{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.010281719103434095,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 22.125,
      "learning_rate": 1e-05,
      "loss": 6.5159,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 19.0,
      "learning_rate": 2e-05,
      "loss": 6.1822,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 8.3125,
      "learning_rate": 3e-05,
      "loss": 5.6575,
      "step": 30
    },
    {
      "epoch": 0.0,
      "grad_norm": 7.21875,
      "learning_rate": 4e-05,
      "loss": 5.0588,
      "step": 40
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.5625,
      "learning_rate": 5e-05,
      "loss": 4.6857,
      "step": 50
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.5625,
      "learning_rate": 6e-05,
      "loss": 4.2879,
      "step": 60
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.09375,
      "learning_rate": 7e-05,
      "loss": 3.8842,
      "step": 70
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.65625,
      "learning_rate": 8e-05,
      "loss": 3.6001,
      "step": 80
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.578125,
      "learning_rate": 9e-05,
      "loss": 3.431,
      "step": 90
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.78125,
      "learning_rate": 0.0001,
      "loss": 3.2564,
      "step": 100
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.6875,
      "learning_rate": 9.99695413509548e-05,
      "loss": 3.1218,
      "step": 110
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.40625,
      "learning_rate": 9.987820251299122e-05,
      "loss": 3.0213,
      "step": 120
    },
    {
      "epoch": 0.0,
      "grad_norm": 5.25,
      "learning_rate": 9.972609476841367e-05,
      "loss": 2.9567,
      "step": 130
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.390625,
      "learning_rate": 9.951340343707852e-05,
      "loss": 2.9283,
      "step": 140
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.453125,
      "learning_rate": 9.924038765061042e-05,
      "loss": 2.9176,
      "step": 150
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.234375,
      "learning_rate": 9.890738003669029e-05,
      "loss": 2.8235,
      "step": 160
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.28125,
      "learning_rate": 9.851478631379982e-05,
      "loss": 2.8968,
      "step": 170
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.4375,
      "learning_rate": 9.806308479691595e-05,
      "loss": 2.8153,
      "step": 180
    },
    {
      "epoch": 0.0,
      "grad_norm": 10.6875,
      "learning_rate": 9.755282581475769e-05,
      "loss": 2.7793,
      "step": 190
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.078125,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.699,
      "step": 200
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.171875,
      "learning_rate": 9.635919272833938e-05,
      "loss": 2.7336,
      "step": 210
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.78125,
      "learning_rate": 9.567727288213005e-05,
      "loss": 2.7361,
      "step": 220
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.328125,
      "learning_rate": 9.493970231495835e-05,
      "loss": 2.7283,
      "step": 230
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.109375,
      "learning_rate": 9.414737964294636e-05,
      "loss": 2.6705,
      "step": 240
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.046875,
      "learning_rate": 9.330127018922194e-05,
      "loss": 2.7163,
      "step": 250
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.96875,
      "learning_rate": 9.24024048078213e-05,
      "loss": 2.6643,
      "step": 260
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.34375,
      "learning_rate": 9.145187862775209e-05,
      "loss": 2.6863,
      "step": 270
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.234375,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.6687,
      "step": 280
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.109375,
      "learning_rate": 8.940053768033609e-05,
      "loss": 2.6537,
      "step": 290
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.890625,
      "learning_rate": 8.83022221559489e-05,
      "loss": 2.6543,
      "step": 300
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.40625,
      "learning_rate": 8.715724127386972e-05,
      "loss": 2.6974,
      "step": 310
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8984375,
      "learning_rate": 8.596699001693255e-05,
      "loss": 2.6437,
      "step": 320
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8671875,
      "learning_rate": 8.473291852294987e-05,
      "loss": 2.6021,
      "step": 330
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.109375,
      "learning_rate": 8.345653031794292e-05,
      "loss": 2.596,
      "step": 340
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.984375,
      "learning_rate": 8.213938048432697e-05,
      "loss": 2.6222,
      "step": 350
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.078125,
      "learning_rate": 8.07830737662829e-05,
      "loss": 2.6384,
      "step": 360
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.09375,
      "learning_rate": 7.938926261462366e-05,
      "loss": 2.6475,
      "step": 370
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.921875,
      "learning_rate": 7.795964517353735e-05,
      "loss": 2.5795,
      "step": 380
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.9296875,
      "learning_rate": 7.649596321166024e-05,
      "loss": 2.5523,
      "step": 390
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.1875,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.5676,
      "step": 400
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8671875,
      "learning_rate": 7.347357813929454e-05,
      "loss": 2.5865,
      "step": 410
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.828125,
      "learning_rate": 7.191855733945387e-05,
      "loss": 2.62,
      "step": 420
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.609375,
      "learning_rate": 7.033683215379002e-05,
      "loss": 2.5757,
      "step": 430
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.9140625,
      "learning_rate": 6.873032967079561e-05,
      "loss": 2.6054,
      "step": 440
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8515625,
      "learning_rate": 6.710100716628344e-05,
      "loss": 2.5444,
      "step": 450
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.25,
      "learning_rate": 6.545084971874738e-05,
      "loss": 2.576,
      "step": 460
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8203125,
      "learning_rate": 6.378186779084995e-05,
      "loss": 2.5477,
      "step": 470
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.859375,
      "learning_rate": 6.209609477998338e-05,
      "loss": 2.5643,
      "step": 480
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8359375,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 2.5724,
      "step": 490
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.0625,
      "learning_rate": 5.868240888334653e-05,
      "loss": 2.5313,
      "step": 500
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.5299240767701169,
      "eval_loss": 2.329106569290161,
      "eval_runtime": 14.8854,
      "eval_samples_per_second": 104.062,
      "eval_steps_per_second": 1.679,
      "step": 500
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.703125,
      "learning_rate": 5.695865504800327e-05,
      "loss": 2.5651,
      "step": 510
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.03125,
      "learning_rate": 5.522642316338268e-05,
      "loss": 2.5242,
      "step": 520
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.9609375,
      "learning_rate": 5.348782368720626e-05,
      "loss": 2.528,
      "step": 530
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.796875,
      "learning_rate": 5.174497483512506e-05,
      "loss": 2.5198,
      "step": 540
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.640625,
      "learning_rate": 5e-05,
      "loss": 2.5374,
      "step": 550
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.984375,
      "learning_rate": 4.825502516487497e-05,
      "loss": 2.5683,
      "step": 560
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.46875,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 2.5292,
      "step": 570
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.71875,
      "learning_rate": 4.477357683661734e-05,
      "loss": 2.5173,
      "step": 580
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.9140625,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 2.5369,
      "step": 590
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.7578125,
      "learning_rate": 4.131759111665349e-05,
      "loss": 2.5048,
      "step": 600
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.703125,
      "learning_rate": 3.960441545911204e-05,
      "loss": 2.5034,
      "step": 610
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6953125,
      "learning_rate": 3.790390522001662e-05,
      "loss": 2.5269,
      "step": 620
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.71875,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 2.5137,
      "step": 630
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6953125,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 2.5282,
      "step": 640
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.734375,
      "learning_rate": 3.289899283371657e-05,
      "loss": 2.525,
      "step": 650
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.625,
      "learning_rate": 3.12696703292044e-05,
      "loss": 2.4879,
      "step": 660
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6953125,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 2.5372,
      "step": 670
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6953125,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 2.5166,
      "step": 680
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.71875,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 2.5106,
      "step": 690
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.640625,
      "learning_rate": 2.500000000000001e-05,
      "loss": 2.516,
      "step": 700
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.875,
      "learning_rate": 2.350403678833976e-05,
      "loss": 2.4967,
      "step": 710
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6640625,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 2.5052,
      "step": 720
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.640625,
      "learning_rate": 2.061073738537635e-05,
      "loss": 2.5026,
      "step": 730
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8671875,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 2.5186,
      "step": 740
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.671875,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 2.4504,
      "step": 750
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6015625,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 2.4817,
      "step": 760
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6015625,
      "learning_rate": 1.526708147705013e-05,
      "loss": 2.5015,
      "step": 770
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8828125,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 2.4757,
      "step": 780
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6640625,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 2.4681,
      "step": 790
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.59375,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 2.4924,
      "step": 800
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.53125,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 2.4658,
      "step": 810
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.5859375,
      "learning_rate": 9.549150281252633e-06,
      "loss": 2.4954,
      "step": 820
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.671875,
      "learning_rate": 8.548121372247918e-06,
      "loss": 2.4989,
      "step": 830
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.65625,
      "learning_rate": 7.597595192178702e-06,
      "loss": 2.5122,
      "step": 840
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.5234375,
      "learning_rate": 6.698729810778065e-06,
      "loss": 2.5014,
      "step": 850
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6015625,
      "learning_rate": 5.852620357053651e-06,
      "loss": 2.5023,
      "step": 860
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.609375,
      "learning_rate": 5.060297685041659e-06,
      "loss": 2.4745,
      "step": 870
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6171875,
      "learning_rate": 4.322727117869951e-06,
      "loss": 2.481,
      "step": 880
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.609375,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 2.4503,
      "step": 890
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.796875,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 2.5154,
      "step": 900
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.625,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 2.5018,
      "step": 910
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6328125,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 2.4622,
      "step": 920
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.578125,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 2.4552,
      "step": 930
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.671875,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 2.4811,
      "step": 940
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.703125,
      "learning_rate": 7.596123493895991e-07,
      "loss": 2.4919,
      "step": 950
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6953125,
      "learning_rate": 4.865965629214819e-07,
      "loss": 2.5005,
      "step": 960
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.65625,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 2.4714,
      "step": 970
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8515625,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 2.463,
      "step": 980
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.640625,
      "learning_rate": 3.04586490452119e-08,
      "loss": 2.4672,
      "step": 990
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.546875,
      "learning_rate": 0.0,
      "loss": 2.4789,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.5390751261968905,
      "eval_loss": 2.2672338485717773,
      "eval_runtime": 14.7852,
      "eval_samples_per_second": 104.767,
      "eval_steps_per_second": 1.691,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "step": 1000,
      "total_flos": 2.7960254276829184e+18,
      "train_loss": 2.7936128311157225,
      "train_runtime": 791.2602,
      "train_samples_per_second": 80.884,
      "train_steps_per_second": 1.264
    }
  ],
  "logging_steps": 10,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10000,
  "total_flos": 2.7960254276829184e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}