{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 704,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 11.881,
      "step": 2
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.985795454545455e-05,
      "loss": 11.0362,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.971590909090909e-05,
      "loss": 4.7507,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9644886363636365e-05,
      "loss": 5.152,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.950284090909092e-05,
      "loss": 4.4355,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.9360795454545455e-05,
      "loss": 4.1403,
      "step": 12
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.921875e-05,
      "loss": 3.8628,
      "step": 14
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.907670454545455e-05,
      "loss": 3.3533,
      "step": 16
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.893465909090909e-05,
      "loss": 3.6164,
      "step": 18
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.8792613636363636e-05,
      "loss": 4.0321,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.865056818181819e-05,
      "loss": 3.5433,
      "step": 22
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.850852272727273e-05,
      "loss": 4.0187,
      "step": 24
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.836647727272727e-05,
      "loss": 3.601,
      "step": 26
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.822443181818182e-05,
      "loss": 3.4055,
      "step": 28
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.808238636363637e-05,
      "loss": 3.4381,
      "step": 30
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.794034090909091e-05,
      "loss": 3.3023,
      "step": 32
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.779829545454546e-05,
      "loss": 3.7605,
      "step": 34
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.765625e-05,
      "loss": 2.0996,
      "step": 36
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.751420454545455e-05,
      "loss": 2.5817,
      "step": 38
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.737215909090909e-05,
      "loss": 2.0194,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.723011363636364e-05,
      "loss": 2.2909,
      "step": 42
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.708806818181818e-05,
      "loss": 1.8493,
      "step": 44
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.694602272727273e-05,
      "loss": 2.2346,
      "step": 46
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.6803977272727274e-05,
      "loss": 2.3284,
      "step": 48
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.666193181818182e-05,
      "loss": 2.1243,
      "step": 50
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.6519886363636364e-05,
      "loss": 2.0376,
      "step": 52
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.6377840909090916e-05,
      "loss": 4.0457,
      "step": 54
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.6235795454545454e-05,
      "loss": 2.4413,
      "step": 56
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.609375e-05,
      "loss": 1.7908,
      "step": 58
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.595170454545455e-05,
      "loss": 2.1725,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.580965909090909e-05,
      "loss": 1.2263,
      "step": 62
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.5667613636363634e-05,
      "loss": 2.2057,
      "step": 64
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.5525568181818186e-05,
      "loss": 1.4206,
      "step": 66
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.538352272727273e-05,
      "loss": 1.6346,
      "step": 68
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.5241477272727276e-05,
      "loss": 1.6167,
      "step": 70
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.509943181818182e-05,
      "loss": 1.3222,
      "step": 72
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.4957386363636366e-05,
      "loss": 1.1639,
      "step": 74
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.481534090909091e-05,
      "loss": 2.0723,
      "step": 76
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.4673295454545457e-05,
      "loss": 1.707,
      "step": 78
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.453125e-05,
      "loss": 1.8604,
      "step": 80
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.438920454545455e-05,
      "loss": 1.9273,
      "step": 82
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.424715909090909e-05,
      "loss": 1.8075,
      "step": 84
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.410511363636364e-05,
      "loss": 1.4449,
      "step": 86
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.396306818181818e-05,
      "loss": 1.0887,
      "step": 88
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.3821022727272734e-05,
      "loss": 1.4518,
      "step": 90
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.367897727272727e-05,
      "loss": 1.3403,
      "step": 92
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.353693181818182e-05,
      "loss": 0.9441,
      "step": 94
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.339488636363637e-05,
      "loss": 1.1635,
      "step": 96
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.3252840909090914e-05,
      "loss": 1.2348,
      "step": 98
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.311079545454545e-05,
      "loss": 1.3606,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.2968750000000004e-05,
      "loss": 1.0958,
      "step": 102
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.282670454545455e-05,
      "loss": 1.6382,
      "step": 104
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.268465909090909e-05,
      "loss": 1.9738,
      "step": 106
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.254261363636364e-05,
      "loss": 1.2341,
      "step": 108
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.2400568181818185e-05,
      "loss": 1.3189,
      "step": 110
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.225852272727273e-05,
      "loss": 1.3616,
      "step": 112
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.2116477272727275e-05,
      "loss": 1.525,
      "step": 114
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.197443181818182e-05,
      "loss": 1.4151,
      "step": 116
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.1832386363636365e-05,
      "loss": 1.6387,
      "step": 118
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.169034090909092e-05,
      "loss": 1.3773,
      "step": 120
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.1548295454545455e-05,
      "loss": 1.4374,
      "step": 122
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.140625e-05,
      "loss": 1.3446,
      "step": 124
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.126420454545455e-05,
      "loss": 1.8676,
      "step": 126
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.112215909090909e-05,
      "loss": 1.8264,
      "step": 128
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.0980113636363635e-05,
      "loss": 1.0708,
      "step": 130
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.083806818181819e-05,
      "loss": 1.4238,
      "step": 132
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.069602272727273e-05,
      "loss": 1.6779,
      "step": 134
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.055397727272727e-05,
      "loss": 1.5807,
      "step": 136
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.041193181818182e-05,
      "loss": 1.1561,
      "step": 138
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.026988636363637e-05,
      "loss": 1.6651,
      "step": 140
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.012784090909091e-05,
      "loss": 1.9554,
      "step": 142
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.998579545454546e-05,
      "loss": 1.27,
      "step": 144
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.984375e-05,
      "loss": 1.9976,
      "step": 146
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.970170454545455e-05,
      "loss": 1.0606,
      "step": 148
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.955965909090909e-05,
      "loss": 1.2185,
      "step": 150
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.941761363636364e-05,
      "loss": 1.6675,
      "step": 152
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.927556818181818e-05,
      "loss": 1.0349,
      "step": 154
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.913352272727273e-05,
      "loss": 0.9453,
      "step": 156
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.899147727272727e-05,
      "loss": 1.0831,
      "step": 158
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.884943181818182e-05,
      "loss": 2.1055,
      "step": 160
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.8707386363636364e-05,
      "loss": 1.5656,
      "step": 162
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.8565340909090915e-05,
      "loss": 1.0042,
      "step": 164
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.8423295454545454e-05,
      "loss": 1.8098,
      "step": 166
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.828125e-05,
      "loss": 1.4084,
      "step": 168
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.813920454545455e-05,
      "loss": 1.5181,
      "step": 170
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.799715909090909e-05,
      "loss": 1.334,
      "step": 172
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.7855113636363634e-05,
      "loss": 1.1288,
      "step": 174
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.7713068181818186e-05,
      "loss": 0.9293,
      "step": 176
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.757102272727273e-05,
      "loss": 1.1451,
      "step": 178
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.742897727272727e-05,
      "loss": 1.1264,
      "step": 180
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.728693181818182e-05,
      "loss": 2.3384,
      "step": 182
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.7144886363636366e-05,
      "loss": 0.8733,
      "step": 184
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.700284090909091e-05,
      "loss": 1.4847,
      "step": 186
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6860795454545456e-05,
      "loss": 1.8384,
      "step": 188
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.671875e-05,
      "loss": 0.973,
      "step": 190
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6576704545454547e-05,
      "loss": 1.5602,
      "step": 192
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.643465909090909e-05,
      "loss": 1.459,
      "step": 194
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.629261363636364e-05,
      "loss": 1.6167,
      "step": 196
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.615056818181818e-05,
      "loss": 0.7387,
      "step": 198
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.6008522727272734e-05,
      "loss": 1.1738,
      "step": 200
    },
    {
      "epoch": 0.28,
      "eval_cer": 0.22616305849838783,
      "eval_loss": 1.2406114339828491,
      "eval_runtime": 381.0451,
      "eval_samples_per_second": 3.698,
      "eval_steps_per_second": 0.465,
      "step": 200
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.586647727272727e-05,
      "loss": 1.1026,
      "step": 202
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.572443181818182e-05,
      "loss": 1.2022,
      "step": 204
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.558238636363637e-05,
      "loss": 1.2101,
      "step": 206
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.5440340909090914e-05,
      "loss": 0.9208,
      "step": 208
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.529829545454545e-05,
      "loss": 1.1579,
      "step": 210
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.5156250000000004e-05,
      "loss": 0.8407,
      "step": 212
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.501420454545455e-05,
      "loss": 1.3193,
      "step": 214
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.487215909090909e-05,
      "loss": 1.7611,
      "step": 216
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.473011363636364e-05,
      "loss": 0.8605,
      "step": 218
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.4588068181818184e-05,
      "loss": 1.2032,
      "step": 220
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.444602272727273e-05,
      "loss": 1.5094,
      "step": 222
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.4303977272727275e-05,
      "loss": 0.8793,
      "step": 224
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.416193181818182e-05,
      "loss": 1.4906,
      "step": 226
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.4019886363636365e-05,
      "loss": 1.5084,
      "step": 228
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.387784090909091e-05,
      "loss": 1.0827,
      "step": 230
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.3735795454545455e-05,
      "loss": 1.9808,
      "step": 232
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.359375e-05,
      "loss": 1.022,
      "step": 234
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.345170454545455e-05,
      "loss": 0.9199,
      "step": 236
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.330965909090909e-05,
      "loss": 1.3656,
      "step": 238
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.3167613636363635e-05,
      "loss": 1.1733,
      "step": 240
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.302556818181819e-05,
      "loss": 1.098,
      "step": 242
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.288352272727273e-05,
      "loss": 1.5639,
      "step": 244
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.274147727272727e-05,
      "loss": 1.2712,
      "step": 246
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.259943181818182e-05,
      "loss": 1.0763,
      "step": 248
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.245738636363637e-05,
      "loss": 0.7826,
      "step": 250
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.231534090909091e-05,
      "loss": 1.7579,
      "step": 252
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.217329545454546e-05,
      "loss": 1.1492,
      "step": 254
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.203125e-05,
      "loss": 0.8802,
      "step": 256
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.188920454545455e-05,
      "loss": 0.9283,
      "step": 258
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.174715909090909e-05,
      "loss": 1.4768,
      "step": 260
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.160511363636364e-05,
      "loss": 1.2097,
      "step": 262
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.146306818181818e-05,
      "loss": 1.083,
      "step": 264
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.132102272727273e-05,
      "loss": 1.5882,
      "step": 266
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.117897727272727e-05,
      "loss": 1.3427,
      "step": 268
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.103693181818182e-05,
      "loss": 1.1064,
      "step": 270
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.089488636363636e-05,
      "loss": 1.3874,
      "step": 272
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0752840909090915e-05,
      "loss": 2.1061,
      "step": 274
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0610795454545454e-05,
      "loss": 0.8418,
      "step": 276
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0468750000000002e-05,
      "loss": 0.897,
      "step": 278
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.0326704545454547e-05,
      "loss": 1.3175,
      "step": 280
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.018465909090909e-05,
      "loss": 0.7876,
      "step": 282
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.0042613636363637e-05,
      "loss": 1.4139,
      "step": 284
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9900568181818182e-05,
      "loss": 0.8398,
      "step": 286
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.975852272727273e-05,
      "loss": 0.7938,
      "step": 288
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9616477272727272e-05,
      "loss": 0.9088,
      "step": 290
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9474431818181818e-05,
      "loss": 1.0675,
      "step": 292
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.9332386363636366e-05,
      "loss": 1.2702,
      "step": 294
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.9190340909090915e-05,
      "loss": 1.0515,
      "step": 296
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.9048295454545453e-05,
      "loss": 0.9444,
      "step": 298
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.890625e-05,
      "loss": 1.4437,
      "step": 300
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.876420454545455e-05,
      "loss": 0.8352,
      "step": 302
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.862215909090909e-05,
      "loss": 0.9627,
      "step": 304
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.8480113636363637e-05,
      "loss": 0.964,
      "step": 306
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.8338068181818185e-05,
      "loss": 1.0729,
      "step": 308
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.819602272727273e-05,
      "loss": 1.094,
      "step": 310
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.8053977272727272e-05,
      "loss": 1.3392,
      "step": 312
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.791193181818182e-05,
      "loss": 0.9211,
      "step": 314
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7769886363636365e-05,
      "loss": 1.0152,
      "step": 316
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7627840909090914e-05,
      "loss": 1.3656,
      "step": 318
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7485795454545455e-05,
      "loss": 1.6998,
      "step": 320
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.734375e-05,
      "loss": 0.9577,
      "step": 322
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.720170454545455e-05,
      "loss": 1.7129,
      "step": 324
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.705965909090909e-05,
      "loss": 1.3892,
      "step": 326
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6917613636363636e-05,
      "loss": 0.8672,
      "step": 328
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6775568181818184e-05,
      "loss": 0.7077,
      "step": 330
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.663352272727273e-05,
      "loss": 1.4751,
      "step": 332
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.649147727272727e-05,
      "loss": 1.221,
      "step": 334
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.634943181818182e-05,
      "loss": 1.01,
      "step": 336
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.6207386363636365e-05,
      "loss": 1.0246,
      "step": 338
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.6065340909090913e-05,
      "loss": 0.9274,
      "step": 340
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.5923295454545455e-05,
      "loss": 1.5181,
      "step": 342
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.578125e-05,
      "loss": 0.6105,
      "step": 344
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.563920454545455e-05,
      "loss": 0.8329,
      "step": 346
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.549715909090909e-05,
      "loss": 0.707,
      "step": 348
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.5355113636363635e-05,
      "loss": 1.2725,
      "step": 350
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.5213068181818184e-05,
      "loss": 1.7603,
      "step": 352
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.5071022727272732e-05,
      "loss": 0.8998,
      "step": 354
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.4928977272727274e-05,
      "loss": 0.8211,
      "step": 356
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.478693181818182e-05,
      "loss": 1.2931,
      "step": 358
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.471590909090909e-05,
      "loss": 1.1298,
      "step": 360
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.4573863636363636e-05,
      "loss": 1.799,
      "step": 362
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4431818181818185e-05,
      "loss": 0.9355,
      "step": 364
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4289772727272727e-05,
      "loss": 0.9555,
      "step": 366
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4147727272727275e-05,
      "loss": 0.5778,
      "step": 368
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.400568181818182e-05,
      "loss": 0.9761,
      "step": 370
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.3863636363636365e-05,
      "loss": 0.9289,
      "step": 372
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.372159090909091e-05,
      "loss": 0.7408,
      "step": 374
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.3579545454545455e-05,
      "loss": 0.6301,
      "step": 376
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.34375e-05,
      "loss": 0.9952,
      "step": 378
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.3295454545454546e-05,
      "loss": 0.8253,
      "step": 380
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.315340909090909e-05,
      "loss": 0.6469,
      "step": 382
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.3011363636363636e-05,
      "loss": 1.0202,
      "step": 384
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.2869318181818184e-05,
      "loss": 0.9247,
      "step": 386
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.272727272727273e-05,
      "loss": 0.6159,
      "step": 388
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.2585227272727274e-05,
      "loss": 0.6967,
      "step": 390
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.244318181818182e-05,
      "loss": 0.6891,
      "step": 392
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.2301136363636365e-05,
      "loss": 0.8713,
      "step": 394
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.215909090909091e-05,
      "loss": 0.7181,
      "step": 396
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.2017045454545458e-05,
      "loss": 0.9892,
      "step": 398
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.1875e-05,
      "loss": 0.9181,
      "step": 400
    },
    {
      "epoch": 0.57,
      "eval_cer": 0.10839858744050361,
      "eval_loss": 0.8294418454170227,
      "eval_runtime": 369.2935,
      "eval_samples_per_second": 3.815,
      "eval_steps_per_second": 0.479,
      "step": 400
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.1732954545454545e-05,
      "loss": 0.8469,
      "step": 402
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.1590909090909093e-05,
      "loss": 1.125,
      "step": 404
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.1448863636363635e-05,
      "loss": 0.65,
      "step": 406
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.1306818181818183e-05,
      "loss": 1.1301,
      "step": 408
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.116477272727273e-05,
      "loss": 0.7355,
      "step": 410
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1022727272727274e-05,
      "loss": 0.8691,
      "step": 412
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.088068181818182e-05,
      "loss": 1.2798,
      "step": 414
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.0738636363636367e-05,
      "loss": 0.9451,
      "step": 416
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.059659090909091e-05,
      "loss": 0.8104,
      "step": 418
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0454545454545457e-05,
      "loss": 0.9429,
      "step": 420
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0312500000000002e-05,
      "loss": 1.0933,
      "step": 422
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0170454545454544e-05,
      "loss": 0.8391,
      "step": 424
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.0028409090909093e-05,
      "loss": 0.5393,
      "step": 426
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9886363636363638e-05,
      "loss": 0.7289,
      "step": 428
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9744318181818183e-05,
      "loss": 1.4527,
      "step": 430
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9602272727272728e-05,
      "loss": 0.6219,
      "step": 432
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9460227272727273e-05,
      "loss": 0.6851,
      "step": 434
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9318181818181818e-05,
      "loss": 0.6631,
      "step": 436
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9176136363636366e-05,
      "loss": 0.947,
      "step": 438
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9034090909090908e-05,
      "loss": 0.5946,
      "step": 440
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8892045454545457e-05,
      "loss": 0.8884,
      "step": 442
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.8167,
      "step": 444
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8607954545454543e-05,
      "loss": 0.5471,
      "step": 446
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8465909090909092e-05,
      "loss": 0.6034,
      "step": 448
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8323863636363637e-05,
      "loss": 0.604,
      "step": 450
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.6084,
      "step": 452
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8039772727272727e-05,
      "loss": 0.6083,
      "step": 454
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7897727272727276e-05,
      "loss": 0.6998,
      "step": 456
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7755681818181817e-05,
      "loss": 0.9948,
      "step": 458
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7613636363636366e-05,
      "loss": 0.7974,
      "step": 460
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.747159090909091e-05,
      "loss": 1.1083,
      "step": 462
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.7329545454545456e-05,
      "loss": 1.481,
      "step": 464
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.71875e-05,
      "loss": 1.5551,
      "step": 466
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.7045454545454546e-05,
      "loss": 0.747,
      "step": 468
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.690340909090909e-05,
      "loss": 0.7262,
      "step": 470
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.6761363636363636e-05,
      "loss": 0.7682,
      "step": 472
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.6619318181818185e-05,
      "loss": 0.8529,
      "step": 474
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.6477272727272726e-05,
      "loss": 0.7041,
      "step": 476
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.6335227272727275e-05,
      "loss": 0.5702,
      "step": 478
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.619318181818182e-05,
      "loss": 1.1103,
      "step": 480
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.6051136363636365e-05,
      "loss": 0.7418,
      "step": 482
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.590909090909091e-05,
      "loss": 1.1425,
      "step": 484
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5767045454545455e-05,
      "loss": 0.7602,
      "step": 486
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5625e-05,
      "loss": 0.6746,
      "step": 488
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5482954545454545e-05,
      "loss": 0.591,
      "step": 490
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.534090909090909e-05,
      "loss": 0.6343,
      "step": 492
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5198863636363636e-05,
      "loss": 0.7024,
      "step": 494
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5056818181818182e-05,
      "loss": 0.8327,
      "step": 496
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4914772727272727e-05,
      "loss": 0.6407,
      "step": 498
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4772727272727274e-05,
      "loss": 0.5521,
      "step": 500
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4630681818181818e-05,
      "loss": 0.6699,
      "step": 502
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.4488636363636366e-05,
      "loss": 0.6511,
      "step": 504
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.434659090909091e-05,
      "loss": 0.5944,
      "step": 506
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.4204545454545456e-05,
      "loss": 0.707,
      "step": 508
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.4062500000000001e-05,
      "loss": 0.8783,
      "step": 510
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3920454545454545e-05,
      "loss": 0.6688,
      "step": 512
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3778409090909091e-05,
      "loss": 0.672,
      "step": 514
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3636363636363637e-05,
      "loss": 0.8398,
      "step": 516
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.3494318181818183e-05,
      "loss": 0.6358,
      "step": 518
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.3352272727272727e-05,
      "loss": 0.8281,
      "step": 520
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.3210227272727273e-05,
      "loss": 0.576,
      "step": 522
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.3068181818181819e-05,
      "loss": 0.7594,
      "step": 524
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2926136363636365e-05,
      "loss": 0.4573,
      "step": 526
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2784090909090909e-05,
      "loss": 0.6455,
      "step": 528
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2642045454545457e-05,
      "loss": 0.5414,
      "step": 530
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.25e-05,
      "loss": 0.6023,
      "step": 532
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.2357954545454546e-05,
      "loss": 0.5274,
      "step": 534
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.2215909090909092e-05,
      "loss": 0.5246,
      "step": 536
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.2073863636363638e-05,
      "loss": 0.54,
      "step": 538
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1931818181818183e-05,
      "loss": 0.6186,
      "step": 540
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1789772727272728e-05,
      "loss": 0.641,
      "step": 542
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1647727272727273e-05,
      "loss": 0.589,
      "step": 544
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.1505681818181818e-05,
      "loss": 0.8244,
      "step": 546
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.1363636363636365e-05,
      "loss": 0.6111,
      "step": 548
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.122159090909091e-05,
      "loss": 1.014,
      "step": 550
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.1079545454545455e-05,
      "loss": 0.6972,
      "step": 552
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.09375e-05,
      "loss": 0.5662,
      "step": 554
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0795454545454547e-05,
      "loss": 1.0225,
      "step": 556
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0653409090909092e-05,
      "loss": 0.6447,
      "step": 558
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.0511363636363637e-05,
      "loss": 0.8831,
      "step": 560
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.0369318181818184e-05,
      "loss": 0.5089,
      "step": 562
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.0227272727272729e-05,
      "loss": 0.88,
      "step": 564
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.0085227272727272e-05,
      "loss": 0.5177,
      "step": 566
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.943181818181819e-06,
      "loss": 0.4923,
      "step": 568
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.801136363636364e-06,
      "loss": 0.6019,
      "step": 570
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.659090909090909e-06,
      "loss": 0.5321,
      "step": 572
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.517045454545454e-06,
      "loss": 0.6639,
      "step": 574
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.375000000000001e-06,
      "loss": 0.5423,
      "step": 576
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.232954545454546e-06,
      "loss": 0.5199,
      "step": 578
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.5719,
      "step": 580
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.948863636363638e-06,
      "loss": 0.5513,
      "step": 582
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.806818181818183e-06,
      "loss": 0.8127,
      "step": 584
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.664772727272728e-06,
      "loss": 0.7982,
      "step": 586
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.522727272727273e-06,
      "loss": 0.4545,
      "step": 588
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.380681818181818e-06,
      "loss": 0.452,
      "step": 590
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.238636363636363e-06,
      "loss": 0.6663,
      "step": 592
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.09659090909091e-06,
      "loss": 0.6574,
      "step": 594
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.954545454545455e-06,
      "loss": 0.4983,
      "step": 596
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.8125e-06,
      "loss": 0.5488,
      "step": 598
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.670454545454545e-06,
      "loss": 0.6552,
      "step": 600
    },
    {
      "epoch": 0.85,
      "eval_cer": 0.05650237985567327,
      "eval_loss": 0.6485163569450378,
      "eval_runtime": 362.1859,
      "eval_samples_per_second": 3.89,
      "eval_steps_per_second": 0.489,
      "step": 600
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.528409090909091e-06,
      "loss": 0.6316,
      "step": 602
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.386363636363637e-06,
      "loss": 0.4354,
      "step": 604
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.244318181818183e-06,
      "loss": 0.7282,
      "step": 606
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.102272727272728e-06,
      "loss": 0.4656,
      "step": 608
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.960227272727272e-06,
      "loss": 0.7131,
      "step": 610
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.818181818181818e-06,
      "loss": 0.5702,
      "step": 612
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.676136363636363e-06,
      "loss": 0.7321,
      "step": 614
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.534090909090909e-06,
      "loss": 0.8692,
      "step": 616
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.392045454545454e-06,
      "loss": 0.6667,
      "step": 618
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.25e-06,
      "loss": 0.5151,
      "step": 620
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.107954545454546e-06,
      "loss": 0.458,
      "step": 622
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.965909090909091e-06,
      "loss": 0.5797,
      "step": 624
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.823863636363636e-06,
      "loss": 0.7189,
      "step": 626
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.681818181818182e-06,
      "loss": 0.6369,
      "step": 628
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.539772727272727e-06,
      "loss": 1.0862,
      "step": 630
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.397727272727273e-06,
      "loss": 0.5919,
      "step": 632
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.255681818181818e-06,
      "loss": 0.4427,
      "step": 634
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.113636363636364e-06,
      "loss": 0.5736,
      "step": 636
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.9715909090909094e-06,
      "loss": 1.3038,
      "step": 638
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.8295454545454545e-06,
      "loss": 0.5233,
      "step": 640
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.6875000000000004e-06,
      "loss": 0.5587,
      "step": 642
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 0.5579,
      "step": 644
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.4034090909090914e-06,
      "loss": 0.501,
      "step": 646
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.2613636363636365e-06,
      "loss": 0.5076,
      "step": 648
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.119318181818182e-06,
      "loss": 0.6053,
      "step": 650
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.9772727272727275e-06,
      "loss": 0.6975,
      "step": 652
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.835227272727273e-06,
      "loss": 0.6589,
      "step": 654
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.6931818181818186e-06,
      "loss": 0.4327,
      "step": 656
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.551136363636364e-06,
      "loss": 0.9146,
      "step": 658
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.409090909090909e-06,
      "loss": 0.4549,
      "step": 660
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2670454545454546e-06,
      "loss": 0.3903,
      "step": 662
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.125e-06,
      "loss": 0.5719,
      "step": 664
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.9829545454545457e-06,
      "loss": 0.4896,
      "step": 666
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.840909090909091e-06,
      "loss": 0.6437,
      "step": 668
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.6988636363636367e-06,
      "loss": 0.4993,
      "step": 670
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.556818181818182e-06,
      "loss": 0.6381,
      "step": 672
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.4147727272727273e-06,
      "loss": 0.7237,
      "step": 674
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.2727272727272728e-06,
      "loss": 0.7229,
      "step": 676
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.1306818181818183e-06,
      "loss": 0.4903,
      "step": 678
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.9886363636363638e-06,
      "loss": 0.602,
      "step": 680
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.8465909090909093e-06,
      "loss": 0.5082,
      "step": 682
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.7045454545454546e-06,
      "loss": 0.5288,
      "step": 684
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.5625e-06,
      "loss": 0.7898,
      "step": 686
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.4204545454545456e-06,
      "loss": 0.5447,
      "step": 688
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.278409090909091e-06,
      "loss": 0.6344,
      "step": 690
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.1363636363636364e-06,
      "loss": 0.9671,
      "step": 692
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.943181818181819e-07,
      "loss": 0.6259,
      "step": 694
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.522727272727273e-07,
      "loss": 0.492,
      "step": 696
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.102272727272728e-07,
      "loss": 0.5994,
      "step": 698
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.681818181818182e-07,
      "loss": 0.4469,
      "step": 700
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.2613636363636364e-07,
      "loss": 0.5537,
      "step": 702
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.840909090909091e-07,
      "loss": 0.5373,
      "step": 704
    },
    {
      "epoch": 1.0,
      "step": 704,
      "total_flos": 4.983731178121986e+18,
      "train_loss": 1.219155483286489,
      "train_runtime": 1812.2679,
      "train_samples_per_second": 3.108,
      "train_steps_per_second": 0.388
    }
  ],
  "max_steps": 704,
  "num_train_epochs": 1,
  "total_flos": 4.983731178121986e+18,
  "trial_name": null,
  "trial_params": null
}