| { | |
| "best_metric": 0.7614678899082569, | |
| "best_model_checkpoint": "swiftformer-xs-dmae-va-U-SF/checkpoint-271", | |
| "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": 0.3119266055045872, | |
| "eval_loss": 1.3887187242507935, | |
| "eval_runtime": 0.6715, | |
| "eval_samples_per_second": 162.322, | |
| "eval_steps_per_second": 5.957, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.4383, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.41284403669724773, | |
| "eval_loss": 1.344041347503662, | |
| "eval_runtime": 0.4343, | |
| "eval_samples_per_second": 250.972, | |
| "eval_steps_per_second": 9.21, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.3956, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.3761467889908257, | |
| "eval_loss": 1.3158775568008423, | |
| "eval_runtime": 0.5875, | |
| "eval_samples_per_second": 185.531, | |
| "eval_steps_per_second": 6.808, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.36, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3853211009174312, | |
| "eval_loss": 1.2906575202941895, | |
| "eval_runtime": 0.416, | |
| "eval_samples_per_second": 262.05, | |
| "eval_steps_per_second": 9.617, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.44036697247706424, | |
| "eval_loss": 1.2488025426864624, | |
| "eval_runtime": 0.4344, | |
| "eval_samples_per_second": 250.911, | |
| "eval_steps_per_second": 9.208, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.2912, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.4036697247706422, | |
| "eval_loss": 1.2129390239715576, | |
| "eval_runtime": 0.4322, | |
| "eval_samples_per_second": 252.176, | |
| "eval_steps_per_second": 9.254, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 1.2387, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.46788990825688076, | |
| "eval_loss": 1.1733514070510864, | |
| "eval_runtime": 0.4452, | |
| "eval_samples_per_second": 244.821, | |
| "eval_steps_per_second": 8.984, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 1.1607, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5137614678899083, | |
| "eval_loss": 1.143617868423462, | |
| "eval_runtime": 0.4224, | |
| "eval_samples_per_second": 258.05, | |
| "eval_steps_per_second": 9.47, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.4954128440366973, | |
| "eval_loss": 1.0990597009658813, | |
| "eval_runtime": 0.459, | |
| "eval_samples_per_second": 237.483, | |
| "eval_steps_per_second": 8.715, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.1224, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.5504587155963303, | |
| "eval_loss": 1.0478596687316895, | |
| "eval_runtime": 0.7035, | |
| "eval_samples_per_second": 154.95, | |
| "eval_steps_per_second": 5.686, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 1.0547, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.5963302752293578, | |
| "eval_loss": 0.9993337392807007, | |
| "eval_runtime": 0.4348, | |
| "eval_samples_per_second": 250.685, | |
| "eval_steps_per_second": 9.199, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 1.0137, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6146788990825688, | |
| "eval_loss": 0.9859614372253418, | |
| "eval_runtime": 0.5726, | |
| "eval_samples_per_second": 190.361, | |
| "eval_steps_per_second": 6.986, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.9652, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.6146788990825688, | |
| "eval_loss": 0.9697992205619812, | |
| "eval_runtime": 0.4313, | |
| "eval_samples_per_second": 252.708, | |
| "eval_steps_per_second": 9.274, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.6055045871559633, | |
| "eval_loss": 0.9519100785255432, | |
| "eval_runtime": 0.4407, | |
| "eval_samples_per_second": 247.344, | |
| "eval_steps_per_second": 9.077, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.9217, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.6055045871559633, | |
| "eval_loss": 0.9242026209831238, | |
| "eval_runtime": 0.4297, | |
| "eval_samples_per_second": 253.648, | |
| "eval_steps_per_second": 9.308, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.9122, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6146788990825688, | |
| "eval_loss": 0.9062366485595703, | |
| "eval_runtime": 0.4329, | |
| "eval_samples_per_second": 251.781, | |
| "eval_steps_per_second": 9.24, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.8763, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.6422018348623854, | |
| "eval_loss": 0.8872672915458679, | |
| "eval_runtime": 0.4224, | |
| "eval_samples_per_second": 258.055, | |
| "eval_steps_per_second": 9.47, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.6513761467889908, | |
| "eval_loss": 0.8477428555488586, | |
| "eval_runtime": 0.5523, | |
| "eval_samples_per_second": 197.365, | |
| "eval_steps_per_second": 7.243, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8471, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.6513761467889908, | |
| "eval_loss": 0.8427405953407288, | |
| "eval_runtime": 0.6098, | |
| "eval_samples_per_second": 178.737, | |
| "eval_steps_per_second": 6.559, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.8331, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6880733944954128, | |
| "eval_loss": 0.8257479667663574, | |
| "eval_runtime": 0.438, | |
| "eval_samples_per_second": 248.852, | |
| "eval_steps_per_second": 9.132, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.8167, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.6880733944954128, | |
| "eval_loss": 0.8025383353233337, | |
| "eval_runtime": 0.4399, | |
| "eval_samples_per_second": 247.761, | |
| "eval_steps_per_second": 9.092, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.8022, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.8010965585708618, | |
| "eval_runtime": 0.4311, | |
| "eval_samples_per_second": 252.864, | |
| "eval_steps_per_second": 9.279, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.8078291416168213, | |
| "eval_runtime": 0.4412, | |
| "eval_samples_per_second": 247.041, | |
| "eval_steps_per_second": 9.066, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.7996, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7064220183486238, | |
| "eval_loss": 0.7920359969139099, | |
| "eval_runtime": 0.4625, | |
| "eval_samples_per_second": 235.682, | |
| "eval_steps_per_second": 8.649, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.7962, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.7247706422018348, | |
| "eval_loss": 0.760365903377533, | |
| "eval_runtime": 0.4325, | |
| "eval_samples_per_second": 251.996, | |
| "eval_steps_per_second": 9.248, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.7268, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.7596898674964905, | |
| "eval_runtime": 0.568, | |
| "eval_samples_per_second": 191.903, | |
| "eval_steps_per_second": 7.042, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.7461942434310913, | |
| "eval_runtime": 0.4265, | |
| "eval_samples_per_second": 255.55, | |
| "eval_steps_per_second": 9.378, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.7477, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7064220183486238, | |
| "eval_loss": 0.7316110730171204, | |
| "eval_runtime": 0.6404, | |
| "eval_samples_per_second": 170.198, | |
| "eval_steps_per_second": 6.246, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.7411, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.7275413870811462, | |
| "eval_runtime": 0.4398, | |
| "eval_samples_per_second": 247.819, | |
| "eval_steps_per_second": 9.094, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.7415, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.7247706422018348, | |
| "eval_loss": 0.7210223078727722, | |
| "eval_runtime": 0.4316, | |
| "eval_samples_per_second": 252.526, | |
| "eval_steps_per_second": 9.267, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.7159, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.7247706422018348, | |
| "eval_loss": 0.727079451084137, | |
| "eval_runtime": 0.4278, | |
| "eval_samples_per_second": 254.812, | |
| "eval_steps_per_second": 9.351, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.700477659702301, | |
| "eval_runtime": 0.4579, | |
| "eval_samples_per_second": 238.037, | |
| "eval_steps_per_second": 8.735, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.7322, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.7012345194816589, | |
| "eval_runtime": 0.4583, | |
| "eval_samples_per_second": 237.855, | |
| "eval_steps_per_second": 8.729, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.7124, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.7052269577980042, | |
| "eval_runtime": 0.4302, | |
| "eval_samples_per_second": 253.362, | |
| "eval_steps_per_second": 9.298, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.7194, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.6964432597160339, | |
| "eval_runtime": 0.6076, | |
| "eval_samples_per_second": 179.38, | |
| "eval_steps_per_second": 6.583, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.7007263898849487, | |
| "eval_runtime": 0.4321, | |
| "eval_samples_per_second": 252.256, | |
| "eval_steps_per_second": 9.257, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 0.0, | |
| "loss": 0.6903, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.6986872553825378, | |
| "eval_runtime": 0.4293, | |
| "eval_samples_per_second": 253.895, | |
| "eval_steps_per_second": 9.317, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "step": 280, | |
| "total_flos": 9.686412043576934e+16, | |
| "train_loss": 0.9347471083913531, | |
| "train_runtime": 250.0799, | |
| "train_samples_per_second": 156.43, | |
| "train_steps_per_second": 1.12 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 280, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 9.686412043576934e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |