| { | |
| "best_metric": 0.9090909090909091, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-85-fold5/checkpoint-90", | |
| "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": 0.3181818181818182, | |
| "eval_loss": 1.1684619188308716, | |
| "eval_runtime": 0.6188, | |
| "eval_samples_per_second": 71.111, | |
| "eval_steps_per_second": 3.232, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.45454545454545453, | |
| "eval_loss": 0.7294825911521912, | |
| "eval_runtime": 0.6083, | |
| "eval_samples_per_second": 72.336, | |
| "eval_steps_per_second": 3.288, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.664056122303009, | |
| "eval_runtime": 0.6283, | |
| "eval_samples_per_second": 70.029, | |
| "eval_steps_per_second": 3.183, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7702975869178772, | |
| "eval_runtime": 0.6109, | |
| "eval_samples_per_second": 72.03, | |
| "eval_steps_per_second": 3.274, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.6766012907028198, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7885, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6750325560569763, | |
| "eval_runtime": 0.6002, | |
| "eval_samples_per_second": 73.305, | |
| "eval_steps_per_second": 3.332, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6446254253387451, | |
| "eval_runtime": 0.5912, | |
| "eval_samples_per_second": 74.428, | |
| "eval_steps_per_second": 3.383, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6919071674346924, | |
| "eval_runtime": 0.5927, | |
| "eval_samples_per_second": 74.236, | |
| "eval_steps_per_second": 3.374, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6488744616508484, | |
| "eval_runtime": 0.6012, | |
| "eval_samples_per_second": 73.189, | |
| "eval_steps_per_second": 3.327, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.5245105028152466, | |
| "eval_runtime": 0.6059, | |
| "eval_samples_per_second": 72.616, | |
| "eval_steps_per_second": 3.301, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.6672229766845703, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4488, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.8494405150413513, | |
| "eval_runtime": 0.6003, | |
| "eval_samples_per_second": 73.291, | |
| "eval_steps_per_second": 3.331, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6818181818181818, | |
| "eval_loss": 0.908623456954956, | |
| "eval_runtime": 0.6003, | |
| "eval_samples_per_second": 73.297, | |
| "eval_steps_per_second": 3.332, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.5250024795532227, | |
| "eval_runtime": 0.6052, | |
| "eval_samples_per_second": 72.707, | |
| "eval_steps_per_second": 3.305, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.5178956985473633, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 71.17, | |
| "eval_steps_per_second": 3.235, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.4423157870769501, | |
| "eval_runtime": 0.5984, | |
| "eval_samples_per_second": 73.524, | |
| "eval_steps_per_second": 3.342, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.9871264696121216, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3387, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.5113910436630249, | |
| "eval_runtime": 0.5985, | |
| "eval_samples_per_second": 73.511, | |
| "eval_steps_per_second": 3.341, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.5047827363014221, | |
| "eval_runtime": 0.606, | |
| "eval_samples_per_second": 72.605, | |
| "eval_steps_per_second": 3.3, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.49968966841697693, | |
| "eval_runtime": 0.6028, | |
| "eval_samples_per_second": 72.998, | |
| "eval_steps_per_second": 3.318, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.47758814692497253, | |
| "eval_runtime": 0.6036, | |
| "eval_samples_per_second": 72.898, | |
| "eval_steps_per_second": 3.314, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.4137561023235321, | |
| "eval_runtime": 0.5985, | |
| "eval_samples_per_second": 73.522, | |
| "eval_steps_per_second": 3.342, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.6528573036193848, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.24, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.32196831703186035, | |
| "eval_runtime": 0.59, | |
| "eval_samples_per_second": 74.576, | |
| "eval_steps_per_second": 3.39, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.33625996112823486, | |
| "eval_runtime": 0.6044, | |
| "eval_samples_per_second": 72.798, | |
| "eval_steps_per_second": 3.309, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.2957504093647003, | |
| "eval_runtime": 0.6018, | |
| "eval_samples_per_second": 73.109, | |
| "eval_steps_per_second": 3.323, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.30981671810150146, | |
| "eval_runtime": 0.606, | |
| "eval_samples_per_second": 72.61, | |
| "eval_steps_per_second": 3.3, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4029864966869354, | |
| "eval_runtime": 0.6199, | |
| "eval_samples_per_second": 70.98, | |
| "eval_steps_per_second": 3.226, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 4.306562900543213, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.1524, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3094272017478943, | |
| "eval_runtime": 0.61, | |
