{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0832,
  "eval_steps": 500,
  "global_step": 1040,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 62.10089111328125,
      "learning_rate": 1e-06,
      "loss": 4.5777,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 39.39016342163086,
      "learning_rate": 2e-06,
      "loss": 4.4077,
      "step": 40
    },
    {
      "epoch": 0.0,
      "grad_norm": 54.24020767211914,
      "learning_rate": 3e-06,
      "loss": 4.4807,
      "step": 60
    },
    {
      "epoch": 0.01,
      "grad_norm": 30.161609649658203,
      "learning_rate": 4e-06,
      "loss": 4.5756,
      "step": 80
    },
    {
      "epoch": 0.01,
      "grad_norm": 40.131675720214844,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 4.4352,
      "step": 100
    },
    {
      "epoch": 0.01,
      "grad_norm": 52.3621940612793,
      "learning_rate": 6e-06,
      "loss": 4.5096,
      "step": 120
    },
    {
      "epoch": 0.01,
      "grad_norm": 49.86561584472656,
      "learning_rate": 7e-06,
      "loss": 4.493,
      "step": 140
    },
    {
      "epoch": 0.01,
      "grad_norm": 20.034923553466797,
      "learning_rate": 8e-06,
      "loss": 4.4088,
      "step": 160
    },
    {
      "epoch": 0.01,
      "grad_norm": 50.790679931640625,
      "learning_rate": 9e-06,
      "loss": 4.4901,
      "step": 180
    },
    {
      "epoch": 0.02,
      "grad_norm": 48.5693473815918,
      "learning_rate": 9.999999999999999e-06,
      "loss": 4.3628,
      "step": 200
    },
    {
      "epoch": 0.02,
      "grad_norm": 37.95353698730469,
      "learning_rate": 1.1e-05,
      "loss": 4.3298,
      "step": 220
    },
    {
      "epoch": 0.02,
      "grad_norm": 35.7153434753418,
      "learning_rate": 1.2e-05,
      "loss": 4.2839,
      "step": 240
    },
    {
      "epoch": 0.02,
      "grad_norm": 91.47773742675781,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 4.1238,
      "step": 260
    },
    {
      "epoch": 0.02,
      "grad_norm": 23.16193389892578,
      "learning_rate": 1.4e-05,
      "loss": 4.1245,
      "step": 280
    },
    {
      "epoch": 0.02,
      "grad_norm": 28.304485321044922,
      "learning_rate": 1.5e-05,
      "loss": 4.2198,
      "step": 300
    },
    {
      "epoch": 0.03,
      "grad_norm": 34.03230285644531,
      "learning_rate": 1.6e-05,
      "loss": 4.0958,
      "step": 320
    },
    {
      "epoch": 0.03,
      "grad_norm": 29.786975860595703,
      "learning_rate": 1.7e-05,
      "loss": 4.024,
      "step": 340
    },
    {
      "epoch": 0.03,
      "grad_norm": 33.04754638671875,
      "learning_rate": 1.8e-05,
      "loss": 4.0832,
      "step": 360
    },
    {
      "epoch": 0.03,
      "grad_norm": 28.68460464477539,
      "learning_rate": 1.9e-05,
      "loss": 3.9827,
      "step": 380
    },
    {
      "epoch": 0.03,
      "grad_norm": 26.463253021240234,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 3.9454,
      "step": 400
    },
    {
      "epoch": 0.03,
      "grad_norm": 19.407127380371094,
      "learning_rate": 2.1e-05,
      "loss": 4.0119,
      "step": 420
    },
    {
      "epoch": 0.04,
      "grad_norm": 26.383380889892578,
      "learning_rate": 2.2e-05,
      "loss": 3.9554,
      "step": 440
    },
    {
      "epoch": 0.04,
      "grad_norm": 33.225223541259766,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 3.8172,
      "step": 460
    },
    {
      "epoch": 0.04,
      "grad_norm": 26.000978469848633,
      "learning_rate": 2.4e-05,
      "loss": 3.8934,
      "step": 480
    },
    {
      "epoch": 0.04,
      "grad_norm": 28.714366912841797,
      "learning_rate": 2.5e-05,
      "loss": 3.9194,
      "step": 500
    },
    {
      "epoch": 0.04,
      "grad_norm": 28.721248626708984,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 3.8144,
      "step": 520
    },
    {
      "epoch": 0.04,
      "grad_norm": 24.934555053710938,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 3.9166,
      "step": 540
    },
    {
      "epoch": 0.04,
      "grad_norm": 23.113840103149414,
      "learning_rate": 2.8e-05,
      "loss": 3.8248,
      "step": 560
    },
    {
      "epoch": 0.05,
      "grad_norm": 21.58758544921875,
      "learning_rate": 2.9e-05,
      "loss": 3.7538,
      "step": 580
    },
    {
      "epoch": 0.05,
      "grad_norm": 22.339618682861328,
      "learning_rate": 3e-05,
      "loss": 3.726,
      "step": 600
    },
    {
      "epoch": 0.05,
      "grad_norm": 49.25693893432617,
      "learning_rate": 2.999992132854894e-05,
      "loss": 3.8692,
      "step": 620
    },
    {
      "epoch": 0.05,
      "grad_norm": 45.1494026184082,
      "learning_rate": 2.999968531502098e-05,
      "loss": 3.7374,
      "step": 640
    },
    {
      "epoch": 0.05,
      "grad_norm": 72.25853729248047,
      "learning_rate": 2.99992919618918e-05,
      "loss": 3.7735,
      "step": 660
    },
    {
      "epoch": 0.05,
      "grad_norm": 39.445220947265625,
      "learning_rate": 2.999874127328748e-05,
      "loss": 3.759,
      "step": 680
    },
    {
      "epoch": 0.06,
      "grad_norm": 21.18370246887207,
      "learning_rate": 2.9998033254984483e-05,
      "loss": 3.7841,
      "step": 700
    },
    {
      "epoch": 0.06,
      "grad_norm": 24.310373306274414,
      "learning_rate": 2.999716791440959e-05,
      "loss": 3.679,
      "step": 720
    },
    {
      "epoch": 0.06,
      "grad_norm": 36.432350158691406,
      "learning_rate": 2.9996145260639812e-05,
      "loss": 3.6796,
      "step": 740
    },
    {
      "epoch": 0.06,
      "grad_norm": 32.12275314331055,
      "learning_rate": 2.9994965304402304e-05,
      "loss": 3.7613,
      "step": 760
    },
    {
      "epoch": 0.06,
      "grad_norm": 38.32442855834961,
      "learning_rate": 2.999362805807425e-05,
      "loss": 3.7586,
      "step": 780
    },
    {
      "epoch": 0.06,
      "grad_norm": 30.289432525634766,
      "learning_rate": 2.9992133535682725e-05,
      "loss": 3.6919,
      "step": 800
    },
    {
      "epoch": 0.07,
      "grad_norm": 32.69138717651367,
      "learning_rate": 2.9990481752904566e-05,
      "loss": 3.6855,
      "step": 820
    },
    {
      "epoch": 0.07,
      "grad_norm": 46.554874420166016,
      "learning_rate": 2.9988672727066197e-05,
      "loss": 3.7201,
      "step": 840
    },
    {
      "epoch": 0.07,
      "grad_norm": 28.671123504638672,
      "learning_rate": 2.9986706477143436e-05,
      "loss": 3.6594,
      "step": 860
    },
    {
      "epoch": 0.07,
      "grad_norm": 49.44480895996094,
      "learning_rate": 2.9984583023761318e-05,
      "loss": 3.7271,
      "step": 880
    },
    {
      "epoch": 0.07,
      "grad_norm": 26.61457061767578,
      "learning_rate": 2.998230238919386e-05,
      "loss": 3.7376,
      "step": 900
    },
    {
      "epoch": 0.07,
      "grad_norm": 27.453275680541992,
      "learning_rate": 2.9979864597363846e-05,
      "loss": 3.6716,
      "step": 920
    },
    {
      "epoch": 0.08,
      "grad_norm": 22.791175842285156,
      "learning_rate": 2.9977269673842554e-05,
      "loss": 3.6172,
      "step": 940
    },
    {
      "epoch": 0.08,
      "grad_norm": 58.2718620300293,
      "learning_rate": 2.997451764584951e-05,
      "loss": 3.7494,
      "step": 960
    },
    {
      "epoch": 0.08,
      "grad_norm": 33.610286712646484,
      "learning_rate": 2.9971608542252175e-05,
      "loss": 3.7077,
      "step": 980
    },
    {
      "epoch": 0.08,
      "grad_norm": 25.48147201538086,
      "learning_rate": 2.9968542393565674e-05,
      "loss": 3.6721,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "grad_norm": 27.07135581970215,
      "learning_rate": 2.996531923195246e-05,
      "loss": 3.7106,
      "step": 1020
    },
    {
      "epoch": 0.08,
      "grad_norm": 37.24673843383789,
      "learning_rate": 2.996193909122197e-05,
      "loss": 3.7447,
      "step": 1040
    }
  ],
  "logging_steps": 20,
  "max_steps": 20000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 20,
  "total_flos": 2473139331366912.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}