{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4,
  "eval_steps": 2000,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 12.253900527954102,
      "learning_rate": 9.8e-07,
      "loss": 0.3335,
      "step": 100
    },
    {
      "epoch": 0.01,
      "grad_norm": 9.193947792053223,
      "learning_rate": 9.9010101010101e-07,
      "loss": 0.2105,
      "step": 200
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.168125152587891,
      "learning_rate": 9.8e-07,
      "loss": 0.1618,
      "step": 300
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.163803815841675,
      "learning_rate": 9.698989898989898e-07,
      "loss": 0.1418,
      "step": 400
    },
    {
      "epoch": 0.03,
      "grad_norm": 10.751598358154297,
      "learning_rate": 9.597979797979797e-07,
      "loss": 0.1215,
      "step": 500
    },
    {
      "epoch": 0.03,
      "grad_norm": 7.5342793464660645,
      "learning_rate": 9.496969696969696e-07,
      "loss": 0.1142,
      "step": 600
    },
    {
      "epoch": 0.04,
      "grad_norm": 8.430835723876953,
      "learning_rate": 9.395959595959596e-07,
      "loss": 0.121,
      "step": 700
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.8440964221954346,
      "learning_rate": 9.295959595959596e-07,
      "loss": 0.0969,
      "step": 800
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.699073553085327,
      "learning_rate": 9.194949494949495e-07,
      "loss": 0.1024,
      "step": 900
    },
    {
      "epoch": 0.05,
      "grad_norm": 11.251789093017578,
      "learning_rate": 9.093939393939394e-07,
      "loss": 0.0899,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "grad_norm": 7.728929042816162,
      "learning_rate": 8.992929292929292e-07,
      "loss": 0.0837,
      "step": 1100
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.7401416301727295,
      "learning_rate": 8.891919191919191e-07,
      "loss": 0.0914,
      "step": 1200
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.5504443645477295,
      "learning_rate": 8.790909090909091e-07,
      "loss": 0.0754,
      "step": 1300
    },
    {
      "epoch": 0.07,
      "grad_norm": 5.9004316329956055,
      "learning_rate": 8.68989898989899e-07,
      "loss": 0.0791,
      "step": 1400
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.9171862602233887,
      "learning_rate": 8.588888888888888e-07,
      "loss": 0.0866,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "grad_norm": 5.907050132751465,
      "learning_rate": 8.487878787878787e-07,
      "loss": 0.0768,
      "step": 1600
    },
    {
      "epoch": 0.09,
      "grad_norm": 4.856807708740234,
      "learning_rate": 8.386868686868687e-07,
      "loss": 0.0745,
      "step": 1700
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.4457929134368896,
      "learning_rate": 8.285858585858585e-07,
      "loss": 0.0808,
      "step": 1800
    },
    {
      "epoch": 0.1,
      "grad_norm": 3.084287643432617,
      "learning_rate": 8.184848484848484e-07,
      "loss": 0.0743,
      "step": 1900
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.936805486679077,
      "learning_rate": 8.083838383838384e-07,
      "loss": 0.0675,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.07925247400999069,
      "eval_runtime": 204.8299,
      "eval_samples_per_second": 4.882,
      "eval_steps_per_second": 1.221,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "grad_norm": 13.029026985168457,
      "learning_rate": 7.982828282828282e-07,
      "loss": 0.0681,
      "step": 2100
    },
    {
      "epoch": 0.11,
      "grad_norm": 10.308150291442871,
      "learning_rate": 7.881818181818182e-07,
      "loss": 0.0798,
      "step": 2200
    },
    {
      "epoch": 0.12,
      "grad_norm": 5.64780330657959,
      "learning_rate": 7.78080808080808e-07,
      "loss": 0.0648,
      "step": 2300
    },
    {
      "epoch": 0.12,
      "grad_norm": 6.982935905456543,
      "learning_rate": 7.679797979797979e-07,
      "loss": 0.0731,
      "step": 2400
    },
    {
      "epoch": 0.12,
      "grad_norm": 3.5423080921173096,
      "learning_rate": 7.578787878787879e-07,
      "loss": 0.0817,
      "step": 2500
    },
    {
      "epoch": 0.13,
      "grad_norm": 3.585963249206543,
      "learning_rate": 7.477777777777778e-07,
      "loss": 0.072,
      "step": 2600
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.086956977844238,
      "learning_rate": 7.376767676767676e-07,
      "loss": 0.0577,
      "step": 2700
    },
    {
      "epoch": 0.14,
      "grad_norm": 4.985733509063721,
      "learning_rate": 7.275757575757575e-07,
      "loss": 0.0672,
      "step": 2800
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.4518052339553833,
      "learning_rate": 7.174747474747475e-07,
      "loss": 0.0616,
      "step": 2900
    },
    {
      "epoch": 0.15,
      "grad_norm": 5.120323657989502,
      "learning_rate": 7.073737373737373e-07,
      "loss": 0.0625,
      "step": 3000
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0982943773269653,
      "learning_rate": 6.972727272727273e-07,
      "loss": 0.0666,
      "step": 3100
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.8519880771636963,
      "learning_rate": 6.871717171717171e-07,
      "loss": 0.0786,
      "step": 3200
    },
    {
      "epoch": 0.17,
      "grad_norm": 3.394327163696289,
      "learning_rate": 6.77070707070707e-07,
      "loss": 0.0786,
      "step": 3300
    },
    {
      "epoch": 0.17,
      "grad_norm": 9.026866912841797,
      "learning_rate": 6.66969696969697e-07,
      "loss": 0.0677,
      "step": 3400
    },
    {
      "epoch": 0.17,
      "grad_norm": 3.718776226043701,
      "learning_rate": 6.568686868686868e-07,
      "loss": 0.0576,
      "step": 3500
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.7873659729957581,
      "learning_rate": 6.467676767676767e-07,
      "loss": 0.0559,
      "step": 3600
    },
    {
      "epoch": 0.18,
      "grad_norm": 5.039899826049805,
      "learning_rate": 6.366666666666667e-07,
      "loss": 0.0668,
      "step": 3700
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.4250056743621826,
      "learning_rate": 6.266666666666667e-07,
      "loss": 0.0524,
      "step": 3800
    },
    {
      "epoch": 0.2,
      "grad_norm": 7.657347202301025,
      "learning_rate": 6.165656565656565e-07,
      "loss": 0.0665,
      "step": 3900
    },
    {
      "epoch": 0.2,
      "grad_norm": 3.9768455028533936,
      "learning_rate": 6.064646464646465e-07,
      "loss": 0.0649,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.06388480961322784,
      "eval_runtime": 200.3994,
      "eval_samples_per_second": 4.99,
      "eval_steps_per_second": 1.248,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "grad_norm": 6.998288154602051,
      "learning_rate": 5.963636363636363e-07,
      "loss": 0.0488,
      "step": 4100
    },
    {
      "epoch": 0.21,
      "grad_norm": 2.007672071456909,
      "learning_rate": 5.862626262626262e-07,
      "loss": 0.0683,
      "step": 4200
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5769347548484802,
      "learning_rate": 5.761616161616162e-07,
      "loss": 0.0544,
      "step": 4300
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.005931401159614325,
      "learning_rate": 5.660606060606061e-07,
      "loss": 0.0603,
      "step": 4400
    },
    {
      "epoch": 0.23,
      "grad_norm": 8.401103973388672,
      "learning_rate": 5.559595959595959e-07,
      "loss": 0.0531,
      "step": 4500
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.780881404876709,
      "learning_rate": 5.459595959595959e-07,
      "loss": 0.0729,
      "step": 4600
    },
    {
      "epoch": 0.23,
      "grad_norm": 8.754217147827148,
      "learning_rate": 5.358585858585858e-07,
      "loss": 0.0766,
      "step": 4700
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.9296135902404785,
      "learning_rate": 5.257575757575757e-07,
      "loss": 0.0651,
      "step": 4800
    },
    {
      "epoch": 0.24,
      "grad_norm": 4.884037494659424,
      "learning_rate": 5.156565656565657e-07,
      "loss": 0.0623,
      "step": 4900
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.247363567352295,
      "learning_rate": 5.055555555555555e-07,
      "loss": 0.0498,
      "step": 5000
    },
    {
      "epoch": 0.26,
      "grad_norm": 5.299038887023926,
      "learning_rate": 4.954545454545454e-07,
      "loss": 0.0558,
      "step": 5100
    },
    {
      "epoch": 0.26,
      "grad_norm": 3.8896191120147705,
      "learning_rate": 4.853535353535353e-07,
      "loss": 0.0733,
      "step": 5200
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.1511714458465576,
      "learning_rate": 4.752525252525252e-07,
      "loss": 0.0647,
      "step": 5300
    },
    {
      "epoch": 0.27,
      "grad_norm": 3.958899736404419,
      "learning_rate": 4.6515151515151513e-07,
      "loss": 0.0594,
      "step": 5400
    },
    {
      "epoch": 0.28,
      "grad_norm": 7.097384929656982,
      "learning_rate": 4.55050505050505e-07,
      "loss": 0.0623,
      "step": 5500
    },
    {
      "epoch": 0.28,
      "grad_norm": 6.70374870300293,
      "learning_rate": 4.449494949494949e-07,
      "loss": 0.0669,
      "step": 5600
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.3009291887283325,
      "learning_rate": 4.3484848484848483e-07,
      "loss": 0.0528,
      "step": 5700
    },
    {
      "epoch": 0.29,
      "grad_norm": 4.70461368560791,
      "learning_rate": 4.2474747474747474e-07,
      "loss": 0.0668,
      "step": 5800
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.03208538889884949,
      "learning_rate": 4.1464646464646466e-07,
      "loss": 0.0641,
      "step": 5900
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.0016973208403214812,
      "learning_rate": 4.045454545454545e-07,
      "loss": 0.064,
      "step": 6000
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.06021637097001076,
      "eval_runtime": 200.1582,
      "eval_samples_per_second": 4.996,
      "eval_steps_per_second": 1.249,
      "step": 6000
    },
    {
      "epoch": 0.3,
      "grad_norm": 9.762524604797363,
      "learning_rate": 3.9444444444444444e-07,
      "loss": 0.0639,
      "step": 6100
    },
    {
      "epoch": 0.31,
      "grad_norm": 6.966504096984863,
      "learning_rate": 3.843434343434343e-07,
      "loss": 0.0541,
      "step": 6200
    },
    {
      "epoch": 0.32,
      "grad_norm": 4.614800930023193,
      "learning_rate": 3.7424242424242427e-07,
      "loss": 0.049,
      "step": 6300
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.268702745437622,
      "learning_rate": 3.6414141414141413e-07,
      "loss": 0.0571,
      "step": 6400
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.18738865852355957,
      "learning_rate": 3.5404040404040405e-07,
      "loss": 0.0701,
      "step": 6500
    },
    {
      "epoch": 0.33,
      "grad_norm": 3.5662453174591064,
      "learning_rate": 3.439393939393939e-07,
      "loss": 0.0513,
      "step": 6600
    },
    {
      "epoch": 0.34,
      "grad_norm": 15.757116317749023,
      "learning_rate": 3.3383838383838383e-07,
      "loss": 0.0542,
      "step": 6700
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.947461485862732,
      "learning_rate": 3.237373737373737e-07,
      "loss": 0.0616,
      "step": 6800
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.4099074602127075,
      "learning_rate": 3.1363636363636366e-07,
      "loss": 0.067,
      "step": 6900
    },
    {
      "epoch": 0.35,
      "grad_norm": 4.870973587036133,
      "learning_rate": 3.035353535353535e-07,
      "loss": 0.0547,
      "step": 7000
    },
    {
      "epoch": 0.35,
      "grad_norm": 5.2733049392700195,
      "learning_rate": 2.9343434343434344e-07,
      "loss": 0.0489,
      "step": 7100
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.496048927307129,
      "learning_rate": 2.833333333333333e-07,
      "loss": 0.0528,
      "step": 7200
    },
    {
      "epoch": 0.36,
      "grad_norm": 5.224077224731445,
      "learning_rate": 2.732323232323232e-07,
      "loss": 0.0587,
      "step": 7300
    },
    {
      "epoch": 0.37,
      "grad_norm": 7.1024699211120605,
      "learning_rate": 2.631313131313131e-07,
      "loss": 0.0564,
      "step": 7400
    },
    {
      "epoch": 0.38,
      "grad_norm": 2.4630818367004395,
      "learning_rate": 2.5303030303030305e-07,
      "loss": 0.0567,
      "step": 7500
    },
    {
      "epoch": 0.38,
      "grad_norm": 5.36046838760376,
      "learning_rate": 2.429292929292929e-07,
      "loss": 0.0453,
      "step": 7600
    },
    {
      "epoch": 0.39,
      "grad_norm": 3.9095382690429688,
      "learning_rate": 2.3282828282828283e-07,
      "loss": 0.0626,
      "step": 7700
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.7284829020500183,
      "learning_rate": 2.2272727272727272e-07,
      "loss": 0.0621,
      "step": 7800
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.15286648273468018,
      "learning_rate": 2.1262626262626264e-07,
      "loss": 0.0535,
      "step": 7900
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.735705852508545,
      "learning_rate": 2.0252525252525253e-07,
      "loss": 0.053,
      "step": 8000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.05654788389801979,
      "eval_runtime": 200.4178,
      "eval_samples_per_second": 4.99,
      "eval_steps_per_second": 1.247,
      "step": 8000
    }
  ],
  "logging_steps": 100,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2000,
  "total_flos": 7.54256789372928e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}