| { |
| "best_global_step": 4266, |
| "best_metric": 0.39381444454193115, |
| "best_model_checkpoint": "distilbert_km_50_v1_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": 1.8509116172790527, |
| "learning_rate": 4.9e-05, |
| "loss": 0.4859, |
| "step": 1422 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7893643334157804, |
| "eval_combined_score": 0.7328794611295348, |
| "eval_f1": 0.6763945888432893, |
| "eval_loss": 0.4412091374397278, |
| "eval_runtime": 16.7192, |
| "eval_samples_per_second": 2418.182, |
| "eval_steps_per_second": 9.45, |
| "step": 1422 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.975377082824707, |
| "learning_rate": 4.8e-05, |
| "loss": 0.3828, |
| "step": 2844 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8177590897848133, |
| "eval_combined_score": 0.7828705149293473, |
| "eval_f1": 0.7479819400738815, |
| "eval_loss": 0.39621418714523315, |
| "eval_runtime": 16.2754, |
| "eval_samples_per_second": 2484.116, |
| "eval_steps_per_second": 9.708, |
| "step": 2844 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.436535120010376, |
| "learning_rate": 4.7e-05, |
| "loss": 0.31, |
| "step": 4266 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8249072470937423, |
| "eval_combined_score": 0.7963854156439363, |
| "eval_f1": 0.7678635841941301, |
| "eval_loss": 0.39381444454193115, |
| "eval_runtime": 16.2433, |
| "eval_samples_per_second": 2489.027, |
| "eval_steps_per_second": 9.727, |
| "step": 4266 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 4.190128803253174, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.2484, |
| "step": 5688 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.829903536977492, |
| "eval_combined_score": 0.7946866500948457, |
| "eval_f1": 0.7594697632121996, |
| "eval_loss": 0.4210052490234375, |
| "eval_runtime": 16.3217, |
| "eval_samples_per_second": 2477.066, |
| "eval_steps_per_second": 9.68, |
| "step": 5688 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.100914239883423, |
| "learning_rate": 4.5e-05, |
| "loss": 0.1964, |
| "step": 7110 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8319812020776651, |
| "eval_combined_score": 0.7991961304222788, |
| "eval_f1": 0.7664110587668925, |
| "eval_loss": 0.4503824710845947, |
| "eval_runtime": 16.3451, |
| "eval_samples_per_second": 2473.524, |
| "eval_steps_per_second": 9.667, |
| "step": 7110 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 3.7162628173828125, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1544, |
| "step": 8532 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8262181548355182, |
| "eval_combined_score": 0.8019452270285432, |
| "eval_f1": 0.7776722992215682, |
| "eval_loss": 0.5133235454559326, |
| "eval_runtime": 16.325, |
| "eval_samples_per_second": 2476.572, |
| "eval_steps_per_second": 9.678, |
| "step": 8532 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 3.272566556930542, |
| "learning_rate": 4.3e-05, |
| "loss": 0.124, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8273806579272818, |
| "eval_combined_score": 0.8023079037346199, |
| "eval_f1": 0.7772351495419579, |
| "eval_loss": 0.5642848610877991, |
| "eval_runtime": 16.3093, |
| "eval_samples_per_second": 2478.951, |
| "eval_steps_per_second": 9.688, |
| "step": 9954 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 5.107132434844971, |
| "learning_rate": 4.2e-05, |
| "loss": 0.1007, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8301508780608459, |
| "eval_combined_score": 0.8036596620242251, |
| "eval_f1": 0.7771684459876043, |
| "eval_loss": 0.6268412470817566, |
| "eval_runtime": 16.3218, |
| "eval_samples_per_second": 2477.052, |
| "eval_steps_per_second": 9.68, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 11376, |
| "total_flos": 1.927909325255639e+17, |
| "train_loss": 0.25033773483941657, |
| "train_runtime": 2767.8279, |
| "train_samples_per_second": 6572.771, |
| "train_steps_per_second": 25.688 |
| } |
| ], |
| "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 |
| } |
|
|
