| { | |
| "best_metric": 0.9534883720930233, | |
| "best_model_checkpoint": "beit-base-patch16-224-75-fold4/checkpoint-144", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.5130478143692017, | |
| "eval_runtime": 0.6812, | |
| "eval_samples_per_second": 63.124, | |
| "eval_steps_per_second": 2.936, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.4860553443431854, | |
| "eval_runtime": 0.6922, | |
| "eval_samples_per_second": 62.12, | |
| "eval_steps_per_second": 2.889, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.4775117039680481, | |
| "eval_runtime": 0.677, | |
| "eval_samples_per_second": 63.512, | |
| "eval_steps_per_second": 2.954, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.44188791513442993, | |
| "eval_runtime": 0.6577, | |
| "eval_samples_per_second": 65.382, | |
| "eval_steps_per_second": 3.041, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 7.749820709228516, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4909, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.36715349555015564, | |
| "eval_runtime": 0.6592, | |
| "eval_samples_per_second": 65.227, | |
| "eval_steps_per_second": 3.034, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3301140367984772, | |
| "eval_runtime": 0.6452, | |
| "eval_samples_per_second": 66.649, | |
| "eval_steps_per_second": 3.1, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3130995035171509, | |
| "eval_runtime": 0.6545, | |
| "eval_samples_per_second": 65.699, | |
| "eval_steps_per_second": 3.056, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4535352885723114, | |
| "eval_runtime": 0.669, | |
| "eval_samples_per_second": 64.28, | |
| "eval_steps_per_second": 2.99, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.30879294872283936, | |
| "eval_runtime": 0.6641, | |
| "eval_samples_per_second": 64.754, | |
| "eval_steps_per_second": 3.012, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 7.826290130615234, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3473, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4452756345272064, | |
| "eval_runtime": 0.6608, | |
| "eval_samples_per_second": 65.074, | |
| "eval_steps_per_second": 3.027, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.42338627576828003, | |
| "eval_runtime": 0.6819, | |
| "eval_samples_per_second": 63.056, | |
| "eval_steps_per_second": 2.933, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3600829541683197, | |
| "eval_runtime": 0.6913, | |
| "eval_samples_per_second": 62.204, | |
| "eval_steps_per_second": 2.893, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.36579060554504395, | |
| "eval_runtime": 0.6967, | |
| "eval_samples_per_second": 61.723, | |
| "eval_steps_per_second": 2.871, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.30810949206352234, | |
| "eval_runtime": 0.6695, | |
| "eval_samples_per_second": 64.231, | |
| "eval_steps_per_second": 2.988, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.6172356605529785, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.2903, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.41277334094047546, | |
| "eval_runtime": 0.6574, | |
| "eval_samples_per_second": 65.405, | |
| "eval_steps_per_second": 3.042, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.2554984986782074, | |
| "eval_runtime": 0.6511, | |
| "eval_samples_per_second": 66.044, | |
| "eval_steps_per_second": 3.072, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3340766429901123, | |
| "eval_runtime": 0.6603, | |
| "eval_samples_per_second": 65.12, | |
| "eval_steps_per_second": 3.029, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24271325767040253, | |
| "eval_runtime": 0.6653, | |
| "eval_samples_per_second": 64.632, | |
| "eval_steps_per_second": 3.006, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4324667751789093, | |
| "eval_runtime": 0.6702, | |
| "eval_samples_per_second": 64.158, | |
| "eval_steps_per_second": 2.984, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 5.741236209869385, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2673, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2636508643627167, | |
| "eval_runtime": 0.6682, | |
| "eval_samples_per_second": 64.354, | |
| "eval_steps_per_second": 2.993, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.29191523790359497, | |
| "eval_runtime": 0.6857, | |
| "eval_samples_per_second": 62.706, | |
| "eval_steps_per_second": 2.917, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.31388339400291443, | |
| "eval_runtime": 0.6588, | |
| "eval_samples_per_second": 65.266, | |
| "eval_steps_per_second": 3.036, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24111181497573853, | |
| "eval_runtime": 0.6531, | |
| "eval_samples_per_second": 65.842, | |
| "eval_steps_per_second": 3.062, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.464458167552948, | |
| "eval_runtime": 0.6616, | |
| "eval_samples_per_second": 64.999, | |
| "eval_steps_per_second": 3.023, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 14.516473770141602, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2103, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.508436918258667, | |
| "eval_runtime": 0.6628, | |
| "eval_samples_per_second": 64.877, | |
| "eval_steps_per_second": 3.018, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.23078864812850952, | |
| "eval_runtime": 0.6774, | |
| "eval_samples_per_second": 63.482, | |
| "eval_steps_per_second": 2.953, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3450285494327545, | |
| "eval_runtime": 0.6606, | |
| "eval_samples_per_second": 65.094, | |
| "eval_steps_per_second": 3.028, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.34442028403282166, | |
| "eval_runtime": 0.6628, | |
| "eval_samples_per_second": 64.873, | |
| "eval_steps_per_second": 3.017, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2546083629131317, | |
| "eval_runtime": 0.6779, | |
| "eval_samples_per_second": 63.434, | |
| "eval_steps_per_second": 2.95, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 5.658456802368164, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1673, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.9117240905761719, | |
| "eval_runtime": 0.6726, | |
| "eval_samples_per_second": 63.93, | |
| "eval_steps_per_second": 2.973, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.8437220454216003, | |
| "eval_runtime": 0.6672, | |
| "eval_samples_per_second": 64.452, | |
| "eval_steps_per_second": 2.998, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.6758245825767517, | |
| "eval_runtime": 0.6742, | |
| "eval_samples_per_second": 63.775, | |
| "eval_steps_per_second": 2.966, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.8019335269927979, | |
| "eval_runtime": 0.6637, | |
| "eval_samples_per_second": 64.793, | |
| "eval_steps_per_second": 3.014, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3364037275314331, | |
| "eval_runtime": 0.6581, | |
| "eval_samples_per_second": 65.339, | |
| "eval_steps_per_second": 3.039, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 3.4963107109069824, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1677, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2928450107574463, | |
| "eval_runtime": 0.6586, | |
| "eval_samples_per_second": 65.287, | |
| "eval_steps_per_second": 3.037, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2546534240245819, | |
| "eval_runtime": 0.6709, | |
| "eval_samples_per_second": 64.096, | |
| "eval_steps_per_second": 2.981, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2968655228614807, | |
| "eval_runtime": 0.6643, | |
| "eval_samples_per_second": 64.731, | |
| "eval_steps_per_second": 3.011, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5706415772438049, | |
| "eval_runtime": 0.6641, | |
| "eval_samples_per_second": 64.754, | |
| "eval_steps_per_second": 3.012, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.700643002986908, | |
| "eval_runtime": 0.6643, | |
| "eval_samples_per_second": 64.734, | |
| "eval_steps_per_second": 3.011, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 7.142909049987793, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1407, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.43211764097213745, | |
| "eval_runtime": 0.6641, | |
| "eval_samples_per_second": 64.75, | |
| "eval_steps_per_second": 3.012, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.43662115931510925, | |
| "eval_runtime": 0.6788, | |
| "eval_samples_per_second": 63.344, | |
| "eval_steps_per_second": 2.946, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.39564797282218933, | |
| "eval_runtime": 0.6577, | |
| "eval_samples_per_second": 65.376, | |
| "eval_steps_per_second": 3.041, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.22897183895111084, | |
| "eval_runtime": 0.6605, | |
| "eval_samples_per_second": 65.097, | |
| "eval_steps_per_second": 3.028, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.36646950244903564, | |
| "eval_runtime": 0.6649, | |
| "eval_samples_per_second": 64.671, | |
| "eval_steps_per_second": 3.008, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 5.7567524909973145, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1474, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4464525580406189, | |
| "eval_runtime": 0.6626, | |
| "eval_samples_per_second": 64.895, | |
| "eval_steps_per_second": 3.018, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.7278981804847717, | |
| "eval_runtime": 0.666, | |
| "eval_samples_per_second": 64.563, | |
| "eval_steps_per_second": 3.003, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.5259287357330322, | |
| "eval_runtime": 0.6797, | |
| "eval_samples_per_second": 63.263, | |
| "eval_steps_per_second": 2.942, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5832454562187195, | |
| "eval_runtime": 0.6814, | |
| "eval_samples_per_second": 63.101, | |
| "eval_steps_per_second": 2.935, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.73279869556427, | |
| "eval_runtime": 0.6725, | |
| "eval_samples_per_second": 63.937, | |
| "eval_steps_per_second": 2.974, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 6.381649971008301, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1344, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3890153169631958, | |
| "eval_runtime": 0.666, | |
| "eval_samples_per_second": 64.569, | |
| "eval_steps_per_second": 3.003, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.26422181725502014, | |
| "eval_runtime": 0.6656, | |
| "eval_samples_per_second": 64.601, | |
| "eval_steps_per_second": 3.005, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3710338771343231, | |
| "eval_runtime": 0.6608, | |
| "eval_samples_per_second": 65.077, | |
| "eval_steps_per_second": 3.027, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.47727954387664795, | |
| "eval_runtime": 0.6628, | |
| "eval_samples_per_second": 64.873, | |
| "eval_steps_per_second": 3.017, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3628208339214325, | |
| "eval_runtime": 0.662, | |
| "eval_samples_per_second": 64.957, | |
| "eval_steps_per_second": 3.021, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 3.1835520267486572, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1166, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.43894824385643005, | |
| "eval_runtime": 0.6643, | |
| "eval_samples_per_second": 64.727, | |
| "eval_steps_per_second": 3.011, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.48127347230911255, | |
| "eval_runtime": 0.6668, | |
| "eval_samples_per_second": 64.488, | |
| "eval_steps_per_second": 2.999, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.5327816009521484, | |
| "eval_runtime": 0.6663, | |
| "eval_samples_per_second": 64.537, | |
| "eval_steps_per_second": 3.002, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.5341907739639282, | |
| "eval_runtime": 0.6645, | |
| "eval_samples_per_second": 64.71, | |
| "eval_steps_per_second": 3.01, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.48921194672584534, | |
| "eval_runtime": 0.6577, | |
| "eval_samples_per_second": 65.38, | |
| "eval_steps_per_second": 3.041, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.6190884113311768, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.097, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.5857459902763367, | |
| "eval_runtime": 0.6713, | |
| "eval_samples_per_second": 64.059, | |
| "eval_steps_per_second": 2.979, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.6681368350982666, | |
| "eval_runtime": 0.6655, | |
| "eval_samples_per_second": 64.61, | |
| "eval_steps_per_second": 3.005, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.5946548581123352, | |
| "eval_runtime": 0.6669, | |
| "eval_samples_per_second": 64.482, | |
| "eval_steps_per_second": 2.999, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.47492873668670654, | |
| "eval_runtime": 0.6758, | |
| "eval_samples_per_second": 63.633, | |
| "eval_steps_per_second": 2.96, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.6091283559799194, | |
| "eval_runtime": 0.6685, | |
| "eval_samples_per_second": 64.325, | |
| "eval_steps_per_second": 2.992, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 4.258938789367676, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1076, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.9725148677825928, | |
| "eval_runtime": 0.6743, | |
| "eval_samples_per_second": 63.766, | |
| "eval_steps_per_second": 2.966, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 1.1372181177139282, | |
| "eval_runtime": 0.6651, | |
| "eval_samples_per_second": 64.656, | |
| "eval_steps_per_second": 3.007, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.7108578085899353, | |
| "eval_runtime": 0.6618, | |
| "eval_samples_per_second": 64.972, | |
| "eval_steps_per_second": 3.022, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.35486963391304016, | |
| "eval_runtime": 0.66, | |
| "eval_samples_per_second": 65.147, | |
| "eval_steps_per_second": 3.03, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.27090850472450256, | |
| "eval_runtime": 0.6636, | |
| "eval_samples_per_second": 64.795, | |
| "eval_steps_per_second": 3.014, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 2.5122122764587402, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0914, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3316062092781067, | |
| "eval_runtime": 0.6617, | |
| "eval_samples_per_second": 64.984, | |
| "eval_steps_per_second": 3.023, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.317620187997818, | |
| "eval_runtime": 0.6589, | |
| "eval_samples_per_second": 65.263, | |
| "eval_steps_per_second": 3.035, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.25086623430252075, | |
| "eval_runtime": 0.661, | |
| "eval_samples_per_second": 65.051, | |
| "eval_steps_per_second": 3.026, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.22562167048454285, | |
| "eval_runtime": 0.6643, | |
| "eval_samples_per_second": 64.732, | |
| "eval_steps_per_second": 3.011, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2569940388202667, | |
| "eval_runtime": 0.6668, | |
| "eval_samples_per_second": 64.488, | |
| "eval_steps_per_second": 2.999, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 4.068800449371338, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0815, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.3081124424934387, | |
| "eval_runtime": 0.6629, | |
| "eval_samples_per_second": 64.862, | |
| "eval_steps_per_second": 3.017, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.4198541045188904, | |
| "eval_runtime": 0.6549, | |
| "eval_samples_per_second": 65.657, | |
| "eval_steps_per_second": 3.054, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.432374507188797, | |
| "eval_runtime": 0.6731, | |
| "eval_samples_per_second": 63.882, | |
| "eval_steps_per_second": 2.971, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.39275258779525757, | |
| "eval_runtime": 0.6597, | |
| "eval_samples_per_second": 65.176, | |
| "eval_steps_per_second": 3.031, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3700079023838043, | |
| "eval_runtime": 0.6605, | |
| "eval_samples_per_second": 65.106, | |
| "eval_steps_per_second": 3.028, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 3.1357898712158203, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0878, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.38116565346717834, | |
| "eval_runtime": 0.7043, | |
| "eval_samples_per_second": 61.055, | |
| "eval_steps_per_second": 2.84, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.4299600422382355, | |
| "eval_runtime": 0.6661, | |
| "eval_samples_per_second": 64.555, | |
| "eval_steps_per_second": 3.003, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.4289122223854065, | |
| "eval_runtime": 0.7043, | |
| "eval_samples_per_second": 61.056, | |
| "eval_steps_per_second": 2.84, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.41245317459106445, | |
| "eval_runtime": 0.662, | |
| "eval_samples_per_second": 64.956, | |
| "eval_steps_per_second": 3.021, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.43505820631980896, | |
| "eval_runtime": 0.6683, | |
| "eval_samples_per_second": 64.34, | |
| "eval_steps_per_second": 2.993, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 3.3373324871063232, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0725, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.5045635104179382, | |
| "eval_runtime": 0.6522, | |
| "eval_samples_per_second": 65.933, | |
| "eval_steps_per_second": 3.067, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.5692218542098999, | |
| "eval_runtime": 0.6625, | |
| "eval_samples_per_second": 64.905, | |
| "eval_steps_per_second": 3.019, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.5486466884613037, | |
| "eval_runtime": 0.6685, | |
| "eval_samples_per_second": 64.327, | |
| "eval_steps_per_second": 2.992, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.5309974551200867, | |
| "eval_runtime": 0.6589, | |
| "eval_samples_per_second": 65.259, | |
| "eval_steps_per_second": 3.035, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.4661545753479004, | |
| "eval_runtime": 0.6663, | |
| "eval_samples_per_second": 64.535, | |
| "eval_steps_per_second": 3.002, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 3.57021427154541, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0944, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.40696418285369873, | |
| "eval_runtime": 0.6586, | |
| "eval_samples_per_second": 65.291, | |
| "eval_steps_per_second": 3.037, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3767591118812561, | |
| "eval_runtime": 0.6608, | |
| "eval_samples_per_second": 65.074, | |
| "eval_steps_per_second": 3.027, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3884381651878357, | |
| "eval_runtime": 0.6611, | |
| "eval_samples_per_second": 65.043, | |
| "eval_steps_per_second": 3.025, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3850724995136261, | |
| "eval_runtime": 0.6655, | |
| "eval_samples_per_second": 64.611, | |
| "eval_steps_per_second": 3.005, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.37587684392929077, | |
| "eval_runtime": 0.6694, | |
| "eval_samples_per_second": 64.238, | |
| "eval_steps_per_second": 2.988, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 2.757040023803711, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0739, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3608282804489136, | |
| "eval_runtime": 0.6628, | |
| "eval_samples_per_second": 64.873, | |
| "eval_steps_per_second": 3.017, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.34555718302726746, | |
| "eval_runtime": 0.6719, | |
| "eval_samples_per_second": 63.999, | |
| "eval_steps_per_second": 2.977, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3360309302806854, | |
| "eval_runtime": 0.6604, | |
| "eval_samples_per_second": 65.116, | |
| "eval_steps_per_second": 3.029, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.33121442794799805, | |
| "eval_runtime": 0.669, | |
| "eval_samples_per_second": 64.28, | |
| "eval_steps_per_second": 2.99, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3320866823196411, | |
| "eval_runtime": 0.6618, | |
| "eval_samples_per_second": 64.972, | |
| "eval_steps_per_second": 3.022, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.8462013006210327, | |
| "learning_rate": 0.0, | |
| "loss": 0.0612, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.33314311504364014, | |
| "eval_runtime": 0.6673, | |
| "eval_samples_per_second": 64.436, | |
| "eval_steps_per_second": 2.997, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.8667587341684736e+18, | |
| "train_loss": 0.16236644983291626, | |
| "train_runtime": 1253.5555, | |
| "train_samples_per_second": 19.225, | |
| "train_steps_per_second": 0.16 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.25086623430252075, | |
| "eval_runtime": 0.7108, | |
| "eval_samples_per_second": 60.499, | |
| "eval_steps_per_second": 2.814, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 1.8667587341684736e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |