{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2,
  "eval_steps": 2000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 39.588829040527344,
      "learning_rate": 9.499999999999999e-07,
      "loss": 1.2125,
      "step": 100
    },
    {
      "epoch": 0.01,
      "grad_norm": 24.040372848510742,
      "learning_rate": 9.904040404040403e-07,
      "loss": 1.0367,
      "step": 200
    },
    {
      "epoch": 0.01,
      "grad_norm": 23.623809814453125,
      "learning_rate": 9.803030303030302e-07,
      "loss": 0.9513,
      "step": 300
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.24086859822273254,
      "learning_rate": 9.7020202020202e-07,
      "loss": 0.8004,
      "step": 400
    },
    {
      "epoch": 0.03,
      "grad_norm": 22.336132049560547,
      "learning_rate": 9.601010101010101e-07,
      "loss": 0.8263,
      "step": 500
    },
    {
      "epoch": 0.03,
      "grad_norm": 42.77682876586914,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.769,
      "step": 600
    },
    {
      "epoch": 0.04,
      "grad_norm": 16.354448318481445,
      "learning_rate": 9.398989898989899e-07,
      "loss": 0.7974,
      "step": 700
    },
    {
      "epoch": 0.04,
      "grad_norm": 40.2994270324707,
      "learning_rate": 9.297979797979798e-07,
      "loss": 0.8112,
      "step": 800
    },
    {
      "epoch": 0.04,
      "grad_norm": 15.487586975097656,
      "learning_rate": 9.196969696969696e-07,
      "loss": 0.7657,
      "step": 900
    },
    {
      "epoch": 0.05,
      "grad_norm": 20.020811080932617,
      "learning_rate": 9.095959595959596e-07,
      "loss": 0.7554,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.6317849159240723,
      "learning_rate": 8.995959595959596e-07,
      "loss": 0.7687,
      "step": 1100
    },
    {
      "epoch": 0.06,
      "grad_norm": 8.649179458618164,
      "learning_rate": 8.894949494949494e-07,
      "loss": 0.7567,
      "step": 1200
    },
    {
      "epoch": 0.07,
      "grad_norm": 27.93072509765625,
      "learning_rate": 8.793939393939394e-07,
      "loss": 0.7014,
      "step": 1300
    },
    {
      "epoch": 0.07,
      "grad_norm": 34.16617202758789,
      "learning_rate": 8.692929292929293e-07,
      "loss": 0.6642,
      "step": 1400
    },
    {
      "epoch": 0.07,
      "grad_norm": 61.1443977355957,
      "learning_rate": 8.591919191919191e-07,
      "loss": 0.7243,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "grad_norm": 69.2451400756836,
      "learning_rate": 8.490909090909091e-07,
      "loss": 0.7365,
      "step": 1600
    },
    {
      "epoch": 0.09,
      "grad_norm": 30.576330184936523,
      "learning_rate": 8.38989898989899e-07,
      "loss": 0.657,
      "step": 1700
    },
    {
      "epoch": 0.09,
      "grad_norm": 15.208847045898438,
      "learning_rate": 8.288888888888888e-07,
      "loss": 0.7884,
      "step": 1800
    },
    {
      "epoch": 0.1,
      "grad_norm": 39.78324508666992,
      "learning_rate": 8.187878787878788e-07,
      "loss": 0.7657,
      "step": 1900
    },
    {
      "epoch": 0.1,
      "grad_norm": 43.55222702026367,
      "learning_rate": 8.086868686868686e-07,
      "loss": 0.6678,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.751860499382019,
      "eval_runtime": 241.5179,
      "eval_samples_per_second": 4.14,
      "eval_steps_per_second": 1.035,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "grad_norm": 26.86771011352539,
      "learning_rate": 7.985858585858586e-07,
      "loss": 0.6361,
      "step": 2100
    },
    {
      "epoch": 0.11,
      "grad_norm": 25.721782684326172,
      "learning_rate": 7.884848484848484e-07,
      "loss": 0.8334,
      "step": 2200
    },
    {
      "epoch": 0.12,
      "grad_norm": 10.995915412902832,
      "learning_rate": 7.783838383838383e-07,
      "loss": 0.7052,
      "step": 2300
    },
    {
      "epoch": 0.12,
      "grad_norm": 33.5030403137207,
      "learning_rate": 7.682828282828282e-07,
      "loss": 0.6972,
      "step": 2400
    },
    {
      "epoch": 0.12,
      "grad_norm": 3.3341565132141113,
      "learning_rate": 7.581818181818182e-07,
      "loss": 0.7283,
      "step": 2500
    },
    {
      "epoch": 0.13,
      "grad_norm": 10.039573669433594,
      "learning_rate": 7.480808080808081e-07,
      "loss": 0.725,
      "step": 2600
    },
    {
      "epoch": 0.14,
      "grad_norm": 41.933143615722656,
      "learning_rate": 7.379797979797979e-07,
      "loss": 0.6081,
      "step": 2700
    },
    {
      "epoch": 0.14,
      "grad_norm": 32.512107849121094,
      "learning_rate": 7.278787878787879e-07,
      "loss": 0.7128,
      "step": 2800
    },
    {
      "epoch": 0.14,
      "grad_norm": 28.775693893432617,
      "learning_rate": 7.177777777777777e-07,
      "loss": 0.6757,
      "step": 2900
    },
    {
      "epoch": 0.15,
      "grad_norm": 47.666160583496094,
      "learning_rate": 7.076767676767677e-07,
      "loss": 0.587,
      "step": 3000
    },
    {
      "epoch": 0.15,
      "grad_norm": 23.707677841186523,
      "learning_rate": 6.975757575757576e-07,
      "loss": 0.7217,
      "step": 3100
    },
    {
      "epoch": 0.16,
      "grad_norm": 38.1664924621582,
      "learning_rate": 6.874747474747474e-07,
      "loss": 0.7153,
      "step": 3200
    },
    {
      "epoch": 0.17,
      "grad_norm": 43.21916198730469,
      "learning_rate": 6.773737373737374e-07,
      "loss": 0.7504,
      "step": 3300
    },
    {
      "epoch": 0.17,
      "grad_norm": 28.080886840820312,
      "learning_rate": 6.674747474747474e-07,
      "loss": 0.6828,
      "step": 3400
    },
    {
      "epoch": 0.17,
      "grad_norm": 24.497976303100586,
      "learning_rate": 6.573737373737374e-07,
      "loss": 0.6419,
      "step": 3500
    },
    {
      "epoch": 0.18,
      "grad_norm": 71.72598266601562,
      "learning_rate": 6.472727272727272e-07,
      "loss": 0.6036,
      "step": 3600
    },
    {
      "epoch": 0.18,
      "grad_norm": 30.938745498657227,
      "learning_rate": 6.371717171717171e-07,
      "loss": 0.705,
      "step": 3700
    },
    {
      "epoch": 0.19,
      "grad_norm": 27.01025390625,
      "learning_rate": 6.270707070707071e-07,
      "loss": 0.5679,
      "step": 3800
    },
    {
      "epoch": 0.2,
      "grad_norm": 24.72421646118164,
      "learning_rate": 6.169696969696969e-07,
      "loss": 0.6707,
      "step": 3900
    },
    {
      "epoch": 0.2,
      "grad_norm": 17.81386375427246,
      "learning_rate": 6.068686868686869e-07,
      "loss": 0.653,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.6371002197265625,
      "eval_runtime": 241.2671,
      "eval_samples_per_second": 4.145,
      "eval_steps_per_second": 1.036,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2000,
  "total_flos": 3.77128394686464e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}