| { | |
| "best_metric": 0.7522935779816514, | |
| "best_model_checkpoint": "swiftformer-xs-dmae-va-U-40/checkpoint-224", | |
| "epoch": 36.12903225806452, | |
| "eval_steps": 500, | |
| "global_step": 280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.29357798165137616, | |
| "eval_loss": 1.3578405380249023, | |
| "eval_runtime": 0.4941, | |
| "eval_samples_per_second": 220.605, | |
| "eval_steps_per_second": 8.096, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.3702, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.29357798165137616, | |
| "eval_loss": 1.3703261613845825, | |
| "eval_runtime": 0.5833, | |
| "eval_samples_per_second": 186.883, | |
| "eval_steps_per_second": 6.858, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.3497, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.3394495412844037, | |
| "eval_loss": 1.3361120223999023, | |
| "eval_runtime": 0.4226, | |
| "eval_samples_per_second": 257.899, | |
| "eval_steps_per_second": 9.464, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.3004, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3669724770642202, | |
| "eval_loss": 1.2852015495300293, | |
| "eval_runtime": 0.4154, | |
| "eval_samples_per_second": 262.426, | |
| "eval_steps_per_second": 9.63, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.43119266055045874, | |
| "eval_loss": 1.231681227684021, | |
| "eval_runtime": 0.4157, | |
| "eval_samples_per_second": 262.204, | |
| "eval_steps_per_second": 9.622, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.2248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.45871559633027525, | |
| "eval_loss": 1.1785550117492676, | |
| "eval_runtime": 0.4211, | |
| "eval_samples_per_second": 258.867, | |
| "eval_steps_per_second": 9.5, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 1.1485, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.5045871559633027, | |
| "eval_loss": 1.123984694480896, | |
| "eval_runtime": 0.4143, | |
| "eval_samples_per_second": 263.124, | |
| "eval_steps_per_second": 9.656, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 1.0759, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5504587155963303, | |
| "eval_loss": 1.0727368593215942, | |
| "eval_runtime": 0.4109, | |
| "eval_samples_per_second": 265.292, | |
| "eval_steps_per_second": 9.735, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.5596330275229358, | |
| "eval_loss": 1.0403714179992676, | |
| "eval_runtime": 0.5431, | |
| "eval_samples_per_second": 200.69, | |
| "eval_steps_per_second": 7.365, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.0244, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.6238532110091743, | |
| "eval_loss": 0.974201500415802, | |
| "eval_runtime": 0.4104, | |
| "eval_samples_per_second": 265.623, | |
| "eval_steps_per_second": 9.748, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.9782, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.6422018348623854, | |
| "eval_loss": 0.9374117851257324, | |
| "eval_runtime": 0.6713, | |
| "eval_samples_per_second": 162.361, | |
| "eval_steps_per_second": 5.958, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.9359, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6788990825688074, | |
| "eval_loss": 0.9196614027023315, | |
| "eval_runtime": 0.4175, | |
| "eval_samples_per_second": 261.061, | |
| "eval_steps_per_second": 9.58, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.9051, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.6880733944954128, | |
| "eval_loss": 0.8753331899642944, | |
| "eval_runtime": 0.41, | |
| "eval_samples_per_second": 265.867, | |
| "eval_steps_per_second": 9.757, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.8679403066635132, | |
| "eval_runtime": 0.4077, | |
| "eval_samples_per_second": 267.322, | |
| "eval_steps_per_second": 9.81, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.8652, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.7155963302752294, | |
| "eval_loss": 0.8316473364830017, | |
| "eval_runtime": 0.4256, | |
| "eval_samples_per_second": 256.117, | |
| "eval_steps_per_second": 9.399, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.8336, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.8222222924232483, | |
| "eval_runtime": 0.4301, | |
| "eval_samples_per_second": 253.434, | |
| "eval_steps_per_second": 9.3, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.8177, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.8177938461303711, | |
| "eval_runtime": 0.4098, | |
| "eval_samples_per_second": 266.011, | |
| "eval_steps_per_second": 9.762, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.7817714810371399, | |
| "eval_runtime": 0.5687, | |
| "eval_samples_per_second": 191.682, | |
| "eval_steps_per_second": 7.034, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8077, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.7627159357070923, | |
| "eval_runtime": 0.4134, | |
| "eval_samples_per_second": 263.676, | |
| "eval_steps_per_second": 9.676, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.7796, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.7478492259979248, | |
| "eval_runtime": 0.4144, | |
| "eval_samples_per_second": 263.034, | |
| "eval_steps_per_second": 9.653, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.7673, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.7414626479148865, | |
| "eval_runtime": 0.408, | |
| "eval_samples_per_second": 267.182, | |
| "eval_steps_per_second": 9.805, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.7445, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.7155963302752294, | |
| "eval_loss": 0.7413556575775146, | |
| "eval_runtime": 0.4043, | |
| "eval_samples_per_second": 269.576, | |
| "eval_steps_per_second": 9.893, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.7155963302752294, | |
| "eval_loss": 0.7375438809394836, | |
| "eval_runtime": 0.4062, | |
| "eval_samples_per_second": 268.371, | |
| "eval_steps_per_second": 9.848, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.7413, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7155963302752294, | |
| "eval_loss": 0.7353999018669128, | |
| "eval_runtime": 0.4171, | |
| "eval_samples_per_second": 261.311, | |
| "eval_steps_per_second": 9.589, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.739, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.71100252866745, | |
| "eval_runtime": 0.4123, | |
| "eval_samples_per_second": 264.367, | |
| "eval_steps_per_second": 9.702, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.6992, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.7120506763458252, | |
| "eval_runtime": 0.4281, | |
| "eval_samples_per_second": 254.6, | |
| "eval_steps_per_second": 9.343, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.7044178247451782, | |
| "eval_runtime": 0.5666, | |
| "eval_samples_per_second": 192.368, | |
| "eval_steps_per_second": 7.059, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.7111, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.6947001218795776, | |
| "eval_runtime": 0.4105, | |
| "eval_samples_per_second": 265.532, | |
| "eval_steps_per_second": 9.744, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.7013, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.7007379531860352, | |
| "eval_runtime": 0.4106, | |
| "eval_samples_per_second": 265.441, | |
| "eval_steps_per_second": 9.741, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.712, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.6792589426040649, | |
| "eval_runtime": 0.4139, | |
| "eval_samples_per_second": 263.354, | |
| "eval_steps_per_second": 9.664, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.671, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.6808269619941711, | |
| "eval_runtime": 0.408, | |
| "eval_samples_per_second": 267.189, | |
| "eval_steps_per_second": 9.805, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.6820599436759949, | |
| "eval_runtime": 0.4244, | |
| "eval_samples_per_second": 256.807, | |
| "eval_steps_per_second": 9.424, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.6862, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.6705361008644104, | |
| "eval_runtime": 0.4159, | |
| "eval_samples_per_second": 262.074, | |
| "eval_steps_per_second": 9.617, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.6606, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.6783888339996338, | |
| "eval_runtime": 0.4197, | |
| "eval_samples_per_second": 259.739, | |
| "eval_steps_per_second": 9.532, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.6667, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.6764441132545471, | |
| "eval_runtime": 0.4166, | |
| "eval_samples_per_second": 261.642, | |
| "eval_steps_per_second": 9.602, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.6716886758804321, | |
| "eval_runtime": 0.5848, | |
| "eval_samples_per_second": 186.395, | |
| "eval_steps_per_second": 6.84, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 0.0, | |
| "loss": 0.6687, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.6736045479774475, | |
| "eval_runtime": 0.4181, | |
| "eval_samples_per_second": 260.729, | |
| "eval_steps_per_second": 9.568, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "step": 280, | |
| "total_flos": 9.686412043576934e+16, | |
| "train_loss": 0.878055340903146, | |
| "train_runtime": 251.6613, | |
| "train_samples_per_second": 155.447, | |
| "train_steps_per_second": 1.113 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 280, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 9.686412043576934e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |