{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 864,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11574074074074074,
      "grad_norm": 0.8536216020584106,
      "learning_rate": 5.6818181818181825e-05,
      "loss": 1.711,
      "step": 25
    },
    {
      "epoch": 0.23148148148148148,
      "grad_norm": 0.9026548862457275,
      "learning_rate": 9.926829268292683e-05,
      "loss": 1.4077,
      "step": 50
    },
    {
      "epoch": 0.3472222222222222,
      "grad_norm": 0.7126691341400146,
      "learning_rate": 9.621951219512195e-05,
      "loss": 1.2191,
      "step": 75
    },
    {
      "epoch": 0.46296296296296297,
      "grad_norm": 0.8097984194755554,
      "learning_rate": 9.317073170731708e-05,
      "loss": 1.1951,
      "step": 100
    },
    {
      "epoch": 0.5787037037037037,
      "grad_norm": 0.6823762655258179,
      "learning_rate": 9.01219512195122e-05,
      "loss": 1.2015,
      "step": 125
    },
    {
      "epoch": 0.6944444444444444,
      "grad_norm": 0.7754731178283691,
      "learning_rate": 8.707317073170732e-05,
      "loss": 1.1927,
      "step": 150
    },
    {
      "epoch": 0.8101851851851852,
      "grad_norm": 0.7315147519111633,
      "learning_rate": 8.402439024390244e-05,
      "loss": 1.1953,
      "step": 175
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 0.6208141446113586,
      "learning_rate": 8.097560975609757e-05,
      "loss": 1.1844,
      "step": 200
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 0.6663596630096436,
      "learning_rate": 7.792682926829269e-05,
      "loss": 1.2056,
      "step": 225
    },
    {
      "epoch": 1.1574074074074074,
      "grad_norm": 0.6997539401054382,
      "learning_rate": 7.487804878048781e-05,
      "loss": 1.1872,
      "step": 250
    },
    {
      "epoch": 1.2731481481481481,
      "grad_norm": 0.7418563365936279,
      "learning_rate": 7.182926829268293e-05,
      "loss": 1.1479,
      "step": 275
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.7975765466690063,
      "learning_rate": 6.878048780487805e-05,
      "loss": 1.1444,
      "step": 300
    },
    {
      "epoch": 1.5046296296296298,
      "grad_norm": 0.759464681148529,
      "learning_rate": 6.573170731707318e-05,
      "loss": 1.1614,
      "step": 325
    },
    {
      "epoch": 1.6203703703703702,
      "grad_norm": 0.7860772609710693,
      "learning_rate": 6.26829268292683e-05,
      "loss": 1.1572,
      "step": 350
    },
    {
      "epoch": 1.7361111111111112,
      "grad_norm": 0.6840630769729614,
      "learning_rate": 5.9634146341463414e-05,
      "loss": 1.1765,
      "step": 375
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 0.6867539882659912,
      "learning_rate": 5.6585365853658533e-05,
      "loss": 1.1464,
      "step": 400
    },
    {
      "epoch": 1.9675925925925926,
      "grad_norm": 0.6985242366790771,
      "learning_rate": 5.3536585365853666e-05,
      "loss": 1.1647,
      "step": 425
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 0.8595041036605835,
      "learning_rate": 5.0487804878048785e-05,
      "loss": 1.1289,
      "step": 450
    },
    {
      "epoch": 2.199074074074074,
      "grad_norm": 0.7451562285423279,
      "learning_rate": 4.7439024390243905e-05,
      "loss": 1.1422,
      "step": 475
    },
    {
      "epoch": 2.314814814814815,
      "grad_norm": 0.8289864659309387,
      "learning_rate": 4.4390243902439024e-05,
      "loss": 1.1268,
      "step": 500
    },
    {
      "epoch": 2.4305555555555554,
      "grad_norm": 0.8567008972167969,
      "learning_rate": 4.134146341463414e-05,
      "loss": 1.1591,
      "step": 525
    },
    {
      "epoch": 2.5462962962962963,
      "grad_norm": 0.8465055823326111,
      "learning_rate": 3.829268292682927e-05,
      "loss": 1.1542,
      "step": 550
    },
    {
      "epoch": 2.662037037037037,
      "grad_norm": 0.8150965571403503,
      "learning_rate": 3.524390243902439e-05,
      "loss": 1.1287,
      "step": 575
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.8683028221130371,
      "learning_rate": 3.2195121951219514e-05,
      "loss": 1.1429,
      "step": 600
    },
    {
      "epoch": 2.8935185185185186,
      "grad_norm": 0.8814287781715393,
      "learning_rate": 2.9146341463414634e-05,
      "loss": 1.1416,
      "step": 625
    },
    {
      "epoch": 3.009259259259259,
      "grad_norm": 0.7834553122520447,
      "learning_rate": 2.609756097560976e-05,
      "loss": 1.1372,
      "step": 650
    },
    {
      "epoch": 3.125,
      "grad_norm": 0.9108631014823914,
      "learning_rate": 2.304878048780488e-05,
      "loss": 1.14,
      "step": 675
    },
    {
      "epoch": 3.240740740740741,
      "grad_norm": 0.8761746287345886,
      "learning_rate": 2e-05,
      "loss": 1.123,
      "step": 700
    },
    {
      "epoch": 3.3564814814814814,
      "grad_norm": 0.8887725472450256,
      "learning_rate": 1.6951219512195124e-05,
      "loss": 1.1083,
      "step": 725
    },
    {
      "epoch": 3.4722222222222223,
      "grad_norm": 0.857607364654541,
      "learning_rate": 1.3902439024390245e-05,
      "loss": 1.1211,
      "step": 750
    },
    {
      "epoch": 3.587962962962963,
      "grad_norm": 0.9724395871162415,
      "learning_rate": 1.0853658536585366e-05,
      "loss": 1.1038,
      "step": 775
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 0.9074528217315674,
      "learning_rate": 7.804878048780489e-06,
      "loss": 1.1196,
      "step": 800
    },
    {
      "epoch": 3.8194444444444446,
      "grad_norm": 0.9636090397834778,
      "learning_rate": 4.75609756097561e-06,
      "loss": 1.134,
      "step": 825
    },
    {
      "epoch": 3.935185185185185,
      "grad_norm": 0.8954392075538635,
      "learning_rate": 1.707317073170732e-06,
      "loss": 1.1378,
      "step": 850
    }
  ],
  "logging_steps": 25,
  "max_steps": 864,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1.5105772588577587e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}