| { | |
| "best_metric": 0.9545454545454546, | |
| "best_model_checkpoint": "beit-base-patch16-224-85-fold4/checkpoint-104", | |
| "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": 0.6818181818181818, | |
| "eval_loss": 0.6873949766159058, | |
| "eval_runtime": 0.7006, | |
| "eval_samples_per_second": 62.799, | |
| "eval_steps_per_second": 2.855, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.70831698179245, | |
| "eval_runtime": 0.7165, | |
| "eval_samples_per_second": 61.411, | |
| "eval_steps_per_second": 2.791, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.8870734572410583, | |
| "eval_runtime": 0.6999, | |
| "eval_samples_per_second": 62.865, | |
| "eval_steps_per_second": 2.857, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7305467128753662, | |
| "eval_runtime": 0.6828, | |
| "eval_samples_per_second": 64.437, | |
| "eval_steps_per_second": 2.929, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.2408814430236816, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6246, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5790721774101257, | |
| "eval_runtime": 0.6488, | |
| "eval_samples_per_second": 67.815, | |
| "eval_steps_per_second": 3.082, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5888476371765137, | |
| "eval_runtime": 0.6667, | |
| "eval_samples_per_second": 66.001, | |
| "eval_steps_per_second": 3.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6050147414207458, | |
| "eval_runtime": 0.669, | |
| "eval_samples_per_second": 65.772, | |
| "eval_steps_per_second": 2.99, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5468326210975647, | |
| "eval_runtime": 0.6807, | |
| "eval_samples_per_second": 64.638, | |
| "eval_steps_per_second": 2.938, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5350865721702576, | |
| "eval_runtime": 0.6994, | |
| "eval_samples_per_second": 62.912, | |
| "eval_steps_per_second": 2.86, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 8.467140197753906, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4453, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4155413508415222, | |
| "eval_runtime": 0.6802, | |
| "eval_samples_per_second": 64.689, | |
| "eval_steps_per_second": 2.94, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.826596200466156, | |
| "eval_runtime": 0.6937, | |
| "eval_samples_per_second": 63.431, | |
| "eval_steps_per_second": 2.883, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.39046552777290344, | |
| "eval_runtime": 0.6717, | |
| "eval_samples_per_second": 65.502, | |
| "eval_steps_per_second": 2.977, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.39417600631713867, | |
| "eval_runtime": 0.6938, | |
| "eval_samples_per_second": 63.421, | |
| "eval_steps_per_second": 2.883, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.40153181552886963, | |
| "eval_runtime": 0.6823, | |
| "eval_samples_per_second": 64.483, | |
| "eval_steps_per_second": 2.931, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 8.223971366882324, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3613, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3473949730396271, | |
| "eval_runtime": 0.7008, | |
| "eval_samples_per_second": 62.787, | |
| "eval_steps_per_second": 2.854, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.47630393505096436, | |
| "eval_runtime": 0.6994, | |
| "eval_samples_per_second": 62.908, | |
| "eval_steps_per_second": 2.859, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.3894023001194, | |
| "eval_runtime": 0.6972, | |
| "eval_samples_per_second": 63.113, | |
| "eval_steps_per_second": 2.869, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.42897891998291016, | |
| "eval_runtime": 0.6873, | |
| "eval_samples_per_second": 64.019, | |
| "eval_steps_per_second": 2.91, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3524658679962158, | |
| "eval_runtime": 0.6741, | |
| "eval_samples_per_second": 65.274, | |
| "eval_steps_per_second": 2.967, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.433536052703857, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2928, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3425785303115845, | |
| "eval_runtime": 0.6911, | |
| "eval_samples_per_second": 63.664, | |
| "eval_steps_per_second": 2.894, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.4060179591178894, | |
| "eval_runtime": 0.6862, | |
| "eval_samples_per_second": 64.121, | |
| "eval_steps_per_second": 2.915, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6961742639541626, | |
| "eval_runtime": 0.6761, | |
| "eval_samples_per_second": 65.079, | |
| "eval_steps_per_second": 2.958, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.351392537355423, | |
| "eval_runtime": 0.6837, | |
| "eval_samples_per_second": 64.357, | |
| "eval_steps_per_second": 2.925, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.5302432775497437, | |
| "eval_runtime": 0.679, | |
| "eval_samples_per_second": 64.799, | |
| "eval_steps_per_second": 2.945, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 4.721464157104492, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2256, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.30937260389328003, | |
| "eval_runtime": 0.6811, | |
| "eval_samples_per_second": 64.598, | |
| "eval_steps_per_second": 2.936, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.29768574237823486, | |
| "eval_runtime": 0.6989, | |
| "eval_samples_per_second": 62.954, | |
| "eval_steps_per_second": 2.862, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.48831918835639954, | |
| "eval_runtime": 0.6775, | |
| "eval_samples_per_second": 64.941, | |
| "eval_steps_per_second": 2.952, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3008198142051697, | |
| "eval_runtime": 0.6834, | |
| "eval_samples_per_second": 64.379, | |
| "eval_steps_per_second": 2.926, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.322564959526062, | |
| "eval_runtime": 0.6771, | |
| "eval_samples_per_second": 64.984, | |
| "eval_steps_per_second": 2.954, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 2.5875678062438965, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2231, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.410068541765213, | |
| "eval_runtime": 0.6799, | |
| "eval_samples_per_second": 64.715, | |
| "eval_steps_per_second": 2.942, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3196963369846344, | |
| "eval_runtime": 0.6863, | |
| "eval_samples_per_second": 64.115, | |
| "eval_steps_per_second": 2.914, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.41331416368484497, | |
| "eval_runtime": 0.6819, | |
| "eval_samples_per_second": 64.523, | |
| "eval_steps_per_second": 2.933, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.2922936677932739, | |
| "eval_runtime": 0.6792, | |
| "eval_samples_per_second": 64.785, | |
| "eval_steps_per_second": 2.945, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4391103982925415, | |
| "eval_runtime": 0.6821, | |
| "eval_samples_per_second": 64.505, | |
| "eval_steps_per_second": 2.932, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 4.117176532745361, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1756, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.30160456895828247, | |
| "eval_runtime": 0.6774, | |
| "eval_samples_per_second": 64.955, | |
| "eval_steps_per_second": 2.953, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2748742997646332, | |
| "eval_runtime": 0.6783, | |
| "eval_samples_per_second": 64.869, | |
| "eval_steps_per_second": 2.949, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3146228492259979, | |
| "eval_runtime": 0.6801, | |
| "eval_samples_per_second": 64.699, | |
| "eval_steps_per_second": 2.941, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.30949413776397705, | |
| "eval_runtime": 0.6754, | |
| "eval_samples_per_second": 65.149, | |
| "eval_steps_per_second": 2.961, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3017137348651886, | |
| "eval_runtime": 0.6787, | |
| "eval_samples_per_second": 64.828, | |
| "eval_steps_per_second": 2.947, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 3.9740469455718994, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1592, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.27615535259246826, | |
| "eval_runtime": 0.6818, | |
| "eval_samples_per_second": 64.537, | |
| "eval_steps_per_second": 2.933, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.40538015961647034, | |
| "eval_runtime": 0.686, | |
| "eval_samples_per_second": 64.137, | |
| "eval_steps_per_second": 2.915, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.27872493863105774, | |
| "eval_runtime": 0.6824, | |
| "eval_samples_per_second": 64.481, | |
| "eval_steps_per_second": 2.931, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.319337397813797, | |
| "eval_runtime": 0.6835, | |
| "eval_samples_per_second": 64.378, | |
| "eval_steps_per_second": 2.926, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2782891094684601, | |
| "eval_runtime": 0.6885, | |
| "eval_samples_per_second": 63.91, | |
| "eval_steps_per_second": 2.905, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 2.1339399814605713, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1857, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2934027314186096, | |
| "eval_runtime": 0.6817, | |
| "eval_samples_per_second": 64.541, | |
| "eval_steps_per_second": 2.934, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3579067289829254, | |
| "eval_runtime": 0.6836, | |
| "eval_samples_per_second": 64.364, | |
| "eval_steps_per_second": 2.926, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.35006046295166016, | |
| "eval_runtime": 0.7004, | |
| "eval_samples_per_second": 62.822, | |
| "eval_steps_per_second": 2.856, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3357532322406769, | |
| "eval_runtime": 0.6991, | |
| "eval_samples_per_second": 62.937, | |
| "eval_steps_per_second": 2.861, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.29814115166664124, | |
| "eval_runtime": 0.6825, | |
| "eval_samples_per_second": 64.469, | |
| "eval_steps_per_second": 2.93, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 2.850813388824463, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1179, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.336421936750412, | |
| "eval_runtime": 0.6808, | |
| "eval_samples_per_second": 64.626, | |
| "eval_steps_per_second": 2.938, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.33238616585731506, | |
| "eval_runtime": 0.6831, | |
| "eval_samples_per_second": 64.416, | |
| "eval_steps_per_second": 2.928, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.17253318428993225, | |
| "eval_runtime": 0.6779, | |
| "eval_samples_per_second": 64.905, | |
| "eval_steps_per_second": 2.95, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.12221997231245041, | |
| "eval_runtime": 0.7025, | |
| "eval_samples_per_second": 62.629, | |
| "eval_steps_per_second": 2.847, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1499694436788559, | |
| "eval_runtime": 0.6826, | |
| "eval_samples_per_second": 64.463, | |
| "eval_steps_per_second": 2.93, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 4.910292625427246, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1448, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.23580028116703033, | |
| "eval_runtime": 0.6775, | |
| "eval_samples_per_second": 64.945, | |
| "eval_steps_per_second": 2.952, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2223827838897705, | |
| "eval_runtime": 0.6842, | |
| "eval_samples_per_second": 64.304, | |
| "eval_steps_per_second": 2.923, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.14569629728794098, | |
| "eval_runtime": 0.6898, | |
| "eval_samples_per_second": 63.788, | |
| "eval_steps_per_second": 2.899, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.17447058856487274, | |
| "eval_runtime": 0.6812, | |
| "eval_samples_per_second": 64.595, | |
| "eval_steps_per_second": 2.936, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1989983469247818, | |
| "eval_runtime": 0.6868, | |
| "eval_samples_per_second": 64.061, | |
| "eval_steps_per_second": 2.912, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 4.1764235496521, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1343, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2904643714427948, | |
| "eval_runtime": 0.6807, | |
| "eval_samples_per_second": 64.644, | |
| "eval_steps_per_second": 2.938, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.38422849774360657, | |
| "eval_runtime": 0.6813, | |
| "eval_samples_per_second": 64.582, | |
| "eval_steps_per_second": 2.936, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3031158149242401, | |
| "eval_runtime": 0.679, | |
| "eval_samples_per_second": 64.804, | |
| "eval_steps_per_second": 2.946, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.26415106654167175, | |
| "eval_runtime": 0.6816, | |
| "eval_samples_per_second": 64.552, | |
| "eval_steps_per_second": 2.934, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2411704957485199, | |
| "eval_runtime": 0.6859, | |
| "eval_samples_per_second": 64.152, | |
| "eval_steps_per_second": 2.916, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 2.9099998474121094, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1109, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.33470064401626587, | |
| "eval_runtime": 0.6891, | |
| "eval_samples_per_second": 63.849, | |
| "eval_steps_per_second": 2.902, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.40046489238739014, | |
| "eval_runtime": 0.7072, | |
| "eval_samples_per_second": 62.215, | |
| "eval_steps_per_second": 2.828, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2905416190624237, | |
| "eval_runtime": 0.6887, | |
| "eval_samples_per_second": 63.888, | |
| "eval_steps_per_second": 2.904, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.316750705242157, | |
| "eval_runtime": 0.6921, | |
| "eval_samples_per_second": 63.573, | |
| "eval_steps_per_second": 2.89, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3844659924507141, | |
| "eval_runtime": 0.7005, | |
| "eval_samples_per_second": 62.815, | |
| "eval_steps_per_second": 2.855, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 4.845411777496338, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1221, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.31782636046409607, | |
| "eval_runtime": 0.6803, | |
| "eval_samples_per_second": 64.673, | |
| "eval_steps_per_second": 2.94, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.26903513073921204, | |
| "eval_runtime": 0.6907, | |
| "eval_samples_per_second": 63.704, | |
| "eval_steps_per_second": 2.896, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.25161102414131165, | |
| "eval_runtime": 0.694, | |
| "eval_samples_per_second": 63.404, | |
| "eval_steps_per_second": 2.882, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.23466776311397552, | |
| "eval_runtime": 0.6925, | |
| "eval_samples_per_second": 63.536, | |
| "eval_steps_per_second": 2.888, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.237599715590477, | |
| "eval_runtime": 0.6802, | |
| "eval_samples_per_second": 64.691, | |
| "eval_steps_per_second": 2.941, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 4.328647613525391, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1191, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.24804998934268951, | |
| "eval_runtime": 0.6995, | |
| "eval_samples_per_second": 62.899, | |
| "eval_steps_per_second": 2.859, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2596595585346222, | |
| "eval_runtime": 0.6888, | |
| "eval_samples_per_second": 63.882, | |
| "eval_steps_per_second": 2.904, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3071075677871704, | |
| "eval_runtime": 0.6787, | |
| "eval_samples_per_second": 64.831, | |
| "eval_steps_per_second": 2.947, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3354208767414093, | |
| "eval_runtime": 0.6923, | |
| "eval_samples_per_second": 63.554, | |
| "eval_steps_per_second": 2.889, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.29882094264030457, | |
| "eval_runtime": 0.673, | |
| "eval_samples_per_second": 65.378, | |
| "eval_steps_per_second": 2.972, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.718787431716919, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1133, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2760173976421356, | |
| "eval_runtime": 0.6737, | |
| "eval_samples_per_second": 65.315, | |
| "eval_steps_per_second": 2.969, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.28321373462677, | |
| "eval_runtime": 0.682, | |
| "eval_samples_per_second": 64.52, | |
| "eval_steps_per_second": 2.933, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.27927514910697937, | |
| "eval_runtime": 0.6754, | |
| "eval_samples_per_second": 65.151, | |
| "eval_steps_per_second": 2.961, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2779335677623749, | |
| "eval_runtime": 0.7063, | |
| "eval_samples_per_second": 62.299, | |
| "eval_steps_per_second": 2.832, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3004065155982971, | |
| "eval_runtime": 0.6888, | |
| "eval_samples_per_second": 63.883, | |
| "eval_steps_per_second": 2.904, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 3.5355427265167236, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.098, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.327462762594223, | |
| "eval_runtime": 0.6835, | |
| "eval_samples_per_second": 64.372, | |
| "eval_steps_per_second": 2.926, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.33943310379981995, | |
| "eval_runtime": 0.6764, | |
| "eval_samples_per_second": 65.054, | |
| "eval_steps_per_second": 2.957, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.325663298368454, | |
| "eval_runtime": 0.6771, | |
| "eval_samples_per_second": 64.983, | |
| "eval_steps_per_second": 2.954, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.31717270612716675, | |
| "eval_runtime": 0.6797, | |
| "eval_samples_per_second": 64.734, | |
| "eval_steps_per_second": 2.942, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.3121934235095978, | |
| "eval_runtime": 0.6739, | |
| "eval_samples_per_second": 65.289, | |
| "eval_steps_per_second": 2.968, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 2.074833631515503, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0917, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.32076776027679443, | |
| "eval_runtime": 0.674, | |
| "eval_samples_per_second": 65.285, | |
| "eval_steps_per_second": 2.968, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.3235606849193573, | |
| "eval_runtime": 0.6852, | |
| "eval_samples_per_second": 64.21, | |
| "eval_steps_per_second": 2.919, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.32737717032432556, | |
| "eval_runtime": 0.6836, | |
| "eval_samples_per_second": 64.363, | |
| "eval_steps_per_second": 2.926, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3330632448196411, | |
| "eval_runtime": 0.695, | |
| "eval_samples_per_second": 63.312, | |
| "eval_steps_per_second": 2.878, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.33794966340065, | |
| "eval_runtime": 0.6851, | |
| "eval_samples_per_second": 64.229, | |
| "eval_steps_per_second": 2.919, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 3.968538999557495, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0989, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.34042808413505554, | |
| "eval_runtime": 0.6761, | |
| "eval_samples_per_second": 65.077, | |
| "eval_steps_per_second": 2.958, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.345187783241272, | |
| "eval_runtime": 0.6775, | |
| "eval_samples_per_second": 64.943, | |
| "eval_steps_per_second": 2.952, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.34910109639167786, | |
| "eval_runtime": 0.6776, | |
| "eval_samples_per_second": 64.935, | |
| "eval_steps_per_second": 2.952, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.34889742732048035, | |
| "eval_runtime": 0.6832, | |
| "eval_samples_per_second": 64.402, | |
| "eval_steps_per_second": 2.927, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.34820887446403503, | |
| "eval_runtime": 0.6865, | |
| "eval_samples_per_second": 64.09, | |
| "eval_steps_per_second": 2.913, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 3.1474955081939697, | |
| "learning_rate": 0.0, | |
| "loss": 0.0916, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.34667694568634033, | |
| "eval_runtime": 0.6814, | |
| "eval_samples_per_second": 64.577, | |
| "eval_steps_per_second": 2.935, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.8822505078960128e+18, | |
| "train_loss": 0.1967863607406616, | |
| "train_runtime": 1219.0063, | |
| "train_samples_per_second": 19.934, | |
| "train_steps_per_second": 0.164 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.17253318428993225, | |
| "eval_runtime": 0.6994, | |
| "eval_samples_per_second": 62.914, | |
| "eval_steps_per_second": 2.86, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 1.8822505078960128e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |