| { | |
| "best_metric": 0.9318181818181818, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-85-fold4/checkpoint-38", | |
| "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": 0.7272727272727273, | |
| "eval_loss": 0.6100477576255798, | |
| "eval_runtime": 0.5982, | |
| "eval_samples_per_second": 73.559, | |
| "eval_steps_per_second": 3.344, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6938337087631226, | |
| "eval_runtime": 0.6086, | |
| "eval_samples_per_second": 72.301, | |
| "eval_steps_per_second": 3.286, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7567586898803711, | |
| "eval_runtime": 0.6143, | |
| "eval_samples_per_second": 71.627, | |
| "eval_steps_per_second": 3.256, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.614034116268158, | |
| "eval_runtime": 0.6056, | |
| "eval_samples_per_second": 72.657, | |
| "eval_steps_per_second": 3.303, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.441150665283203, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.5388, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.4976235330104828, | |
| "eval_runtime": 0.6035, | |
| "eval_samples_per_second": 72.903, | |
| "eval_steps_per_second": 3.314, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.4809054434299469, | |
| "eval_runtime": 0.609, | |
| "eval_samples_per_second": 72.253, | |
| "eval_steps_per_second": 3.284, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.5276241898536682, | |
| "eval_runtime": 0.5835, | |
| "eval_samples_per_second": 75.412, | |
| "eval_steps_per_second": 3.428, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.4454606771469116, | |
| "eval_runtime": 0.5958, | |
| "eval_samples_per_second": 73.854, | |
| "eval_steps_per_second": 3.357, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.39154812693595886, | |
| "eval_runtime": 0.5916, | |
| "eval_samples_per_second": 74.371, | |
| "eval_steps_per_second": 3.381, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.7959985733032227, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4154, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.5070433020591736, | |
| "eval_runtime": 0.6, | |
| "eval_samples_per_second": 73.334, | |
| "eval_steps_per_second": 3.333, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3746650516986847, | |
| "eval_runtime": 0.5985, | |
| "eval_samples_per_second": 73.522, | |
| "eval_steps_per_second": 3.342, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3026847541332245, | |
| "eval_runtime": 0.6061, | |
| "eval_samples_per_second": 72.598, | |
| "eval_steps_per_second": 3.3, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3053370714187622, | |
| "eval_runtime": 0.5986, | |
| "eval_samples_per_second": 73.507, | |
| "eval_steps_per_second": 3.341, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3194069564342499, | |
| "eval_runtime": 0.598, | |
| "eval_samples_per_second": 73.582, | |
| "eval_steps_per_second": 3.345, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.3912155628204346, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3258, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.31341588497161865, | |
| "eval_runtime": 0.5971, | |
| "eval_samples_per_second": 73.684, | |
| "eval_steps_per_second": 3.349, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.29249703884124756, | |
| "eval_runtime": 0.5964, | |
| "eval_samples_per_second": 73.771, | |
| "eval_steps_per_second": 3.353, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2448718547821045, | |
| "eval_runtime": 0.6033, | |
| "eval_samples_per_second": 72.932, | |
| "eval_steps_per_second": 3.315, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.23076215386390686, | |
| "eval_runtime": 0.5984, | |
| "eval_samples_per_second": 73.526, | |
| "eval_steps_per_second": 3.342, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21408408880233765, | |
| "eval_runtime": 0.5956, | |
| "eval_samples_per_second": 73.872, | |
| "eval_steps_per_second": 3.358, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.007518291473389, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2528, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.23295708000659943, | |
| "eval_runtime": 0.5971, | |
| "eval_samples_per_second": 73.696, | |
| "eval_steps_per_second": 3.35, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2172897756099701, | |
| "eval_runtime": 0.5979, | |
| "eval_samples_per_second": 73.596, | |
| "eval_steps_per_second": 3.345, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24502836167812347, | |
| "eval_runtime": 0.606, | |
| "eval_samples_per_second": 72.609, | |
| "eval_steps_per_second": 3.3, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2548504173755646, | |
| "eval_runtime": 0.5981, | |
| "eval_samples_per_second": 73.562, | |
| "eval_steps_per_second": 3.344, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.43413516879081726, | |
| "eval_runtime": 0.6039, | |
| "eval_samples_per_second": 72.864, | |
| "eval_steps_per_second": 3.312, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 4.777149200439453, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.175, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.23583923280239105, | |
| "eval_runtime": 0.6153, | |
| "eval_samples_per_second": 71.511, | |
| "eval_steps_per_second": 3.251, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.28283193707466125, | |
| "eval_runtime": 0.6059, | |
| "eval_samples_per_second": 72.619, | |
| "eval_steps_per_second": 3.301, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.22355598211288452, | |
| "eval_runtime": 0.603, | |
| "eval_samples_per_second": 72.972, | |
| "eval_steps_per_second": 3.317, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.25913283228874207, | |
| "eval_runtime": 0.6015, | |
| "eval_samples_per_second": 73.148, | |
| "eval_steps_per_second": 3.325, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2702433168888092, | |
| "eval_runtime": 0.605, | |
| "eval_samples_per_second": 72.733, | |
| "eval_steps_per_second": 3.306, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 2.15972900390625, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.169, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.29102349281311035, | |
| "eval_runtime": 0.6164, | |
| "eval_samples_per_second": 71.387, | |
| "eval_steps_per_second": 3.245, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3593917489051819, | |
| "eval_runtime": 0.6149, | |
| "eval_samples_per_second": 71.557, | |
| "eval_steps_per_second": 3.253, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4245869815349579, | |
| "eval_runtime": 0.6021, | |
| "eval_samples_per_second": 73.075, | |
| "eval_steps_per_second": 3.322, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2654532194137573, | |
| "eval_runtime": 0.6, | |
| "eval_samples_per_second": 73.338, | |
| "eval_steps_per_second": 3.334, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2581041157245636, | |
| "eval_runtime": 0.5943, | |
| "eval_samples_per_second": 74.04, | |
| "eval_steps_per_second": 3.365, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 2.121372699737549, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1336, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.24943043291568756, | |
| "eval_runtime": 0.5899, | |
| "eval_samples_per_second": 74.591, | |
| "eval_steps_per_second": 3.39, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.24377375841140747, | |
| "eval_runtime": 0.6083, | |
| "eval_samples_per_second": 72.327, | |
| "eval_steps_per_second": 3.288, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3245680630207062, | |
| "eval_runtime": 0.5951, | |
| "eval_samples_per_second": 73.933, | |
| "eval_steps_per_second": 3.361, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.2887042760848999, | |
| "eval_runtime": 0.5941, | |
| "eval_samples_per_second": 74.068, | |
| "eval_steps_per_second": 3.367, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3559167981147766, | |
| "eval_runtime": 0.612, | |
| "eval_samples_per_second": 71.894, | |
| "eval_steps_per_second": 3.268, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.0831546783447266, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1281, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3273710608482361, | |
| "eval_runtime": 0.5971, | |
| "eval_samples_per_second": 73.696, | |
| "eval_steps_per_second": 3.35, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3370969891548157, | |
| "eval_runtime": 0.5941, | |
| "eval_samples_per_second": 74.064, | |
| "eval_steps_per_second": 3.367, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3902018368244171, | |
| "eval_runtime": 0.5919, | |
| "eval_samples_per_second": 74.341, | |
| "eval_steps_per_second": 3.379, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3100315034389496, | |
| "eval_runtime": 0.5984, | |
| "eval_samples_per_second": 73.529, | |
| "eval_steps_per_second": 3.342, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.31125500798225403, | |
| "eval_runtime": 0.5953, | |
| "eval_samples_per_second": 73.917, | |
| "eval_steps_per_second": 3.36, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 3.431731700897217, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.136, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3244210481643677, | |
| "eval_runtime": 0.5945, | |
| "eval_samples_per_second": 74.013, | |
| "eval_steps_per_second": 3.364, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.376524955034256, | |
| "eval_runtime": 0.602, | |
| "eval_samples_per_second": 73.085, | |
| "eval_steps_per_second": 3.322, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.38375338912010193, | |
| "eval_runtime": 0.6159, | |
| "eval_samples_per_second": 71.437, | |
| "eval_steps_per_second": 3.247, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.38451334834098816, | |
| "eval_runtime": 0.6046, | |
| "eval_samples_per_second": 72.778, | |
| "eval_steps_per_second": 3.308, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.3910039961338043, | |
| "eval_runtime": 0.604, | |
| "eval_samples_per_second": 72.849, | |
| "eval_steps_per_second": 3.311, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 1.7765434980392456, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0934, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4889444410800934, | |
| "eval_runtime": 0.6053, | |
| "eval_samples_per_second": 72.694, | |
| "eval_steps_per_second": 3.304, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.6680083870887756, | |
| "eval_runtime": 0.5923, | |
| "eval_samples_per_second": 74.282, | |
| "eval_steps_per_second": 3.376, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.42642828822135925, | |
| "eval_runtime": 0.6032, | |
| "eval_samples_per_second": 72.944, | |
| "eval_steps_per_second": 3.316, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.32662203907966614, | |
| "eval_runtime": 0.6399, | |
| "eval_samples_per_second": 68.762, | |
| "eval_steps_per_second": 3.126, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3167722523212433, | |
| "eval_runtime": 0.609, | |
| "eval_samples_per_second": 72.245, | |
| "eval_steps_per_second": 3.284, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 1.9470126628875732, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0999, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.36713549494743347, | |
| "eval_runtime": 0.6162, | |
| "eval_samples_per_second": 71.405, | |
| "eval_steps_per_second": 3.246, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.4684106111526489, | |
| "eval_runtime": 0.6017, | |
| "eval_samples_per_second": 73.127, | |
| "eval_steps_per_second": 3.324, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.4253728687763214, | |
| "eval_runtime": 0.5945, | |
| "eval_samples_per_second": 74.015, | |
| "eval_steps_per_second": 3.364, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3194880187511444, | |
| "eval_runtime": 0.5884, | |
| "eval_samples_per_second": 74.775, | |
| "eval_steps_per_second": 3.399, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.38602694869041443, | |
| "eval_runtime": 0.6193, | |
| "eval_samples_per_second": 71.047, | |
| "eval_steps_per_second": 3.229, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.181431293487549, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1145, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.480492502450943, | |
| "eval_runtime": 0.5978, | |
| "eval_samples_per_second": 73.606, | |
| "eval_steps_per_second": 3.346, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.38636693358421326, | |
| "eval_runtime": 0.6014, | |
| "eval_samples_per_second": 73.167, | |
| "eval_steps_per_second": 3.326, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.33465278148651123, | |
| "eval_runtime": 0.5938, | |
| "eval_samples_per_second": 74.093, | |
| "eval_steps_per_second": 3.368, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.31437891721725464, | |
| "eval_runtime": 0.6035, | |
| "eval_samples_per_second": 72.904, | |
| "eval_steps_per_second": 3.314, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.32668235898017883, | |
| "eval_runtime": 0.6028, | |
| "eval_samples_per_second": 72.997, | |
| "eval_steps_per_second": 3.318, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 1.6601436138153076, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0769, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.35917937755584717, | |
| "eval_runtime": 0.6038, | |
| "eval_samples_per_second": 72.874, | |
| "eval_steps_per_second": 3.312, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.35203883051872253, | |
| "eval_runtime": 0.6143, | |
| "eval_samples_per_second": 71.623, | |
| "eval_steps_per_second": 3.256, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.363180011510849, | |
| "eval_runtime": 0.6023, | |
| "eval_samples_per_second": 73.058, | |
| "eval_steps_per_second": 3.321, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.39549723267555237, | |
| "eval_runtime": 0.6117, | |
| "eval_samples_per_second": 71.928, | |
| "eval_steps_per_second": 3.269, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.40525779128074646, | |
| "eval_runtime": 0.6028, | |
| "eval_samples_per_second": 72.992, | |
| "eval_steps_per_second": 3.318, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 2.5139784812927246, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0976, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4272138774394989, | |
| "eval_runtime": 0.5874, | |
| "eval_samples_per_second": 74.912, | |
| "eval_steps_per_second": 3.405, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.43452635407447815, | |
| "eval_runtime": 0.5921, | |
| "eval_samples_per_second": 74.314, | |
| "eval_steps_per_second": 3.378, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3943205177783966, | |
| "eval_runtime": 0.6032, | |
| "eval_samples_per_second": 72.942, | |
| "eval_steps_per_second": 3.316, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.38274791836738586, | |
| "eval_runtime": 0.6139, | |
| "eval_samples_per_second": 71.67, | |
| "eval_steps_per_second": 3.258, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.41329869627952576, | |
| "eval_runtime": 0.6031, | |
| "eval_samples_per_second": 72.959, | |
| "eval_steps_per_second": 3.316, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 2.924851894378662, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0981, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4310622215270996, | |
| "eval_runtime": 0.6091, | |
| "eval_samples_per_second": 72.241, | |
| "eval_steps_per_second": 3.284, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.41259530186653137, | |
| "eval_runtime": 0.6322, | |
| "eval_samples_per_second": 69.601, | |
| "eval_steps_per_second": 3.164, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3651290237903595, | |
| "eval_runtime": 0.6073, | |
| "eval_samples_per_second": 72.456, | |
| "eval_steps_per_second": 3.293, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.35113096237182617, | |
| "eval_runtime": 0.6127, | |
| "eval_samples_per_second": 71.808, | |
| "eval_steps_per_second": 3.264, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.36248430609703064, | |
| "eval_runtime": 0.6026, | |
| "eval_samples_per_second": 73.021, | |
| "eval_steps_per_second": 3.319, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.581064224243164, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.085, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.36069193482398987, | |
| "eval_runtime": 0.6083, | |
| "eval_samples_per_second": 72.335, | |
| "eval_steps_per_second": 3.288, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3470054566860199, | |
| "eval_runtime": 0.6093, | |
| "eval_samples_per_second": 72.22, | |
| "eval_steps_per_second": 3.283, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3638794720172882, | |
| "eval_runtime": 0.609, | |
| "eval_samples_per_second": 72.247, | |
| "eval_steps_per_second": 3.284, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.37499523162841797, | |
| "eval_runtime": 0.5889, | |
| "eval_samples_per_second": 74.721, | |
| "eval_steps_per_second": 3.396, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.37264367938041687, | |
| "eval_runtime": 0.5983, | |
| "eval_samples_per_second": 73.54, | |
| "eval_steps_per_second": 3.343, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 2.4498684406280518, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0831, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3740333318710327, | |
| "eval_runtime": 0.6045, | |
| "eval_samples_per_second": 72.782, | |
| "eval_steps_per_second": 3.308, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3807307183742523, | |
| "eval_runtime": 0.5999, | |
| "eval_samples_per_second": 73.346, | |
| "eval_steps_per_second": 3.334, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3874521255493164, | |
| "eval_runtime": 0.603, | |
| "eval_samples_per_second": 72.974, | |
| "eval_steps_per_second": 3.317, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.38860151171684265, | |
| "eval_runtime": 0.6003, | |
| "eval_samples_per_second": 73.301, | |
| "eval_steps_per_second": 3.332, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.40169432759284973, | |
| "eval_runtime": 0.6004, | |
| "eval_samples_per_second": 73.284, | |
| "eval_steps_per_second": 3.331, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 1.2369847297668457, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0811, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.42713773250579834, | |
| "eval_runtime": 0.595, | |
| "eval_samples_per_second": 73.954, | |
| "eval_steps_per_second": 3.362, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.4293382465839386, | |
| "eval_runtime": 0.6009, | |
| "eval_samples_per_second": 73.225, | |
| "eval_steps_per_second": 3.328, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.4243467450141907, | |
| "eval_runtime": 0.5998, | |
| "eval_samples_per_second": 73.355, | |
| "eval_steps_per_second": 3.334, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.40882861614227295, | |
| "eval_runtime": 0.6167, | |
| "eval_samples_per_second": 71.343, | |
| "eval_steps_per_second": 3.243, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.3986482620239258, | |
| "eval_runtime": 0.6026, | |
| "eval_samples_per_second": 73.016, | |
| "eval_steps_per_second": 3.319, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 3.285630226135254, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0692, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3963455259799957, | |
| "eval_runtime": 0.5999, | |
| "eval_samples_per_second": 73.347, | |
| "eval_steps_per_second": 3.334, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.398687481880188, | |
| "eval_runtime": 0.6088, | |
| "eval_samples_per_second": 72.273, | |
| "eval_steps_per_second": 3.285, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4020019769668579, | |
| "eval_runtime": 0.6171, | |
| "eval_samples_per_second": 71.306, | |
| "eval_steps_per_second": 3.241, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4014514684677124, | |
| "eval_runtime": 0.6001, | |
| "eval_samples_per_second": 73.32, | |
| "eval_steps_per_second": 3.333, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.40093833208084106, | |
| "eval_runtime": 0.6022, | |
| "eval_samples_per_second": 73.068, | |
| "eval_steps_per_second": 3.321, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.9313592910766602, | |
| "learning_rate": 0.0, | |
| "loss": 0.0644, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4002065360546112, | |
| "eval_runtime": 0.608, | |
| "eval_samples_per_second": 72.372, | |
| "eval_steps_per_second": 3.29, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_loss": 0.16189448922872543, | |
| "train_runtime": 1112.6178, | |
| "train_samples_per_second": 21.84, | |
| "train_steps_per_second": 0.18 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21408408880233765, | |
| "eval_runtime": 0.6455, | |
| "eval_samples_per_second": 68.166, | |
| "eval_steps_per_second": 3.098, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 1.883089058199552e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |