{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9697377269670477,
  "eval_steps": 50,
  "global_step": 276,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.01,
      "step": 3
    },
    {
      "epoch": 0.06,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.0093,
      "step": 6
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.2857142857142859e-05,
      "loss": 0.0089,
      "step": 9
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 0.0091,
      "step": 12
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9923664122137406e-05,
      "loss": 0.0095,
      "step": 15
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.969465648854962e-05,
      "loss": 0.0081,
      "step": 18
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.9465648854961833e-05,
      "loss": 0.0073,
      "step": 21
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.923664122137405e-05,
      "loss": 0.0082,
      "step": 24
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.900763358778626e-05,
      "loss": 0.0063,
      "step": 27
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8778625954198473e-05,
      "loss": 0.0077,
      "step": 30
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.854961832061069e-05,
      "loss": 0.006,
      "step": 33
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.83206106870229e-05,
      "loss": 0.0064,
      "step": 36
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.8091603053435117e-05,
      "loss": 0.0056,
      "step": 39
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.786259541984733e-05,
      "loss": 0.0055,
      "step": 42
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.7633587786259544e-05,
      "loss": 0.0052,
      "step": 45
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.7404580152671757e-05,
      "loss": 0.0064,
      "step": 48
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.000551095581613481,
      "eval_mse": 0.0005510955593948523,
      "eval_runtime": 8.5427,
      "eval_samples_per_second": 36.64,
      "eval_steps_per_second": 9.248,
      "step": 50
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.717557251908397e-05,
      "loss": 0.0061,
      "step": 51
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6946564885496184e-05,
      "loss": 0.0052,
      "step": 54
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6717557251908398e-05,
      "loss": 0.0053,
      "step": 57
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.648854961832061e-05,
      "loss": 0.0043,
      "step": 60
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.6259541984732825e-05,
      "loss": 0.0035,
      "step": 63
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.6030534351145038e-05,
      "loss": 0.0041,
      "step": 66
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5801526717557255e-05,
      "loss": 0.0046,
      "step": 69
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5572519083969465e-05,
      "loss": 0.0036,
      "step": 72
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.5343511450381682e-05,
      "loss": 0.0042,
      "step": 75
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.5114503816793895e-05,
      "loss": 0.0037,
      "step": 78
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4885496183206107e-05,
      "loss": 0.0039,
      "step": 81
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4656488549618322e-05,
      "loss": 0.0027,
      "step": 84
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.4427480916030536e-05,
      "loss": 0.003,
      "step": 87
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4198473282442749e-05,
      "loss": 0.0031,
      "step": 90
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3969465648854963e-05,
      "loss": 0.0032,
      "step": 93
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3740458015267178e-05,
      "loss": 0.0031,
      "step": 96
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.351145038167939e-05,
      "loss": 0.0043,
      "step": 99
    },
    {
      "epoch": 1.08,
      "eval_loss": 0.0004679011180996895,
      "eval_mse": 0.00046790109512442456,
      "eval_runtime": 8.5901,
      "eval_samples_per_second": 36.437,
      "eval_steps_per_second": 9.197,
      "step": 100
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.3282442748091605e-05,
      "loss": 0.0032,
      "step": 102
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3053435114503818e-05,
      "loss": 0.0037,
      "step": 105
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2824427480916032e-05,
      "loss": 0.0026,
      "step": 108
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2595419847328245e-05,
      "loss": 0.003,
      "step": 111
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.236641221374046e-05,
      "loss": 0.0027,
      "step": 114
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.2137404580152672e-05,
      "loss": 0.0034,
      "step": 117
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1908396946564887e-05,
      "loss": 0.0031,
      "step": 120
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.16793893129771e-05,
      "loss": 0.0038,
      "step": 123
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1450381679389312e-05,
      "loss": 0.003,
      "step": 126
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.1221374045801527e-05,
      "loss": 0.003,
      "step": 129
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0992366412213743e-05,
      "loss": 0.0032,
      "step": 132
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0763358778625954e-05,
      "loss": 0.0028,
      "step": 135
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0534351145038168e-05,
      "loss": 0.003,
      "step": 138
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.0305343511450383e-05,
      "loss": 0.0025,
      "step": 141
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.0076335877862595e-05,
      "loss": 0.0033,
      "step": 144
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.84732824427481e-06,
      "loss": 0.0031,
      "step": 147
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.618320610687025e-06,
      "loss": 0.0028,
      "step": 150
    },
    {
      "epoch": 1.61,
      "eval_loss": 0.0006046506459824741,
      "eval_mse": 0.0006046506115009975,
      "eval_runtime": 8.4067,
      "eval_samples_per_second": 37.232,
      "eval_steps_per_second": 9.397,
      "step": 150
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.389312977099237e-06,
      "loss": 0.0029,
      "step": 153
    },
    {
      "epoch": 1.68,
      "learning_rate": 9.16030534351145e-06,
      "loss": 0.0029,
      "step": 156
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.931297709923665e-06,
      "loss": 0.0021,
      "step": 159
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.702290076335879e-06,
      "loss": 0.0029,
      "step": 162
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.473282442748092e-06,
      "loss": 0.0026,
      "step": 165
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.244274809160306e-06,
      "loss": 0.0028,
      "step": 168
    },
    {
      "epoch": 1.84,
      "learning_rate": 8.015267175572519e-06,
      "loss": 0.0031,
      "step": 171
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.786259541984733e-06,
      "loss": 0.0024,
      "step": 174
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.557251908396948e-06,
      "loss": 0.0021,
      "step": 177
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.328244274809161e-06,
      "loss": 0.0024,
      "step": 180
    },
    {
      "epoch": 1.97,
      "learning_rate": 7.0992366412213746e-06,
      "loss": 0.0024,
      "step": 183
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.870229007633589e-06,
      "loss": 0.0025,
      "step": 186
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.641221374045802e-06,
      "loss": 0.0024,
      "step": 189
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.412213740458016e-06,
      "loss": 0.0027,
      "step": 192
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.18320610687023e-06,
      "loss": 0.0029,
      "step": 195
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.9541984732824435e-06,
      "loss": 0.0025,
      "step": 198
    },
    {
      "epoch": 2.15,
      "eval_loss": 0.0004669851914513856,
      "eval_mse": 0.0004669852106191058,
      "eval_runtime": 8.5435,
      "eval_samples_per_second": 36.636,
      "eval_steps_per_second": 9.247,
      "step": 200
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.725190839694656e-06,
      "loss": 0.002,
      "step": 201
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.496183206106871e-06,
      "loss": 0.0026,
      "step": 204
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.267175572519084e-06,
      "loss": 0.0024,
      "step": 207
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.038167938931297e-06,
      "loss": 0.0025,
      "step": 210
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.8091603053435125e-06,
      "loss": 0.0022,
      "step": 213
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.580152671755725e-06,
      "loss": 0.002,
      "step": 216
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.351145038167939e-06,
      "loss": 0.0024,
      "step": 219
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.122137404580153e-06,
      "loss": 0.0021,
      "step": 222
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.893129770992366e-06,
      "loss": 0.0019,
      "step": 225
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6641221374045806e-06,
      "loss": 0.0021,
      "step": 228
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4351145038167944e-06,
      "loss": 0.0027,
      "step": 231
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.206106870229008e-06,
      "loss": 0.0024,
      "step": 234
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.9770992366412218e-06,
      "loss": 0.0024,
      "step": 237
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.7480916030534356e-06,
      "loss": 0.0021,
      "step": 240
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5190839694656487e-06,
      "loss": 0.0019,
      "step": 243
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.2900763358778625e-06,
      "loss": 0.0019,
      "step": 246
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.0610687022900764e-06,
      "loss": 0.0025,
      "step": 249
    },
    {
      "epoch": 2.69,
      "eval_loss": 0.00047420692862942815,
      "eval_mse": 0.00047420695769180793,
      "eval_runtime": 8.5496,
      "eval_samples_per_second": 36.61,
      "eval_steps_per_second": 9.24,
      "step": 250
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.8320610687022903e-06,
      "loss": 0.0021,
      "step": 252
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.603053435114504e-06,
      "loss": 0.0024,
      "step": 255
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.3740458015267178e-06,
      "loss": 0.0024,
      "step": 258
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.1450381679389313e-06,
      "loss": 0.0021,
      "step": 261
    },
    {
      "epoch": 2.84,
      "learning_rate": 9.160305343511451e-07,
      "loss": 0.0026,
      "step": 264
    },
    {
      "epoch": 2.87,
      "learning_rate": 6.870229007633589e-07,
      "loss": 0.0023,
      "step": 267
    },
    {
      "epoch": 2.91,
      "learning_rate": 4.5801526717557257e-07,
      "loss": 0.0021,
      "step": 270
    },
    {
      "epoch": 2.94,
      "learning_rate": 2.2900763358778629e-07,
      "loss": 0.0026,
      "step": 273
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0,
      "loss": 0.0024,
      "step": 276
    },
    {
      "epoch": 2.97,
      "step": 276,
      "total_flos": 4678301863342080.0,
      "train_loss": 0.0036955964937131257,
      "train_runtime": 2019.4918,
      "train_samples_per_second": 8.831,
      "train_steps_per_second": 0.137
    }
  ],
  "logging_steps": 3,
  "max_steps": 276,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 4678301863342080.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}