| { |
| "best_global_step": 30, |
| "best_metric": 0.591094434261322, |
| "best_model_checkpoint": "bert_base_rand_10_v1_mrpc/checkpoint-30", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 105, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.8852230906486511, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6362, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6985294117647058, |
| "eval_combined_score": 0.7550840014413116, |
| "eval_f1": 0.8116385911179173, |
| "eval_loss": 0.5929145812988281, |
| "eval_runtime": 0.2694, |
| "eval_samples_per_second": 1514.386, |
| "eval_steps_per_second": 7.423, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.5255680084228516, |
| "learning_rate": 4.8e-05, |
| "loss": 0.5718, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.7156862745098039, |
| "eval_combined_score": 0.7614976223047358, |
| "eval_f1": 0.8073089700996677, |
| "eval_loss": 0.591094434261322, |
| "eval_runtime": 0.2892, |
| "eval_samples_per_second": 1410.999, |
| "eval_steps_per_second": 6.917, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.7048187255859375, |
| "learning_rate": 4.7e-05, |
| "loss": 0.4802, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6617647058823529, |
| "eval_combined_score": 0.7094034797017399, |
| "eval_f1": 0.7570422535211268, |
| "eval_loss": 0.6726020574569702, |
| "eval_runtime": 0.2783, |
| "eval_samples_per_second": 1466.223, |
| "eval_steps_per_second": 7.187, |
| "step": 45 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 5.6311116218566895, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.3577, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6666666666666666, |
| "eval_combined_score": 0.7136150234741784, |
| "eval_f1": 0.7605633802816901, |
| "eval_loss": 0.7456569075584412, |
| "eval_runtime": 0.3148, |
| "eval_samples_per_second": 1296.242, |
| "eval_steps_per_second": 6.354, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 4.9386162757873535, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2272, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6372549019607843, |
| "eval_combined_score": 0.6840819964349376, |
| "eval_f1": 0.730909090909091, |
| "eval_loss": 1.0270462036132812, |
| "eval_runtime": 0.2766, |
| "eval_samples_per_second": 1474.94, |
| "eval_steps_per_second": 7.23, |
| "step": 75 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.5608620643615723, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1378, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6421568627450981, |
| "eval_combined_score": 0.6888320545609549, |
| "eval_f1": 0.7355072463768116, |
| "eval_loss": 1.1775949001312256, |
| "eval_runtime": 0.2748, |
| "eval_samples_per_second": 1484.885, |
| "eval_steps_per_second": 7.279, |
| "step": 90 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 5.652944087982178, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1022, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6274509803921569, |
| "eval_combined_score": 0.6740196078431373, |
| "eval_f1": 0.7205882352941176, |
| "eval_loss": 1.371445894241333, |
| "eval_runtime": 0.2791, |
| "eval_samples_per_second": 1461.721, |
| "eval_steps_per_second": 7.165, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 105, |
| "total_flos": 3377819728711680.0, |
| "train_loss": 0.35898870059422083, |
| "train_runtime": 57.9855, |
| "train_samples_per_second": 3162.86, |
| "train_steps_per_second": 12.934 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 750, |
| "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": 3377819728711680.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
