{
  "best_metric": 0.9180364705120548,
  "best_model_checkpoint": "../save/marinda-default-ep0.4-26m-ti-all-O2-shuffle/checkpoint-700",
  "epoch": 4.761904761904762,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 9.972789115646259e-05,
      "loss": 3.7949,
      "step": 2
    },
    {
      "epoch": 0.03,
      "learning_rate": 9.945578231292518e-05,
      "loss": 3.3775,
      "step": 4
    },
    {
      "epoch": 0.04,
      "learning_rate": 9.918367346938776e-05,
      "loss": 2.5674,
      "step": 6
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.891156462585035e-05,
      "loss": 1.3071,
      "step": 8
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.863945578231294e-05,
      "loss": 0.5747,
      "step": 10
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.836734693877552e-05,
      "loss": 0.4116,
      "step": 12
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.80952380952381e-05,
      "loss": 0.387,
      "step": 14
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.782312925170068e-05,
      "loss": 0.3709,
      "step": 16
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.755102040816328e-05,
      "loss": 0.3384,
      "step": 18
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.727891156462585e-05,
      "loss": 0.3449,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.700680272108844e-05,
      "loss": 0.31,
      "step": 22
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.673469387755102e-05,
      "loss": 0.2991,
      "step": 24
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.646258503401361e-05,
      "loss": 0.2726,
      "step": 26
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.61904761904762e-05,
      "loss": 0.2695,
      "step": 28
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.591836734693878e-05,
      "loss": 0.2748,
      "step": 30
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.564625850340137e-05,
      "loss": 0.2599,
      "step": 32
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.537414965986395e-05,
      "loss": 0.237,
      "step": 34
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.510204081632653e-05,
      "loss": 0.2351,
      "step": 36
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.482993197278913e-05,
      "loss": 0.2282,
      "step": 38
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.455782312925171e-05,
      "loss": 0.2113,
      "step": 40
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.428571428571429e-05,
      "loss": 0.2064,
      "step": 42
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.401360544217687e-05,
      "loss": 0.2039,
      "step": 44
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.374149659863946e-05,
      "loss": 0.1884,
      "step": 46
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.346938775510204e-05,
      "loss": 0.1805,
      "step": 48
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.319727891156463e-05,
      "loss": 0.1864,
      "step": 50
    },
    {
      "epoch": 0.34,
      "eval_f1": 0.34935683381798105,
      "eval_loss": 0.16132193803787231,
      "eval_precision": 0.4228036171158445,
      "eval_recall": 0.29765784863362554,
      "eval_runtime": 11.4327,
      "eval_samples_per_second": 129.628,
      "eval_steps_per_second": 0.525,
      "step": 50
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.292517006802722e-05,
      "loss": 0.1711,
      "step": 52
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.26530612244898e-05,
      "loss": 0.1747,
      "step": 54
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.238095238095239e-05,
      "loss": 0.1688,
      "step": 56
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.210884353741497e-05,
      "loss": 0.1676,
      "step": 58
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.183673469387756e-05,
      "loss": 0.1564,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.156462585034013e-05,
      "loss": 0.1585,
      "step": 62
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.129251700680272e-05,
      "loss": 0.1452,
      "step": 64
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.102040816326532e-05,
      "loss": 0.1384,
      "step": 66
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.074829931972789e-05,
      "loss": 0.1508,
      "step": 68
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.047619047619048e-05,
      "loss": 0.1359,
      "step": 70
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.020408163265308e-05,
      "loss": 0.1475,
      "step": 72
    },
    {
      "epoch": 0.5,
      "learning_rate": 8.993197278911565e-05,
      "loss": 0.1259,
      "step": 74
    },
    {
      "epoch": 0.52,
      "learning_rate": 8.965986394557823e-05,
      "loss": 0.1328,
      "step": 76
    },
    {
      "epoch": 0.53,
      "learning_rate": 8.938775510204082e-05,
      "loss": 0.1334,
      "step": 78
    },
    {
      "epoch": 0.54,
      "learning_rate": 8.911564625850341e-05,
      "loss": 0.1244,
      "step": 80
    },
    {
      "epoch": 0.56,
      "learning_rate": 8.884353741496599e-05,
      "loss": 0.1247,
      "step": 82
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.857142857142857e-05,
      "loss": 0.1168,
      "step": 84
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.829931972789117e-05,
      "loss": 0.1216,
      "step": 86
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.802721088435375e-05,
      "loss": 0.1245,
      "step": 88
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.775510204081632e-05,
      "loss": 0.1167,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.748299319727892e-05,
      "loss": 0.0986,
      "step": 92
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.72108843537415e-05,
      "loss": 0.1065,
      "step": 94
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.693877551020408e-05,
      "loss": 0.1121,
      "step": 96
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.666666666666667e-05,
      "loss": 0.1101,
      "step": 98
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.639455782312925e-05,
      "loss": 0.0995,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_f1": 0.651423571104831,
      "eval_loss": 0.09403873980045319,
      "eval_precision": 0.6548211682202488,
      "eval_recall": 0.6480709462229433,
      "eval_runtime": 11.2473,
      "eval_samples_per_second": 131.765,
      "eval_steps_per_second": 0.533,
      "step": 100
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.612244897959184e-05,
      "loss": 0.1104,
      "step": 102
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.585034013605443e-05,
      "loss": 0.1175,
      "step": 104
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.557823129251701e-05,
      "loss": 0.1007,
      "step": 106
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.53061224489796e-05,
      "loss": 0.0911,
      "step": 108
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.503401360544217e-05,
      "loss": 0.0969,
      "step": 110
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.476190476190477e-05,
      "loss": 0.0924,
      "step": 112
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.448979591836736e-05,
      "loss": 0.0949,
      "step": 114
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.421768707482993e-05,
      "loss": 0.0858,
      "step": 116
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.394557823129252e-05,
      "loss": 0.0945,
      "step": 118
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.367346938775511e-05,
      "loss": 0.0961,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.340136054421769e-05,
      "loss": 0.0913,
      "step": 122
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.312925170068027e-05,
      "loss": 0.0845,
      "step": 124
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.285714285714287e-05,
      "loss": 0.0873,
      "step": 126
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.258503401360545e-05,
      "loss": 0.0837,
      "step": 128
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.231292517006803e-05,
      "loss": 0.0779,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.204081632653062e-05,
      "loss": 0.0839,
      "step": 132
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.17687074829932e-05,
      "loss": 0.0872,
      "step": 134
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.149659863945579e-05,
      "loss": 0.0882,
      "step": 136
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.122448979591836e-05,
      "loss": 0.0829,
      "step": 138
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.095238095238096e-05,
      "loss": 0.0726,
      "step": 140
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.068027210884355e-05,
      "loss": 0.0828,
      "step": 142
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.040816326530612e-05,
      "loss": 0.0872,
      "step": 144
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.013605442176872e-05,
      "loss": 0.0838,
      "step": 146
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.986394557823129e-05,
      "loss": 0.0763,
      "step": 148
    },
    {
      "epoch": 1.02,
      "learning_rate": 7.959183673469388e-05,
      "loss": 0.076,
      "step": 150
    },
    {
      "epoch": 1.02,
      "eval_f1": 0.7706927100252173,
      "eval_loss": 0.06672389805316925,
      "eval_precision": 0.7788096890497564,
      "eval_recall": 0.7627529744843834,
      "eval_runtime": 11.0605,
      "eval_samples_per_second": 133.991,
      "eval_steps_per_second": 0.542,
      "step": 150
    },
    {
      "epoch": 1.03,
      "learning_rate": 7.931972789115646e-05,
      "loss": 0.0737,
      "step": 152
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.904761904761905e-05,
      "loss": 0.0597,
      "step": 154
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.877551020408164e-05,
      "loss": 0.0746,
      "step": 156
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.850340136054422e-05,
      "loss": 0.0739,
      "step": 158
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.823129251700681e-05,
      "loss": 0.0791,
      "step": 160
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.79591836734694e-05,
      "loss": 0.0784,
      "step": 162
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.768707482993197e-05,
      "loss": 0.0731,
      "step": 164
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.741496598639457e-05,
      "loss": 0.0684,
      "step": 166
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.714285714285715e-05,
      "loss": 0.0576,
      "step": 168
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.687074829931973e-05,
      "loss": 0.0682,
      "step": 170
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.659863945578231e-05,
      "loss": 0.0725,
      "step": 172
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.632653061224491e-05,
      "loss": 0.0692,
      "step": 174
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.605442176870748e-05,
      "loss": 0.069,
      "step": 176
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.578231292517007e-05,
      "loss": 0.0702,
      "step": 178
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.551020408163266e-05,
      "loss": 0.0629,
      "step": 180
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.523809523809524e-05,
      "loss": 0.0679,
      "step": 182
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.496598639455783e-05,
      "loss": 0.0601,
      "step": 184
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.469387755102041e-05,
      "loss": 0.0681,
      "step": 186
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.4421768707483e-05,
      "loss": 0.0596,
      "step": 188
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.414965986394559e-05,
      "loss": 0.0693,
      "step": 190
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.387755102040816e-05,
      "loss": 0.0653,
      "step": 192
    },
    {
      "epoch": 1.32,
      "learning_rate": 7.360544217687076e-05,
      "loss": 0.0541,
      "step": 194
    },
    {
      "epoch": 1.33,
      "learning_rate": 7.333333333333333e-05,
      "loss": 0.0639,
      "step": 196
    },
    {
      "epoch": 1.35,
      "learning_rate": 7.306122448979592e-05,
      "loss": 0.0568,
      "step": 198
    },
    {
      "epoch": 1.36,
      "learning_rate": 7.27891156462585e-05,
      "loss": 0.0613,
      "step": 200
    },
    {
      "epoch": 1.36,
      "eval_f1": 0.8183980412867451,
      "eval_loss": 0.053818028420209885,
      "eval_precision": 0.8210879676347845,
      "eval_recall": 0.8157356167545399,
      "eval_runtime": 10.9366,
      "eval_samples_per_second": 135.509,
      "eval_steps_per_second": 0.549,
      "step": 200
    },
    {
      "epoch": 1.37,
      "learning_rate": 7.251700680272109e-05,
      "loss": 0.0601,
      "step": 202
    },
    {
      "epoch": 1.39,
      "learning_rate": 7.224489795918368e-05,
      "loss": 0.071,
      "step": 204
    },
    {
      "epoch": 1.4,
      "learning_rate": 7.197278911564626e-05,
      "loss": 0.0623,
      "step": 206
    },
    {
      "epoch": 1.41,
      "learning_rate": 7.170068027210885e-05,
      "loss": 0.0579,
      "step": 208
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.142857142857143e-05,
      "loss": 0.0622,
      "step": 210
    },
    {
      "epoch": 1.44,
      "learning_rate": 7.1156462585034e-05,
      "loss": 0.0558,
      "step": 212
    },
    {
      "epoch": 1.46,
      "learning_rate": 7.08843537414966e-05,
      "loss": 0.0584,
      "step": 214
    },
    {
      "epoch": 1.47,
      "learning_rate": 7.061224489795919e-05,
      "loss": 0.0537,
      "step": 216
    },
    {
      "epoch": 1.48,
      "learning_rate": 7.034013605442176e-05,
      "loss": 0.0485,
      "step": 218
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.006802721088435e-05,
      "loss": 0.0617,
      "step": 220
    },
    {
      "epoch": 1.51,
      "learning_rate": 6.979591836734695e-05,
      "loss": 0.0548,
      "step": 222
    },
    {
      "epoch": 1.52,
      "learning_rate": 6.952380952380952e-05,
      "loss": 0.0497,
      "step": 224
    },
    {
      "epoch": 1.54,
      "learning_rate": 6.925170068027211e-05,
      "loss": 0.0519,
      "step": 226
    },
    {
      "epoch": 1.55,
      "learning_rate": 6.897959183673471e-05,
      "loss": 0.0531,
      "step": 228
    },
    {
      "epoch": 1.56,
      "learning_rate": 6.870748299319728e-05,
      "loss": 0.053,
      "step": 230
    },
    {
      "epoch": 1.58,
      "learning_rate": 6.843537414965987e-05,
      "loss": 0.0558,
      "step": 232
    },
    {
      "epoch": 1.59,
      "learning_rate": 6.816326530612245e-05,
      "loss": 0.0508,
      "step": 234
    },
    {
      "epoch": 1.61,
      "learning_rate": 6.789115646258504e-05,
      "loss": 0.059,
      "step": 236
    },
    {
      "epoch": 1.62,
      "learning_rate": 6.761904761904763e-05,
      "loss": 0.0566,
      "step": 238
    },
    {
      "epoch": 1.63,
      "learning_rate": 6.73469387755102e-05,
      "loss": 0.0504,
      "step": 240
    },
    {
      "epoch": 1.65,
      "learning_rate": 6.70748299319728e-05,
      "loss": 0.0569,
      "step": 242
    },
    {
      "epoch": 1.66,
      "learning_rate": 6.680272108843538e-05,
      "loss": 0.0504,
      "step": 244
    },
    {
      "epoch": 1.67,
      "learning_rate": 6.653061224489796e-05,
      "loss": 0.0492,
      "step": 246
    },
    {
      "epoch": 1.69,
      "learning_rate": 6.625850340136056e-05,
      "loss": 0.0477,
      "step": 248
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.598639455782313e-05,
      "loss": 0.0531,
      "step": 250
    },
    {
      "epoch": 1.7,
      "eval_f1": 0.8437027128569305,
      "eval_loss": 0.0455990694463253,
      "eval_precision": 0.836850346106293,
      "eval_recall": 0.8506783894105371,
      "eval_runtime": 11.2699,
      "eval_samples_per_second": 131.501,
      "eval_steps_per_second": 0.532,
      "step": 250
    },
    {
      "epoch": 1.71,
      "learning_rate": 6.571428571428571e-05,
      "loss": 0.053,
      "step": 252
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.54421768707483e-05,
      "loss": 0.0451,
      "step": 254
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.517006802721089e-05,
      "loss": 0.0366,
      "step": 256
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.489795918367347e-05,
      "loss": 0.0465,
      "step": 258
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.462585034013606e-05,
      "loss": 0.051,
      "step": 260
    },
    {
      "epoch": 1.78,
      "learning_rate": 6.435374149659864e-05,
      "loss": 0.0484,
      "step": 262
    },
    {
      "epoch": 1.8,
      "learning_rate": 6.408163265306123e-05,
      "loss": 0.0503,
      "step": 264
    },
    {
      "epoch": 1.81,
      "learning_rate": 6.38095238095238e-05,
      "loss": 0.0518,
      "step": 266
    },
    {
      "epoch": 1.82,
      "learning_rate": 6.35374149659864e-05,
      "loss": 0.0471,
      "step": 268
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.326530612244899e-05,
      "loss": 0.0519,
      "step": 270
    },
    {
      "epoch": 1.85,
      "learning_rate": 6.299319727891156e-05,
      "loss": 0.0475,
      "step": 272
    },
    {
      "epoch": 1.86,
      "learning_rate": 6.272108843537415e-05,
      "loss": 0.0532,
      "step": 274
    },
    {
      "epoch": 1.88,
      "learning_rate": 6.244897959183675e-05,
      "loss": 0.0572,
      "step": 276
    },
    {
      "epoch": 1.89,
      "learning_rate": 6.217687074829932e-05,
      "loss": 0.0521,
      "step": 278
    },
    {
      "epoch": 1.9,
      "learning_rate": 6.19047619047619e-05,
      "loss": 0.0499,
      "step": 280
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.163265306122449e-05,
      "loss": 0.0453,
      "step": 282
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.136054421768708e-05,
      "loss": 0.0484,
      "step": 284
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.108843537414966e-05,
      "loss": 0.0477,
      "step": 286
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.081632653061224e-05,
      "loss": 0.0457,
      "step": 288
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.0544217687074836e-05,
      "loss": 0.0425,
      "step": 290
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.027210884353742e-05,
      "loss": 0.0487,
      "step": 292
    },
    {
      "epoch": 2.0,
      "learning_rate": 6e-05,
      "loss": 0.0499,
      "step": 294
    },
    {
      "epoch": 2.01,
      "learning_rate": 5.972789115646259e-05,
      "loss": 0.0428,
      "step": 296
    },
    {
      "epoch": 2.03,
      "learning_rate": 5.945578231292517e-05,
      "loss": 0.047,
      "step": 298
    },
    {
      "epoch": 2.04,
      "learning_rate": 5.918367346938776e-05,
      "loss": 0.0414,
      "step": 300
    },
    {
      "epoch": 2.04,
      "eval_f1": 0.8653020451693569,
      "eval_loss": 0.039749592542648315,
      "eval_precision": 0.8607319627563133,
      "eval_recall": 0.8699310233685053,
      "eval_runtime": 11.2277,
      "eval_samples_per_second": 131.995,
      "eval_steps_per_second": 0.534,
      "step": 300
    },
    {
      "epoch": 2.05,
      "learning_rate": 5.8911564625850346e-05,
      "loss": 0.0381,
      "step": 302
    },
    {
      "epoch": 2.07,
      "learning_rate": 5.8639455782312925e-05,
      "loss": 0.0445,
      "step": 304
    },
    {
      "epoch": 2.08,
      "learning_rate": 5.836734693877551e-05,
      "loss": 0.0413,
      "step": 306
    },
    {
      "epoch": 2.1,
      "learning_rate": 5.8095238095238104e-05,
      "loss": 0.0444,
      "step": 308
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.782312925170068e-05,
      "loss": 0.0489,
      "step": 310
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.755102040816327e-05,
      "loss": 0.0465,
      "step": 312
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.727891156462585e-05,
      "loss": 0.0379,
      "step": 314
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7006802721088435e-05,
      "loss": 0.0307,
      "step": 316
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.673469387755103e-05,
      "loss": 0.0484,
      "step": 318
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.646258503401361e-05,
      "loss": 0.04,
      "step": 320
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.619047619047619e-05,
      "loss": 0.0368,
      "step": 322
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.5918367346938786e-05,
      "loss": 0.0387,
      "step": 324
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.564625850340136e-05,
      "loss": 0.0357,
      "step": 326
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.537414965986395e-05,
      "loss": 0.0364,
      "step": 328
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.510204081632653e-05,
      "loss": 0.0404,
      "step": 330
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.4829931972789117e-05,
      "loss": 0.0379,
      "step": 332
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.455782312925171e-05,
      "loss": 0.0472,
      "step": 334
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.428571428571428e-05,
      "loss": 0.0394,
      "step": 336
    },
    {
      "epoch": 2.3,
      "learning_rate": 5.4013605442176875e-05,
      "loss": 0.0389,
      "step": 338
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.374149659863946e-05,
      "loss": 0.0362,
      "step": 340
    },
    {
      "epoch": 2.33,
      "learning_rate": 5.346938775510204e-05,
      "loss": 0.0359,
      "step": 342
    },
    {
      "epoch": 2.34,
      "learning_rate": 5.319727891156463e-05,
      "loss": 0.0356,
      "step": 344
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.292517006802722e-05,
      "loss": 0.0408,
      "step": 346
    },
    {
      "epoch": 2.37,
      "learning_rate": 5.26530612244898e-05,
      "loss": 0.0382,
      "step": 348
    },
    {
      "epoch": 2.38,
      "learning_rate": 5.2380952380952384e-05,
      "loss": 0.0315,
      "step": 350
    },
    {
      "epoch": 2.38,
      "eval_f1": 0.8847392138215896,
      "eval_loss": 0.0348842516541481,
      "eval_precision": 0.8886000451956572,
      "eval_recall": 0.8809217002342745,
      "eval_runtime": 11.2848,
      "eval_samples_per_second": 131.327,
      "eval_steps_per_second": 0.532,
      "step": 350
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.2108843537414964e-05,
      "loss": 0.0381,
      "step": 352
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.1836734693877557e-05,
      "loss": 0.0356,
      "step": 354
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.156462585034014e-05,
      "loss": 0.041,
      "step": 356
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.129251700680272e-05,
      "loss": 0.0387,
      "step": 358
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.102040816326531e-05,
      "loss": 0.0364,
      "step": 360
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.07482993197279e-05,
      "loss": 0.0342,
      "step": 362
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.047619047619048e-05,
      "loss": 0.04,
      "step": 364
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.0204081632653066e-05,
      "loss": 0.039,
      "step": 366
    },
    {
      "epoch": 2.5,
      "learning_rate": 4.993197278911565e-05,
      "loss": 0.0434,
      "step": 368
    },
    {
      "epoch": 2.52,
      "learning_rate": 4.965986394557823e-05,
      "loss": 0.0346,
      "step": 370
    },
    {
      "epoch": 2.53,
      "learning_rate": 4.938775510204082e-05,
      "loss": 0.0423,
      "step": 372
    },
    {
      "epoch": 2.54,
      "learning_rate": 4.9115646258503404e-05,
      "loss": 0.0368,
      "step": 374
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.884353741496599e-05,
      "loss": 0.0382,
      "step": 376
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.8571428571428576e-05,
      "loss": 0.0387,
      "step": 378
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.8299319727891155e-05,
      "loss": 0.0394,
      "step": 380
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.802721088435375e-05,
      "loss": 0.0364,
      "step": 382
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.775510204081633e-05,
      "loss": 0.0465,
      "step": 384
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.7482993197278913e-05,
      "loss": 0.0269,
      "step": 386
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.72108843537415e-05,
      "loss": 0.0391,
      "step": 388
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.6938775510204086e-05,
      "loss": 0.0343,
      "step": 390
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.0358,
      "step": 392
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.639455782312925e-05,
      "loss": 0.0409,
      "step": 394
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.612244897959184e-05,
      "loss": 0.0273,
      "step": 396
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.585034013605442e-05,
      "loss": 0.0336,
      "step": 398
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.557823129251701e-05,
      "loss": 0.0318,
      "step": 400
    },
    {
      "epoch": 2.72,
      "eval_f1": 0.8934494072355922,
      "eval_loss": 0.03241937607526779,
      "eval_precision": 0.8919694788154646,
      "eval_recall": 0.8949442879595663,
      "eval_runtime": 11.1708,
      "eval_samples_per_second": 132.667,
      "eval_steps_per_second": 0.537,
      "step": 400
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.5306122448979595e-05,
      "loss": 0.0377,
      "step": 402
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.5034013605442174e-05,
      "loss": 0.0335,
      "step": 404
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.476190476190477e-05,
      "loss": 0.0326,
      "step": 406
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.448979591836735e-05,
      "loss": 0.0422,
      "step": 408
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.421768707482993e-05,
      "loss": 0.0329,
      "step": 410
    },
    {
      "epoch": 2.8,
      "learning_rate": 4.394557823129252e-05,
      "loss": 0.0452,
      "step": 412
    },
    {
      "epoch": 2.82,
      "learning_rate": 4.3673469387755105e-05,
      "loss": 0.0325,
      "step": 414
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.340136054421769e-05,
      "loss": 0.0355,
      "step": 416
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.312925170068027e-05,
      "loss": 0.0333,
      "step": 418
    },
    {
      "epoch": 2.86,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.0348,
      "step": 420
    },
    {
      "epoch": 2.87,
      "learning_rate": 4.258503401360545e-05,
      "loss": 0.0332,
      "step": 422
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.231292517006803e-05,
      "loss": 0.0336,
      "step": 424
    },
    {
      "epoch": 2.9,
      "learning_rate": 4.2040816326530615e-05,
      "loss": 0.0363,
      "step": 426
    },
    {
      "epoch": 2.91,
      "learning_rate": 4.1768707482993194e-05,
      "loss": 0.0303,
      "step": 428
    },
    {
      "epoch": 2.93,
      "learning_rate": 4.149659863945579e-05,
      "loss": 0.0308,
      "step": 430
    },
    {
      "epoch": 2.94,
      "learning_rate": 4.122448979591837e-05,
      "loss": 0.0362,
      "step": 432
    },
    {
      "epoch": 2.95,
      "learning_rate": 4.095238095238095e-05,
      "loss": 0.0385,
      "step": 434
    },
    {
      "epoch": 2.97,
      "learning_rate": 4.068027210884354e-05,
      "loss": 0.0341,
      "step": 436
    },
    {
      "epoch": 2.98,
      "learning_rate": 4.0408163265306124e-05,
      "loss": 0.0327,
      "step": 438
    },
    {
      "epoch": 2.99,
      "learning_rate": 4.013605442176871e-05,
      "loss": 0.0365,
      "step": 440
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.9863945578231296e-05,
      "loss": 0.0373,
      "step": 442
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.9591836734693876e-05,
      "loss": 0.0305,
      "step": 444
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.931972789115647e-05,
      "loss": 0.0248,
      "step": 446
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.904761904761905e-05,
      "loss": 0.0333,
      "step": 448
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.8775510204081634e-05,
      "loss": 0.0335,
      "step": 450
    },
    {
      "epoch": 3.06,
      "eval_f1": 0.9027128432459335,
      "eval_loss": 0.03037634864449501,
      "eval_precision": 0.9016257082029295,
      "eval_recall": 0.9038126272236696,
      "eval_runtime": 10.8474,
      "eval_samples_per_second": 136.623,
      "eval_steps_per_second": 0.553,
      "step": 450
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.850340136054422e-05,
      "loss": 0.0303,
      "step": 452
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.8231292517006806e-05,
      "loss": 0.0337,
      "step": 454
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.795918367346939e-05,
      "loss": 0.0298,
      "step": 456
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.768707482993197e-05,
      "loss": 0.0274,
      "step": 458
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.7414965986394564e-05,
      "loss": 0.0282,
      "step": 460
    },
    {
      "epoch": 3.14,
      "learning_rate": 3.7142857142857143e-05,
      "loss": 0.0279,
      "step": 462
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.687074829931973e-05,
      "loss": 0.035,
      "step": 464
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.6598639455782316e-05,
      "loss": 0.0396,
      "step": 466
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.63265306122449e-05,
      "loss": 0.0291,
      "step": 468
    },
    {
      "epoch": 3.2,
      "learning_rate": 3.605442176870749e-05,
      "loss": 0.0323,
      "step": 470
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.578231292517007e-05,
      "loss": 0.0303,
      "step": 472
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.551020408163265e-05,
      "loss": 0.0304,
      "step": 474
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.523809523809524e-05,
      "loss": 0.0339,
      "step": 476
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.4965986394557825e-05,
      "loss": 0.0326,
      "step": 478
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.469387755102041e-05,
      "loss": 0.0287,
      "step": 480
    },
    {
      "epoch": 3.28,
      "learning_rate": 3.442176870748299e-05,
      "loss": 0.0273,
      "step": 482
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.4149659863945583e-05,
      "loss": 0.031,
      "step": 484
    },
    {
      "epoch": 3.31,
      "learning_rate": 3.387755102040816e-05,
      "loss": 0.0274,
      "step": 486
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.360544217687075e-05,
      "loss": 0.0314,
      "step": 488
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0238,
      "step": 490
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.306122448979592e-05,
      "loss": 0.0301,
      "step": 492
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.278911564625851e-05,
      "loss": 0.0326,
      "step": 494
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.2517006802721086e-05,
      "loss": 0.0322,
      "step": 496
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.224489795918367e-05,
      "loss": 0.0265,
      "step": 498
    },
    {
      "epoch": 3.4,
      "learning_rate": 3.1972789115646265e-05,
      "loss": 0.0302,
      "step": 500
    },
    {
      "epoch": 3.4,
      "eval_f1": 0.9087960923681575,
      "eval_loss": 0.02837410755455494,
      "eval_precision": 0.9100790506916395,
      "eval_recall": 0.9075267180265848,
      "eval_runtime": 10.84,
      "eval_samples_per_second": 136.716,
      "eval_steps_per_second": 0.554,
      "step": 500
    },
    {
      "epoch": 3.41,
      "learning_rate": 3.1700680272108845e-05,
      "loss": 0.03,
      "step": 502
    },
    {
      "epoch": 3.43,
      "learning_rate": 3.142857142857143e-05,
      "loss": 0.0302,
      "step": 504
    },
    {
      "epoch": 3.44,
      "learning_rate": 3.115646258503401e-05,
      "loss": 0.0257,
      "step": 506
    },
    {
      "epoch": 3.46,
      "learning_rate": 3.08843537414966e-05,
      "loss": 0.0288,
      "step": 508
    },
    {
      "epoch": 3.47,
      "learning_rate": 3.061224489795919e-05,
      "loss": 0.0349,
      "step": 510
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.0340136054421768e-05,
      "loss": 0.0303,
      "step": 512
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.0068027210884354e-05,
      "loss": 0.0309,
      "step": 514
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.9795918367346944e-05,
      "loss": 0.0271,
      "step": 516
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.9523809523809526e-05,
      "loss": 0.0291,
      "step": 518
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.925170068027211e-05,
      "loss": 0.0304,
      "step": 520
    },
    {
      "epoch": 3.55,
      "learning_rate": 2.8979591836734692e-05,
      "loss": 0.0345,
      "step": 522
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.870748299319728e-05,
      "loss": 0.0276,
      "step": 524
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.8435374149659867e-05,
      "loss": 0.0299,
      "step": 526
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.816326530612245e-05,
      "loss": 0.0269,
      "step": 528
    },
    {
      "epoch": 3.61,
      "learning_rate": 2.7891156462585033e-05,
      "loss": 0.0281,
      "step": 530
    },
    {
      "epoch": 3.62,
      "learning_rate": 2.7619047619047622e-05,
      "loss": 0.0329,
      "step": 532
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.7346938775510205e-05,
      "loss": 0.028,
      "step": 534
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.707482993197279e-05,
      "loss": 0.0283,
      "step": 536
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.6802721088435374e-05,
      "loss": 0.0313,
      "step": 538
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.6530612244897963e-05,
      "loss": 0.027,
      "step": 540
    },
    {
      "epoch": 3.69,
      "learning_rate": 2.6258503401360546e-05,
      "loss": 0.029,
      "step": 542
    },
    {
      "epoch": 3.7,
      "learning_rate": 2.598639455782313e-05,
      "loss": 0.0262,
      "step": 544
    },
    {
      "epoch": 3.71,
      "learning_rate": 2.5714285714285714e-05,
      "loss": 0.0289,
      "step": 546
    },
    {
      "epoch": 3.73,
      "learning_rate": 2.54421768707483e-05,
      "loss": 0.0326,
      "step": 548
    },
    {
      "epoch": 3.74,
      "learning_rate": 2.5170068027210887e-05,
      "loss": 0.0289,
      "step": 550
    },
    {
      "epoch": 3.74,
      "eval_f1": 0.912944557830135,
      "eval_loss": 0.027326099574565887,
      "eval_precision": 0.9129149606541497,
      "eval_recall": 0.9129841575737253,
      "eval_runtime": 11.1164,
      "eval_samples_per_second": 133.317,
      "eval_steps_per_second": 0.54,
      "step": 550
    },
    {
      "epoch": 3.76,
      "learning_rate": 2.489795918367347e-05,
      "loss": 0.0277,
      "step": 552
    },
    {
      "epoch": 3.77,
      "learning_rate": 2.4625850340136055e-05,
      "loss": 0.0269,
      "step": 554
    },
    {
      "epoch": 3.78,
      "learning_rate": 2.4353741496598638e-05,
      "loss": 0.0263,
      "step": 556
    },
    {
      "epoch": 3.8,
      "learning_rate": 2.4081632653061224e-05,
      "loss": 0.0342,
      "step": 558
    },
    {
      "epoch": 3.81,
      "learning_rate": 2.380952380952381e-05,
      "loss": 0.0234,
      "step": 560
    },
    {
      "epoch": 3.82,
      "learning_rate": 2.3537414965986396e-05,
      "loss": 0.0255,
      "step": 562
    },
    {
      "epoch": 3.84,
      "learning_rate": 2.326530612244898e-05,
      "loss": 0.0329,
      "step": 564
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.2993197278911565e-05,
      "loss": 0.0287,
      "step": 566
    },
    {
      "epoch": 3.86,
      "learning_rate": 2.272108843537415e-05,
      "loss": 0.0272,
      "step": 568
    },
    {
      "epoch": 3.88,
      "learning_rate": 2.2448979591836737e-05,
      "loss": 0.0248,
      "step": 570
    },
    {
      "epoch": 3.89,
      "learning_rate": 2.2176870748299323e-05,
      "loss": 0.0275,
      "step": 572
    },
    {
      "epoch": 3.9,
      "learning_rate": 2.1904761904761906e-05,
      "loss": 0.0269,
      "step": 574
    },
    {
      "epoch": 3.92,
      "learning_rate": 2.1632653061224492e-05,
      "loss": 0.0288,
      "step": 576
    },
    {
      "epoch": 3.93,
      "learning_rate": 2.1360544217687075e-05,
      "loss": 0.0288,
      "step": 578
    },
    {
      "epoch": 3.95,
      "learning_rate": 2.108843537414966e-05,
      "loss": 0.0244,
      "step": 580
    },
    {
      "epoch": 3.96,
      "learning_rate": 2.0816326530612247e-05,
      "loss": 0.0255,
      "step": 582
    },
    {
      "epoch": 3.97,
      "learning_rate": 2.0544217687074833e-05,
      "loss": 0.028,
      "step": 584
    },
    {
      "epoch": 3.99,
      "learning_rate": 2.0272108843537416e-05,
      "loss": 0.0209,
      "step": 586
    },
    {
      "epoch": 4.0,
      "learning_rate": 2e-05,
      "loss": 0.0288,
      "step": 588
    },
    {
      "epoch": 4.01,
      "learning_rate": 1.9727891156462584e-05,
      "loss": 0.0266,
      "step": 590
    },
    {
      "epoch": 4.03,
      "learning_rate": 1.945578231292517e-05,
      "loss": 0.025,
      "step": 592
    },
    {
      "epoch": 4.04,
      "learning_rate": 1.9183673469387756e-05,
      "loss": 0.0242,
      "step": 594
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.8911564625850343e-05,
      "loss": 0.0296,
      "step": 596
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.8639455782312925e-05,
      "loss": 0.0253,
      "step": 598
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.836734693877551e-05,
      "loss": 0.0206,
      "step": 600
    },
    {
      "epoch": 4.08,
      "eval_f1": 0.9121926864008709,
      "eval_loss": 0.027346264570951462,
      "eval_precision": 0.9039821355385321,
      "eval_recall": 0.9205639347225317,
      "eval_runtime": 11.0662,
      "eval_samples_per_second": 133.921,
      "eval_steps_per_second": 0.542,
      "step": 600
    },
    {
      "epoch": 4.1,
      "learning_rate": 1.8095238095238094e-05,
      "loss": 0.0298,
      "step": 602
    },
    {
      "epoch": 4.11,
      "learning_rate": 1.7823129251700683e-05,
      "loss": 0.0273,
      "step": 604
    },
    {
      "epoch": 4.12,
      "learning_rate": 1.7551020408163266e-05,
      "loss": 0.0275,
      "step": 606
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.7278911564625852e-05,
      "loss": 0.0229,
      "step": 608
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.7006802721088435e-05,
      "loss": 0.026,
      "step": 610
    },
    {
      "epoch": 4.16,
      "learning_rate": 1.673469387755102e-05,
      "loss": 0.0235,
      "step": 612
    },
    {
      "epoch": 4.18,
      "learning_rate": 1.6462585034013607e-05,
      "loss": 0.0255,
      "step": 614
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.6190476190476193e-05,
      "loss": 0.0279,
      "step": 616
    },
    {
      "epoch": 4.2,
      "learning_rate": 1.5918367346938776e-05,
      "loss": 0.0262,
      "step": 618
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.5646258503401362e-05,
      "loss": 0.0228,
      "step": 620
    },
    {
      "epoch": 4.23,
      "learning_rate": 1.5374149659863945e-05,
      "loss": 0.0238,
      "step": 622
    },
    {
      "epoch": 4.24,
      "learning_rate": 1.5102040816326532e-05,
      "loss": 0.0251,
      "step": 624
    },
    {
      "epoch": 4.26,
      "learning_rate": 1.4829931972789115e-05,
      "loss": 0.0264,
      "step": 626
    },
    {
      "epoch": 4.27,
      "learning_rate": 1.4557823129251703e-05,
      "loss": 0.0271,
      "step": 628
    },
    {
      "epoch": 4.29,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.0269,
      "step": 630
    },
    {
      "epoch": 4.3,
      "learning_rate": 1.4013605442176872e-05,
      "loss": 0.0301,
      "step": 632
    },
    {
      "epoch": 4.31,
      "learning_rate": 1.3741496598639456e-05,
      "loss": 0.0261,
      "step": 634
    },
    {
      "epoch": 4.33,
      "learning_rate": 1.3469387755102042e-05,
      "loss": 0.0273,
      "step": 636
    },
    {
      "epoch": 4.34,
      "learning_rate": 1.3197278911564626e-05,
      "loss": 0.0236,
      "step": 638
    },
    {
      "epoch": 4.35,
      "learning_rate": 1.2925170068027212e-05,
      "loss": 0.0289,
      "step": 640
    },
    {
      "epoch": 4.37,
      "learning_rate": 1.2653061224489795e-05,
      "loss": 0.0234,
      "step": 642
    },
    {
      "epoch": 4.38,
      "learning_rate": 1.2380952380952381e-05,
      "loss": 0.0237,
      "step": 644
    },
    {
      "epoch": 4.39,
      "learning_rate": 1.2108843537414967e-05,
      "loss": 0.0291,
      "step": 646
    },
    {
      "epoch": 4.41,
      "learning_rate": 1.1836734693877552e-05,
      "loss": 0.0229,
      "step": 648
    },
    {
      "epoch": 4.42,
      "learning_rate": 1.1564625850340138e-05,
      "loss": 0.0287,
      "step": 650
    },
    {
      "epoch": 4.42,
      "eval_f1": 0.9175770006975295,
      "eval_loss": 0.025734370574355125,
      "eval_precision": 0.917026269273203,
      "eval_recall": 0.9181384060349137,
      "eval_runtime": 10.957,
      "eval_samples_per_second": 135.256,
      "eval_steps_per_second": 0.548,
      "step": 650
    },
    {
      "epoch": 4.44,
      "learning_rate": 1.1292517006802722e-05,
      "loss": 0.0277,
      "step": 652
    },
    {
      "epoch": 4.45,
      "learning_rate": 1.1020408163265306e-05,
      "loss": 0.0259,
      "step": 654
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.0748299319727893e-05,
      "loss": 0.0263,
      "step": 656
    },
    {
      "epoch": 4.48,
      "learning_rate": 1.0476190476190477e-05,
      "loss": 0.0292,
      "step": 658
    },
    {
      "epoch": 4.49,
      "learning_rate": 1.0204081632653061e-05,
      "loss": 0.0303,
      "step": 660
    },
    {
      "epoch": 4.5,
      "learning_rate": 9.931972789115647e-06,
      "loss": 0.0262,
      "step": 662
    },
    {
      "epoch": 4.52,
      "learning_rate": 9.659863945578232e-06,
      "loss": 0.0245,
      "step": 664
    },
    {
      "epoch": 4.53,
      "learning_rate": 9.387755102040816e-06,
      "loss": 0.0222,
      "step": 666
    },
    {
      "epoch": 4.54,
      "learning_rate": 9.115646258503402e-06,
      "loss": 0.0314,
      "step": 668
    },
    {
      "epoch": 4.56,
      "learning_rate": 8.843537414965987e-06,
      "loss": 0.0261,
      "step": 670
    },
    {
      "epoch": 4.57,
      "learning_rate": 8.571428571428573e-06,
      "loss": 0.0235,
      "step": 672
    },
    {
      "epoch": 4.59,
      "learning_rate": 8.299319727891157e-06,
      "loss": 0.023,
      "step": 674
    },
    {
      "epoch": 4.6,
      "learning_rate": 8.027210884353741e-06,
      "loss": 0.0263,
      "step": 676
    },
    {
      "epoch": 4.61,
      "learning_rate": 7.755102040816327e-06,
      "loss": 0.026,
      "step": 678
    },
    {
      "epoch": 4.63,
      "learning_rate": 7.482993197278912e-06,
      "loss": 0.0238,
      "step": 680
    },
    {
      "epoch": 4.64,
      "learning_rate": 7.210884353741497e-06,
      "loss": 0.0278,
      "step": 682
    },
    {
      "epoch": 4.65,
      "learning_rate": 6.938775510204082e-06,
      "loss": 0.0283,
      "step": 684
    },
    {
      "epoch": 4.67,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.0274,
      "step": 686
    },
    {
      "epoch": 4.68,
      "learning_rate": 6.394557823129252e-06,
      "loss": 0.0273,
      "step": 688
    },
    {
      "epoch": 4.69,
      "learning_rate": 6.122448979591837e-06,
      "loss": 0.0238,
      "step": 690
    },
    {
      "epoch": 4.71,
      "learning_rate": 5.8503401360544215e-06,
      "loss": 0.0316,
      "step": 692
    },
    {
      "epoch": 4.72,
      "learning_rate": 5.578231292517007e-06,
      "loss": 0.0239,
      "step": 694
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.306122448979592e-06,
      "loss": 0.0251,
      "step": 696
    },
    {
      "epoch": 4.75,
      "learning_rate": 5.034013605442177e-06,
      "loss": 0.0195,
      "step": 698
    },
    {
      "epoch": 4.76,
      "learning_rate": 4.7619047619047615e-06,
      "loss": 0.0228,
      "step": 700
    },
    {
      "epoch": 4.76,
      "eval_f1": 0.9180364705120548,
      "eval_loss": 0.02530401200056076,
      "eval_precision": 0.9165848123032982,
      "eval_recall": 0.9195027659216988,
      "eval_runtime": 11.9163,
      "eval_samples_per_second": 124.368,
      "eval_steps_per_second": 0.504,
      "step": 700
    }
  ],
  "max_steps": 735,
  "num_train_epochs": 5,
  "total_flos": 3.51335241220096e+16,
  "trial_name": null,
  "trial_params": null
}