| "eval_samples_per_second": 72.127, | |
| "eval_steps_per_second": 3.278, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2721233367919922, | |
| "eval_runtime": 0.614, | |
| "eval_samples_per_second": 71.667, | |
| "eval_steps_per_second": 3.258, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3363112211227417, | |
| "eval_runtime": 0.6016, | |
| "eval_samples_per_second": 73.137, | |
| "eval_steps_per_second": 3.324, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.2730633020401001, | |
| "eval_runtime": 0.6012, | |
| "eval_samples_per_second": 73.192, | |
| "eval_steps_per_second": 3.327, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.5659830570220947, | |
| "eval_runtime": 0.5929, | |
| "eval_samples_per_second": 74.212, | |
| "eval_steps_per_second": 3.373, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 5.7067646980285645, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1646, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4948980212211609, | |
| "eval_runtime": 0.6025, | |
| "eval_samples_per_second": 73.033, | |
| "eval_steps_per_second": 3.32, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.40872201323509216, | |
| "eval_runtime": 0.601, | |
| "eval_samples_per_second": 73.207, | |
| "eval_steps_per_second": 3.328, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4466763734817505, | |
| "eval_runtime": 0.6055, | |
| "eval_samples_per_second": 72.662, | |
| "eval_steps_per_second": 3.303, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.4129643738269806, | |
| "eval_runtime": 0.6118, | |
| "eval_samples_per_second": 71.919, | |
| "eval_steps_per_second": 3.269, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.37272271513938904, | |
| "eval_runtime": 0.6041, | |
| "eval_samples_per_second": 72.832, | |
| "eval_steps_per_second": 3.311, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 3.3331470489501953, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.136, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.5894367694854736, | |
| "eval_runtime": 0.5976, | |
| "eval_samples_per_second": 73.632, | |
| "eval_steps_per_second": 3.347, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.9461824893951416, | |
| "eval_runtime": 0.6004, | |
| "eval_samples_per_second": 73.285, | |
| "eval_steps_per_second": 3.331, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.592619001865387, | |
| "eval_runtime": 0.5974, | |
| "eval_samples_per_second": 73.654, | |
| "eval_steps_per_second": 3.348, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.31382161378860474, | |
| "eval_runtime": 0.6063, | |
| "eval_samples_per_second": 72.571, | |
| "eval_steps_per_second": 3.299, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4172639846801758, | |
| "eval_runtime": 0.603, | |
| "eval_samples_per_second": 72.968, | |
| "eval_steps_per_second": 3.317, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.0806145668029785, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.163, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3153996765613556, | |
| "eval_runtime": 0.6087, | |
| "eval_samples_per_second": 72.291, | |
| "eval_steps_per_second": 3.286, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3235173523426056, | |
| "eval_runtime": 0.6301, | |
| "eval_samples_per_second": 69.828, | |
| "eval_steps_per_second": 3.174, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.39018821716308594, | |
| "eval_runtime": 0.6096, | |
| "eval_samples_per_second": 72.183, | |
| "eval_steps_per_second": 3.281, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.36994898319244385, | |
| "eval_runtime": 0.6141, | |
| "eval_samples_per_second": 71.644, | |
| "eval_steps_per_second": 3.257, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.43112871050834656, | |
| "eval_runtime": 0.6217, | |
| "eval_samples_per_second": 70.773, | |
| "eval_steps_per_second": 3.217, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 3.939944267272949, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1018, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.30710768699645996, | |
| "eval_runtime": 0.6006, | |
| "eval_samples_per_second": 73.262, | |
| "eval_steps_per_second": 3.33, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2848808467388153, | |
| "eval_runtime": 0.6258, | |
| "eval_samples_per_second": 70.311, | |
| "eval_steps_per_second": 3.196, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.32258015871047974, | |
| "eval_runtime": 0.623, | |
| "eval_samples_per_second": 70.625, | |
| "eval_steps_per_second": 3.21, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.29670485854148865, | |
| "eval_runtime": 0.6047, | |
| "eval_samples_per_second": 72.766, | |
| "eval_steps_per_second": 3.308, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.2936011254787445, | |
| "eval_runtime": 0.5979, | |
| "eval_samples_per_second": 73.591, | |
| "eval_steps_per_second": 3.345, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 2.3803811073303223, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0957, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2737404704093933, | |
| "eval_runtime": 0.5981, | |
| "eval_samples_per_second": 73.564, | |
| "eval_steps_per_second": 3.344, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.28448688983917236, | |
| "eval_runtime": 0.5965, | |
| "eval_samples_per_second": 73.764, | |
| "eval_steps_per_second": 3.353, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.32999566197395325, | |
| "eval_runtime": 0.6153, | |
| "eval_samples_per_second": 71.516, | |
| "eval_steps_per_second": 3.251, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4029237627983093, | |
| "eval_runtime": 0.6071, | |
| "eval_samples_per_second": 72.475, | |
| "eval_steps_per_second": 3.294, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.42787179350852966, | |
| "eval_runtime": 0.6207, | |
| "eval_samples_per_second": 70.892, | |
| "eval_steps_per_second": 3.222, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 1.518283724784851, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1036, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3899739384651184, | |
| "eval_runtime": 0.5942, | |
| "eval_samples_per_second": 74.046, | |
| "eval_steps_per_second": 3.366, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4038306176662445, | |
| "eval_runtime": 0.5987, | |
| "eval_samples_per_second": 73.488, | |
| "eval_steps_per_second": 3.34, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.35685035586357117, | |
| "eval_runtime": 0.5979, | |
| "eval_samples_per_second": 73.589, | |
| "eval_steps_per_second": 3.345, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.5610812902450562, | |
| "eval_runtime": 0.5989, | |
| "eval_samples_per_second": 73.469, | |
| "eval_steps_per_second": 3.339, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.6900007724761963, | |
| "eval_runtime": 0.6016, | |
| "eval_samples_per_second": 73.144, | |
| "eval_steps_per_second": 3.325, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 2.332082986831665, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1048, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.5678878426551819, | |
| "eval_runtime": 0.598, | |
| "eval_samples_per_second": 73.579, | |
| "eval_steps_per_second": 3.344, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.45673760771751404, | |
| "eval_runtime": 0.6139, | |
| "eval_samples_per_second": 71.676, | |
| "eval_steps_per_second": 3.258, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.38148048520088196, | |
| "eval_runtime": 0.6034, | |
| "eval_samples_per_second": 72.917, | |
| "eval_steps_per_second": 3.314, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.3546294867992401, | |
| "eval_runtime": 0.5975, | |
| "eval_samples_per_second": 73.644, | |
| "eval_steps_per_second": 3.347, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.3653978109359741, | |
| "eval_runtime": 0.5948, | |
| "eval_samples_per_second": 73.976, | |
| "eval_steps_per_second": 3.363, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 1.0287319421768188, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0928, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3336702585220337, | |
| "eval_runtime": 0.6029, | |
| "eval_samples_per_second": 72.981, | |
| "eval_steps_per_second": 3.317, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4160974621772766, | |
| "eval_runtime": 0.6025, | |
| "eval_samples_per_second": 73.023, | |
| "eval_steps_per_second": 3.319, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3615375757217407, | |
| "eval_runtime": 0.6048, | |
| "eval_samples_per_second": 72.757, | |
| "eval_steps_per_second": 3.307, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.4061291813850403, | |
| "eval_runtime": 0.5989, | |
| "eval_samples_per_second": 73.465, | |
| "eval_steps_per_second": 3.339, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.419089674949646, | |
| "eval_runtime": 0.62, | |
| "eval_samples_per_second": 70.967, | |
| "eval_steps_per_second": 3.226, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 3.327659845352173, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1091, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.39782723784446716, | |
| "eval_runtime": 0.6071, | |
| "eval_samples_per_second": 72.477, | |
| "eval_steps_per_second": 3.294, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.5167635083198547, | |
| "eval_runtime": 0.6009, | |
| "eval_samples_per_second": 73.218, | |
| "eval_steps_per_second": 3.328, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.5268294811248779, | |
| "eval_runtime": 0.6017, | |
| "eval_samples_per_second": 73.131, | |
| "eval_steps_per_second": 3.324, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.5666711330413818, | |
| "eval_runtime": 0.6046, | |
| "eval_samples_per_second": 72.779, | |
| "eval_steps_per_second": 3.308, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.5395712852478027, | |
| "eval_runtime": 0.6056, | |
| "eval_samples_per_second": 72.651, | |
| "eval_steps_per_second": 3.302, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 2.171034097671509, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1009, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.4806671142578125, | |
| "eval_runtime": 0.6064, | |
| "eval_samples_per_second": 72.558, | |
| "eval_steps_per_second": 3.298, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.39570510387420654, | |
| "eval_runtime": 0.6022, | |
| "eval_samples_per_second": 73.069, | |
| "eval_steps_per_second": 3.321, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3518943190574646, | |
| "eval_runtime": 0.6137, | |
| "eval_samples_per_second": 71.698, | |
| "eval_steps_per_second": 3.259, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.36540624499320984, | |
| "eval_runtime": 0.6075, | |
| "eval_samples_per_second": 72.431, | |
| "eval_steps_per_second": 3.292, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.35774800181388855, | |
| "eval_runtime": 0.6077, | |
| "eval_samples_per_second": 72.405, | |
| "eval_steps_per_second": 3.291, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.3086776733398438, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0836, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3215782046318054, | |
| "eval_runtime": 0.6004, | |
| "eval_samples_per_second": 73.282, | |
| "eval_steps_per_second": 3.331, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3132060170173645, | |
| "eval_runtime": 0.5986, | |
| "eval_samples_per_second": 73.501, | |
| "eval_steps_per_second": 3.341, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3002861440181732, | |
| "eval_runtime": 0.6006, | |
| "eval_samples_per_second": 73.261, | |
| "eval_steps_per_second": 3.33, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.30236127972602844, | |
| "eval_runtime": 0.6034, | |
| "eval_samples_per_second": 72.915, | |
| "eval_steps_per_second": 3.314, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3213507831096649, | |
| "eval_runtime": 0.6075, | |
| "eval_samples_per_second": 72.423, | |
| "eval_steps_per_second": 3.292, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 3.096367120742798, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0928, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.33060044050216675, | |
| "eval_runtime": 0.6133, | |
| "eval_samples_per_second": 71.747, | |
| "eval_steps_per_second": 3.261, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.32838213443756104, | |
| "eval_runtime": 0.643, | |
| "eval_samples_per_second": 68.429, | |
| "eval_steps_per_second": 3.11, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3271595537662506, | |
| "eval_runtime": 0.6058, | |
| "eval_samples_per_second": 72.627, | |
| "eval_steps_per_second": 3.301, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.32607272267341614, | |
| "eval_runtime": 0.6115, | |
| "eval_samples_per_second": 71.954, | |
| "eval_steps_per_second": 3.271, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3098549544811249, | |
| "eval_runtime": 0.603, | |
| "eval_samples_per_second": 72.966, | |
| "eval_steps_per_second": 3.317, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 1.7827409505844116, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0915, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.2928163707256317, | |
| "eval_runtime": 0.6039, | |
| "eval_samples_per_second": 72.857, | |
| "eval_steps_per_second": 3.312, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.28483426570892334, | |
| "eval_runtime": 0.6039, | |
| "eval_samples_per_second": 72.855, | |
| "eval_steps_per_second": 3.312, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.282673716545105, | |
| "eval_runtime": 0.6125, | |
| "eval_samples_per_second": 71.84, | |
| "eval_steps_per_second": 3.265, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.2885366380214691, | |
| "eval_runtime": 0.6004, | |
| "eval_samples_per_second": 73.285, | |
| "eval_steps_per_second": 3.331, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3083859980106354, | |
| "eval_runtime": 0.5955, | |
| "eval_samples_per_second": 73.888, | |
| "eval_steps_per_second": 3.359, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.1581565141677856, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0775, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3321191966533661, | |
| "eval_runtime": 0.6031, | |
| "eval_samples_per_second": 72.961, | |
| "eval_steps_per_second": 3.316, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.33584651350975037, | |
| "eval_runtime": 0.6044, | |
| "eval_samples_per_second": 72.805, | |
| "eval_steps_per_second": 3.309, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3220769166946411, | |
| "eval_runtime": 0.6004, | |
| "eval_samples_per_second": 73.281, | |
| "eval_steps_per_second": 3.331, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.30963456630706787, | |
| "eval_runtime": 0.5998, | |
| "eval_samples_per_second": 73.363, | |
| "eval_steps_per_second": 3.335, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.30070677399635315, | |
| "eval_runtime": 0.6112, | |
| "eval_samples_per_second": 71.984, | |
| "eval_steps_per_second": 3.272, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.9102116823196411, | |
| "learning_rate": 0.0, | |
| "loss": 0.091, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.29760289192199707, | |
| "eval_runtime": 0.6091, | |
| "eval_samples_per_second": 72.242, | |
| "eval_steps_per_second": 3.284, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_loss": 0.17884186565876006, | |
| "train_runtime": 1123.4343, | |
| "train_samples_per_second": 21.63, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.30710768699645996, | |
| "eval_runtime": 0.6743, | |
| "eval_samples_per_second": 65.249, | |
| "eval_steps_per_second": 2.966, | |
| "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 | |
| } | |