{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4342058085335245,
  "eval_steps": 2000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 2.6033709049224854,
      "learning_rate": 9.9e-07,
      "loss": 1.0789,
      "step": 100
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.7718708515167236,
      "learning_rate": 9.9e-07,
      "loss": 0.8366,
      "step": 200
    },
    {
      "epoch": 0.11,
      "grad_norm": 4.209937572479248,
      "learning_rate": 9.8e-07,
      "loss": 0.6863,
      "step": 300
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.8079888820648193,
      "learning_rate": 9.698989898989898e-07,
      "loss": 0.6659,
      "step": 400
    },
    {
      "epoch": 0.18,
      "grad_norm": 4.12679386138916,
      "learning_rate": 9.598989898989899e-07,
      "loss": 0.6546,
      "step": 500
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.909221887588501,
      "learning_rate": 9.497979797979798e-07,
      "loss": 0.6581,
      "step": 600
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.234121322631836,
      "learning_rate": 9.396969696969696e-07,
      "loss": 0.6224,
      "step": 700
    },
    {
      "epoch": 0.29,
      "grad_norm": 3.9791226387023926,
      "learning_rate": 9.295959595959596e-07,
      "loss": 0.6139,
      "step": 800
    },
    {
      "epoch": 0.32,
      "grad_norm": 3.557487726211548,
      "learning_rate": 9.194949494949495e-07,
      "loss": 0.6272,
      "step": 900
    },
    {
      "epoch": 0.36,
      "grad_norm": 3.946579694747925,
      "learning_rate": 9.093939393939394e-07,
      "loss": 0.6312,
      "step": 1000
    },
    {
      "epoch": 0.39,
      "grad_norm": 3.4085192680358887,
      "learning_rate": 8.992929292929292e-07,
      "loss": 0.6139,
      "step": 1100
    },
    {
      "epoch": 0.43,
      "grad_norm": 3.036348342895508,
      "learning_rate": 8.891919191919191e-07,
      "loss": 0.637,
      "step": 1200
    },
    {
      "epoch": 0.47,
      "grad_norm": 5.595344066619873,
      "learning_rate": 8.790909090909091e-07,
      "loss": 0.6447,
      "step": 1300
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.76248025894165,
      "learning_rate": 8.68989898989899e-07,
      "loss": 0.6419,
      "step": 1400
    },
    {
      "epoch": 0.54,
      "grad_norm": 2.9645845890045166,
      "learning_rate": 8.588888888888888e-07,
      "loss": 0.6569,
      "step": 1500
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.357501268386841,
      "learning_rate": 8.487878787878787e-07,
      "loss": 0.6345,
      "step": 1600
    },
    {
      "epoch": 0.61,
      "grad_norm": 4.0495171546936035,
      "learning_rate": 8.386868686868687e-07,
      "loss": 0.6004,
      "step": 1700
    },
    {
      "epoch": 0.65,
      "grad_norm": 3.0601418018341064,
      "learning_rate": 8.285858585858585e-07,
      "loss": 0.6444,
      "step": 1800
    },
    {
      "epoch": 0.68,
      "grad_norm": 4.655466079711914,
      "learning_rate": 8.184848484848484e-07,
      "loss": 0.621,
      "step": 1900
    },
    {
      "epoch": 0.72,
      "grad_norm": 4.7475666999816895,
      "learning_rate": 8.083838383838384e-07,
      "loss": 0.6554,
      "step": 2000
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.7412441372871399,
      "eval_runtime": 565.2186,
      "eval_samples_per_second": 1.769,
      "eval_steps_per_second": 0.442,
      "step": 2000
    },
    {
      "epoch": 0.75,
      "grad_norm": 4.251164436340332,
      "learning_rate": 7.982828282828282e-07,
      "loss": 0.5995,
      "step": 2100
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.217769145965576,
      "learning_rate": 7.881818181818182e-07,
      "loss": 0.5995,
      "step": 2200
    },
    {
      "epoch": 0.82,
      "grad_norm": 3.5968048572540283,
      "learning_rate": 7.78080808080808e-07,
      "loss": 0.5888,
      "step": 2300
    },
    {
      "epoch": 0.86,
      "grad_norm": 3.394106388092041,
      "learning_rate": 7.679797979797979e-07,
      "loss": 0.5945,
      "step": 2400
    },
    {
      "epoch": 0.9,
      "grad_norm": 2.829554796218872,
      "learning_rate": 7.578787878787879e-07,
      "loss": 0.6102,
      "step": 2500
    },
    {
      "epoch": 0.93,
      "grad_norm": 6.449063777923584,
      "learning_rate": 7.477777777777778e-07,
      "loss": 0.6231,
      "step": 2600
    },
    {
      "epoch": 0.97,
      "grad_norm": 5.725988388061523,
      "learning_rate": 7.376767676767676e-07,
      "loss": 0.5908,
      "step": 2700
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.436861515045166,
      "learning_rate": 7.276767676767677e-07,
      "loss": 0.6021,
      "step": 2800
    },
    {
      "epoch": 1.04,
      "grad_norm": 5.17264986038208,
      "learning_rate": 7.175757575757575e-07,
      "loss": 0.6142,
      "step": 2900
    },
    {
      "epoch": 1.08,
      "grad_norm": 4.872317790985107,
      "learning_rate": 7.074747474747474e-07,
      "loss": 0.572,
      "step": 3000
    },
    {
      "epoch": 1.11,
      "grad_norm": 2.839364767074585,
      "learning_rate": 6.973737373737374e-07,
      "loss": 0.5839,
      "step": 3100
    },
    {
      "epoch": 1.15,
      "grad_norm": 4.395648956298828,
      "learning_rate": 6.872727272727273e-07,
      "loss": 0.5789,
      "step": 3200
    },
    {
      "epoch": 1.18,
      "grad_norm": 3.2688417434692383,
      "learning_rate": 6.771717171717171e-07,
      "loss": 0.5866,
      "step": 3300
    },
    {
      "epoch": 1.22,
      "grad_norm": 3.7224113941192627,
      "learning_rate": 6.67070707070707e-07,
      "loss": 0.5899,
      "step": 3400
    },
    {
      "epoch": 1.25,
      "grad_norm": 5.710260391235352,
      "learning_rate": 6.56969696969697e-07,
      "loss": 0.5904,
      "step": 3500
    },
    {
      "epoch": 1.29,
      "grad_norm": 2.4542274475097656,
      "learning_rate": 6.468686868686868e-07,
      "loss": 0.576,
      "step": 3600
    },
    {
      "epoch": 1.33,
      "grad_norm": 3.639458417892456,
      "learning_rate": 6.367676767676767e-07,
      "loss": 0.5973,
      "step": 3700
    },
    {
      "epoch": 1.36,
      "grad_norm": 5.416239261627197,
      "learning_rate": 6.267676767676767e-07,
      "loss": 0.5855,
      "step": 3800
    },
    {
      "epoch": 1.4,
      "grad_norm": 4.7683491706848145,
      "learning_rate": 6.166666666666667e-07,
      "loss": 0.601,
      "step": 3900
    },
    {
      "epoch": 1.43,
      "grad_norm": 4.035972595214844,
      "learning_rate": 6.065656565656565e-07,
      "loss": 0.5937,
      "step": 4000
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.7268955111503601,
      "eval_runtime": 200.0217,
      "eval_samples_per_second": 4.999,
      "eval_steps_per_second": 1.25,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 2000,
  "total_flos": 3.770812536371282e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}