| { | |
| "best_metric": 0.4766906339372134, | |
| "best_model_checkpoint": "trained_models/training_1696386400_imdb_t5-large/model/checkpoint-1250", | |
| "epoch": 8.0, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2083, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0004, | |
| "loss": 0.0893, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_acc": 0.9126, | |
| "eval_f1": 0.4766906339372134, | |
| "eval_loss": 0.08332405984401703, | |
| "eval_runtime": 2242.501, | |
| "eval_samples_per_second": 11.148, | |
| "eval_steps_per_second": 1.394, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0006, | |
| "loss": 0.1071, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0008, | |
| "loss": 0.1357, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.1427, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_acc": 0.8804, | |
| "eval_f1": 0.4681979775093732, | |
| "eval_loss": 0.1213899627327919, | |
| "eval_runtime": 2199.3043, | |
| "eval_samples_per_second": 11.367, | |
| "eval_steps_per_second": 1.421, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0009777777777777777, | |
| "loss": 0.1386, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0009555555555555556, | |
| "loss": 0.1408, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_acc": 0.90072, | |
| "eval_f1": 0.47388185764694557, | |
| "eval_loss": 0.10105577856302261, | |
| "eval_runtime": 2213.3186, | |
| "eval_samples_per_second": 11.295, | |
| "eval_steps_per_second": 1.412, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0009333333333333333, | |
| "loss": 0.1132, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0009111111111111111, | |
| "loss": 0.1031, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0008888888888888888, | |
| "loss": 0.1014, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_acc": 0.87984, | |
| "eval_f1": 0.4667470127189438, | |
| "eval_loss": 0.10490511357784271, | |
| "eval_runtime": 2197.6464, | |
| "eval_samples_per_second": 11.376, | |
| "eval_steps_per_second": 1.422, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0008666666666666667, | |
| "loss": 0.0786, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0008444444444444444, | |
| "loss": 0.0884, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_acc": 0.86384, | |
| "eval_f1": 0.4614100111141227, | |
| "eval_loss": 0.1329798400402069, | |
| "eval_runtime": 2272.8013, | |
| "eval_samples_per_second": 11.0, | |
| "eval_steps_per_second": 1.375, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.0008222222222222222, | |
| "loss": 0.0797, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0008, | |
| "loss": 0.078, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0007777777777777778, | |
| "loss": 0.0787, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_acc": 0.88716, | |
| "eval_f1": 0.4692667461450598, | |
| "eval_loss": 0.10306421667337418, | |
| "eval_runtime": 2288.8555, | |
| "eval_samples_per_second": 10.922, | |
| "eval_steps_per_second": 1.365, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0007555555555555555, | |
| "loss": 0.0599, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.0007333333333333333, | |
| "loss": 0.0662, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_acc": 0.8984, | |
| "eval_f1": 0.4730434700951143, | |
| "eval_loss": 0.10054909437894821, | |
| "eval_runtime": 2312.3779, | |
| "eval_samples_per_second": 10.811, | |
| "eval_steps_per_second": 1.351, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.0007111111111111111, | |
| "loss": 0.057, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 0.000688888888888889, | |
| "loss": 0.0553, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0006666666666666666, | |
| "loss": 0.0637, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_acc": 0.903, | |
| "eval_f1": 0.47421149544208796, | |
| "eval_loss": 0.0991751179099083, | |
| "eval_runtime": 2251.6698, | |
| "eval_samples_per_second": 11.103, | |
| "eval_steps_per_second": 1.388, | |
| "step": 10000 | |
| } | |
| ], | |
| "max_steps": 25000, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.4640756736e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |