| { |
| "best_global_step": 4266, |
| "best_metric": 0.3996587097644806, |
| "best_model_checkpoint": "distilbert_km_20_v2_qqp/checkpoint-4266", |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 11376, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.645755648612976, |
| "learning_rate": 4.9e-05, |
| "loss": 0.4873, |
| "step": 1422 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7871382636655948, |
| "eval_combined_score": 0.7330094303402601, |
| "eval_f1": 0.6788805970149254, |
| "eval_loss": 0.44391870498657227, |
| "eval_runtime": 17.1295, |
| "eval_samples_per_second": 2360.254, |
| "eval_steps_per_second": 9.224, |
| "step": 1422 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.4969005584716797, |
| "learning_rate": 4.8e-05, |
| "loss": 0.3842, |
| "step": 2844 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8114271580509522, |
| "eval_combined_score": 0.7733984106811512, |
| "eval_f1": 0.7353696633113502, |
| "eval_loss": 0.40490370988845825, |
| "eval_runtime": 16.1565, |
| "eval_samples_per_second": 2502.405, |
| "eval_steps_per_second": 9.779, |
| "step": 2844 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.8454954624176025, |
| "learning_rate": 4.7e-05, |
| "loss": 0.3094, |
| "step": 4266 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8217412812268118, |
| "eval_combined_score": 0.7907980005814177, |
| "eval_f1": 0.7598547199360235, |
| "eval_loss": 0.3996587097644806, |
| "eval_runtime": 16.1929, |
| "eval_samples_per_second": 2496.774, |
| "eval_steps_per_second": 9.757, |
| "step": 4266 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 4.840691566467285, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.2464, |
| "step": 5688 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8266881028938907, |
| "eval_combined_score": 0.7929198944332307, |
| "eval_f1": 0.7591516859725708, |
| "eval_loss": 0.4325845241546631, |
| "eval_runtime": 17.0075, |
| "eval_samples_per_second": 2377.192, |
| "eval_steps_per_second": 9.29, |
| "step": 5688 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.780235767364502, |
| "learning_rate": 4.5e-05, |
| "loss": 0.1952, |
| "step": 7110 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8320801385110067, |
| "eval_combined_score": 0.8014577907744906, |
| "eval_f1": 0.7708354430379747, |
| "eval_loss": 0.44030794501304626, |
| "eval_runtime": 16.2856, |
| "eval_samples_per_second": 2482.564, |
| "eval_steps_per_second": 9.702, |
| "step": 7110 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 4.330679416656494, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1541, |
| "step": 8532 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8302745486025229, |
| "eval_combined_score": 0.7976131255035632, |
| "eval_f1": 0.7649517024046036, |
| "eval_loss": 0.5289042592048645, |
| "eval_runtime": 16.4617, |
| "eval_samples_per_second": 2456.0, |
| "eval_steps_per_second": 9.598, |
| "step": 8532 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 4.10150146484375, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1242, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8327974276527331, |
| "eval_combined_score": 0.8015811311101495, |
| "eval_f1": 0.7703648345675658, |
| "eval_loss": 0.6010123491287231, |
| "eval_runtime": 16.3775, |
| "eval_samples_per_second": 2468.636, |
| "eval_steps_per_second": 9.647, |
| "step": 9954 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 3.78735089302063, |
| "learning_rate": 4.2e-05, |
| "loss": 0.1003, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8328716299777393, |
| "eval_combined_score": 0.8030899566334497, |
| "eval_f1": 0.7733082832891602, |
| "eval_loss": 0.6165689826011658, |
| "eval_runtime": 16.263, |
| "eval_samples_per_second": 2486.012, |
| "eval_steps_per_second": 9.715, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 11376, |
| "total_flos": 1.927909325255639e+17, |
| "train_loss": 0.2501452965072439, |
| "train_runtime": 2776.6461, |
| "train_samples_per_second": 6551.897, |
| "train_steps_per_second": 25.606 |
| } |
| ], |
| "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": 1.927909325255639e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
