|
{ |
|
"best_metric": 0.7729083665338645, |
|
"best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-qc45moa5/checkpoint-200", |
|
"epoch": 2.0, |
|
"eval_steps": 500, |
|
"global_step": 200, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 2.348206043243408, |
|
"learning_rate": 4.68218942679356e-05, |
|
"loss": 0.5991, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 3.71054744720459, |
|
"learning_rate": 4.562133287645007e-05, |
|
"loss": 0.4764, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 4.863480567932129, |
|
"learning_rate": 4.442077148496454e-05, |
|
"loss": 0.4552, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 6.876228332519531, |
|
"learning_rate": 4.3220210093479014e-05, |
|
"loss": 0.4601, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"grad_norm": 4.546775817871094, |
|
"learning_rate": 4.201964870199349e-05, |
|
"loss": 0.4314, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 5.100883483886719, |
|
"learning_rate": 4.0819087310507955e-05, |
|
"loss": 0.3505, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 7.4187774658203125, |
|
"learning_rate": 3.961852591902243e-05, |
|
"loss": 0.459, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 3.1828725337982178, |
|
"learning_rate": 3.84179645275369e-05, |
|
"loss": 0.3651, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"grad_norm": 3.504612445831299, |
|
"learning_rate": 3.7217403136051376e-05, |
|
"loss": 0.3359, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 2.794691801071167, |
|
"learning_rate": 3.601684174456584e-05, |
|
"loss": 0.2915, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.8275, |
|
"eval_f1": 0.7434944237918215, |
|
"eval_loss": 0.3887926936149597, |
|
"eval_precision": 0.684931506849315, |
|
"eval_recall": 0.8130081300813008, |
|
"eval_runtime": 1.5334, |
|
"eval_samples_per_second": 260.864, |
|
"eval_steps_per_second": 16.304, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"grad_norm": 5.653812408447266, |
|
"learning_rate": 3.4816280353080317e-05, |
|
"loss": 0.2709, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"grad_norm": 4.2549967765808105, |
|
"learning_rate": 3.3615718961594784e-05, |
|
"loss": 0.3661, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"grad_norm": 6.278103351593018, |
|
"learning_rate": 3.2415157570109264e-05, |
|
"loss": 0.2245, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"grad_norm": 5.513023376464844, |
|
"learning_rate": 3.121459617862373e-05, |
|
"loss": 0.3022, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"grad_norm": 3.467862844467163, |
|
"learning_rate": 3.0014034787138205e-05, |
|
"loss": 0.1837, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 5.904586315155029, |
|
"learning_rate": 2.8813473395652675e-05, |
|
"loss": 0.2116, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"grad_norm": 7.647643566131592, |
|
"learning_rate": 2.7612912004167145e-05, |
|
"loss": 0.2362, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"grad_norm": 4.498129367828369, |
|
"learning_rate": 2.6412350612681622e-05, |
|
"loss": 0.3255, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"grad_norm": 5.15887451171875, |
|
"learning_rate": 2.5211789221196093e-05, |
|
"loss": 0.2468, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 2.0280911922454834, |
|
"learning_rate": 2.4011227829710563e-05, |
|
"loss": 0.241, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8575, |
|
"eval_f1": 0.7729083665338645, |
|
"eval_loss": 0.3814562261104584, |
|
"eval_precision": 0.7578125, |
|
"eval_recall": 0.7886178861788617, |
|
"eval_runtime": 1.5262, |
|
"eval_samples_per_second": 262.083, |
|
"eval_steps_per_second": 16.38, |
|
"step": 200 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 400, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 423630740901888.0, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": { |
|
"_wandb": {}, |
|
"assignments": {}, |
|
"learning_rate": 4.8022455659421126e-05, |
|
"metric": "eval/loss", |
|
"num_train_epochs": 4, |
|
"per_device_train_batch_size": 16, |
|
"seed": 21 |
|
} |
|
} |
|
|