| { | |
| "best_metric": 0.8415765069551777, | |
| "best_model_checkpoint": "cvt-13-384-22k-fv-finetuned-memes/checkpoint-380", | |
| "epoch": 19.987654320987655, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6042, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3821, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.49690880989180836, | |
| "eval_f1": 0.4458280336574581, | |
| "eval_loss": 1.2779731750488281, | |
| "eval_precision": 0.5083478748744594, | |
| "eval_recall": 0.49690880989180836, | |
| "eval_runtime": 11.7189, | |
| "eval_samples_per_second": 110.42, | |
| "eval_steps_per_second": 1.792, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2697, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0785, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.6669242658423493, | |
| "eval_f1": 0.6499601011830624, | |
| "eval_loss": 0.8633419275283813, | |
| "eval_precision": 0.6658463652527506, | |
| "eval_recall": 0.6669242658423493, | |
| "eval_runtime": 10.8339, | |
| "eval_samples_per_second": 119.44, | |
| "eval_steps_per_second": 1.938, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 0.8908, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00011333333333333333, | |
| "loss": 0.8862, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7217928902627512, | |
| "eval_f1": 0.701310570825714, | |
| "eval_loss": 0.71103435754776, | |
| "eval_precision": 0.7257543548866834, | |
| "eval_recall": 0.7217928902627512, | |
| "eval_runtime": 11.1739, | |
| "eval_samples_per_second": 115.806, | |
| "eval_steps_per_second": 1.879, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.00011, | |
| "loss": 0.7375, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 0.665, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_accuracy": 0.8044822256568779, | |
| "eval_f1": 0.805017946518536, | |
| "eval_loss": 0.5514629483222961, | |
| "eval_precision": 0.8136720242079513, | |
| "eval_recall": 0.8044822256568779, | |
| "eval_runtime": 10.847, | |
| "eval_samples_per_second": 119.295, | |
| "eval_steps_per_second": 1.936, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.00010333333333333334, | |
| "loss": 0.6256, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6056, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.7959814528593508, | |
| "eval_f1": 0.7846029654723703, | |
| "eval_loss": 0.5956416726112366, | |
| "eval_precision": 0.8040705084962495, | |
| "eval_recall": 0.7959814528593508, | |
| "eval_runtime": 11.1412, | |
| "eval_samples_per_second": 116.145, | |
| "eval_steps_per_second": 1.885, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 9.666666666666667e-05, | |
| "loss": 0.5889, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.4779, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.7936630602782071, | |
| "eval_f1": 0.7856708002513016, | |
| "eval_loss": 0.622891366481781, | |
| "eval_precision": 0.7944777820954576, | |
| "eval_recall": 0.7936630602782071, | |
| "eval_runtime": 12.2285, | |
| "eval_samples_per_second": 105.818, | |
| "eval_steps_per_second": 1.717, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 9e-05, | |
| "loss": 0.4969, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.4554, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.8098918083462133, | |
| "eval_f1": 0.8086497823932973, | |
| "eval_loss": 0.5355423092842102, | |
| "eval_precision": 0.8125537937361387, | |
| "eval_recall": 0.8098918083462133, | |
| "eval_runtime": 11.0611, | |
| "eval_samples_per_second": 116.986, | |
| "eval_steps_per_second": 1.899, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.475, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 7.999999999999999e-05, | |
| "loss": 0.4249, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_accuracy": 0.8268933539412674, | |
| "eval_f1": 0.8235690575721668, | |
| "eval_loss": 0.5447452664375305, | |
| "eval_precision": 0.8274671013460899, | |
| "eval_recall": 0.8268933539412674, | |
| "eval_runtime": 10.9569, | |
| "eval_samples_per_second": 118.1, | |
| "eval_steps_per_second": 1.917, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 7.666666666666667e-05, | |
| "loss": 0.4241, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 7.333333333333334e-05, | |
| "loss": 0.4313, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.8153013910355487, | |
| "eval_f1": 0.8132391167433421, | |
| "eval_loss": 0.5530120730400085, | |
| "eval_precision": 0.8139736548329274, | |
| "eval_recall": 0.8153013910355487, | |
| "eval_runtime": 10.8836, | |
| "eval_samples_per_second": 118.895, | |
| "eval_steps_per_second": 1.93, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 0.4092, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.423, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.8238021638330757, | |
| "eval_f1": 0.8222730304295096, | |
| "eval_loss": 0.5345684885978699, | |
| "eval_precision": 0.8230247813283141, | |
| "eval_recall": 0.8238021638330757, | |
| "eval_runtime": 10.942, | |
| "eval_samples_per_second": 118.259, | |
| "eval_steps_per_second": 1.919, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 6.333333333333333e-05, | |
| "loss": 0.3905, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3997, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.8338485316846986, | |
| "eval_f1": 0.8338050888395913, | |
| "eval_loss": 0.5413401126861572, | |
| "eval_precision": 0.8346663333915662, | |
| "eval_recall": 0.8338485316846986, | |
| "eval_runtime": 11.954, | |
| "eval_samples_per_second": 108.248, | |
| "eval_steps_per_second": 1.757, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 5.6666666666666664e-05, | |
| "loss": 0.3824, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.4095, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "eval_accuracy": 0.8207109737248841, | |
| "eval_f1": 0.8177361362413736, | |
| "eval_loss": 0.5999411344528198, | |
| "eval_precision": 0.8230816829155458, | |
| "eval_recall": 0.8207109737248841, | |
| "eval_runtime": 11.094, | |
| "eval_samples_per_second": 116.639, | |
| "eval_steps_per_second": 1.893, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3849, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.3979, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.8284389489953632, | |
| "eval_f1": 0.825049010410594, | |
| "eval_loss": 0.563246488571167, | |
| "eval_precision": 0.8254511686180086, | |
| "eval_recall": 0.8284389489953632, | |
| "eval_runtime": 11.1372, | |
| "eval_samples_per_second": 116.187, | |
| "eval_steps_per_second": 1.886, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.3571, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 0.3408, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.8207109737248841, | |
| "eval_f1": 0.819562228886707, | |
| "eval_loss": 0.5724937319755554, | |
| "eval_precision": 0.8198440242244688, | |
| "eval_recall": 0.8207109737248841, | |
| "eval_runtime": 10.994, | |
| "eval_samples_per_second": 117.7, | |
| "eval_steps_per_second": 1.91, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 3.666666666666667e-05, | |
| "loss": 0.3746, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3828, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.8276661514683153, | |
| "eval_f1": 0.8260480131366693, | |
| "eval_loss": 0.5630865693092346, | |
| "eval_precision": 0.8257927020758725, | |
| "eval_recall": 0.8276661514683153, | |
| "eval_runtime": 10.9146, | |
| "eval_samples_per_second": 118.557, | |
| "eval_steps_per_second": 1.924, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3506, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.3595, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "eval_accuracy": 0.8307573415765069, | |
| "eval_f1": 0.8275327108334436, | |
| "eval_loss": 0.6005313396453857, | |
| "eval_precision": 0.8296678468094887, | |
| "eval_recall": 0.8307573415765069, | |
| "eval_runtime": 11.7668, | |
| "eval_samples_per_second": 109.971, | |
| "eval_steps_per_second": 1.785, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.3535, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 0.3789, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.8299845440494591, | |
| "eval_f1": 0.8273370558566532, | |
| "eval_loss": 0.5840371251106262, | |
| "eval_precision": 0.827140388432744, | |
| "eval_recall": 0.8299845440494591, | |
| "eval_runtime": 10.837, | |
| "eval_samples_per_second": 119.405, | |
| "eval_steps_per_second": 1.938, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.389, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.3545, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.8245749613601236, | |
| "eval_f1": 0.8222146970644545, | |
| "eval_loss": 0.5983098745346069, | |
| "eval_precision": 0.8226259868849637, | |
| "eval_recall": 0.8245749613601236, | |
| "eval_runtime": 10.7069, | |
| "eval_samples_per_second": 120.857, | |
| "eval_steps_per_second": 1.961, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 0.3661, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.3472, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.8415765069551777, | |
| "eval_f1": 0.8390361807029366, | |
| "eval_loss": 0.5795101523399353, | |
| "eval_precision": 0.838184001941316, | |
| "eval_recall": 0.8415765069551777, | |
| "eval_runtime": 10.7681, | |
| "eval_samples_per_second": 120.17, | |
| "eval_steps_per_second": 1.95, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.3654, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.355, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "eval_accuracy": 0.8315301391035549, | |
| "eval_f1": 0.8292026505769348, | |
| "eval_loss": 0.5761234760284424, | |
| "eval_precision": 0.8302128280229624, | |
| "eval_recall": 0.8315301391035549, | |
| "eval_runtime": 11.7079, | |
| "eval_samples_per_second": 110.524, | |
| "eval_steps_per_second": 1.794, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "step": 400, | |
| "total_flos": 1.832202236159447e+18, | |
| "train_loss": 0.5447965228557586, | |
| "train_runtime": 1430.1645, | |
| "train_samples_per_second": 72.355, | |
| "train_steps_per_second": 0.28 | |
| } | |
| ], | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.832202236159447e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |