| { | |
| "best_metric": 0.05566494166851044, | |
| "best_model_checkpoint": "output/bxw3-9299-s4ce-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": 1.3095238095238096e-05, | |
| "loss": 1.6774, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.7380952380952383e-05, | |
| "loss": 1.4563, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.8852, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.905913978494624e-05, | |
| "loss": 0.4858, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.731182795698925e-05, | |
| "loss": 0.3189, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.556451612903226e-05, | |
| "loss": 0.1496, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.381720430107527e-05, | |
| "loss": 0.2137, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.206989247311828e-05, | |
| "loss": 0.1153, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.032258064516129e-05, | |
| "loss": 0.1762, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.8575268817204305e-05, | |
| "loss": 0.1755, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9727272727272728, | |
| "eval_f1_macro": 0.972396986523243, | |
| "eval_f1_micro": 0.9727272727272728, | |
| "eval_f1_weighted": 0.9726060520073766, | |
| "eval_loss": 0.08732765913009644, | |
| "eval_precision_macro": 0.9722370619929125, | |
| "eval_precision_micro": 0.9727272727272728, | |
| "eval_precision_weighted": 0.9728649069611348, | |
| "eval_recall_macro": 0.9729742025670353, | |
| "eval_recall_micro": 0.9727272727272728, | |
| "eval_recall_weighted": 0.9727272727272728, | |
| "eval_runtime": 1.0879, | |
| "eval_samples_per_second": 505.581, | |
| "eval_steps_per_second": 16.546, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.682795698924731e-05, | |
| "loss": 0.1259, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.508064516129033e-05, | |
| "loss": 0.1674, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1101, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.158602150537634e-05, | |
| "loss": 0.1049, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.9838709677419357e-05, | |
| "loss": 0.0704, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.8091397849462365e-05, | |
| "loss": 0.0785, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.6344086021505376e-05, | |
| "loss": 0.0617, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.4596774193548387e-05, | |
| "loss": 0.1625, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.28494623655914e-05, | |
| "loss": 0.0238, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.110215053763441e-05, | |
| "loss": 0.0708, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.935483870967742e-05, | |
| "loss": 0.0856, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9818181818181818, | |
| "eval_f1_macro": 0.9820711118316969, | |
| "eval_f1_micro": 0.9818181818181818, | |
| "eval_f1_weighted": 0.9817660734040653, | |
| "eval_loss": 0.0563533715903759, | |
| "eval_precision_macro": 0.9816901091597003, | |
| "eval_precision_micro": 0.9818181818181818, | |
| "eval_precision_weighted": 0.9818836715959347, | |
| "eval_recall_macro": 0.9826135468293302, | |
| "eval_recall_micro": 0.9818181818181818, | |
| "eval_recall_weighted": 0.9818181818181818, | |
| "eval_runtime": 1.0889, | |
| "eval_samples_per_second": 505.086, | |
| "eval_steps_per_second": 16.53, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.7607526881720432e-05, | |
| "loss": 0.0957, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.586021505376344e-05, | |
| "loss": 0.0648, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.4112903225806454e-05, | |
| "loss": 0.1037, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1.2365591397849464e-05, | |
| "loss": 0.0823, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.0618279569892473e-05, | |
| "loss": 0.1222, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 8.870967741935484e-06, | |
| "loss": 0.0418, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 7.1236559139784956e-06, | |
| "loss": 0.0513, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 5.376344086021506e-06, | |
| "loss": 0.0941, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 3.6290322580645166e-06, | |
| "loss": 0.023, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.881720430107527e-06, | |
| "loss": 0.0559, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9818181818181818, | |
| "eval_f1_macro": 0.9822762607526727, | |
| "eval_f1_micro": 0.9818181818181818, | |
| "eval_f1_weighted": 0.9817521242226859, | |
| "eval_loss": 0.05566494166851044, | |
| "eval_precision_macro": 0.9820894227603638, | |
| "eval_precision_micro": 0.9818181818181818, | |
| "eval_precision_weighted": 0.981850377976082, | |
| "eval_recall_macro": 0.9826135468293302, | |
| "eval_recall_micro": 0.9818181818181818, | |
| "eval_recall_weighted": 0.9818181818181818, | |
| "eval_runtime": 1.0922, | |
| "eval_samples_per_second": 503.548, | |
| "eval_steps_per_second": 16.48, | |
| "step": 414 | |
| } | |
| ], | |
| "logging_steps": 13, | |
| "max_steps": 414, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 5.133207467857019e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |