| { |
| "best_global_step": 2844, |
| "best_metric": 0.40847912430763245, |
| "best_model_checkpoint": "distilbert_km_100_v1_qqp/checkpoint-2844", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 9954, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.6529167890548706, |
| "learning_rate": 4.9e-05, |
| "loss": 0.4919, |
| "step": 1422 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7821419737818451, |
| "eval_combined_score": 0.7231844361132465, |
| "eval_f1": 0.6642268984446478, |
| "eval_loss": 0.453318327665329, |
| "eval_runtime": 16.5453, |
| "eval_samples_per_second": 2443.594, |
| "eval_steps_per_second": 9.55, |
| "step": 1422 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.8776006698608398, |
| "learning_rate": 4.8e-05, |
| "loss": 0.3921, |
| "step": 2844 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8093000247341083, |
| "eval_combined_score": 0.7758494777863658, |
| "eval_f1": 0.7423989308386234, |
| "eval_loss": 0.40847912430763245, |
| "eval_runtime": 16.2004, |
| "eval_samples_per_second": 2495.612, |
| "eval_steps_per_second": 9.753, |
| "step": 2844 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.3093512058258057, |
| "learning_rate": 4.7e-05, |
| "loss": 0.3188, |
| "step": 4266 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8208013851100667, |
| "eval_combined_score": 0.7885499744010422, |
| "eval_f1": 0.7562985636920179, |
| "eval_loss": 0.41204676032066345, |
| "eval_runtime": 16.2098, |
| "eval_samples_per_second": 2494.176, |
| "eval_steps_per_second": 9.747, |
| "step": 4266 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.299015522003174, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.2569, |
| "step": 5688 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8241652238436804, |
| "eval_combined_score": 0.7853308378605772, |
| "eval_f1": 0.7464964518774739, |
| "eval_loss": 0.44257310032844543, |
| "eval_runtime": 16.2338, |
| "eval_samples_per_second": 2490.488, |
| "eval_steps_per_second": 9.733, |
| "step": 5688 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.1120574474334717, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2054, |
| "step": 7110 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8259955478604997, |
| "eval_combined_score": 0.7966821759010841, |
| "eval_f1": 0.7673688039416686, |
| "eval_loss": 0.45467233657836914, |
| "eval_runtime": 16.2003, |
| "eval_samples_per_second": 2495.63, |
| "eval_steps_per_second": 9.753, |
| "step": 7110 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 5.509576320648193, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1647, |
| "step": 8532 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8284442245857037, |
| "eval_combined_score": 0.800248140950667, |
| "eval_f1": 0.7720520573156303, |
| "eval_loss": 0.4839842915534973, |
| "eval_runtime": 16.2584, |
| "eval_samples_per_second": 2486.719, |
| "eval_steps_per_second": 9.718, |
| "step": 8532 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 4.307466506958008, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1329, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.824956715310413, |
| "eval_combined_score": 0.7981377107427432, |
| "eval_f1": 0.7713187061750735, |
| "eval_loss": 0.5799120664596558, |
| "eval_runtime": 16.2265, |
| "eval_samples_per_second": 2491.607, |
| "eval_steps_per_second": 9.737, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 9954, |
| "total_flos": 1.686920659598684e+17, |
| "train_loss": 0.2803885850988767, |
| "train_runtime": 2417.6003, |
| "train_samples_per_second": 7524.941, |
| "train_steps_per_second": 29.409 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 71100, |
| "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": 1.686920659598684e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
