| { | |
| "best_metric": 0.3964671194553375, | |
| "best_model_checkpoint": "limb_classification_person_crop_seq/t2_4heads_1layers_5e-4lr/checkpoint-2520", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 2700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1388888888888889, | |
| "grad_norm": 275926.53125, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2686, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 328248.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9051, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 368871.40625, | |
| "learning_rate": 0.00015, | |
| "loss": 0.7478, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 98371.5390625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5482, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6944444444444444, | |
| "grad_norm": 116783.9921875, | |
| "learning_rate": 0.00025, | |
| "loss": 0.5521, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 390853.9375, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4814, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9722222222222222, | |
| "grad_norm": 156001.8125, | |
| "learning_rate": 0.00035, | |
| "loss": 0.5406, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8883399209486166, | |
| "eval_loss": 0.48044389486312866, | |
| "eval_runtime": 37.7276, | |
| "eval_samples_per_second": 26.824, | |
| "eval_steps_per_second": 0.848, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 191408.703125, | |
| "learning_rate": 0.0004, | |
| "loss": 0.5133, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 313042.46875, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.4485, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 164934.65625, | |
| "learning_rate": 0.0005, | |
| "loss": 0.5473, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.5277777777777777, | |
| "grad_norm": 266101.15625, | |
| "learning_rate": 0.0004948979591836735, | |
| "loss": 0.4893, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 216736.703125, | |
| "learning_rate": 0.0004897959183673469, | |
| "loss": 0.4952, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.8055555555555556, | |
| "grad_norm": 130781.34375, | |
| "learning_rate": 0.0004846938775510204, | |
| "loss": 0.5304, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 83755.078125, | |
| "learning_rate": 0.00047959183673469387, | |
| "loss": 0.4852, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8794466403162056, | |
| "eval_loss": 0.5456948280334473, | |
| "eval_runtime": 36.26, | |
| "eval_samples_per_second": 27.91, | |
| "eval_steps_per_second": 0.883, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 140542.96875, | |
| "learning_rate": 0.0004744897959183674, | |
| "loss": 0.5491, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 275681.375, | |
| "learning_rate": 0.00046938775510204083, | |
| "loss": 0.4691, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.361111111111111, | |
| "grad_norm": 119873.5625, | |
| "learning_rate": 0.00046428571428571433, | |
| "loss": 0.4607, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 147333.703125, | |
| "learning_rate": 0.0004591836734693878, | |
| "loss": 0.5189, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.638888888888889, | |
| "grad_norm": 152648.078125, | |
| "learning_rate": 0.00045408163265306124, | |
| "loss": 0.4508, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 129466.5078125, | |
| "learning_rate": 0.0004489795918367347, | |
| "loss": 0.4649, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.9166666666666665, | |
| "grad_norm": 39402.28515625, | |
| "learning_rate": 0.00044387755102040814, | |
| "loss": 0.4664, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9051383399209486, | |
| "eval_loss": 0.42033523321151733, | |
| "eval_runtime": 36.9286, | |
| "eval_samples_per_second": 27.404, | |
| "eval_steps_per_second": 0.867, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.0555555555555554, | |
| "grad_norm": 167756.265625, | |
| "learning_rate": 0.00043877551020408165, | |
| "loss": 0.4957, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.1944444444444446, | |
| "grad_norm": 178619.546875, | |
| "learning_rate": 0.0004336734693877551, | |
| "loss": 0.4743, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 109380.5, | |
| "learning_rate": 0.00042857142857142855, | |
| "loss": 0.5037, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.4722222222222223, | |
| "grad_norm": 161733.875, | |
| "learning_rate": 0.00042346938775510206, | |
| "loss": 0.4998, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.611111111111111, | |
| "grad_norm": 139706.953125, | |
| "learning_rate": 0.00041836734693877556, | |
| "loss": 0.4484, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 150799.125, | |
| "learning_rate": 0.000413265306122449, | |
| "loss": 0.5218, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 233782.0625, | |
| "learning_rate": 0.00040816326530612246, | |
| "loss": 0.4929, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.900197628458498, | |
| "eval_loss": 0.43486830592155457, | |
| "eval_runtime": 35.8273, | |
| "eval_samples_per_second": 28.247, | |
| "eval_steps_per_second": 0.893, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.027777777777778, | |
| "grad_norm": 124010.9140625, | |
| "learning_rate": 0.0004030612244897959, | |
| "loss": 0.4506, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 142697.78125, | |
| "learning_rate": 0.00039795918367346937, | |
| "loss": 0.4781, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.305555555555555, | |
| "grad_norm": 167024.359375, | |
| "learning_rate": 0.0003928571428571429, | |
| "loss": 0.4571, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 91725.46875, | |
| "learning_rate": 0.0003877551020408163, | |
| "loss": 0.45, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.583333333333333, | |
| "grad_norm": 99782.9921875, | |
| "learning_rate": 0.0003826530612244898, | |
| "loss": 0.4521, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 4.722222222222222, | |
| "grad_norm": 178208.5625, | |
| "learning_rate": 0.00037755102040816323, | |
| "loss": 0.4598, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.861111111111111, | |
| "grad_norm": 189875.09375, | |
| "learning_rate": 0.0003724489795918368, | |
| "loss": 0.4492, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 31828.625, | |
| "learning_rate": 0.00036734693877551024, | |
| "loss": 0.4334, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8764822134387352, | |
| "eval_loss": 0.4815811514854431, | |
| "eval_runtime": 36.4103, | |
| "eval_samples_per_second": 27.794, | |
| "eval_steps_per_second": 0.879, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.138888888888889, | |
| "grad_norm": 131415.0, | |
| "learning_rate": 0.0003622448979591837, | |
| "loss": 0.4699, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 5.277777777777778, | |
| "grad_norm": 42832.09765625, | |
| "learning_rate": 0.00035714285714285714, | |
| "loss": 0.4448, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.416666666666667, | |
| "grad_norm": 202659.328125, | |
| "learning_rate": 0.00035204081632653065, | |
| "loss": 0.4595, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "grad_norm": 95209.203125, | |
| "learning_rate": 0.0003469387755102041, | |
| "loss": 0.4968, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.694444444444445, | |
| "grad_norm": 121417.4140625, | |
| "learning_rate": 0.00034183673469387755, | |
| "loss": 0.4662, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 103243.7734375, | |
| "learning_rate": 0.000336734693877551, | |
| "loss": 0.4726, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.972222222222222, | |
| "grad_norm": 79883.8671875, | |
| "learning_rate": 0.00033163265306122445, | |
| "loss": 0.4709, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8932806324110671, | |
| "eval_loss": 0.45736971497535706, | |
| "eval_runtime": 36.8482, | |
| "eval_samples_per_second": 27.464, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 6.111111111111111, | |
| "grad_norm": 70588.4609375, | |
| "learning_rate": 0.00032653061224489796, | |
| "loss": 0.4736, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 235716.375, | |
| "learning_rate": 0.00032142857142857147, | |
| "loss": 0.4877, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 6.388888888888889, | |
| "grad_norm": 118860.515625, | |
| "learning_rate": 0.0003163265306122449, | |
| "loss": 0.4801, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 6.527777777777778, | |
| "grad_norm": 96496.296875, | |
| "learning_rate": 0.00031122448979591837, | |
| "loss": 0.4387, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 122115.0859375, | |
| "learning_rate": 0.0003061224489795919, | |
| "loss": 0.3712, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.805555555555555, | |
| "grad_norm": 87567.2265625, | |
| "learning_rate": 0.0003010204081632653, | |
| "loss": 0.443, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 102648.7265625, | |
| "learning_rate": 0.0002959183673469388, | |
| "loss": 0.4525, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8883399209486166, | |
| "eval_loss": 0.465226411819458, | |
| "eval_runtime": 36.2286, | |
| "eval_samples_per_second": 27.934, | |
| "eval_steps_per_second": 0.883, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 7.083333333333333, | |
| "grad_norm": 54489.8828125, | |
| "learning_rate": 0.00029081632653061223, | |
| "loss": 0.4306, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 7.222222222222222, | |
| "grad_norm": 81611.7890625, | |
| "learning_rate": 0.0002857142857142857, | |
| "loss": 0.4381, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.361111111111111, | |
| "grad_norm": 139663.171875, | |
| "learning_rate": 0.0002806122448979592, | |
| "loss": 0.457, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 73111.8828125, | |
| "learning_rate": 0.00027551020408163264, | |
| "loss": 0.4403, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 7.638888888888889, | |
| "grad_norm": 56448.75, | |
| "learning_rate": 0.00027040816326530614, | |
| "loss": 0.4308, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 7.777777777777778, | |
| "grad_norm": 59582.625, | |
| "learning_rate": 0.0002653061224489796, | |
| "loss": 0.4565, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.916666666666667, | |
| "grad_norm": 167981.71875, | |
| "learning_rate": 0.0002602040816326531, | |
| "loss": 0.4601, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.900197628458498, | |
| "eval_loss": 0.4387129545211792, | |
| "eval_runtime": 36.2307, | |
| "eval_samples_per_second": 27.932, | |
| "eval_steps_per_second": 0.883, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 8.055555555555555, | |
| "grad_norm": 146485.015625, | |
| "learning_rate": 0.00025510204081632655, | |
| "loss": 0.4442, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 8.194444444444445, | |
| "grad_norm": 77854.203125, | |
| "learning_rate": 0.00025, | |
| "loss": 0.4185, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 123192.34375, | |
| "learning_rate": 0.00024489795918367346, | |
| "loss": 0.4692, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.472222222222221, | |
| "grad_norm": 74983.546875, | |
| "learning_rate": 0.00023979591836734694, | |
| "loss": 0.4189, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 8.61111111111111, | |
| "grad_norm": 136541.65625, | |
| "learning_rate": 0.00023469387755102041, | |
| "loss": 0.4135, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 90828.046875, | |
| "learning_rate": 0.0002295918367346939, | |
| "loss": 0.419, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 92607.2109375, | |
| "learning_rate": 0.00022448979591836734, | |
| "loss": 0.4361, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9071146245059288, | |
| "eval_loss": 0.4137505292892456, | |
| "eval_runtime": 36.7818, | |
| "eval_samples_per_second": 27.514, | |
| "eval_steps_per_second": 0.87, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 9.027777777777779, | |
| "grad_norm": 206206.59375, | |
| "learning_rate": 0.00021938775510204082, | |
| "loss": 0.4229, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "grad_norm": 61973.49609375, | |
| "learning_rate": 0.00021428571428571427, | |
| "loss": 0.4488, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 9.305555555555555, | |
| "grad_norm": 134002.28125, | |
| "learning_rate": 0.00020918367346938778, | |
| "loss": 0.4099, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 9.444444444444445, | |
| "grad_norm": 162290.21875, | |
| "learning_rate": 0.00020408163265306123, | |
| "loss": 0.4352, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.583333333333334, | |
| "grad_norm": 108405.265625, | |
| "learning_rate": 0.00019897959183673468, | |
| "loss": 0.4241, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 9.722222222222221, | |
| "grad_norm": 66163.1875, | |
| "learning_rate": 0.00019387755102040816, | |
| "loss": 0.4053, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 9.86111111111111, | |
| "grad_norm": 35489.41796875, | |
| "learning_rate": 0.00018877551020408161, | |
| "loss": 0.4023, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 148451.015625, | |
| "learning_rate": 0.00018367346938775512, | |
| "loss": 0.4297, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9140316205533597, | |
| "eval_loss": 0.4089234173297882, | |
| "eval_runtime": 36.486, | |
| "eval_samples_per_second": 27.737, | |
| "eval_steps_per_second": 0.877, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.13888888888889, | |
| "grad_norm": 132313.015625, | |
| "learning_rate": 0.00017857142857142857, | |
| "loss": 0.4157, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 10.277777777777779, | |
| "grad_norm": 79190.890625, | |
| "learning_rate": 0.00017346938775510205, | |
| "loss": 0.4155, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 10.416666666666666, | |
| "grad_norm": 59022.73828125, | |
| "learning_rate": 0.0001683673469387755, | |
| "loss": 0.3802, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 10.555555555555555, | |
| "grad_norm": 113138.1640625, | |
| "learning_rate": 0.00016326530612244898, | |
| "loss": 0.4633, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.694444444444445, | |
| "grad_norm": 80191.265625, | |
| "learning_rate": 0.00015816326530612246, | |
| "loss": 0.4034, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 10.833333333333334, | |
| "grad_norm": 80256.8984375, | |
| "learning_rate": 0.00015306122448979594, | |
| "loss": 0.4033, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 10.972222222222221, | |
| "grad_norm": 92320.3359375, | |
| "learning_rate": 0.0001479591836734694, | |
| "loss": 0.4, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9199604743083004, | |
| "eval_loss": 0.3999524712562561, | |
| "eval_runtime": 36.413, | |
| "eval_samples_per_second": 27.792, | |
| "eval_steps_per_second": 0.879, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 11.11111111111111, | |
| "grad_norm": 83646.3828125, | |
| "learning_rate": 0.00014285714285714284, | |
| "loss": 0.4017, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "grad_norm": 47865.28515625, | |
| "learning_rate": 0.00013775510204081632, | |
| "loss": 0.4172, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 11.38888888888889, | |
| "grad_norm": 100859.5859375, | |
| "learning_rate": 0.0001326530612244898, | |
| "loss": 0.3601, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 11.527777777777779, | |
| "grad_norm": 89678.1796875, | |
| "learning_rate": 0.00012755102040816328, | |
| "loss": 0.459, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 11.666666666666666, | |
| "grad_norm": 97468.703125, | |
| "learning_rate": 0.00012244897959183673, | |
| "loss": 0.3659, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.805555555555555, | |
| "grad_norm": 116296.359375, | |
| "learning_rate": 0.00011734693877551021, | |
| "loss": 0.4134, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 11.944444444444445, | |
| "grad_norm": 66697.9296875, | |
| "learning_rate": 0.00011224489795918367, | |
| "loss": 0.4035, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9071146245059288, | |
| "eval_loss": 0.42599722743034363, | |
| "eval_runtime": 36.752, | |
| "eval_samples_per_second": 27.536, | |
| "eval_steps_per_second": 0.871, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 12.083333333333334, | |
| "grad_norm": 139186.21875, | |
| "learning_rate": 0.00010714285714285714, | |
| "loss": 0.3609, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 12.222222222222221, | |
| "grad_norm": 69709.2109375, | |
| "learning_rate": 0.00010204081632653062, | |
| "loss": 0.4146, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.36111111111111, | |
| "grad_norm": 84500.0859375, | |
| "learning_rate": 9.693877551020408e-05, | |
| "loss": 0.4013, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 45239.5703125, | |
| "learning_rate": 9.183673469387756e-05, | |
| "loss": 0.3918, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 12.63888888888889, | |
| "grad_norm": 49387.7421875, | |
| "learning_rate": 8.673469387755102e-05, | |
| "loss": 0.388, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 12.777777777777779, | |
| "grad_norm": 98527.546875, | |
| "learning_rate": 8.163265306122449e-05, | |
| "loss": 0.3941, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.916666666666666, | |
| "grad_norm": 75106.1640625, | |
| "learning_rate": 7.653061224489797e-05, | |
| "loss": 0.3875, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9100790513833992, | |
| "eval_loss": 0.40881994366645813, | |
| "eval_runtime": 36.7096, | |
| "eval_samples_per_second": 27.568, | |
| "eval_steps_per_second": 0.872, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 13.055555555555555, | |
| "grad_norm": 94333.46875, | |
| "learning_rate": 7.142857142857142e-05, | |
| "loss": 0.395, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 13.194444444444445, | |
| "grad_norm": 150090.71875, | |
| "learning_rate": 6.63265306122449e-05, | |
| "loss": 0.3972, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 86562.6015625, | |
| "learning_rate": 6.122448979591836e-05, | |
| "loss": 0.3347, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.472222222222221, | |
| "grad_norm": 205886.484375, | |
| "learning_rate": 5.6122448979591836e-05, | |
| "loss": 0.4316, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 13.61111111111111, | |
| "grad_norm": 90394.1640625, | |
| "learning_rate": 5.102040816326531e-05, | |
| "loss": 0.3521, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "grad_norm": 118663.3359375, | |
| "learning_rate": 4.591836734693878e-05, | |
| "loss": 0.4071, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 13.88888888888889, | |
| "grad_norm": 91543.4765625, | |
| "learning_rate": 4.0816326530612245e-05, | |
| "loss": 0.4117, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9179841897233202, | |
| "eval_loss": 0.3964671194553375, | |
| "eval_runtime": 36.2893, | |
| "eval_samples_per_second": 27.887, | |
| "eval_steps_per_second": 0.882, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 14.027777777777779, | |
| "grad_norm": 48705.08203125, | |
| "learning_rate": 3.571428571428571e-05, | |
| "loss": 0.3965, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 14.166666666666666, | |
| "grad_norm": 71733.8046875, | |
| "learning_rate": 3.061224489795918e-05, | |
| "loss": 0.4043, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 14.305555555555555, | |
| "grad_norm": 113618.6484375, | |
| "learning_rate": 2.5510204081632654e-05, | |
| "loss": 0.3868, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 14.444444444444445, | |
| "grad_norm": 90760.4609375, | |
| "learning_rate": 2.0408163265306123e-05, | |
| "loss": 0.3633, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.583333333333334, | |
| "grad_norm": 58063.44921875, | |
| "learning_rate": 1.530612244897959e-05, | |
| "loss": 0.3651, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 14.722222222222221, | |
| "grad_norm": 66486.3984375, | |
| "learning_rate": 1.0204081632653061e-05, | |
| "loss": 0.3904, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 14.86111111111111, | |
| "grad_norm": 68429.1484375, | |
| "learning_rate": 5.102040816326531e-06, | |
| "loss": 0.4017, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 489841.0625, | |
| "learning_rate": 0.0, | |
| "loss": 0.3518, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.91600790513834, | |
| "eval_loss": 0.3987027406692505, | |
| "eval_runtime": 36.8038, | |
| "eval_samples_per_second": 27.497, | |
| "eval_steps_per_second": 0.869, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 2700, | |
| "total_flos": 0.0, | |
| "train_loss": 0.45526096591243037, | |
| "train_runtime": 5324.9693, | |
| "train_samples_per_second": 16.141, | |
| "train_steps_per_second": 0.507 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |