| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 4390, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5694760820045558, | |
| "grad_norm": 1.5650427341461182, | |
| "learning_rate": 1.7722095671981778e-05, | |
| "loss": 0.5173, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9492747867253404, | |
| "eval_f1": 0.7589075584259208, | |
| "eval_loss": 0.21155649423599243, | |
| "eval_precision": 0.7429275610801543, | |
| "eval_recall": 0.7755901107506432, | |
| "eval_runtime": 3.0816, | |
| "eval_samples_per_second": 1054.64, | |
| "eval_steps_per_second": 66.199, | |
| "step": 878 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 3.1316566467285156, | |
| "learning_rate": 1.5444191343963555e-05, | |
| "loss": 0.2538, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "grad_norm": 3.542625665664673, | |
| "learning_rate": 1.3166287015945332e-05, | |
| "loss": 0.196, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9620474367324893, | |
| "eval_f1": 0.8322505413959687, | |
| "eval_loss": 0.15278467535972595, | |
| "eval_precision": 0.8262403528114663, | |
| "eval_recall": 0.838348808591565, | |
| "eval_runtime": 4.0623, | |
| "eval_samples_per_second": 800.036, | |
| "eval_steps_per_second": 50.218, | |
| "step": 1756 | |
| }, | |
| { | |
| "epoch": 2.277904328018223, | |
| "grad_norm": 5.83428955078125, | |
| "learning_rate": 1.0888382687927108e-05, | |
| "loss": 0.1682, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.847380410022779, | |
| "grad_norm": 4.558180332183838, | |
| "learning_rate": 8.610478359908885e-06, | |
| "loss": 0.1444, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9651611673312469, | |
| "eval_f1": 0.8525989138867338, | |
| "eval_loss": 0.13545359671115875, | |
| "eval_precision": 0.8447348193697156, | |
| "eval_recall": 0.860610806577917, | |
| "eval_runtime": 2.9602, | |
| "eval_samples_per_second": 1097.888, | |
| "eval_steps_per_second": 68.914, | |
| "step": 2634 | |
| }, | |
| { | |
| "epoch": 3.416856492027335, | |
| "grad_norm": 2.630763292312622, | |
| "learning_rate": 6.3325740318906616e-06, | |
| "loss": 0.1302, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.9863325740318905, | |
| "grad_norm": 5.151127338409424, | |
| "learning_rate": 4.054669703872437e-06, | |
| "loss": 0.116, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9662573275930545, | |
| "eval_f1": 0.85545364128633, | |
| "eval_loss": 0.12546123564243317, | |
| "eval_precision": 0.845179604760345, | |
| "eval_recall": 0.865980534735429, | |
| "eval_runtime": 3.1188, | |
| "eval_samples_per_second": 1042.072, | |
| "eval_steps_per_second": 65.41, | |
| "step": 3512 | |
| }, | |
| { | |
| "epoch": 4.555808656036446, | |
| "grad_norm": 3.605297803878784, | |
| "learning_rate": 1.7767653758542143e-06, | |
| "loss": 0.1116, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4390, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 91500454459296.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |