| { | |
| "best_metric": 0.3234252333641052, | |
| "best_model_checkpoint": "bert_base_lda_100_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": 1.8851810693740845, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.4474, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8283947563690329, | |
| "eval_combined_score": 0.7949221753238371, | |
| "eval_f1": 0.7614495942786413, | |
| "eval_loss": 0.3685195744037628, | |
| "eval_runtime": 26.2526, | |
| "eval_samples_per_second": 1540.036, | |
| "eval_steps_per_second": 6.018, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.972749710083008, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.3271, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.847588424437299, | |
| "eval_combined_score": 0.8289650340099699, | |
| "eval_f1": 0.8103416435826409, | |
| "eval_loss": 0.3386009633541107, | |
| "eval_runtime": 26.2178, | |
| "eval_samples_per_second": 1542.084, | |
| "eval_steps_per_second": 6.026, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.082465887069702, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.2564, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8609448429384121, | |
| "eval_combined_score": 0.8397833002922801, | |
| "eval_f1": 0.818621757646148, | |
| "eval_loss": 0.3234252333641052, | |
| "eval_runtime": 26.1778, | |
| "eval_samples_per_second": 1544.441, | |
| "eval_steps_per_second": 6.036, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.214107036590576, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.1978, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.865322780113777, | |
| "eval_combined_score": 0.845827594247875, | |
| "eval_f1": 0.8263324083819731, | |
| "eval_loss": 0.3627840280532837, | |
| "eval_runtime": 26.0619, | |
| "eval_samples_per_second": 1551.309, | |
| "eval_steps_per_second": 6.062, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.68060302734375, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1516, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.869527578530794, | |
| "eval_combined_score": 0.8474033176016561, | |
| "eval_f1": 0.8252790566725182, | |
| "eval_loss": 0.40143004059791565, | |
| "eval_runtime": 26.1999, | |
| "eval_samples_per_second": 1543.138, | |
| "eval_steps_per_second": 6.031, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.411165237426758, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.1169, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.867276774672273, | |
| "eval_combined_score": 0.8475178249685433, | |
| "eval_f1": 0.8277588752648135, | |
| "eval_loss": 0.3963766396045685, | |
| "eval_runtime": 26.1361, | |
| "eval_samples_per_second": 1546.904, | |
| "eval_steps_per_second": 6.045, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.608915090560913, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.093, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8676230521889686, | |
| "eval_combined_score": 0.8477775754772003, | |
| "eval_f1": 0.8279320987654322, | |
| "eval_loss": 0.4813096225261688, | |
| "eval_runtime": 25.9104, | |
| "eval_samples_per_second": 1560.379, | |
| "eval_steps_per_second": 6.098, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.501567840576172, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.076, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8693297056641108, | |
| "eval_combined_score": 0.8489380492102074, | |
| "eval_f1": 0.8285463927563042, | |
| "eval_loss": 0.43455520272254944, | |
| "eval_runtime": 26.0711, | |
| "eval_samples_per_second": 1550.762, | |
| "eval_steps_per_second": 6.06, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 11376, | |
| "total_flos": 3.8292762019405824e+17, | |
| "train_loss": 0.2082672454469147, | |
| "train_runtime": 4927.5674, | |
| "train_samples_per_second": 3691.943, | |
| "train_steps_per_second": 14.429 | |
| } | |
| ], | |
| "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": 3.8292762019405824e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |