sign_language_classification_v1 / trainer_state.json
joseluhf11's picture
End of training
5e2a998
{
"best_metric": 0.8611111111111112,
"best_model_checkpoint": "sign_language_classification_v1/checkpoint-264",
"epoch": 50.0,
"eval_steps": 500,
"global_step": 400,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.12,
"learning_rate": 1.25e-06,
"loss": 3.2873,
"step": 1
},
{
"epoch": 0.25,
"learning_rate": 2.5e-06,
"loss": 3.2978,
"step": 2
},
{
"epoch": 0.38,
"learning_rate": 3.75e-06,
"loss": 3.28,
"step": 3
},
{
"epoch": 0.5,
"learning_rate": 5e-06,
"loss": 3.2786,
"step": 4
},
{
"epoch": 0.62,
"learning_rate": 6.25e-06,
"loss": 3.2834,
"step": 5
},
{
"epoch": 0.75,
"learning_rate": 7.5e-06,
"loss": 3.2949,
"step": 6
},
{
"epoch": 0.88,
"learning_rate": 8.75e-06,
"loss": 3.3025,
"step": 7
},
{
"epoch": 1.0,
"learning_rate": 1e-05,
"loss": 3.2889,
"step": 8
},
{
"epoch": 1.0,
"eval_accuracy": 0.05555555555555555,
"eval_loss": 3.2713797092437744,
"eval_runtime": 1.1489,
"eval_samples_per_second": 62.668,
"eval_steps_per_second": 1.741,
"step": 8
},
{
"epoch": 1.12,
"learning_rate": 1.125e-05,
"loss": 3.2724,
"step": 9
},
{
"epoch": 1.25,
"learning_rate": 1.25e-05,
"loss": 3.2712,
"step": 10
},
{
"epoch": 1.38,
"learning_rate": 1.3750000000000002e-05,
"loss": 3.2872,
"step": 11
},
{
"epoch": 1.5,
"learning_rate": 1.5e-05,
"loss": 3.2629,
"step": 12
},
{
"epoch": 1.62,
"learning_rate": 1.6250000000000002e-05,
"loss": 3.2719,
"step": 13
},
{
"epoch": 1.75,
"learning_rate": 1.75e-05,
"loss": 3.2452,
"step": 14
},
{
"epoch": 1.88,
"learning_rate": 1.8750000000000002e-05,
"loss": 3.256,
"step": 15
},
{
"epoch": 2.0,
"learning_rate": 2e-05,
"loss": 3.2492,
"step": 16
},
{
"epoch": 2.0,
"eval_accuracy": 0.125,
"eval_loss": 3.261528730392456,
"eval_runtime": 1.0965,
"eval_samples_per_second": 65.664,
"eval_steps_per_second": 1.824,
"step": 16
},
{
"epoch": 2.12,
"learning_rate": 2.125e-05,
"loss": 3.2441,
"step": 17
},
{
"epoch": 2.25,
"learning_rate": 2.25e-05,
"loss": 3.2413,
"step": 18
},
{
"epoch": 2.38,
"learning_rate": 2.375e-05,
"loss": 3.2249,
"step": 19
},
{
"epoch": 2.5,
"learning_rate": 2.5e-05,
"loss": 3.2309,
"step": 20
},
{
"epoch": 2.62,
"learning_rate": 2.625e-05,
"loss": 3.2292,
"step": 21
},
{
"epoch": 2.75,
"learning_rate": 2.7500000000000004e-05,
"loss": 3.1908,
"step": 22
},
{
"epoch": 2.88,
"learning_rate": 2.8749999999999997e-05,
"loss": 3.1873,
"step": 23
},
{
"epoch": 3.0,
"learning_rate": 3e-05,
"loss": 3.2263,
"step": 24
},
{
"epoch": 3.0,
"eval_accuracy": 0.125,
"eval_loss": 3.203394889831543,
"eval_runtime": 1.1276,
"eval_samples_per_second": 63.851,
"eval_steps_per_second": 1.774,
"step": 24
},
{
"epoch": 3.12,
"learning_rate": 3.125e-05,
"loss": 3.1968,
"step": 25
},
{
"epoch": 3.25,
"learning_rate": 3.2500000000000004e-05,
"loss": 3.1653,
"step": 26
},
{
"epoch": 3.38,
"learning_rate": 3.375000000000001e-05,
"loss": 3.1591,
"step": 27
},
{
"epoch": 3.5,
"learning_rate": 3.5e-05,
"loss": 3.1429,
"step": 28
},
{
"epoch": 3.62,
"learning_rate": 3.625e-05,
"loss": 3.1298,
"step": 29
},
{
"epoch": 3.75,
"learning_rate": 3.7500000000000003e-05,
"loss": 3.1157,
"step": 30
},
{
"epoch": 3.88,
"learning_rate": 3.875e-05,
"loss": 3.075,
"step": 31
},
{
"epoch": 4.0,
"learning_rate": 4e-05,
"loss": 3.1271,
"step": 32
},
{
"epoch": 4.0,
"eval_accuracy": 0.20833333333333334,
"eval_loss": 3.1297194957733154,
"eval_runtime": 1.124,
"eval_samples_per_second": 64.06,
"eval_steps_per_second": 1.779,
"step": 32
},
{
"epoch": 4.12,
"learning_rate": 4.125e-05,
"loss": 3.0596,
"step": 33
},
{
"epoch": 4.25,
"learning_rate": 4.25e-05,
"loss": 3.0475,
"step": 34
},
{
"epoch": 4.38,
"learning_rate": 4.375e-05,
"loss": 3.0853,
"step": 35
},
{
"epoch": 4.5,
"learning_rate": 4.5e-05,
"loss": 3.0794,
"step": 36
},
{
"epoch": 4.62,
"learning_rate": 4.6250000000000006e-05,
"loss": 3.0746,
"step": 37
},
{
"epoch": 4.75,
"learning_rate": 4.75e-05,
"loss": 2.9772,
"step": 38
},
{
"epoch": 4.88,
"learning_rate": 4.875e-05,
"loss": 3.0311,
"step": 39
},
{
"epoch": 5.0,
"learning_rate": 5e-05,
"loss": 2.9592,
"step": 40
},
{
"epoch": 5.0,
"eval_accuracy": 0.2638888888888889,
"eval_loss": 3.06551194190979,
"eval_runtime": 1.1056,
"eval_samples_per_second": 65.122,
"eval_steps_per_second": 1.809,
"step": 40
},
{
"epoch": 5.12,
"learning_rate": 4.986111111111111e-05,
"loss": 2.8884,
"step": 41
},
{
"epoch": 5.25,
"learning_rate": 4.972222222222223e-05,
"loss": 2.9681,
"step": 42
},
{
"epoch": 5.38,
"learning_rate": 4.958333333333334e-05,
"loss": 2.9507,
"step": 43
},
{
"epoch": 5.5,
"learning_rate": 4.9444444444444446e-05,
"loss": 2.9259,
"step": 44
},
{
"epoch": 5.62,
"learning_rate": 4.930555555555556e-05,
"loss": 2.9088,
"step": 45
},
{
"epoch": 5.75,
"learning_rate": 4.9166666666666665e-05,
"loss": 2.8482,
"step": 46
},
{
"epoch": 5.88,
"learning_rate": 4.902777777777778e-05,
"loss": 2.9336,
"step": 47
},
{
"epoch": 6.0,
"learning_rate": 4.888888888888889e-05,
"loss": 2.9414,
"step": 48
},
{
"epoch": 6.0,
"eval_accuracy": 0.3472222222222222,
"eval_loss": 2.9281809329986572,
"eval_runtime": 1.1366,
"eval_samples_per_second": 63.344,
"eval_steps_per_second": 1.76,
"step": 48
},
{
"epoch": 6.12,
"learning_rate": 4.875e-05,
"loss": 2.9047,
"step": 49
},
{
"epoch": 6.25,
"learning_rate": 4.8611111111111115e-05,
"loss": 2.8374,
"step": 50
},
{
"epoch": 6.38,
"learning_rate": 4.8472222222222224e-05,
"loss": 2.8785,
"step": 51
},
{
"epoch": 6.5,
"learning_rate": 4.8333333333333334e-05,
"loss": 2.734,
"step": 52
},
{
"epoch": 6.62,
"learning_rate": 4.819444444444445e-05,
"loss": 2.7564,
"step": 53
},
{
"epoch": 6.75,
"learning_rate": 4.805555555555556e-05,
"loss": 2.7417,
"step": 54
},
{
"epoch": 6.88,
"learning_rate": 4.791666666666667e-05,
"loss": 2.6786,
"step": 55
},
{
"epoch": 7.0,
"learning_rate": 4.7777777777777784e-05,
"loss": 2.7337,
"step": 56
},
{
"epoch": 7.0,
"eval_accuracy": 0.4027777777777778,
"eval_loss": 2.825361490249634,
"eval_runtime": 1.2892,
"eval_samples_per_second": 55.85,
"eval_steps_per_second": 1.551,
"step": 56
},
{
"epoch": 7.12,
"learning_rate": 4.7638888888888887e-05,
"loss": 2.7761,
"step": 57
},
{
"epoch": 7.25,
"learning_rate": 4.75e-05,
"loss": 2.7389,
"step": 58
},
{
"epoch": 7.38,
"learning_rate": 4.736111111111111e-05,
"loss": 2.6867,
"step": 59
},
{
"epoch": 7.5,
"learning_rate": 4.722222222222222e-05,
"loss": 2.6275,
"step": 60
},
{
"epoch": 7.62,
"learning_rate": 4.708333333333334e-05,
"loss": 2.6247,
"step": 61
},
{
"epoch": 7.75,
"learning_rate": 4.6944444444444446e-05,
"loss": 2.7138,
"step": 62
},
{
"epoch": 7.88,
"learning_rate": 4.6805555555555556e-05,
"loss": 2.5862,
"step": 63
},
{
"epoch": 8.0,
"learning_rate": 4.666666666666667e-05,
"loss": 2.6683,
"step": 64
},
{
"epoch": 8.0,
"eval_accuracy": 0.4583333333333333,
"eval_loss": 2.690866231918335,
"eval_runtime": 1.3259,
"eval_samples_per_second": 54.303,
"eval_steps_per_second": 1.508,
"step": 64
},
{
"epoch": 8.12,
"learning_rate": 4.652777777777778e-05,
"loss": 2.5625,
"step": 65
},
{
"epoch": 8.25,
"learning_rate": 4.638888888888889e-05,
"loss": 2.5487,
"step": 66
},
{
"epoch": 8.38,
"learning_rate": 4.6250000000000006e-05,
"loss": 2.5947,
"step": 67
},
{
"epoch": 8.5,
"learning_rate": 4.6111111111111115e-05,
"loss": 2.5763,
"step": 68
},
{
"epoch": 8.62,
"learning_rate": 4.5972222222222225e-05,
"loss": 2.4386,
"step": 69
},
{
"epoch": 8.75,
"learning_rate": 4.5833333333333334e-05,
"loss": 2.434,
"step": 70
},
{
"epoch": 8.88,
"learning_rate": 4.569444444444444e-05,
"loss": 2.4864,
"step": 71
},
{
"epoch": 9.0,
"learning_rate": 4.555555555555556e-05,
"loss": 2.5837,
"step": 72
},
{
"epoch": 9.0,
"eval_accuracy": 0.5416666666666666,
"eval_loss": 2.59035587310791,
"eval_runtime": 1.1612,
"eval_samples_per_second": 62.005,
"eval_steps_per_second": 1.722,
"step": 72
},
{
"epoch": 9.12,
"learning_rate": 4.541666666666667e-05,
"loss": 2.5574,
"step": 73
},
{
"epoch": 9.25,
"learning_rate": 4.527777777777778e-05,
"loss": 2.3011,
"step": 74
},
{
"epoch": 9.38,
"learning_rate": 4.5138888888888894e-05,
"loss": 2.4267,
"step": 75
},
{
"epoch": 9.5,
"learning_rate": 4.5e-05,
"loss": 2.4231,
"step": 76
},
{
"epoch": 9.62,
"learning_rate": 4.486111111111111e-05,
"loss": 2.469,
"step": 77
},
{
"epoch": 9.75,
"learning_rate": 4.472222222222223e-05,
"loss": 2.3926,
"step": 78
},
{
"epoch": 9.88,
"learning_rate": 4.458333333333334e-05,
"loss": 2.3579,
"step": 79
},
{
"epoch": 10.0,
"learning_rate": 4.4444444444444447e-05,
"loss": 2.4566,
"step": 80
},
{
"epoch": 10.0,
"eval_accuracy": 0.5833333333333334,
"eval_loss": 2.53802227973938,
"eval_runtime": 1.3374,
"eval_samples_per_second": 53.836,
"eval_steps_per_second": 1.495,
"step": 80
},
{
"epoch": 10.12,
"learning_rate": 4.4305555555555556e-05,
"loss": 2.3805,
"step": 81
},
{
"epoch": 10.25,
"learning_rate": 4.4166666666666665e-05,
"loss": 2.3258,
"step": 82
},
{
"epoch": 10.38,
"learning_rate": 4.402777777777778e-05,
"loss": 2.3485,
"step": 83
},
{
"epoch": 10.5,
"learning_rate": 4.388888888888889e-05,
"loss": 2.3897,
"step": 84
},
{
"epoch": 10.62,
"learning_rate": 4.375e-05,
"loss": 2.2521,
"step": 85
},
{
"epoch": 10.75,
"learning_rate": 4.3611111111111116e-05,
"loss": 2.2908,
"step": 86
},
{
"epoch": 10.88,
"learning_rate": 4.3472222222222225e-05,
"loss": 2.2622,
"step": 87
},
{
"epoch": 11.0,
"learning_rate": 4.3333333333333334e-05,
"loss": 2.2188,
"step": 88
},
{
"epoch": 11.0,
"eval_accuracy": 0.5416666666666666,
"eval_loss": 2.4681763648986816,
"eval_runtime": 1.1411,
"eval_samples_per_second": 63.1,
"eval_steps_per_second": 1.753,
"step": 88
},
{
"epoch": 11.12,
"learning_rate": 4.319444444444445e-05,
"loss": 2.3194,
"step": 89
},
{
"epoch": 11.25,
"learning_rate": 4.305555555555556e-05,
"loss": 2.2072,
"step": 90
},
{
"epoch": 11.38,
"learning_rate": 4.291666666666667e-05,
"loss": 2.172,
"step": 91
},
{
"epoch": 11.5,
"learning_rate": 4.277777777777778e-05,
"loss": 2.2478,
"step": 92
},
{
"epoch": 11.62,
"learning_rate": 4.263888888888889e-05,
"loss": 2.1903,
"step": 93
},
{
"epoch": 11.75,
"learning_rate": 4.25e-05,
"loss": 2.1867,
"step": 94
},
{
"epoch": 11.88,
"learning_rate": 4.236111111111111e-05,
"loss": 2.1636,
"step": 95
},
{
"epoch": 12.0,
"learning_rate": 4.222222222222222e-05,
"loss": 2.2885,
"step": 96
},
{
"epoch": 12.0,
"eval_accuracy": 0.5833333333333334,
"eval_loss": 2.3195643424987793,
"eval_runtime": 1.3304,
"eval_samples_per_second": 54.118,
"eval_steps_per_second": 1.503,
"step": 96
},
{
"epoch": 12.12,
"learning_rate": 4.208333333333334e-05,
"loss": 2.077,
"step": 97
},
{
"epoch": 12.25,
"learning_rate": 4.194444444444445e-05,
"loss": 2.1554,
"step": 98
},
{
"epoch": 12.38,
"learning_rate": 4.1805555555555556e-05,
"loss": 2.0482,
"step": 99
},
{
"epoch": 12.5,
"learning_rate": 4.166666666666667e-05,
"loss": 2.107,
"step": 100
},
{
"epoch": 12.62,
"learning_rate": 4.152777777777778e-05,
"loss": 2.1134,
"step": 101
},
{
"epoch": 12.75,
"learning_rate": 4.138888888888889e-05,
"loss": 2.1261,
"step": 102
},
{
"epoch": 12.88,
"learning_rate": 4.125e-05,
"loss": 2.0718,
"step": 103
},
{
"epoch": 13.0,
"learning_rate": 4.111111111111111e-05,
"loss": 2.005,
"step": 104
},
{
"epoch": 13.0,
"eval_accuracy": 0.6666666666666666,
"eval_loss": 2.282362699508667,
"eval_runtime": 1.3009,
"eval_samples_per_second": 55.348,
"eval_steps_per_second": 1.537,
"step": 104
},
{
"epoch": 13.12,
"learning_rate": 4.0972222222222225e-05,
"loss": 2.0269,
"step": 105
},
{
"epoch": 13.25,
"learning_rate": 4.0833333333333334e-05,
"loss": 1.9191,
"step": 106
},
{
"epoch": 13.38,
"learning_rate": 4.0694444444444444e-05,
"loss": 2.0349,
"step": 107
},
{
"epoch": 13.5,
"learning_rate": 4.055555555555556e-05,
"loss": 2.0988,
"step": 108
},
{
"epoch": 13.62,
"learning_rate": 4.041666666666667e-05,
"loss": 2.0798,
"step": 109
},
{
"epoch": 13.75,
"learning_rate": 4.027777777777778e-05,
"loss": 1.8602,
"step": 110
},
{
"epoch": 13.88,
"learning_rate": 4.0138888888888894e-05,
"loss": 1.9826,
"step": 111
},
{
"epoch": 14.0,
"learning_rate": 4e-05,
"loss": 1.9293,
"step": 112
},
{
"epoch": 14.0,
"eval_accuracy": 0.6388888888888888,
"eval_loss": 2.196725606918335,
"eval_runtime": 1.2726,
"eval_samples_per_second": 56.577,
"eval_steps_per_second": 1.572,
"step": 112
},
{
"epoch": 14.12,
"learning_rate": 3.986111111111111e-05,
"loss": 1.9408,
"step": 113
},
{
"epoch": 14.25,
"learning_rate": 3.972222222222222e-05,
"loss": 1.8799,
"step": 114
},
{
"epoch": 14.38,
"learning_rate": 3.958333333333333e-05,
"loss": 1.842,
"step": 115
},
{
"epoch": 14.5,
"learning_rate": 3.944444444444445e-05,
"loss": 2.0401,
"step": 116
},
{
"epoch": 14.62,
"learning_rate": 3.9305555555555556e-05,
"loss": 1.8972,
"step": 117
},
{
"epoch": 14.75,
"learning_rate": 3.9166666666666665e-05,
"loss": 1.9211,
"step": 118
},
{
"epoch": 14.88,
"learning_rate": 3.902777777777778e-05,
"loss": 1.8864,
"step": 119
},
{
"epoch": 15.0,
"learning_rate": 3.888888888888889e-05,
"loss": 1.8396,
"step": 120
},
{
"epoch": 15.0,
"eval_accuracy": 0.7361111111111112,
"eval_loss": 2.0286779403686523,
"eval_runtime": 1.165,
"eval_samples_per_second": 61.8,
"eval_steps_per_second": 1.717,
"step": 120
},
{
"epoch": 15.12,
"learning_rate": 3.875e-05,
"loss": 1.9551,
"step": 121
},
{
"epoch": 15.25,
"learning_rate": 3.8611111111111116e-05,
"loss": 1.8472,
"step": 122
},
{
"epoch": 15.38,
"learning_rate": 3.8472222222222225e-05,
"loss": 1.7747,
"step": 123
},
{
"epoch": 15.5,
"learning_rate": 3.8333333333333334e-05,
"loss": 1.7947,
"step": 124
},
{
"epoch": 15.62,
"learning_rate": 3.8194444444444444e-05,
"loss": 1.7818,
"step": 125
},
{
"epoch": 15.75,
"learning_rate": 3.805555555555555e-05,
"loss": 1.7632,
"step": 126
},
{
"epoch": 15.88,
"learning_rate": 3.791666666666667e-05,
"loss": 1.773,
"step": 127
},
{
"epoch": 16.0,
"learning_rate": 3.777777777777778e-05,
"loss": 1.7066,
"step": 128
},
{
"epoch": 16.0,
"eval_accuracy": 0.7361111111111112,
"eval_loss": 2.035660743713379,
"eval_runtime": 1.1579,
"eval_samples_per_second": 62.183,
"eval_steps_per_second": 1.727,
"step": 128
},
{
"epoch": 16.12,
"learning_rate": 3.763888888888889e-05,
"loss": 1.8366,
"step": 129
},
{
"epoch": 16.25,
"learning_rate": 3.7500000000000003e-05,
"loss": 1.7713,
"step": 130
},
{
"epoch": 16.38,
"learning_rate": 3.736111111111111e-05,
"loss": 1.8229,
"step": 131
},
{
"epoch": 16.5,
"learning_rate": 3.722222222222222e-05,
"loss": 1.7652,
"step": 132
},
{
"epoch": 16.62,
"learning_rate": 3.708333333333334e-05,
"loss": 1.7239,
"step": 133
},
{
"epoch": 16.75,
"learning_rate": 3.694444444444445e-05,
"loss": 1.6065,
"step": 134
},
{
"epoch": 16.88,
"learning_rate": 3.6805555555555556e-05,
"loss": 1.5874,
"step": 135
},
{
"epoch": 17.0,
"learning_rate": 3.6666666666666666e-05,
"loss": 1.6911,
"step": 136
},
{
"epoch": 17.0,
"eval_accuracy": 0.7361111111111112,
"eval_loss": 1.9670071601867676,
"eval_runtime": 1.1584,
"eval_samples_per_second": 62.153,
"eval_steps_per_second": 1.726,
"step": 136
},
{
"epoch": 17.12,
"learning_rate": 3.6527777777777775e-05,
"loss": 1.5983,
"step": 137
},
{
"epoch": 17.25,
"learning_rate": 3.638888888888889e-05,
"loss": 1.7529,
"step": 138
},
{
"epoch": 17.38,
"learning_rate": 3.625e-05,
"loss": 1.6416,
"step": 139
},
{
"epoch": 17.5,
"learning_rate": 3.611111111111111e-05,
"loss": 1.6631,
"step": 140
},
{
"epoch": 17.62,
"learning_rate": 3.5972222222222225e-05,
"loss": 1.7048,
"step": 141
},
{
"epoch": 17.75,
"learning_rate": 3.5833333333333335e-05,
"loss": 1.6884,
"step": 142
},
{
"epoch": 17.88,
"learning_rate": 3.5694444444444444e-05,
"loss": 1.6636,
"step": 143
},
{
"epoch": 18.0,
"learning_rate": 3.555555555555556e-05,
"loss": 1.6285,
"step": 144
},
{
"epoch": 18.0,
"eval_accuracy": 0.7361111111111112,
"eval_loss": 1.9186458587646484,
"eval_runtime": 1.1572,
"eval_samples_per_second": 62.22,
"eval_steps_per_second": 1.728,
"step": 144
},
{
"epoch": 18.12,
"learning_rate": 3.541666666666667e-05,
"loss": 1.5803,
"step": 145
},
{
"epoch": 18.25,
"learning_rate": 3.527777777777778e-05,
"loss": 1.6862,
"step": 146
},
{
"epoch": 18.38,
"learning_rate": 3.513888888888889e-05,
"loss": 1.6057,
"step": 147
},
{
"epoch": 18.5,
"learning_rate": 3.5e-05,
"loss": 1.7233,
"step": 148
},
{
"epoch": 18.62,
"learning_rate": 3.486111111111111e-05,
"loss": 1.628,
"step": 149
},
{
"epoch": 18.75,
"learning_rate": 3.472222222222222e-05,
"loss": 1.4979,
"step": 150
},
{
"epoch": 18.88,
"learning_rate": 3.458333333333333e-05,
"loss": 1.5487,
"step": 151
},
{
"epoch": 19.0,
"learning_rate": 3.444444444444445e-05,
"loss": 1.6064,
"step": 152
},
{
"epoch": 19.0,
"eval_accuracy": 0.6944444444444444,
"eval_loss": 1.9239338636398315,
"eval_runtime": 1.1488,
"eval_samples_per_second": 62.674,
"eval_steps_per_second": 1.741,
"step": 152
},
{
"epoch": 19.12,
"learning_rate": 3.430555555555556e-05,
"loss": 1.5404,
"step": 153
},
{
"epoch": 19.25,
"learning_rate": 3.4166666666666666e-05,
"loss": 1.4113,
"step": 154
},
{
"epoch": 19.38,
"learning_rate": 3.402777777777778e-05,
"loss": 1.5838,
"step": 155
},
{
"epoch": 19.5,
"learning_rate": 3.388888888888889e-05,
"loss": 1.5107,
"step": 156
},
{
"epoch": 19.62,
"learning_rate": 3.375000000000001e-05,
"loss": 1.4749,
"step": 157
},
{
"epoch": 19.75,
"learning_rate": 3.3611111111111116e-05,
"loss": 1.6153,
"step": 158
},
{
"epoch": 19.88,
"learning_rate": 3.347222222222222e-05,
"loss": 1.6072,
"step": 159
},
{
"epoch": 20.0,
"learning_rate": 3.3333333333333335e-05,
"loss": 1.6067,
"step": 160
},
{
"epoch": 20.0,
"eval_accuracy": 0.7777777777777778,
"eval_loss": 1.7723278999328613,
"eval_runtime": 1.158,
"eval_samples_per_second": 62.175,
"eval_steps_per_second": 1.727,
"step": 160
},
{
"epoch": 20.12,
"learning_rate": 3.3194444444444444e-05,
"loss": 1.5211,
"step": 161
},
{
"epoch": 20.25,
"learning_rate": 3.3055555555555553e-05,
"loss": 1.4849,
"step": 162
},
{
"epoch": 20.38,
"learning_rate": 3.291666666666667e-05,
"loss": 1.4824,
"step": 163
},
{
"epoch": 20.5,
"learning_rate": 3.277777777777778e-05,
"loss": 1.3954,
"step": 164
},
{
"epoch": 20.62,
"learning_rate": 3.263888888888889e-05,
"loss": 1.4999,
"step": 165
},
{
"epoch": 20.75,
"learning_rate": 3.2500000000000004e-05,
"loss": 1.4035,
"step": 166
},
{
"epoch": 20.88,
"learning_rate": 3.236111111111111e-05,
"loss": 1.4078,
"step": 167
},
{
"epoch": 21.0,
"learning_rate": 3.222222222222223e-05,
"loss": 1.4094,
"step": 168
},
{
"epoch": 21.0,
"eval_accuracy": 0.75,
"eval_loss": 1.7701455354690552,
"eval_runtime": 1.1486,
"eval_samples_per_second": 62.684,
"eval_steps_per_second": 1.741,
"step": 168
},
{
"epoch": 21.12,
"learning_rate": 3.208333333333334e-05,
"loss": 1.4196,
"step": 169
},
{
"epoch": 21.25,
"learning_rate": 3.194444444444444e-05,
"loss": 1.4107,
"step": 170
},
{
"epoch": 21.38,
"learning_rate": 3.180555555555556e-05,
"loss": 1.3853,
"step": 171
},
{
"epoch": 21.5,
"learning_rate": 3.1666666666666666e-05,
"loss": 1.32,
"step": 172
},
{
"epoch": 21.62,
"learning_rate": 3.1527777777777775e-05,
"loss": 1.3411,
"step": 173
},
{
"epoch": 21.75,
"learning_rate": 3.138888888888889e-05,
"loss": 1.377,
"step": 174
},
{
"epoch": 21.88,
"learning_rate": 3.125e-05,
"loss": 1.3504,
"step": 175
},
{
"epoch": 22.0,
"learning_rate": 3.111111111111111e-05,
"loss": 1.4664,
"step": 176
},
{
"epoch": 22.0,
"eval_accuracy": 0.75,
"eval_loss": 1.7453147172927856,
"eval_runtime": 1.1609,
"eval_samples_per_second": 62.02,
"eval_steps_per_second": 1.723,
"step": 176
},
{
"epoch": 22.12,
"learning_rate": 3.0972222222222226e-05,
"loss": 1.279,
"step": 177
},
{
"epoch": 22.25,
"learning_rate": 3.0833333333333335e-05,
"loss": 1.3256,
"step": 178
},
{
"epoch": 22.38,
"learning_rate": 3.069444444444445e-05,
"loss": 1.3083,
"step": 179
},
{
"epoch": 22.5,
"learning_rate": 3.055555555555556e-05,
"loss": 1.3774,
"step": 180
},
{
"epoch": 22.62,
"learning_rate": 3.0416666666666666e-05,
"loss": 1.2658,
"step": 181
},
{
"epoch": 22.75,
"learning_rate": 3.0277777777777776e-05,
"loss": 1.2999,
"step": 182
},
{
"epoch": 22.88,
"learning_rate": 3.0138888888888888e-05,
"loss": 1.4756,
"step": 183
},
{
"epoch": 23.0,
"learning_rate": 3e-05,
"loss": 1.3255,
"step": 184
},
{
"epoch": 23.0,
"eval_accuracy": 0.7083333333333334,
"eval_loss": 1.7102550268173218,
"eval_runtime": 1.1441,
"eval_samples_per_second": 62.932,
"eval_steps_per_second": 1.748,
"step": 184
},
{
"epoch": 23.12,
"learning_rate": 2.9861111111111113e-05,
"loss": 1.2007,
"step": 185
},
{
"epoch": 23.25,
"learning_rate": 2.9722222222222223e-05,
"loss": 1.3202,
"step": 186
},
{
"epoch": 23.38,
"learning_rate": 2.9583333333333335e-05,
"loss": 1.3656,
"step": 187
},
{
"epoch": 23.5,
"learning_rate": 2.9444444444444448e-05,
"loss": 1.3709,
"step": 188
},
{
"epoch": 23.62,
"learning_rate": 2.9305555555555557e-05,
"loss": 1.2755,
"step": 189
},
{
"epoch": 23.75,
"learning_rate": 2.916666666666667e-05,
"loss": 1.2904,
"step": 190
},
{
"epoch": 23.88,
"learning_rate": 2.9027777777777782e-05,
"loss": 1.3189,
"step": 191
},
{
"epoch": 24.0,
"learning_rate": 2.8888888888888888e-05,
"loss": 1.3253,
"step": 192
},
{
"epoch": 24.0,
"eval_accuracy": 0.7777777777777778,
"eval_loss": 1.721591591835022,
"eval_runtime": 1.1527,
"eval_samples_per_second": 62.462,
"eval_steps_per_second": 1.735,
"step": 192
},
{
"epoch": 24.12,
"learning_rate": 2.8749999999999997e-05,
"loss": 1.2694,
"step": 193
},
{
"epoch": 24.25,
"learning_rate": 2.861111111111111e-05,
"loss": 1.4625,
"step": 194
},
{
"epoch": 24.38,
"learning_rate": 2.8472222222222223e-05,
"loss": 1.1592,
"step": 195
},
{
"epoch": 24.5,
"learning_rate": 2.8333333333333335e-05,
"loss": 1.1885,
"step": 196
},
{
"epoch": 24.62,
"learning_rate": 2.8194444444444445e-05,
"loss": 1.1877,
"step": 197
},
{
"epoch": 24.75,
"learning_rate": 2.8055555555555557e-05,
"loss": 1.282,
"step": 198
},
{
"epoch": 24.88,
"learning_rate": 2.791666666666667e-05,
"loss": 1.2798,
"step": 199
},
{
"epoch": 25.0,
"learning_rate": 2.777777777777778e-05,
"loss": 1.2416,
"step": 200
},
{
"epoch": 25.0,
"eval_accuracy": 0.7777777777777778,
"eval_loss": 1.5769885778427124,
"eval_runtime": 1.1503,
"eval_samples_per_second": 62.594,
"eval_steps_per_second": 1.739,
"step": 200
},
{
"epoch": 25.12,
"learning_rate": 2.7638888888888892e-05,
"loss": 1.2002,
"step": 201
},
{
"epoch": 25.25,
"learning_rate": 2.7500000000000004e-05,
"loss": 1.2582,
"step": 202
},
{
"epoch": 25.38,
"learning_rate": 2.7361111111111114e-05,
"loss": 1.3235,
"step": 203
},
{
"epoch": 25.5,
"learning_rate": 2.7222222222222223e-05,
"loss": 1.3069,
"step": 204
},
{
"epoch": 25.62,
"learning_rate": 2.7083333333333332e-05,
"loss": 1.2221,
"step": 205
},
{
"epoch": 25.75,
"learning_rate": 2.6944444444444445e-05,
"loss": 1.1905,
"step": 206
},
{
"epoch": 25.88,
"learning_rate": 2.6805555555555557e-05,
"loss": 1.1395,
"step": 207
},
{
"epoch": 26.0,
"learning_rate": 2.6666666666666667e-05,
"loss": 1.1696,
"step": 208
},
{
"epoch": 26.0,
"eval_accuracy": 0.7916666666666666,
"eval_loss": 1.5098954439163208,
"eval_runtime": 1.1834,
"eval_samples_per_second": 60.841,
"eval_steps_per_second": 1.69,
"step": 208
},
{
"epoch": 26.12,
"learning_rate": 2.652777777777778e-05,
"loss": 1.1661,
"step": 209
},
{
"epoch": 26.25,
"learning_rate": 2.6388888888888892e-05,
"loss": 1.267,
"step": 210
},
{
"epoch": 26.38,
"learning_rate": 2.625e-05,
"loss": 1.334,
"step": 211
},
{
"epoch": 26.5,
"learning_rate": 2.6111111111111114e-05,
"loss": 1.1817,
"step": 212
},
{
"epoch": 26.62,
"learning_rate": 2.5972222222222226e-05,
"loss": 1.1527,
"step": 213
},
{
"epoch": 26.75,
"learning_rate": 2.5833333333333336e-05,
"loss": 1.1049,
"step": 214
},
{
"epoch": 26.88,
"learning_rate": 2.5694444444444445e-05,
"loss": 1.2063,
"step": 215
},
{
"epoch": 27.0,
"learning_rate": 2.5555555555555554e-05,
"loss": 1.1645,
"step": 216
},
{
"epoch": 27.0,
"eval_accuracy": 0.7916666666666666,
"eval_loss": 1.4630253314971924,
"eval_runtime": 1.3424,
"eval_samples_per_second": 53.635,
"eval_steps_per_second": 1.49,
"step": 216
},
{
"epoch": 27.12,
"learning_rate": 2.5416666666666667e-05,
"loss": 1.0659,
"step": 217
},
{
"epoch": 27.25,
"learning_rate": 2.527777777777778e-05,
"loss": 1.2332,
"step": 218
},
{
"epoch": 27.38,
"learning_rate": 2.513888888888889e-05,
"loss": 1.2253,
"step": 219
},
{
"epoch": 27.5,
"learning_rate": 2.5e-05,
"loss": 1.089,
"step": 220
},
{
"epoch": 27.62,
"learning_rate": 2.4861111111111114e-05,
"loss": 1.088,
"step": 221
},
{
"epoch": 27.75,
"learning_rate": 2.4722222222222223e-05,
"loss": 1.0119,
"step": 222
},
{
"epoch": 27.88,
"learning_rate": 2.4583333333333332e-05,
"loss": 1.1187,
"step": 223
},
{
"epoch": 28.0,
"learning_rate": 2.4444444444444445e-05,
"loss": 1.0646,
"step": 224
},
{
"epoch": 28.0,
"eval_accuracy": 0.7916666666666666,
"eval_loss": 1.4988899230957031,
"eval_runtime": 1.1779,
"eval_samples_per_second": 61.127,
"eval_steps_per_second": 1.698,
"step": 224
},
{
"epoch": 28.12,
"learning_rate": 2.4305555555555558e-05,
"loss": 1.1824,
"step": 225
},
{
"epoch": 28.25,
"learning_rate": 2.4166666666666667e-05,
"loss": 1.0541,
"step": 226
},
{
"epoch": 28.38,
"learning_rate": 2.402777777777778e-05,
"loss": 1.1322,
"step": 227
},
{
"epoch": 28.5,
"learning_rate": 2.3888888888888892e-05,
"loss": 1.1507,
"step": 228
},
{
"epoch": 28.62,
"learning_rate": 2.375e-05,
"loss": 1.1121,
"step": 229
},
{
"epoch": 28.75,
"learning_rate": 2.361111111111111e-05,
"loss": 1.1407,
"step": 230
},
{
"epoch": 28.88,
"learning_rate": 2.3472222222222223e-05,
"loss": 0.99,
"step": 231
},
{
"epoch": 29.0,
"learning_rate": 2.3333333333333336e-05,
"loss": 1.0149,
"step": 232
},
{
"epoch": 29.0,
"eval_accuracy": 0.7222222222222222,
"eval_loss": 1.5569297075271606,
"eval_runtime": 1.154,
"eval_samples_per_second": 62.394,
"eval_steps_per_second": 1.733,
"step": 232
},
{
"epoch": 29.12,
"learning_rate": 2.3194444444444445e-05,
"loss": 1.1034,
"step": 233
},
{
"epoch": 29.25,
"learning_rate": 2.3055555555555558e-05,
"loss": 1.1994,
"step": 234
},
{
"epoch": 29.38,
"learning_rate": 2.2916666666666667e-05,
"loss": 1.2542,
"step": 235
},
{
"epoch": 29.5,
"learning_rate": 2.277777777777778e-05,
"loss": 1.0511,
"step": 236
},
{
"epoch": 29.62,
"learning_rate": 2.263888888888889e-05,
"loss": 1.0576,
"step": 237
},
{
"epoch": 29.75,
"learning_rate": 2.25e-05,
"loss": 1.1146,
"step": 238
},
{
"epoch": 29.88,
"learning_rate": 2.2361111111111114e-05,
"loss": 1.1259,
"step": 239
},
{
"epoch": 30.0,
"learning_rate": 2.2222222222222223e-05,
"loss": 1.0799,
"step": 240
},
{
"epoch": 30.0,
"eval_accuracy": 0.8333333333333334,
"eval_loss": 1.360217571258545,
"eval_runtime": 1.1747,
"eval_samples_per_second": 61.292,
"eval_steps_per_second": 1.703,
"step": 240
},
{
"epoch": 30.12,
"learning_rate": 2.2083333333333333e-05,
"loss": 0.99,
"step": 241
},
{
"epoch": 30.25,
"learning_rate": 2.1944444444444445e-05,
"loss": 1.1745,
"step": 242
},
{
"epoch": 30.38,
"learning_rate": 2.1805555555555558e-05,
"loss": 1.0998,
"step": 243
},
{
"epoch": 30.5,
"learning_rate": 2.1666666666666667e-05,
"loss": 1.0923,
"step": 244
},
{
"epoch": 30.62,
"learning_rate": 2.152777777777778e-05,
"loss": 1.0057,
"step": 245
},
{
"epoch": 30.75,
"learning_rate": 2.138888888888889e-05,
"loss": 1.0278,
"step": 246
},
{
"epoch": 30.88,
"learning_rate": 2.125e-05,
"loss": 0.9082,
"step": 247
},
{
"epoch": 31.0,
"learning_rate": 2.111111111111111e-05,
"loss": 0.9528,
"step": 248
},
{
"epoch": 31.0,
"eval_accuracy": 0.8472222222222222,
"eval_loss": 1.3781746625900269,
"eval_runtime": 1.1414,
"eval_samples_per_second": 63.083,
"eval_steps_per_second": 1.752,
"step": 248
},
{
"epoch": 31.12,
"learning_rate": 2.0972222222222223e-05,
"loss": 1.0382,
"step": 249
},
{
"epoch": 31.25,
"learning_rate": 2.0833333333333336e-05,
"loss": 1.0538,
"step": 250
},
{
"epoch": 31.38,
"learning_rate": 2.0694444444444445e-05,
"loss": 1.0172,
"step": 251
},
{
"epoch": 31.5,
"learning_rate": 2.0555555555555555e-05,
"loss": 1.0461,
"step": 252
},
{
"epoch": 31.62,
"learning_rate": 2.0416666666666667e-05,
"loss": 1.1038,
"step": 253
},
{
"epoch": 31.75,
"learning_rate": 2.027777777777778e-05,
"loss": 1.014,
"step": 254
},
{
"epoch": 31.88,
"learning_rate": 2.013888888888889e-05,
"loss": 1.1257,
"step": 255
},
{
"epoch": 32.0,
"learning_rate": 2e-05,
"loss": 1.0461,
"step": 256
},
{
"epoch": 32.0,
"eval_accuracy": 0.8333333333333334,
"eval_loss": 1.369834065437317,
"eval_runtime": 1.158,
"eval_samples_per_second": 62.174,
"eval_steps_per_second": 1.727,
"step": 256
},
{
"epoch": 32.12,
"learning_rate": 1.986111111111111e-05,
"loss": 0.9561,
"step": 257
},
{
"epoch": 32.25,
"learning_rate": 1.9722222222222224e-05,
"loss": 1.165,
"step": 258
},
{
"epoch": 32.38,
"learning_rate": 1.9583333333333333e-05,
"loss": 1.1076,
"step": 259
},
{
"epoch": 32.5,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.8628,
"step": 260
},
{
"epoch": 32.62,
"learning_rate": 1.9305555555555558e-05,
"loss": 0.9574,
"step": 261
},
{
"epoch": 32.75,
"learning_rate": 1.9166666666666667e-05,
"loss": 1.0838,
"step": 262
},
{
"epoch": 32.88,
"learning_rate": 1.9027777777777776e-05,
"loss": 0.9627,
"step": 263
},
{
"epoch": 33.0,
"learning_rate": 1.888888888888889e-05,
"loss": 0.9019,
"step": 264
},
{
"epoch": 33.0,
"eval_accuracy": 0.8611111111111112,
"eval_loss": 1.3250584602355957,
"eval_runtime": 1.1822,
"eval_samples_per_second": 60.901,
"eval_steps_per_second": 1.692,
"step": 264
},
{
"epoch": 33.12,
"learning_rate": 1.8750000000000002e-05,
"loss": 1.0657,
"step": 265
},
{
"epoch": 33.25,
"learning_rate": 1.861111111111111e-05,
"loss": 0.986,
"step": 266
},
{
"epoch": 33.38,
"learning_rate": 1.8472222222222224e-05,
"loss": 1.053,
"step": 267
},
{
"epoch": 33.5,
"learning_rate": 1.8333333333333333e-05,
"loss": 0.8547,
"step": 268
},
{
"epoch": 33.62,
"learning_rate": 1.8194444444444445e-05,
"loss": 0.9879,
"step": 269
},
{
"epoch": 33.75,
"learning_rate": 1.8055555555555555e-05,
"loss": 1.1052,
"step": 270
},
{
"epoch": 33.88,
"learning_rate": 1.7916666666666667e-05,
"loss": 1.0943,
"step": 271
},
{
"epoch": 34.0,
"learning_rate": 1.777777777777778e-05,
"loss": 0.9494,
"step": 272
},
{
"epoch": 34.0,
"eval_accuracy": 0.8472222222222222,
"eval_loss": 1.3585959672927856,
"eval_runtime": 1.1745,
"eval_samples_per_second": 61.303,
"eval_steps_per_second": 1.703,
"step": 272
},
{
"epoch": 34.12,
"learning_rate": 1.763888888888889e-05,
"loss": 1.0655,
"step": 273
},
{
"epoch": 34.25,
"learning_rate": 1.75e-05,
"loss": 1.0317,
"step": 274
},
{
"epoch": 34.38,
"learning_rate": 1.736111111111111e-05,
"loss": 0.9261,
"step": 275
},
{
"epoch": 34.5,
"learning_rate": 1.7222222222222224e-05,
"loss": 0.8874,
"step": 276
},
{
"epoch": 34.62,
"learning_rate": 1.7083333333333333e-05,
"loss": 0.8992,
"step": 277
},
{
"epoch": 34.75,
"learning_rate": 1.6944444444444446e-05,
"loss": 0.8334,
"step": 278
},
{
"epoch": 34.88,
"learning_rate": 1.6805555555555558e-05,
"loss": 1.0757,
"step": 279
},
{
"epoch": 35.0,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.9439,
"step": 280
},
{
"epoch": 35.0,
"eval_accuracy": 0.8333333333333334,
"eval_loss": 1.3526421785354614,
"eval_runtime": 1.1674,
"eval_samples_per_second": 61.674,
"eval_steps_per_second": 1.713,
"step": 280
},
{
"epoch": 35.12,
"learning_rate": 1.6527777777777777e-05,
"loss": 0.9677,
"step": 281
},
{
"epoch": 35.25,
"learning_rate": 1.638888888888889e-05,
"loss": 0.8912,
"step": 282
},
{
"epoch": 35.38,
"learning_rate": 1.6250000000000002e-05,
"loss": 0.9065,
"step": 283
},
{
"epoch": 35.5,
"learning_rate": 1.6111111111111115e-05,
"loss": 0.8201,
"step": 284
},
{
"epoch": 35.62,
"learning_rate": 1.597222222222222e-05,
"loss": 0.9935,
"step": 285
},
{
"epoch": 35.75,
"learning_rate": 1.5833333333333333e-05,
"loss": 0.8867,
"step": 286
},
{
"epoch": 35.88,
"learning_rate": 1.5694444444444446e-05,
"loss": 0.8713,
"step": 287
},
{
"epoch": 36.0,
"learning_rate": 1.5555555555555555e-05,
"loss": 0.9089,
"step": 288
},
{
"epoch": 36.0,
"eval_accuracy": 0.8333333333333334,
"eval_loss": 1.2727890014648438,
"eval_runtime": 1.1882,
"eval_samples_per_second": 60.598,
"eval_steps_per_second": 1.683,
"step": 288
},
{
"epoch": 36.12,
"learning_rate": 1.5416666666666668e-05,
"loss": 1.026,
"step": 289
},
{
"epoch": 36.25,
"learning_rate": 1.527777777777778e-05,
"loss": 1.0595,
"step": 290
},
{
"epoch": 36.38,
"learning_rate": 1.5138888888888888e-05,
"loss": 1.0047,
"step": 291
},
{
"epoch": 36.5,
"learning_rate": 1.5e-05,
"loss": 0.8895,
"step": 292
},
{
"epoch": 36.62,
"learning_rate": 1.4861111111111111e-05,
"loss": 0.889,
"step": 293
},
{
"epoch": 36.75,
"learning_rate": 1.4722222222222224e-05,
"loss": 0.9788,
"step": 294
},
{
"epoch": 36.88,
"learning_rate": 1.4583333333333335e-05,
"loss": 1.0459,
"step": 295
},
{
"epoch": 37.0,
"learning_rate": 1.4444444444444444e-05,
"loss": 0.8962,
"step": 296
},
{
"epoch": 37.0,
"eval_accuracy": 0.7916666666666666,
"eval_loss": 1.30055570602417,
"eval_runtime": 1.1986,
"eval_samples_per_second": 60.068,
"eval_steps_per_second": 1.669,
"step": 296
},
{
"epoch": 37.12,
"learning_rate": 1.4305555555555555e-05,
"loss": 0.8617,
"step": 297
},
{
"epoch": 37.25,
"learning_rate": 1.4166666666666668e-05,
"loss": 0.8941,
"step": 298
},
{
"epoch": 37.38,
"learning_rate": 1.4027777777777779e-05,
"loss": 0.8506,
"step": 299
},
{
"epoch": 37.5,
"learning_rate": 1.388888888888889e-05,
"loss": 0.8558,
"step": 300
},
{
"epoch": 37.62,
"learning_rate": 1.3750000000000002e-05,
"loss": 1.0617,
"step": 301
},
{
"epoch": 37.75,
"learning_rate": 1.3611111111111111e-05,
"loss": 0.9328,
"step": 302
},
{
"epoch": 37.88,
"learning_rate": 1.3472222222222222e-05,
"loss": 1.0107,
"step": 303
},
{
"epoch": 38.0,
"learning_rate": 1.3333333333333333e-05,
"loss": 0.9482,
"step": 304
},
{
"epoch": 38.0,
"eval_accuracy": 0.8611111111111112,
"eval_loss": 1.2592272758483887,
"eval_runtime": 1.1755,
"eval_samples_per_second": 61.252,
"eval_steps_per_second": 1.701,
"step": 304
},
{
"epoch": 38.12,
"learning_rate": 1.3194444444444446e-05,
"loss": 0.9592,
"step": 305
},
{
"epoch": 38.25,
"learning_rate": 1.3055555555555557e-05,
"loss": 1.0206,
"step": 306
},
{
"epoch": 38.38,
"learning_rate": 1.2916666666666668e-05,
"loss": 0.8312,
"step": 307
},
{
"epoch": 38.5,
"learning_rate": 1.2777777777777777e-05,
"loss": 0.9144,
"step": 308
},
{
"epoch": 38.62,
"learning_rate": 1.263888888888889e-05,
"loss": 0.8125,
"step": 309
},
{
"epoch": 38.75,
"learning_rate": 1.25e-05,
"loss": 0.8706,
"step": 310
},
{
"epoch": 38.88,
"learning_rate": 1.2361111111111112e-05,
"loss": 0.9144,
"step": 311
},
{
"epoch": 39.0,
"learning_rate": 1.2222222222222222e-05,
"loss": 0.8804,
"step": 312
},
{
"epoch": 39.0,
"eval_accuracy": 0.7777777777777778,
"eval_loss": 1.3526850938796997,
"eval_runtime": 1.163,
"eval_samples_per_second": 61.908,
"eval_steps_per_second": 1.72,
"step": 312
},
{
"epoch": 39.12,
"learning_rate": 1.2083333333333333e-05,
"loss": 0.9029,
"step": 313
},
{
"epoch": 39.25,
"learning_rate": 1.1944444444444446e-05,
"loss": 1.0177,
"step": 314
},
{
"epoch": 39.38,
"learning_rate": 1.1805555555555555e-05,
"loss": 0.9169,
"step": 315
},
{
"epoch": 39.5,
"learning_rate": 1.1666666666666668e-05,
"loss": 0.9165,
"step": 316
},
{
"epoch": 39.62,
"learning_rate": 1.1527777777777779e-05,
"loss": 0.8733,
"step": 317
},
{
"epoch": 39.75,
"learning_rate": 1.138888888888889e-05,
"loss": 0.8487,
"step": 318
},
{
"epoch": 39.88,
"learning_rate": 1.125e-05,
"loss": 0.9189,
"step": 319
},
{
"epoch": 40.0,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.8348,
"step": 320
},
{
"epoch": 40.0,
"eval_accuracy": 0.8055555555555556,
"eval_loss": 1.275934100151062,
"eval_runtime": 1.1713,
"eval_samples_per_second": 61.468,
"eval_steps_per_second": 1.707,
"step": 320
},
{
"epoch": 40.12,
"learning_rate": 1.0972222222222223e-05,
"loss": 0.9,
"step": 321
},
{
"epoch": 40.25,
"learning_rate": 1.0833333333333334e-05,
"loss": 0.7773,
"step": 322
},
{
"epoch": 40.38,
"learning_rate": 1.0694444444444444e-05,
"loss": 0.8709,
"step": 323
},
{
"epoch": 40.5,
"learning_rate": 1.0555555555555555e-05,
"loss": 0.8861,
"step": 324
},
{
"epoch": 40.62,
"learning_rate": 1.0416666666666668e-05,
"loss": 0.9732,
"step": 325
},
{
"epoch": 40.75,
"learning_rate": 1.0277777777777777e-05,
"loss": 0.7803,
"step": 326
},
{
"epoch": 40.88,
"learning_rate": 1.013888888888889e-05,
"loss": 0.9102,
"step": 327
},
{
"epoch": 41.0,
"learning_rate": 1e-05,
"loss": 0.7823,
"step": 328
},
{
"epoch": 41.0,
"eval_accuracy": 0.8194444444444444,
"eval_loss": 1.307101845741272,
"eval_runtime": 1.1847,
"eval_samples_per_second": 60.774,
"eval_steps_per_second": 1.688,
"step": 328
},
{
"epoch": 41.12,
"learning_rate": 9.861111111111112e-06,
"loss": 0.9279,
"step": 329
},
{
"epoch": 41.25,
"learning_rate": 9.722222222222223e-06,
"loss": 0.8138,
"step": 330
},
{
"epoch": 41.38,
"learning_rate": 9.583333333333334e-06,
"loss": 0.8626,
"step": 331
},
{
"epoch": 41.5,
"learning_rate": 9.444444444444445e-06,
"loss": 0.8504,
"step": 332
},
{
"epoch": 41.62,
"learning_rate": 9.305555555555555e-06,
"loss": 0.7799,
"step": 333
},
{
"epoch": 41.75,
"learning_rate": 9.166666666666666e-06,
"loss": 0.8985,
"step": 334
},
{
"epoch": 41.88,
"learning_rate": 9.027777777777777e-06,
"loss": 0.9914,
"step": 335
},
{
"epoch": 42.0,
"learning_rate": 8.88888888888889e-06,
"loss": 0.8944,
"step": 336
},
{
"epoch": 42.0,
"eval_accuracy": 0.8194444444444444,
"eval_loss": 1.2427630424499512,
"eval_runtime": 1.1748,
"eval_samples_per_second": 61.287,
"eval_steps_per_second": 1.702,
"step": 336
},
{
"epoch": 42.12,
"learning_rate": 8.75e-06,
"loss": 0.8499,
"step": 337
},
{
"epoch": 42.25,
"learning_rate": 8.611111111111112e-06,
"loss": 0.9826,
"step": 338
},
{
"epoch": 42.38,
"learning_rate": 8.472222222222223e-06,
"loss": 0.7537,
"step": 339
},
{
"epoch": 42.5,
"learning_rate": 8.333333333333334e-06,
"loss": 0.8476,
"step": 340
},
{
"epoch": 42.62,
"learning_rate": 8.194444444444445e-06,
"loss": 0.8031,
"step": 341
},
{
"epoch": 42.75,
"learning_rate": 8.055555555555557e-06,
"loss": 0.9647,
"step": 342
},
{
"epoch": 42.88,
"learning_rate": 7.916666666666667e-06,
"loss": 0.8912,
"step": 343
},
{
"epoch": 43.0,
"learning_rate": 7.777777777777777e-06,
"loss": 0.9677,
"step": 344
},
{
"epoch": 43.0,
"eval_accuracy": 0.7777777777777778,
"eval_loss": 1.29029381275177,
"eval_runtime": 1.16,
"eval_samples_per_second": 62.067,
"eval_steps_per_second": 1.724,
"step": 344
},
{
"epoch": 43.12,
"learning_rate": 7.63888888888889e-06,
"loss": 1.0013,
"step": 345
},
{
"epoch": 43.25,
"learning_rate": 7.5e-06,
"loss": 0.934,
"step": 346
},
{
"epoch": 43.38,
"learning_rate": 7.361111111111112e-06,
"loss": 0.9126,
"step": 347
},
{
"epoch": 43.5,
"learning_rate": 7.222222222222222e-06,
"loss": 0.8916,
"step": 348
},
{
"epoch": 43.62,
"learning_rate": 7.083333333333334e-06,
"loss": 0.8839,
"step": 349
},
{
"epoch": 43.75,
"learning_rate": 6.944444444444445e-06,
"loss": 0.8895,
"step": 350
},
{
"epoch": 43.88,
"learning_rate": 6.805555555555556e-06,
"loss": 0.8558,
"step": 351
},
{
"epoch": 44.0,
"learning_rate": 6.666666666666667e-06,
"loss": 0.9584,
"step": 352
},
{
"epoch": 44.0,
"eval_accuracy": 0.7638888888888888,
"eval_loss": 1.311880350112915,
"eval_runtime": 1.1671,
"eval_samples_per_second": 61.689,
"eval_steps_per_second": 1.714,
"step": 352
},
{
"epoch": 44.12,
"learning_rate": 6.5277777777777784e-06,
"loss": 0.81,
"step": 353
},
{
"epoch": 44.25,
"learning_rate": 6.3888888888888885e-06,
"loss": 0.8161,
"step": 354
},
{
"epoch": 44.38,
"learning_rate": 6.25e-06,
"loss": 0.7856,
"step": 355
},
{
"epoch": 44.5,
"learning_rate": 6.111111111111111e-06,
"loss": 0.6869,
"step": 356
},
{
"epoch": 44.62,
"learning_rate": 5.972222222222223e-06,
"loss": 0.841,
"step": 357
},
{
"epoch": 44.75,
"learning_rate": 5.833333333333334e-06,
"loss": 0.8728,
"step": 358
},
{
"epoch": 44.88,
"learning_rate": 5.694444444444445e-06,
"loss": 1.0075,
"step": 359
},
{
"epoch": 45.0,
"learning_rate": 5.555555555555556e-06,
"loss": 0.8342,
"step": 360
},
{
"epoch": 45.0,
"eval_accuracy": 0.7777777777777778,
"eval_loss": 1.3501887321472168,
"eval_runtime": 1.1764,
"eval_samples_per_second": 61.203,
"eval_steps_per_second": 1.7,
"step": 360
},
{
"epoch": 45.12,
"learning_rate": 5.416666666666667e-06,
"loss": 0.8549,
"step": 361
},
{
"epoch": 45.25,
"learning_rate": 5.277777777777778e-06,
"loss": 0.8518,
"step": 362
},
{
"epoch": 45.38,
"learning_rate": 5.138888888888889e-06,
"loss": 0.8303,
"step": 363
},
{
"epoch": 45.5,
"learning_rate": 5e-06,
"loss": 0.8378,
"step": 364
},
{
"epoch": 45.62,
"learning_rate": 4.861111111111111e-06,
"loss": 0.8211,
"step": 365
},
{
"epoch": 45.75,
"learning_rate": 4.722222222222222e-06,
"loss": 0.8094,
"step": 366
},
{
"epoch": 45.88,
"learning_rate": 4.583333333333333e-06,
"loss": 0.7646,
"step": 367
},
{
"epoch": 46.0,
"learning_rate": 4.444444444444445e-06,
"loss": 0.7878,
"step": 368
},
{
"epoch": 46.0,
"eval_accuracy": 0.8333333333333334,
"eval_loss": 1.1940851211547852,
"eval_runtime": 1.3426,
"eval_samples_per_second": 53.626,
"eval_steps_per_second": 1.49,
"step": 368
},
{
"epoch": 46.12,
"learning_rate": 4.305555555555556e-06,
"loss": 0.8782,
"step": 369
},
{
"epoch": 46.25,
"learning_rate": 4.166666666666667e-06,
"loss": 0.7709,
"step": 370
},
{
"epoch": 46.38,
"learning_rate": 4.027777777777779e-06,
"loss": 0.8922,
"step": 371
},
{
"epoch": 46.5,
"learning_rate": 3.888888888888889e-06,
"loss": 0.8539,
"step": 372
},
{
"epoch": 46.62,
"learning_rate": 3.75e-06,
"loss": 0.7969,
"step": 373
},
{
"epoch": 46.75,
"learning_rate": 3.611111111111111e-06,
"loss": 0.8109,
"step": 374
},
{
"epoch": 46.88,
"learning_rate": 3.4722222222222224e-06,
"loss": 0.8346,
"step": 375
},
{
"epoch": 47.0,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.7817,
"step": 376
},
{
"epoch": 47.0,
"eval_accuracy": 0.8055555555555556,
"eval_loss": 1.2670094966888428,
"eval_runtime": 1.1931,
"eval_samples_per_second": 60.348,
"eval_steps_per_second": 1.676,
"step": 376
},
{
"epoch": 47.12,
"learning_rate": 3.1944444444444443e-06,
"loss": 0.7965,
"step": 377
},
{
"epoch": 47.25,
"learning_rate": 3.0555555555555556e-06,
"loss": 0.7273,
"step": 378
},
{
"epoch": 47.38,
"learning_rate": 2.916666666666667e-06,
"loss": 0.8002,
"step": 379
},
{
"epoch": 47.5,
"learning_rate": 2.777777777777778e-06,
"loss": 0.8555,
"step": 380
},
{
"epoch": 47.62,
"learning_rate": 2.638888888888889e-06,
"loss": 0.8853,
"step": 381
},
{
"epoch": 47.75,
"learning_rate": 2.5e-06,
"loss": 0.9409,
"step": 382
},
{
"epoch": 47.88,
"learning_rate": 2.361111111111111e-06,
"loss": 0.8717,
"step": 383
},
{
"epoch": 48.0,
"learning_rate": 2.2222222222222225e-06,
"loss": 0.812,
"step": 384
},
{
"epoch": 48.0,
"eval_accuracy": 0.8194444444444444,
"eval_loss": 1.2068408727645874,
"eval_runtime": 1.1942,
"eval_samples_per_second": 60.292,
"eval_steps_per_second": 1.675,
"step": 384
},
{
"epoch": 48.12,
"learning_rate": 2.0833333333333334e-06,
"loss": 0.8121,
"step": 385
},
{
"epoch": 48.25,
"learning_rate": 1.9444444444444444e-06,
"loss": 0.8815,
"step": 386
},
{
"epoch": 48.38,
"learning_rate": 1.8055555555555555e-06,
"loss": 0.8755,
"step": 387
},
{
"epoch": 48.5,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.7822,
"step": 388
},
{
"epoch": 48.62,
"learning_rate": 1.5277777777777778e-06,
"loss": 0.827,
"step": 389
},
{
"epoch": 48.75,
"learning_rate": 1.388888888888889e-06,
"loss": 0.7554,
"step": 390
},
{
"epoch": 48.88,
"learning_rate": 1.25e-06,
"loss": 0.9329,
"step": 391
},
{
"epoch": 49.0,
"learning_rate": 1.1111111111111112e-06,
"loss": 0.9714,
"step": 392
},
{
"epoch": 49.0,
"eval_accuracy": 0.75,
"eval_loss": 1.3479645252227783,
"eval_runtime": 1.1626,
"eval_samples_per_second": 61.932,
"eval_steps_per_second": 1.72,
"step": 392
},
{
"epoch": 49.12,
"learning_rate": 9.722222222222222e-07,
"loss": 1.0015,
"step": 393
},
{
"epoch": 49.25,
"learning_rate": 8.333333333333333e-07,
"loss": 0.814,
"step": 394
},
{
"epoch": 49.38,
"learning_rate": 6.944444444444445e-07,
"loss": 1.0259,
"step": 395
},
{
"epoch": 49.5,
"learning_rate": 5.555555555555556e-07,
"loss": 0.7199,
"step": 396
},
{
"epoch": 49.62,
"learning_rate": 4.1666666666666667e-07,
"loss": 0.7806,
"step": 397
},
{
"epoch": 49.75,
"learning_rate": 2.777777777777778e-07,
"loss": 0.915,
"step": 398
},
{
"epoch": 49.88,
"learning_rate": 1.388888888888889e-07,
"loss": 0.8828,
"step": 399
},
{
"epoch": 50.0,
"learning_rate": 0.0,
"loss": 0.9362,
"step": 400
},
{
"epoch": 50.0,
"eval_accuracy": 0.7083333333333334,
"eval_loss": 1.4028462171554565,
"eval_runtime": 1.1672,
"eval_samples_per_second": 61.685,
"eval_steps_per_second": 1.713,
"step": 400
},
{
"epoch": 50.0,
"step": 400,
"total_flos": 1.9532356960886784e+18,
"train_loss": 1.5779063104093074,
"train_runtime": 1340.4309,
"train_samples_per_second": 18.8,
"train_steps_per_second": 0.298
}
],
"logging_steps": 1,
"max_steps": 400,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 1.9532356960886784e+18,
"trial_name": null,
"trial_params": null
}