| { | |
| "best_global_step": 5688, | |
| "best_metric": 0.31825923919677734, | |
| "best_model_checkpoint": "tinybert_base_train_book_ent_15p_s_init_qqp/checkpoint-5688", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 12798, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7183722257614136, | |
| "learning_rate": 4.900070323488045e-05, | |
| "loss": 0.4585, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8173138758347761, | |
| "eval_combined_score": 0.7915378208299848, | |
| "eval_f1": 0.7657617658251935, | |
| "eval_loss": 0.38073402643203735, | |
| "eval_runtime": 21.6007, | |
| "eval_samples_per_second": 1871.695, | |
| "eval_steps_per_second": 7.315, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.8384530544281006, | |
| "learning_rate": 4.800070323488045e-05, | |
| "loss": 0.3589, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8233737323769478, | |
| "eval_combined_score": 0.8073442495135272, | |
| "eval_f1": 0.7913147666501067, | |
| "eval_loss": 0.371241956949234, | |
| "eval_runtime": 21.5593, | |
| "eval_samples_per_second": 1875.294, | |
| "eval_steps_per_second": 7.329, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.3306589126586914, | |
| "learning_rate": 4.700070323488045e-05, | |
| "loss": 0.3115, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8476873608706406, | |
| "eval_combined_score": 0.8294305809627335, | |
| "eval_f1": 0.8111738010548264, | |
| "eval_loss": 0.3361490070819855, | |
| "eval_runtime": 21.4495, | |
| "eval_samples_per_second": 1884.89, | |
| "eval_steps_per_second": 7.366, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.4020464420318604, | |
| "learning_rate": 4.600070323488046e-05, | |
| "loss": 0.2742, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8579520158298294, | |
| "eval_combined_score": 0.8305796812863981, | |
| "eval_f1": 0.8032073467429668, | |
| "eval_loss": 0.31825923919677734, | |
| "eval_runtime": 21.6362, | |
| "eval_samples_per_second": 1868.627, | |
| "eval_steps_per_second": 7.303, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.4398610591888428, | |
| "learning_rate": 4.500070323488045e-05, | |
| "loss": 0.2431, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8610685134800891, | |
| "eval_combined_score": 0.8386481124573683, | |
| "eval_f1": 0.8162277114346475, | |
| "eval_loss": 0.3296893835067749, | |
| "eval_runtime": 21.5093, | |
| "eval_samples_per_second": 1879.655, | |
| "eval_steps_per_second": 7.346, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.022575378417969, | |
| "learning_rate": 4.400070323488045e-05, | |
| "loss": 0.2161, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8658174622804848, | |
| "eval_combined_score": 0.8453324550287905, | |
| "eval_f1": 0.8248474477770962, | |
| "eval_loss": 0.32380661368370056, | |
| "eval_runtime": 21.4879, | |
| "eval_samples_per_second": 1881.524, | |
| "eval_steps_per_second": 7.353, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.2843637466430664, | |
| "learning_rate": 4.300070323488045e-05, | |
| "loss": 0.1931, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8625772940885481, | |
| "eval_combined_score": 0.843382818299258, | |
| "eval_f1": 0.8241883425099678, | |
| "eval_loss": 0.34378689527511597, | |
| "eval_runtime": 21.4885, | |
| "eval_samples_per_second": 1881.471, | |
| "eval_steps_per_second": 7.353, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.7418906688690186, | |
| "learning_rate": 4.200070323488045e-05, | |
| "loss": 0.171, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8671531041305961, | |
| "eval_combined_score": 0.8482060462955171, | |
| "eval_f1": 0.8292589884604381, | |
| "eval_loss": 0.37025919556617737, | |
| "eval_runtime": 21.5924, | |
| "eval_samples_per_second": 1872.419, | |
| "eval_steps_per_second": 7.317, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.6632304191589355, | |
| "learning_rate": 4.100070323488045e-05, | |
| "loss": 0.1516, | |
| "step": 12798 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.862824635171902, | |
| "eval_combined_score": 0.845481112131969, | |
| "eval_f1": 0.8281375890920359, | |
| "eval_loss": 0.39689913392066956, | |
| "eval_runtime": 21.8663, | |
| "eval_samples_per_second": 1848.96, | |
| "eval_steps_per_second": 7.226, | |
| "step": 12798 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 12798, | |
| "total_flos": 6.475495387021517e+16, | |
| "train_loss": 0.26421997885831466, | |
| "train_runtime": 2604.7894, | |
| "train_samples_per_second": 6984.173, | |
| "train_steps_per_second": 27.296 | |
| } | |
| ], | |
| "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": 6.475495387021517e+16, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |