| { | |
| "best_metric": 0.8165137614678899, | |
| "best_model_checkpoint": "swiftformer-xs-dmae-va-U-80/checkpoint-255", | |
| "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": 0.3486238532110092, | |
| "eval_loss": 1.3780734539031982, | |
| "eval_runtime": 0.4072, | |
| "eval_samples_per_second": 267.678, | |
| "eval_steps_per_second": 9.823, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.3901, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.3302752293577982, | |
| "eval_loss": 1.353822112083435, | |
| "eval_runtime": 0.5603, | |
| "eval_samples_per_second": 194.555, | |
| "eval_steps_per_second": 7.14, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.3561, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.3761467889908257, | |
| "eval_loss": 1.3090553283691406, | |
| "eval_runtime": 0.4075, | |
| "eval_samples_per_second": 267.483, | |
| "eval_steps_per_second": 9.816, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.2933, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.42201834862385323, | |
| "eval_loss": 1.2622544765472412, | |
| "eval_runtime": 0.5245, | |
| "eval_samples_per_second": 207.799, | |
| "eval_steps_per_second": 7.626, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.5137614678899083, | |
| "eval_loss": 1.206465721130371, | |
| "eval_runtime": 0.4369, | |
| "eval_samples_per_second": 249.483, | |
| "eval_steps_per_second": 9.155, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.2274, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.5412844036697247, | |
| "eval_loss": 1.1352602243423462, | |
| "eval_runtime": 0.4157, | |
| "eval_samples_per_second": 262.219, | |
| "eval_steps_per_second": 9.623, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 1.1686, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.6146788990825688, | |
| "eval_loss": 1.0853842496871948, | |
| "eval_runtime": 0.4114, | |
| "eval_samples_per_second": 264.933, | |
| "eval_steps_per_second": 9.722, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 1.097, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6330275229357798, | |
| "eval_loss": 1.0488643646240234, | |
| "eval_runtime": 0.4198, | |
| "eval_samples_per_second": 259.629, | |
| "eval_steps_per_second": 9.528, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.6605504587155964, | |
| "eval_loss": 1.0184704065322876, | |
| "eval_runtime": 0.4158, | |
| "eval_samples_per_second": 262.135, | |
| "eval_steps_per_second": 9.62, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.0349, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.6422018348623854, | |
| "eval_loss": 0.968228280544281, | |
| "eval_runtime": 0.4229, | |
| "eval_samples_per_second": 257.766, | |
| "eval_steps_per_second": 9.459, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 1.0161, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.6055045871559633, | |
| "eval_loss": 0.9511241912841797, | |
| "eval_runtime": 0.5485, | |
| "eval_samples_per_second": 198.707, | |
| "eval_steps_per_second": 7.292, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.9633, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6605504587155964, | |
| "eval_loss": 0.9009455442428589, | |
| "eval_runtime": 0.42, | |
| "eval_samples_per_second": 259.508, | |
| "eval_steps_per_second": 9.523, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.939, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.6513761467889908, | |
| "eval_loss": 0.9054697155952454, | |
| "eval_runtime": 0.457, | |
| "eval_samples_per_second": 238.519, | |
| "eval_steps_per_second": 8.753, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.6697247706422018, | |
| "eval_loss": 0.8781045079231262, | |
| "eval_runtime": 0.4171, | |
| "eval_samples_per_second": 261.315, | |
| "eval_steps_per_second": 9.59, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.9036, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.7247706422018348, | |
| "eval_loss": 0.8493882417678833, | |
| "eval_runtime": 0.4124, | |
| "eval_samples_per_second": 264.29, | |
| "eval_steps_per_second": 9.699, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.8687, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6788990825688074, | |
| "eval_loss": 0.850315272808075, | |
| "eval_runtime": 0.4239, | |
| "eval_samples_per_second": 257.162, | |
| "eval_steps_per_second": 9.437, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.8535, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.7247706422018348, | |
| "eval_loss": 0.8164355158805847, | |
| "eval_runtime": 0.4095, | |
| "eval_samples_per_second": 266.199, | |
| "eval_steps_per_second": 9.769, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.7882992029190063, | |
| "eval_runtime": 0.4137, | |
| "eval_samples_per_second": 263.479, | |
| "eval_steps_per_second": 9.669, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8306, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.7667210698127747, | |
| "eval_runtime": 0.4184, | |
| "eval_samples_per_second": 260.52, | |
| "eval_steps_per_second": 9.56, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.8047, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.7600359320640564, | |
| "eval_runtime": 0.5597, | |
| "eval_samples_per_second": 194.752, | |
| "eval_steps_per_second": 7.147, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.7735, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.7331238985061646, | |
| "eval_runtime": 0.4102, | |
| "eval_samples_per_second": 265.702, | |
| "eval_steps_per_second": 9.751, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.784, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.7294867038726807, | |
| "eval_runtime": 0.4762, | |
| "eval_samples_per_second": 228.901, | |
| "eval_steps_per_second": 8.4, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.7431192660550459, | |
| "eval_loss": 0.7281246185302734, | |
| "eval_runtime": 0.4298, | |
| "eval_samples_per_second": 253.584, | |
| "eval_steps_per_second": 9.306, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.7596, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.7044692635536194, | |
| "eval_runtime": 0.4192, | |
| "eval_samples_per_second": 260.022, | |
| "eval_steps_per_second": 9.542, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.7609, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.7706422018348624, | |
| "eval_loss": 0.6915290951728821, | |
| "eval_runtime": 0.4079, | |
| "eval_samples_per_second": 267.227, | |
| "eval_steps_per_second": 9.806, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.7307, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.8073394495412844, | |
| "eval_loss": 0.6969889998435974, | |
| "eval_runtime": 0.4136, | |
| "eval_samples_per_second": 263.538, | |
| "eval_steps_per_second": 9.671, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.6796095371246338, | |
| "eval_runtime": 0.4183, | |
| "eval_samples_per_second": 260.58, | |
| "eval_steps_per_second": 9.563, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.7263, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7706422018348624, | |
| "eval_loss": 0.6614620089530945, | |
| "eval_runtime": 0.4203, | |
| "eval_samples_per_second": 259.327, | |
| "eval_steps_per_second": 9.517, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.6933, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.7798165137614679, | |
| "eval_loss": 0.662841796875, | |
| "eval_runtime": 0.5565, | |
| "eval_samples_per_second": 195.855, | |
| "eval_steps_per_second": 7.187, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.6914, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.8073394495412844, | |
| "eval_loss": 0.6596092581748962, | |
| "eval_runtime": 0.4186, | |
| "eval_samples_per_second": 260.4, | |
| "eval_steps_per_second": 9.556, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.7192, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.6452586054801941, | |
| "eval_runtime": 0.6385, | |
| "eval_samples_per_second": 170.723, | |
| "eval_steps_per_second": 6.265, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7798165137614679, | |
| "eval_loss": 0.6569339632987976, | |
| "eval_runtime": 0.4062, | |
| "eval_samples_per_second": 268.371, | |
| "eval_steps_per_second": 9.848, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.6956, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.8165137614678899, | |
| "eval_loss": 0.6494290828704834, | |
| "eval_runtime": 0.4328, | |
| "eval_samples_per_second": 251.861, | |
| "eval_steps_per_second": 9.243, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.7037, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.8073394495412844, | |
| "eval_loss": 0.6478080153465271, | |
| "eval_runtime": 0.4185, | |
| "eval_samples_per_second": 260.424, | |
| "eval_steps_per_second": 9.557, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.669, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.7798165137614679, | |
| "eval_loss": 0.641533613204956, | |
| "eval_runtime": 0.4124, | |
| "eval_samples_per_second": 264.279, | |
| "eval_steps_per_second": 9.698, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7889908256880734, | |
| "eval_loss": 0.6440734267234802, | |
| "eval_runtime": 0.4379, | |
| "eval_samples_per_second": 248.897, | |
| "eval_steps_per_second": 9.134, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 0.0, | |
| "loss": 0.6715, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_accuracy": 0.7798165137614679, | |
| "eval_loss": 0.6445415616035461, | |
| "eval_runtime": 0.5619, | |
| "eval_samples_per_second": 193.991, | |
| "eval_steps_per_second": 7.119, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "step": 280, | |
| "total_flos": 9.686412043576934e+16, | |
| "train_loss": 0.8973502908434187, | |
| "train_runtime": 243.9108, | |
| "train_samples_per_second": 160.386, | |
| "train_steps_per_second": 1.148 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 280, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 9.686412043576934e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |