| { | |
| "best_metric": 0.7156354784965515, | |
| "best_model_checkpoint": "final_models/structroberta_sx2_final/finetune/mnli/checkpoint-8000", | |
| "epoch": 4.249422632794457, | |
| "global_step": 9200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.5428223013877869, | |
| "eval_loss": 0.9369295835494995, | |
| "eval_runtime": 13.9292, | |
| "eval_samples_per_second": 471.095, | |
| "eval_steps_per_second": 58.941, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.884526558891455e-05, | |
| "loss": 1.0082, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.6206949353218079, | |
| "eval_loss": 0.8501443266868591, | |
| "eval_runtime": 13.7492, | |
| "eval_samples_per_second": 477.264, | |
| "eval_steps_per_second": 59.713, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.7690531177829104e-05, | |
| "loss": 0.871, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.6496495008468628, | |
| "eval_loss": 0.7984079122543335, | |
| "eval_runtime": 13.7319, | |
| "eval_samples_per_second": 477.866, | |
| "eval_steps_per_second": 59.788, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.653579676674365e-05, | |
| "loss": 0.815, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.6677842140197754, | |
| "eval_loss": 0.762497067451477, | |
| "eval_runtime": 13.762, | |
| "eval_samples_per_second": 476.82, | |
| "eval_steps_per_second": 59.657, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.53810623556582e-05, | |
| "loss": 0.7787, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.6645839810371399, | |
| "eval_loss": 0.7623239159584045, | |
| "eval_runtime": 13.7401, | |
| "eval_samples_per_second": 477.58, | |
| "eval_steps_per_second": 59.752, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.6699177026748657, | |
| "eval_loss": 0.7520171999931335, | |
| "eval_runtime": 13.7455, | |
| "eval_samples_per_second": 477.394, | |
| "eval_steps_per_second": 59.729, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.422632794457275e-05, | |
| "loss": 0.7272, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.6799756288528442, | |
| "eval_loss": 0.7413595914840698, | |
| "eval_runtime": 13.7587, | |
| "eval_samples_per_second": 476.935, | |
| "eval_steps_per_second": 59.671, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.30715935334873e-05, | |
| "loss": 0.6926, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.6915574669837952, | |
| "eval_loss": 0.7350733876228333, | |
| "eval_runtime": 13.7303, | |
| "eval_samples_per_second": 477.921, | |
| "eval_steps_per_second": 59.795, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.1916859122401844e-05, | |
| "loss": 0.6859, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_accuracy": 0.7016153335571289, | |
| "eval_loss": 0.7069934606552124, | |
| "eval_runtime": 13.7469, | |
| "eval_samples_per_second": 477.343, | |
| "eval_steps_per_second": 59.722, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.07621247113164e-05, | |
| "loss": 0.6807, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.7028344869613647, | |
| "eval_loss": 0.709619402885437, | |
| "eval_runtime": 13.7492, | |
| "eval_samples_per_second": 477.263, | |
| "eval_steps_per_second": 59.712, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 0.7000914216041565, | |
| "eval_loss": 0.7379443049430847, | |
| "eval_runtime": 13.7346, | |
| "eval_samples_per_second": 477.772, | |
| "eval_steps_per_second": 59.776, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.960739030023095e-05, | |
| "loss": 0.6457, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.7045108079910278, | |
| "eval_loss": 0.7176894545555115, | |
| "eval_runtime": 13.7295, | |
| "eval_samples_per_second": 477.95, | |
| "eval_steps_per_second": 59.798, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.84526558891455e-05, | |
| "loss": 0.5912, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.7087777853012085, | |
| "eval_loss": 0.7219710946083069, | |
| "eval_runtime": 13.7644, | |
| "eval_samples_per_second": 476.736, | |
| "eval_steps_per_second": 59.647, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.729792147806005e-05, | |
| "loss": 0.5847, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_accuracy": 0.7104541063308716, | |
| "eval_loss": 0.7230660915374756, | |
| "eval_runtime": 13.7736, | |
| "eval_samples_per_second": 476.419, | |
| "eval_steps_per_second": 59.607, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.61431870669746e-05, | |
| "loss": 0.5924, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.7127400040626526, | |
| "eval_loss": 0.7147745490074158, | |
| "eval_runtime": 13.7316, | |
| "eval_samples_per_second": 477.876, | |
| "eval_steps_per_second": 59.789, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.7083206176757812, | |
| "eval_loss": 0.7208989858627319, | |
| "eval_runtime": 13.7467, | |
| "eval_samples_per_second": 477.352, | |
| "eval_steps_per_second": 59.724, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.498845265588915e-05, | |
| "loss": 0.5888, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_accuracy": 0.7095397710800171, | |
| "eval_loss": 0.7584365010261536, | |
| "eval_runtime": 13.7775, | |
| "eval_samples_per_second": 476.283, | |
| "eval_steps_per_second": 59.59, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.38337182448037e-05, | |
| "loss": 0.4934, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_accuracy": 0.7148734927177429, | |
| "eval_loss": 0.7487167716026306, | |
| "eval_runtime": 13.7416, | |
| "eval_samples_per_second": 477.528, | |
| "eval_steps_per_second": 59.746, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.2678983833718243e-05, | |
| "loss": 0.4967, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_accuracy": 0.7141115665435791, | |
| "eval_loss": 0.7581225037574768, | |
| "eval_runtime": 13.7553, | |
| "eval_samples_per_second": 477.051, | |
| "eval_steps_per_second": 59.686, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 3.1524249422632794e-05, | |
| "loss": 0.5027, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_accuracy": 0.7156354784965515, | |
| "eval_loss": 0.7501972913742065, | |
| "eval_runtime": 13.7448, | |
| "eval_samples_per_second": 477.416, | |
| "eval_steps_per_second": 59.732, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "eval_accuracy": 0.7136543989181519, | |
| "eval_loss": 0.7529350519180298, | |
| "eval_runtime": 13.7694, | |
| "eval_samples_per_second": 476.565, | |
| "eval_steps_per_second": 59.625, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.0369515011547345e-05, | |
| "loss": 0.513, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "eval_accuracy": 0.7093873620033264, | |
| "eval_loss": 0.840041995048523, | |
| "eval_runtime": 13.753, | |
| "eval_samples_per_second": 477.131, | |
| "eval_steps_per_second": 59.696, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 2.9214780600461896e-05, | |
| "loss": 0.44, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_accuracy": 0.7103017568588257, | |
| "eval_loss": 0.8139531016349792, | |
| "eval_runtime": 13.8049, | |
| "eval_samples_per_second": 475.34, | |
| "eval_steps_per_second": 59.472, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "step": 9200, | |
| "total_flos": 1.0112408227780608e+17, | |
| "train_loss": 0.6453767793074898, | |
| "train_runtime": 5369.95, | |
| "train_samples_per_second": 483.766, | |
| "train_steps_per_second": 4.032 | |
| } | |
| ], | |
| "max_steps": 21650, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.0112408227780608e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |