| { | |
| "best_metric": 1.29217529296875, | |
| "best_model_checkpoint": "output/owi6-m9m1-h027-0/checkpoint-414", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 414, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.5476190476190476e-05, | |
| "loss": 1.6091, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.095238095238095e-05, | |
| "loss": 1.6068, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.642857142857143e-05, | |
| "loss": 1.6044, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.865591397849463e-05, | |
| "loss": 1.5946, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.690860215053764e-05, | |
| "loss": 1.5918, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.516129032258064e-05, | |
| "loss": 1.5766, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.341397849462366e-05, | |
| "loss": 1.5641, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.5567, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.991935483870968e-05, | |
| "loss": 1.5416, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.817204301075269e-05, | |
| "loss": 1.5457, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3890909090909091, | |
| "eval_f1_macro": 0.2871727910969339, | |
| "eval_f1_micro": 0.3890909090909091, | |
| "eval_f1_weighted": 0.3073118698887861, | |
| "eval_loss": 1.5030215978622437, | |
| "eval_precision_macro": 0.5867425092932165, | |
| "eval_precision_micro": 0.3890909090909091, | |
| "eval_precision_weighted": 0.5745601059619471, | |
| "eval_recall_macro": 0.34872131147540986, | |
| "eval_recall_micro": 0.3890909090909091, | |
| "eval_recall_weighted": 0.3890909090909091, | |
| "eval_runtime": 1.3431, | |
| "eval_samples_per_second": 409.495, | |
| "eval_steps_per_second": 13.402, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.64247311827957e-05, | |
| "loss": 1.5157, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.467741935483872e-05, | |
| "loss": 1.4911, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.293010752688172e-05, | |
| "loss": 1.4966, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.118279569892473e-05, | |
| "loss": 1.4792, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.9435483870967743e-05, | |
| "loss": 1.4843, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.768817204301075e-05, | |
| "loss": 1.4431, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.5940860215053762e-05, | |
| "loss": 1.4175, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.4327956989247314e-05, | |
| "loss": 1.4219, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.258064516129032e-05, | |
| "loss": 1.4021, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.3641, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.9086021505376344e-05, | |
| "loss": 1.4188, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4909090909090909, | |
| "eval_f1_macro": 0.41370213457202976, | |
| "eval_f1_micro": 0.4909090909090909, | |
| "eval_f1_weighted": 0.4294625784062733, | |
| "eval_loss": 1.3412450551986694, | |
| "eval_precision_macro": 0.7931712110046671, | |
| "eval_precision_micro": 0.4909090909090909, | |
| "eval_precision_weighted": 0.7625187021281347, | |
| "eval_recall_macro": 0.45226759397251204, | |
| "eval_recall_micro": 0.4909090909090909, | |
| "eval_recall_weighted": 0.4909090909090909, | |
| "eval_runtime": 1.2753, | |
| "eval_samples_per_second": 431.276, | |
| "eval_steps_per_second": 14.114, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.7473118279569895e-05, | |
| "loss": 1.36, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.5725806451612903e-05, | |
| "loss": 1.3683, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.3978494623655914e-05, | |
| "loss": 1.3277, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1.2231182795698925e-05, | |
| "loss": 1.3308, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.0483870967741936e-05, | |
| "loss": 1.3453, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 8.736559139784948e-06, | |
| "loss": 1.3551, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 7.1236559139784956e-06, | |
| "loss": 1.3131, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 5.376344086021506e-06, | |
| "loss": 1.3615, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 3.6290322580645166e-06, | |
| "loss": 1.3472, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.881720430107527e-06, | |
| "loss": 1.3738, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.58, | |
| "eval_f1_macro": 0.5376292484253957, | |
| "eval_f1_micro": 0.58, | |
| "eval_f1_weighted": 0.5466008675953137, | |
| "eval_loss": 1.29217529296875, | |
| "eval_precision_macro": 0.7771323529411764, | |
| "eval_precision_micro": 0.58, | |
| "eval_precision_weighted": 0.7513901515151514, | |
| "eval_recall_macro": 0.5474661367776121, | |
| "eval_recall_micro": 0.58, | |
| "eval_recall_weighted": 0.58, | |
| "eval_runtime": 1.2722, | |
| "eval_samples_per_second": 432.333, | |
| "eval_steps_per_second": 14.149, | |
| "step": 414 | |
| } | |
| ], | |
| "logging_steps": 13, | |
| "max_steps": 414, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.4070008986048922e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |