| { | |
| "best_metric": 0.8227848101265823, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-55-fold5/checkpoint-150", | |
| "epoch": 85.71428571428571, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_accuracy": 0.5189873417721519, | |
| "eval_loss": 0.7318889498710632, | |
| "eval_runtime": 1.1213, | |
| "eval_samples_per_second": 70.455, | |
| "eval_steps_per_second": 2.676, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.569620253164557, | |
| "eval_loss": 0.675592303276062, | |
| "eval_runtime": 1.1562, | |
| "eval_samples_per_second": 68.326, | |
| "eval_steps_per_second": 2.595, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.9025880098342896, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.6544, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.6455696202531646, | |
| "eval_loss": 0.619909405708313, | |
| "eval_runtime": 1.2043, | |
| "eval_samples_per_second": 65.598, | |
| "eval_steps_per_second": 2.491, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.620253164556962, | |
| "eval_loss": 0.5987372994422913, | |
| "eval_runtime": 1.1134, | |
| "eval_samples_per_second": 70.952, | |
| "eval_steps_per_second": 2.694, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_accuracy": 0.6708860759493671, | |
| "eval_loss": 0.5676239728927612, | |
| "eval_runtime": 1.1193, | |
| "eval_samples_per_second": 70.581, | |
| "eval_steps_per_second": 2.68, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 2.5250775814056396, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6173, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5822784810126582, | |
| "eval_loss": 0.6542880535125732, | |
| "eval_runtime": 1.1319, | |
| "eval_samples_per_second": 69.791, | |
| "eval_steps_per_second": 2.65, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.5309708118438721, | |
| "eval_runtime": 1.1635, | |
| "eval_samples_per_second": 67.9, | |
| "eval_steps_per_second": 2.578, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6075949367088608, | |
| "eval_loss": 0.6724244356155396, | |
| "eval_runtime": 1.1519, | |
| "eval_samples_per_second": 68.58, | |
| "eval_steps_per_second": 2.604, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 3.0099055767059326, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5245, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.857142857142858, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.6443866491317749, | |
| "eval_runtime": 1.1308, | |
| "eval_samples_per_second": 69.861, | |
| "eval_steps_per_second": 2.653, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.5026643872261047, | |
| "eval_runtime": 1.1348, | |
| "eval_samples_per_second": 69.614, | |
| "eval_steps_per_second": 2.644, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.6327927708625793, | |
| "eval_runtime": 1.1376, | |
| "eval_samples_per_second": 69.443, | |
| "eval_steps_per_second": 2.637, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 3.429919719696045, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.4554, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.48831707239151, | |
| "eval_runtime": 1.1619, | |
| "eval_samples_per_second": 67.99, | |
| "eval_steps_per_second": 2.582, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.6736446619033813, | |
| "eval_runtime": 1.1393, | |
| "eval_samples_per_second": 69.338, | |
| "eval_steps_per_second": 2.633, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.4583931863307953, | |
| "eval_runtime": 1.1376, | |
| "eval_samples_per_second": 69.444, | |
| "eval_steps_per_second": 2.637, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 3.318418264389038, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.4575, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.6455696202531646, | |
| "eval_loss": 0.8099328875541687, | |
| "eval_runtime": 1.1512, | |
| "eval_samples_per_second": 68.622, | |
| "eval_steps_per_second": 2.606, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.47672876715660095, | |
| "eval_runtime": 1.1261, | |
| "eval_samples_per_second": 70.154, | |
| "eval_steps_per_second": 2.664, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.857142857142858, | |
| "eval_accuracy": 0.6835443037974683, | |
| "eval_loss": 0.6058729290962219, | |
| "eval_runtime": 1.1442, | |
| "eval_samples_per_second": 69.041, | |
| "eval_steps_per_second": 2.622, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 4.997932434082031, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3798, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.4863268733024597, | |
| "eval_runtime": 1.1309, | |
| "eval_samples_per_second": 69.854, | |
| "eval_steps_per_second": 2.653, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.5635764598846436, | |
| "eval_runtime": 1.1455, | |
| "eval_samples_per_second": 68.967, | |
| "eval_steps_per_second": 2.619, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.959798574447632, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.3419, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.4677181839942932, | |
| "eval_runtime": 1.1321, | |
| "eval_samples_per_second": 69.782, | |
| "eval_steps_per_second": 2.65, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.857142857142858, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.4883308410644531, | |
| "eval_runtime": 1.1375, | |
| "eval_samples_per_second": 69.453, | |
| "eval_steps_per_second": 2.637, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.5549487471580505, | |
| "eval_runtime": 1.1347, | |
| "eval_samples_per_second": 69.623, | |
| "eval_steps_per_second": 2.644, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 6.16960334777832, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.3079, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.4324057400226593, | |
| "eval_runtime": 1.1312, | |
| "eval_samples_per_second": 69.84, | |
| "eval_steps_per_second": 2.652, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6708860759493671, | |
| "eval_loss": 0.6183649897575378, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 67.446, | |
| "eval_steps_per_second": 2.561, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.857142857142858, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.6148533821105957, | |
| "eval_runtime": 1.147, | |
| "eval_samples_per_second": 68.878, | |
| "eval_steps_per_second": 2.616, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 3.098267078399658, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2616, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.44876629114151, | |
| "eval_runtime": 1.1303, | |
| "eval_samples_per_second": 69.895, | |
| "eval_steps_per_second": 2.654, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.43682238459587097, | |
| "eval_runtime": 1.1557, | |
| "eval_samples_per_second": 68.354, | |
| "eval_steps_per_second": 2.596, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.45659732818603516, | |
| "eval_runtime": 1.1429, | |
| "eval_samples_per_second": 69.124, | |
| "eval_steps_per_second": 2.625, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 4.5599517822265625, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2157, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.4657270312309265, | |
| "eval_runtime": 1.1397, | |
| "eval_samples_per_second": 69.319, | |
| "eval_steps_per_second": 2.632, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.4514474868774414, | |
| "eval_runtime": 1.1483, | |
| "eval_samples_per_second": 68.795, | |
| "eval_steps_per_second": 2.612, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5082857608795166, | |
| "eval_runtime": 1.1218, | |
| "eval_samples_per_second": 70.425, | |
| "eval_steps_per_second": 2.674, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.428571428571427, | |
| "grad_norm": 2.26061749458313, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.2258, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5260893106460571, | |
| "eval_runtime": 1.1489, | |
| "eval_samples_per_second": 68.759, | |
| "eval_steps_per_second": 2.611, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.556735634803772, | |
| "eval_runtime": 1.1778, | |
| "eval_samples_per_second": 67.076, | |
| "eval_steps_per_second": 2.547, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.5566364526748657, | |
| "eval_runtime": 1.1382, | |
| "eval_samples_per_second": 69.406, | |
| "eval_steps_per_second": 2.636, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 2.7529988288879395, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1972, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.5494687557220459, | |
| "eval_runtime": 1.1486, | |
| "eval_samples_per_second": 68.78, | |
| "eval_steps_per_second": 2.612, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.49917009472846985, | |
| "eval_runtime": 1.1413, | |
| "eval_samples_per_second": 69.221, | |
| "eval_steps_per_second": 2.629, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 36.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.566094696521759, | |
| "eval_runtime": 1.1393, | |
| "eval_samples_per_second": 69.341, | |
| "eval_steps_per_second": 2.633, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 37.142857142857146, | |
| "grad_norm": 3.6283090114593506, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.1709, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.7325516939163208, | |
| "eval_runtime": 1.1565, | |
| "eval_samples_per_second": 68.311, | |
| "eval_steps_per_second": 2.594, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.5634633898735046, | |
| "eval_runtime": 1.1297, | |
| "eval_samples_per_second": 69.93, | |
| "eval_steps_per_second": 2.656, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.0195701122283936, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.1537, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8130123019218445, | |
| "eval_runtime": 1.1502, | |
| "eval_samples_per_second": 68.685, | |
| "eval_steps_per_second": 2.608, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.857142857142854, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6983793377876282, | |
| "eval_runtime": 1.1474, | |
| "eval_samples_per_second": 68.849, | |
| "eval_steps_per_second": 2.615, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7777484655380249, | |
| "eval_runtime": 1.1447, | |
| "eval_samples_per_second": 69.011, | |
| "eval_steps_per_second": 2.621, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "grad_norm": 3.92864727973938, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1687, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.645160436630249, | |
| "eval_runtime": 1.1266, | |
| "eval_samples_per_second": 70.123, | |
| "eval_steps_per_second": 2.663, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.8527082800865173, | |
| "eval_runtime": 1.162, | |
| "eval_samples_per_second": 67.988, | |
| "eval_steps_per_second": 2.582, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 44.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6483196020126343, | |
| "eval_runtime": 1.1395, | |
| "eval_samples_per_second": 69.329, | |
| "eval_steps_per_second": 2.633, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 2.570295810699463, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1588, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.8185186982154846, | |
| "eval_runtime": 1.1438, | |
| "eval_samples_per_second": 69.068, | |
| "eval_steps_per_second": 2.623, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6821473836898804, | |
| "eval_runtime": 1.174, | |
| "eval_samples_per_second": 67.29, | |
| "eval_steps_per_second": 2.555, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.7593724131584167, | |
| "eval_runtime": 1.1514, | |
| "eval_samples_per_second": 68.611, | |
| "eval_steps_per_second": 2.605, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "grad_norm": 2.7910444736480713, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.144, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 48.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 1.0232322216033936, | |
| "eval_runtime": 1.1447, | |
| "eval_samples_per_second": 69.012, | |
| "eval_steps_per_second": 2.621, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6178007125854492, | |
| "eval_runtime": 1.1535, | |
| "eval_samples_per_second": 68.485, | |
| "eval_steps_per_second": 2.601, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.6243430376052856, | |
| "eval_runtime": 1.149, | |
| "eval_samples_per_second": 68.756, | |
| "eval_steps_per_second": 2.611, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 2.441843271255493, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1449, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.8159094452857971, | |
| "eval_runtime": 1.1388, | |
| "eval_samples_per_second": 69.372, | |
| "eval_steps_per_second": 2.634, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 52.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6664446592330933, | |
| "eval_runtime": 1.1472, | |
| "eval_samples_per_second": 68.865, | |
| "eval_steps_per_second": 2.615, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.7069984078407288, | |
| "eval_runtime": 1.1371, | |
| "eval_samples_per_second": 69.475, | |
| "eval_steps_per_second": 2.638, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 54.285714285714285, | |
| "grad_norm": 1.9345264434814453, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.144, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.7360506057739258, | |
| "eval_runtime": 1.1264, | |
| "eval_samples_per_second": 70.132, | |
| "eval_steps_per_second": 2.663, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.6655969023704529, | |
| "eval_runtime": 1.1372, | |
| "eval_samples_per_second": 69.471, | |
| "eval_steps_per_second": 2.638, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 56.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.7487228512763977, | |
| "eval_runtime": 1.158, | |
| "eval_samples_per_second": 68.222, | |
| "eval_steps_per_second": 2.591, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 2.869379758834839, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1199, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.7992926239967346, | |
| "eval_runtime": 1.1371, | |
| "eval_samples_per_second": 69.472, | |
| "eval_steps_per_second": 2.638, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 58.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7426438927650452, | |
| "eval_runtime": 1.1515, | |
| "eval_samples_per_second": 68.608, | |
| "eval_steps_per_second": 2.605, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.277022123336792, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1258, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7530637979507446, | |
| "eval_runtime": 1.1459, | |
| "eval_samples_per_second": 68.943, | |
| "eval_steps_per_second": 2.618, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.857142857142854, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7388187646865845, | |
| "eval_runtime": 1.1538, | |
| "eval_samples_per_second": 68.468, | |
| "eval_steps_per_second": 2.6, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7394750714302063, | |
| "eval_runtime": 1.1338, | |
| "eval_samples_per_second": 69.677, | |
| "eval_steps_per_second": 2.646, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "grad_norm": 2.4016706943511963, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.1392, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8237771391868591, | |
| "eval_runtime": 1.1382, | |
| "eval_samples_per_second": 69.41, | |
| "eval_steps_per_second": 2.636, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.9302475452423096, | |
| "eval_runtime": 1.1601, | |
| "eval_samples_per_second": 68.098, | |
| "eval_steps_per_second": 2.586, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 64.85714285714286, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7539066076278687, | |
| "eval_runtime": 1.142, | |
| "eval_samples_per_second": 69.178, | |
| "eval_steps_per_second": 2.627, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 65.71428571428571, | |
| "grad_norm": 2.6353225708007812, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1303, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6738837361335754, | |
| "eval_runtime": 1.127, | |
| "eval_samples_per_second": 70.095, | |
| "eval_steps_per_second": 2.662, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 66.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.662686824798584, | |
| "eval_runtime": 1.1663, | |
| "eval_samples_per_second": 67.734, | |
| "eval_steps_per_second": 2.572, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.640326201915741, | |
| "eval_runtime": 1.133, | |
| "eval_samples_per_second": 69.724, | |
| "eval_steps_per_second": 2.648, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 68.57142857142857, | |
| "grad_norm": 3.1281940937042236, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1423, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 68.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6378774642944336, | |
| "eval_runtime": 1.1326, | |
| "eval_samples_per_second": 69.753, | |
| "eval_steps_per_second": 2.649, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7658242583274841, | |
| "eval_runtime": 1.1438, | |
| "eval_samples_per_second": 69.068, | |
| "eval_steps_per_second": 2.623, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 70.85714285714286, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.9195311069488525, | |
| "eval_runtime": 1.1256, | |
| "eval_samples_per_second": 70.185, | |
| "eval_steps_per_second": 2.665, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 71.42857142857143, | |
| "grad_norm": 2.4990944862365723, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1019, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7286852598190308, | |
| "eval_runtime": 1.1291, | |
| "eval_samples_per_second": 69.969, | |
| "eval_steps_per_second": 2.657, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 72.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6548206210136414, | |
| "eval_runtime": 1.127, | |
| "eval_samples_per_second": 70.098, | |
| "eval_steps_per_second": 2.662, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.653405487537384, | |
| "eval_runtime": 1.1305, | |
| "eval_samples_per_second": 69.88, | |
| "eval_steps_per_second": 2.654, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 74.28571428571429, | |
| "grad_norm": 2.0475404262542725, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1286, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 74.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7330575585365295, | |
| "eval_runtime": 1.1366, | |
| "eval_samples_per_second": 69.508, | |
| "eval_steps_per_second": 2.64, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7845055460929871, | |
| "eval_runtime": 1.1356, | |
| "eval_samples_per_second": 69.566, | |
| "eval_steps_per_second": 2.642, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 76.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7187694311141968, | |
| "eval_runtime": 1.1547, | |
| "eval_samples_per_second": 68.416, | |
| "eval_steps_per_second": 2.598, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 77.14285714285714, | |
| "grad_norm": 3.98937726020813, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1054, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6595200896263123, | |
| "eval_runtime": 1.1479, | |
| "eval_samples_per_second": 68.823, | |
| "eval_steps_per_second": 2.614, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 78.85714285714286, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.662256121635437, | |
| "eval_runtime": 1.1304, | |
| "eval_samples_per_second": 69.885, | |
| "eval_steps_per_second": 2.654, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.8091683387756348, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.1053, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7336876392364502, | |
| "eval_runtime": 1.1418, | |
| "eval_samples_per_second": 69.186, | |
| "eval_steps_per_second": 2.627, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.85714285714286, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8085371851921082, | |
| "eval_runtime": 1.1388, | |
| "eval_samples_per_second": 69.372, | |
| "eval_steps_per_second": 2.634, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8200588226318359, | |
| "eval_runtime": 1.1478, | |
| "eval_samples_per_second": 68.825, | |
| "eval_steps_per_second": 2.614, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "grad_norm": 3.0363099575042725, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.1086, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.7947056293487549, | |
| "eval_runtime": 1.1218, | |
| "eval_samples_per_second": 70.419, | |
| "eval_steps_per_second": 2.674, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7668984532356262, | |
| "eval_runtime": 1.1428, | |
| "eval_samples_per_second": 69.129, | |
| "eval_steps_per_second": 2.625, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 84.85714285714286, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7581708431243896, | |
| "eval_runtime": 1.1285, | |
| "eval_samples_per_second": 70.006, | |
| "eval_steps_per_second": 2.658, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "grad_norm": 2.2897379398345947, | |
| "learning_rate": 0.0, | |
| "loss": 0.1186, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7541030645370483, | |
| "eval_runtime": 1.1594, | |
| "eval_samples_per_second": 68.138, | |
| "eval_steps_per_second": 2.588, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "step": 300, | |
| "total_flos": 2.9362240500074496e+18, | |
| "train_loss": 0.23499208807945252, | |
| "train_runtime": 1665.1689, | |
| "train_samples_per_second": 26.544, | |
| "train_steps_per_second": 0.18 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.645160436630249, | |
| "eval_runtime": 1.1931, | |
| "eval_samples_per_second": 66.215, | |
| "eval_steps_per_second": 2.514, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "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": 2.9362240500074496e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |