7107725739479065, |
|
"eval_runtime": 0.0282, |
|
"eval_samples_per_second": 2516.037, |
|
"eval_steps_per_second": 35.437, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 1.2788060903549194, |
|
"learning_rate": 4.8e-05, |
|
"loss": 0.7024, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.5633802816901409, |
|
"eval_loss": 0.7025748491287231, |
|
"eval_runtime": 0.0293, |
|
"eval_samples_per_second": 2424.71, |
|
"eval_steps_per_second": 34.151, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 0.7909756302833557, |
|
"learning_rate": 4.7e-05, |
|
"loss": 0.7001, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.352112676056338, |
|
"eval_loss": 0.723371684551239, |
|
"eval_runtime": 0.0314, |
|
"eval_samples_per_second": 2260.411, |
|
"eval_steps_per_second": 31.837, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 0.8608013987541199, |
|
"learning_rate": 4.600000000000001e-05, |
|
"loss": 0.6938, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.5070422535211268, |
|
"eval_loss": 0.7065912485122681, |
|
"eval_runtime": 0.0296, |
|
"eval_samples_per_second": 2398.618, |
|
"eval_steps_per_second": 33.783, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"grad_norm": 1.210613489151001, |
|
"learning_rate": 4.5e-05, |
|
"loss": 0.6922, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.4225352112676056, |
|
"eval_loss": 0.7106072902679443, |
|
"eval_runtime": 0.0321, |
|
"eval_samples_per_second": 2208.56, |
|
"eval_steps_per_second": 31.106, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"grad_norm": 1.0442590713500977, |
|
"learning_rate": 4.4000000000000006e-05, |
|
"loss": 0.6958, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.36619718309859156, |
|
"eval_loss": 0.7257372736930847, |
|
"eval_runtime": 0.0321, |
|
"eval_samples_per_second": 2214.884, |
|
"eval_steps_per_second": 31.196, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"grad_norm": 0.6264722943305969, |
|
"learning_rate": 4.3e-05, |
|
"loss": 0.6891, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.29577464788732394, |
|
"eval_loss": 0.721610963344574, |
|
"eval_runtime": 0.0298, |
|
"eval_samples_per_second": 2382.593, |
|
"eval_steps_per_second": 33.558, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"step": 21, |
|
"total_flos": 116563804707840.0, |
|
"train_loss": 0.6973756040845599, |
|
"train_runtime": 8.699, |
|
"train_samples_per_second": 3649.84, |
|
"train_steps_per_second": 17.243 |
|
} |
|
], |
|
"logging_steps": 1, |
|
"max_steps": 150, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 50, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"EarlyStoppingCallback": { |
|
"args": { |
|
"early_stopping_patience": 5, |
|
"early_stopping_threshold": 0.0 |
|
}, |
|
"attributes": { |
|
"early_stopping_patience_counter": 5 |
|
} |
|
}, |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 116563804707840.0, |
|
"train_batch_size": 256, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|