| { | |
| "best_metric": 0.930276087973795, | |
| "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-2016", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 5040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9920634920634921, | |
| "grad_norm": 0.06281786412000656, | |
| "learning_rate": 4.503968253968254e-05, | |
| "loss": 0.0189, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9983953339100828, | |
| "eval_f1": 0.9040358744394619, | |
| "eval_loss": 0.00518822530284524, | |
| "eval_precision": 0.8712186689714779, | |
| "eval_recall": 0.9394221808014911, | |
| "eval_runtime": 13.2131, | |
| "eval_samples_per_second": 525.689, | |
| "eval_steps_per_second": 65.768, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 1.9841269841269842, | |
| "grad_norm": 0.20689290761947632, | |
| "learning_rate": 4.007936507936508e-05, | |
| "loss": 0.0047, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9986883598917199, | |
| "eval_f1": 0.9244402985074627, | |
| "eval_loss": 0.004834321793168783, | |
| "eval_precision": 0.9253034547152195, | |
| "eval_recall": 0.923578751164958, | |
| "eval_runtime": 13.2434, | |
| "eval_samples_per_second": 524.487, | |
| "eval_steps_per_second": 65.618, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 2.9761904761904763, | |
| "grad_norm": 0.02256501279771328, | |
| "learning_rate": 3.511904761904762e-05, | |
| "loss": 0.0027, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9986255686099406, | |
| "eval_f1": 0.9239384041063929, | |
| "eval_loss": 0.005881821736693382, | |
| "eval_precision": 0.9252336448598131, | |
| "eval_recall": 0.9226467847157502, | |
| "eval_runtime": 13.1715, | |
| "eval_samples_per_second": 527.351, | |
| "eval_steps_per_second": 65.976, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "grad_norm": 0.0326182059943676, | |
| "learning_rate": 3.0158730158730158e-05, | |
| "loss": 0.0015, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9987162671280663, | |
| "eval_f1": 0.930276087973795, | |
| "eval_loss": 0.0064844791777431965, | |
| "eval_precision": 0.9342105263157895, | |
| "eval_recall": 0.9263746505125815, | |
| "eval_runtime": 13.5939, | |
| "eval_samples_per_second": 510.963, | |
| "eval_steps_per_second": 63.926, | |
| "step": 2016 | |
| }, | |
| { | |
| "epoch": 4.9603174603174605, | |
| "grad_norm": 0.1600140929222107, | |
| "learning_rate": 2.5198412698412697e-05, | |
| "loss": 0.0011, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9986116149917673, | |
| "eval_f1": 0.923076923076923, | |
| "eval_loss": 0.007346163038164377, | |
| "eval_precision": 0.9072907290729073, | |
| "eval_recall": 0.9394221808014911, | |
| "eval_runtime": 13.3438, | |
| "eval_samples_per_second": 520.543, | |
| "eval_steps_per_second": 65.124, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 5.9523809523809526, | |
| "grad_norm": 0.09641193598508835, | |
| "learning_rate": 2.023809523809524e-05, | |
| "loss": 0.0005, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.998444171573689, | |
| "eval_f1": 0.9204281060958585, | |
| "eval_loss": 0.009004838764667511, | |
| "eval_precision": 0.9191449814126395, | |
| "eval_recall": 0.9217148182665424, | |
| "eval_runtime": 13.3195, | |
| "eval_samples_per_second": 521.491, | |
| "eval_steps_per_second": 65.243, | |
| "step": 3024 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 0.029784763231873512, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.0007, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9985767309463344, | |
| "eval_f1": 0.9190432382704691, | |
| "eval_loss": 0.008385799825191498, | |
| "eval_precision": 0.9073569482288828, | |
| "eval_recall": 0.9310344827586207, | |
| "eval_runtime": 13.4485, | |
| "eval_samples_per_second": 516.487, | |
| "eval_steps_per_second": 64.617, | |
| "step": 3528 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "grad_norm": 0.0010318646673113108, | |
| "learning_rate": 1.0317460317460318e-05, | |
| "loss": 0.0004, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9985558005190746, | |
| "eval_f1": 0.9213793103448276, | |
| "eval_loss": 0.008501913398504257, | |
| "eval_precision": 0.9092558983666061, | |
| "eval_recall": 0.9338303821062441, | |
| "eval_runtime": 13.3613, | |
| "eval_samples_per_second": 519.86, | |
| "eval_steps_per_second": 65.039, | |
| "step": 4032 | |
| }, | |
| { | |
| "epoch": 8.928571428571429, | |
| "grad_norm": 0.0005677491426467896, | |
| "learning_rate": 5.357142857142857e-06, | |
| "loss": 0.0003, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9987023135098931, | |
| "eval_f1": 0.9270544783010157, | |
| "eval_loss": 0.008021777495741844, | |
| "eval_precision": 0.918572735590119, | |
| "eval_recall": 0.9356943150046598, | |
| "eval_runtime": 13.487, | |
| "eval_samples_per_second": 515.014, | |
| "eval_steps_per_second": 64.432, | |
| "step": 4536 | |
| }, | |
| { | |
| "epoch": 9.920634920634921, | |
| "grad_norm": 0.0005077613168396056, | |
| "learning_rate": 3.9682539682539683e-07, | |
| "loss": 0.0002, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9986883598917199, | |
| "eval_f1": 0.9278445883441258, | |
| "eval_loss": 0.008253143168985844, | |
| "eval_precision": 0.921028466483012, | |
| "eval_recall": 0.934762348555452, | |
| "eval_runtime": 13.7257, | |
| "eval_samples_per_second": 506.057, | |
| "eval_steps_per_second": 63.312, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 5040, | |
| "total_flos": 1.394320679130096e+16, | |
| "train_loss": 0.0030765269683407886, | |
| "train_runtime": 1249.6681, | |
| "train_samples_per_second": 257.924, | |
| "train_steps_per_second": 4.033 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5040, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.394320679130096e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |