| { | |
| "best_metric": 0.8354430379746836, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-55-fold1/checkpoint-101", | |
| "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": 0.43037974683544306, | |
| "eval_loss": 0.7879665493965149, | |
| "eval_runtime": 1.0027, | |
| "eval_samples_per_second": 78.785, | |
| "eval_steps_per_second": 2.992, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5443037974683544, | |
| "eval_loss": 0.833726704120636, | |
| "eval_runtime": 0.9971, | |
| "eval_samples_per_second": 79.229, | |
| "eval_steps_per_second": 3.009, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 4.414129734039307, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.776, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.5443037974683544, | |
| "eval_loss": 0.704697847366333, | |
| "eval_runtime": 1.0453, | |
| "eval_samples_per_second": 75.575, | |
| "eval_steps_per_second": 2.87, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.6729037761688232, | |
| "eval_runtime": 1.0381, | |
| "eval_samples_per_second": 76.102, | |
| "eval_steps_per_second": 2.89, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_accuracy": 0.569620253164557, | |
| "eval_loss": 0.6784361600875854, | |
| "eval_runtime": 1.0714, | |
| "eval_samples_per_second": 73.736, | |
| "eval_steps_per_second": 2.8, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 1.1414231061935425, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6852, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6329113924050633, | |
| "eval_loss": 0.6250560879707336, | |
| "eval_runtime": 1.0536, | |
| "eval_samples_per_second": 74.978, | |
| "eval_steps_per_second": 2.847, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.6329113924050633, | |
| "eval_loss": 0.624561607837677, | |
| "eval_runtime": 1.0579, | |
| "eval_samples_per_second": 74.676, | |
| "eval_steps_per_second": 2.836, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6835443037974683, | |
| "eval_loss": 0.5865054726600647, | |
| "eval_runtime": 1.0641, | |
| "eval_samples_per_second": 74.244, | |
| "eval_steps_per_second": 2.819, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 4.807124614715576, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6223, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.857142857142858, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.5573264956474304, | |
| "eval_runtime": 1.0677, | |
| "eval_samples_per_second": 73.988, | |
| "eval_steps_per_second": 2.81, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.5841125249862671, | |
| "eval_runtime": 1.0817, | |
| "eval_samples_per_second": 73.035, | |
| "eval_steps_per_second": 2.773, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.5790536403656006, | |
| "eval_runtime": 1.067, | |
| "eval_samples_per_second": 74.04, | |
| "eval_steps_per_second": 2.812, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 2.0765061378479004, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.5573, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.5282906889915466, | |
| "eval_runtime": 1.1116, | |
| "eval_samples_per_second": 71.069, | |
| "eval_steps_per_second": 2.699, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.5125677585601807, | |
| "eval_runtime": 1.1021, | |
| "eval_samples_per_second": 71.681, | |
| "eval_steps_per_second": 2.722, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.5642557740211487, | |
| "eval_runtime": 1.1087, | |
| "eval_samples_per_second": 71.253, | |
| "eval_steps_per_second": 2.706, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 5.382793426513672, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.4772, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.6582278481012658, | |
| "eval_loss": 0.6735838055610657, | |
| "eval_runtime": 1.1489, | |
| "eval_samples_per_second": 68.759, | |
| "eval_steps_per_second": 2.611, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.570650041103363, | |
| "eval_runtime": 1.1118, | |
| "eval_samples_per_second": 71.054, | |
| "eval_steps_per_second": 2.698, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.857142857142858, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.5198583602905273, | |
| "eval_runtime": 1.1171, | |
| "eval_samples_per_second": 70.72, | |
| "eval_steps_per_second": 2.686, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 5.891899108886719, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.4656, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.5285176634788513, | |
| "eval_runtime": 1.1106, | |
| "eval_samples_per_second": 71.133, | |
| "eval_steps_per_second": 2.701, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.4534749984741211, | |
| "eval_runtime": 1.1235, | |
| "eval_samples_per_second": 70.319, | |
| "eval_steps_per_second": 2.67, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.8716275691986084, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.4147, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.45567476749420166, | |
| "eval_runtime": 1.1234, | |
| "eval_samples_per_second": 70.324, | |
| "eval_steps_per_second": 2.671, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.857142857142858, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.44825637340545654, | |
| "eval_runtime": 1.1316, | |
| "eval_samples_per_second": 69.813, | |
| "eval_steps_per_second": 2.651, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.5025177001953125, | |
| "eval_runtime": 1.1367, | |
| "eval_samples_per_second": 69.497, | |
| "eval_steps_per_second": 2.639, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 3.101086378097534, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.342, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.4716118574142456, | |
| "eval_runtime": 1.176, | |
| "eval_samples_per_second": 67.18, | |
| "eval_steps_per_second": 2.551, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.5310951471328735, | |
| "eval_runtime": 1.1575, | |
| "eval_samples_per_second": 68.251, | |
| "eval_steps_per_second": 2.592, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.45601552724838257, | |
| "eval_runtime": 1.1329, | |
| "eval_samples_per_second": 69.732, | |
| "eval_steps_per_second": 2.648, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 5.771670818328857, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2993, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.511889636516571, | |
| "eval_runtime": 1.1386, | |
| "eval_samples_per_second": 69.386, | |
| "eval_steps_per_second": 2.635, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.532077968120575, | |
| "eval_runtime": 1.1122, | |
| "eval_samples_per_second": 71.03, | |
| "eval_steps_per_second": 2.697, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.49367329478263855, | |
| "eval_runtime": 1.1193, | |
| "eval_samples_per_second": 70.578, | |
| "eval_steps_per_second": 2.68, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 3.30381441116333, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2506, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.857142857142858, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.4563353955745697, | |
| "eval_runtime": 1.1319, | |
| "eval_samples_per_second": 69.793, | |
| "eval_steps_per_second": 2.65, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.5234401822090149, | |
| "eval_runtime": 1.1604, | |
| "eval_samples_per_second": 68.081, | |
| "eval_steps_per_second": 2.585, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5359364151954651, | |
| "eval_runtime": 1.1233, | |
| "eval_samples_per_second": 70.328, | |
| "eval_steps_per_second": 2.671, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.428571428571427, | |
| "grad_norm": 2.6743521690368652, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.2201, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.5144768357276917, | |
| "eval_runtime": 1.1392, | |
| "eval_samples_per_second": 69.347, | |
| "eval_steps_per_second": 2.633, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.857142857142854, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.534345269203186, | |
| "eval_runtime": 1.1264, | |
| "eval_samples_per_second": 70.134, | |
| "eval_steps_per_second": 2.663, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.468890905380249, | |
| "eval_runtime": 1.1486, | |
| "eval_samples_per_second": 68.782, | |
| "eval_steps_per_second": 2.612, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 3.1685876846313477, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2098, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6464852690696716, | |
| "eval_runtime": 1.1773, | |
| "eval_samples_per_second": 67.104, | |
| "eval_steps_per_second": 2.548, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5002541542053223, | |
| "eval_runtime": 1.1346, | |
| "eval_samples_per_second": 69.626, | |
| "eval_steps_per_second": 2.644, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 36.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.611255943775177, | |
| "eval_runtime": 1.139, | |
| "eval_samples_per_second": 69.361, | |
| "eval_steps_per_second": 2.634, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 37.142857142857146, | |
| "grad_norm": 5.094492435455322, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.1808, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.8215884566307068, | |
| "eval_runtime": 1.1601, | |
| "eval_samples_per_second": 68.096, | |
| "eval_steps_per_second": 2.586, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.5603389739990234, | |
| "eval_runtime": 1.1334, | |
| "eval_samples_per_second": 69.699, | |
| "eval_steps_per_second": 2.647, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 3.0100135803222656, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.1892, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.613608181476593, | |
| "eval_runtime": 1.1451, | |
| "eval_samples_per_second": 68.989, | |
| "eval_steps_per_second": 2.62, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6074010133743286, | |
| "eval_runtime": 1.181, | |
| "eval_samples_per_second": 66.893, | |
| "eval_steps_per_second": 2.54, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6503015756607056, | |
| "eval_runtime": 1.1394, | |
| "eval_samples_per_second": 69.334, | |
| "eval_steps_per_second": 2.633, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "grad_norm": 2.3642420768737793, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.154, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7923038601875305, | |
| "eval_runtime": 1.1209, | |
| "eval_samples_per_second": 70.482, | |
| "eval_steps_per_second": 2.677, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7790768146514893, | |
| "eval_runtime": 1.1483, | |
| "eval_samples_per_second": 68.795, | |
| "eval_steps_per_second": 2.612, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 44.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7947573661804199, | |
| "eval_runtime": 1.1487, | |
| "eval_samples_per_second": 68.772, | |
| "eval_steps_per_second": 2.612, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 3.1455414295196533, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1613, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.727028489112854, | |
| "eval_runtime": 1.1467, | |
| "eval_samples_per_second": 68.895, | |
| "eval_steps_per_second": 2.616, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7283490300178528, | |
| "eval_runtime": 1.1446, | |
| "eval_samples_per_second": 69.021, | |
| "eval_steps_per_second": 2.621, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7057384848594666, | |
| "eval_runtime": 1.1324, | |
| "eval_samples_per_second": 69.764, | |
| "eval_steps_per_second": 2.649, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "grad_norm": 3.1506268978118896, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.141, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 48.857142857142854, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6691973209381104, | |
| "eval_runtime": 1.1153, | |
| "eval_samples_per_second": 70.834, | |
| "eval_steps_per_second": 2.69, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6390013098716736, | |
| "eval_runtime": 1.1382, | |
| "eval_samples_per_second": 69.408, | |
| "eval_steps_per_second": 2.636, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6543086171150208, | |
| "eval_runtime": 1.1235, | |
| "eval_samples_per_second": 70.319, | |
| "eval_steps_per_second": 2.67, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 2.4540371894836426, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1434, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.773634135723114, | |
| "eval_runtime": 1.1471, | |
| "eval_samples_per_second": 68.869, | |
| "eval_steps_per_second": 2.615, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 52.857142857142854, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6425884366035461, | |
| "eval_runtime": 1.1427, | |
| "eval_samples_per_second": 69.136, | |
| "eval_steps_per_second": 2.625, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6890805959701538, | |
| "eval_runtime": 1.1358, | |
| "eval_samples_per_second": 69.556, | |
| "eval_steps_per_second": 2.641, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 54.285714285714285, | |
| "grad_norm": 3.488741636276245, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.1583, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7520738244056702, | |
| "eval_runtime": 1.1294, | |
| "eval_samples_per_second": 69.949, | |
| "eval_steps_per_second": 2.656, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6495163440704346, | |
| "eval_runtime": 1.1339, | |
| "eval_samples_per_second": 69.67, | |
| "eval_steps_per_second": 2.646, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 56.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.704879641532898, | |
| "eval_runtime": 1.1387, | |
| "eval_samples_per_second": 69.379, | |
| "eval_steps_per_second": 2.635, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 2.114783525466919, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1418, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7533851861953735, | |
| "eval_runtime": 1.1446, | |
| "eval_samples_per_second": 69.022, | |
| "eval_steps_per_second": 2.621, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 58.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.68922358751297, | |
| "eval_runtime": 1.1411, | |
| "eval_samples_per_second": 69.23, | |
| "eval_steps_per_second": 2.629, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.576995611190796, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1488, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.752753734588623, | |
| "eval_runtime": 1.1437, | |
| "eval_samples_per_second": 69.076, | |
| "eval_steps_per_second": 2.623, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6920117735862732, | |
| "eval_runtime": 1.1413, | |
| "eval_samples_per_second": 69.217, | |
| "eval_steps_per_second": 2.629, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6767023801803589, | |
| "eval_runtime": 1.1202, | |
| "eval_samples_per_second": 70.526, | |
| "eval_steps_per_second": 2.678, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "grad_norm": 3.9229702949523926, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.1481, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7510101795196533, | |
| "eval_runtime": 1.1464, | |
| "eval_samples_per_second": 68.911, | |
| "eval_steps_per_second": 2.617, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.6074531674385071, | |
| "eval_runtime": 1.1404, | |
| "eval_samples_per_second": 69.276, | |
| "eval_steps_per_second": 2.631, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 64.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.5857882499694824, | |
| "eval_runtime": 1.1393, | |
| "eval_samples_per_second": 69.341, | |
| "eval_steps_per_second": 2.633, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 65.71428571428571, | |
| "grad_norm": 1.5416897535324097, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1014, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.666824460029602, | |
| "eval_runtime": 1.124, | |
| "eval_samples_per_second": 70.287, | |
| "eval_steps_per_second": 2.669, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 66.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6127079129219055, | |
| "eval_runtime": 1.1537, | |
| "eval_samples_per_second": 68.477, | |
| "eval_steps_per_second": 2.6, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6295149922370911, | |
| "eval_runtime": 1.1392, | |
| "eval_samples_per_second": 69.346, | |
| "eval_steps_per_second": 2.633, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 68.57142857142857, | |
| "grad_norm": 3.536512851715088, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1147, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 68.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6722728610038757, | |
| "eval_runtime": 1.1376, | |
| "eval_samples_per_second": 69.442, | |
| "eval_steps_per_second": 2.637, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7166734933853149, | |
| "eval_runtime": 1.1521, | |
| "eval_samples_per_second": 68.57, | |
| "eval_steps_per_second": 2.604, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 70.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6914483904838562, | |
| "eval_runtime": 1.1378, | |
| "eval_samples_per_second": 69.429, | |
| "eval_steps_per_second": 2.637, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 71.42857142857143, | |
| "grad_norm": 2.5358872413635254, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1289, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6676318645477295, | |
| "eval_runtime": 1.1313, | |
| "eval_samples_per_second": 69.83, | |
| "eval_steps_per_second": 2.652, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 72.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.6874268651008606, | |
| "eval_runtime": 1.1388, | |
| "eval_samples_per_second": 69.372, | |
| "eval_steps_per_second": 2.634, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.7485886216163635, | |
| "eval_runtime": 1.1881, | |
| "eval_samples_per_second": 66.491, | |
| "eval_steps_per_second": 2.525, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 74.28571428571429, | |
| "grad_norm": 2.7322018146514893, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1084, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 74.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.7192913293838501, | |
| "eval_runtime": 1.1432, | |
| "eval_samples_per_second": 69.106, | |
| "eval_steps_per_second": 2.624, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.7053883671760559, | |
| "eval_runtime": 1.1734, | |
| "eval_samples_per_second": 67.328, | |
| "eval_steps_per_second": 2.557, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 76.85714285714286, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.7052078247070312, | |
| "eval_runtime": 1.1272, | |
| "eval_samples_per_second": 70.087, | |
| "eval_steps_per_second": 2.662, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 77.14285714285714, | |
| "grad_norm": 2.65423583984375, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.11, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6885020136833191, | |
| "eval_runtime": 1.1436, | |
| "eval_samples_per_second": 69.078, | |
| "eval_steps_per_second": 2.623, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 78.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.7162737846374512, | |
| "eval_runtime": 1.1462, | |
| "eval_samples_per_second": 68.925, | |
| "eval_steps_per_second": 2.617, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 3.0879228115081787, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.1144, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6901566982269287, | |
| "eval_runtime": 1.1477, | |
| "eval_samples_per_second": 68.833, | |
| "eval_steps_per_second": 2.614, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.6886458396911621, | |
| "eval_runtime": 1.1476, | |
| "eval_samples_per_second": 68.837, | |
| "eval_steps_per_second": 2.614, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.7062325477600098, | |
| "eval_runtime": 1.1538, | |
| "eval_samples_per_second": 68.471, | |
| "eval_steps_per_second": 2.6, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "grad_norm": 2.7569634914398193, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.1026, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "eval_accuracy": 0.810126582278481, | |
| "eval_loss": 0.7195845246315002, | |
| "eval_runtime": 1.1374, | |
| "eval_samples_per_second": 69.459, | |
| "eval_steps_per_second": 2.638, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7332103252410889, | |
| "eval_runtime": 1.1551, | |
| "eval_samples_per_second": 68.392, | |
| "eval_steps_per_second": 2.597, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 84.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7225062847137451, | |
| "eval_runtime": 1.1807, | |
| "eval_samples_per_second": 66.909, | |
| "eval_steps_per_second": 2.541, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "grad_norm": 2.9502005577087402, | |
| "learning_rate": 0.0, | |
| "loss": 0.1143, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7162460684776306, | |
| "eval_runtime": 1.2487, | |
| "eval_samples_per_second": 63.268, | |
| "eval_steps_per_second": 2.403, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "step": 300, | |
| "total_flos": 2.9362240500074496e+18, | |
| "train_loss": 0.25938483238220217, | |
| "train_runtime": 1661.2834, | |
| "train_samples_per_second": 26.606, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.8354430379746836, | |
| "eval_loss": 0.4563353955745697, | |
| "eval_runtime": 1.1716, | |
| "eval_samples_per_second": 67.429, | |
| "eval_steps_per_second": 2.561, | |
| "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 | |
| } | |