{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 48.184357541899445,
  "eval_steps": 500,
  "global_step": 34500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7,
      "grad_norm": 4.0421319007873535,
      "learning_rate": 1.9720670391061455e-05,
      "loss": 0.408,
      "step": 500
    },
    {
      "epoch": 1.4,
      "grad_norm": 4.579340934753418,
      "learning_rate": 1.9441340782122907e-05,
      "loss": 0.303,
      "step": 1000
    },
    {
      "epoch": 2.09,
      "grad_norm": 5.2814459800720215,
      "learning_rate": 1.9162011173184357e-05,
      "loss": 0.2487,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "grad_norm": 4.263625144958496,
      "learning_rate": 1.888268156424581e-05,
      "loss": 0.1642,
      "step": 2000
    },
    {
      "epoch": 3.49,
      "grad_norm": 16.92409896850586,
      "learning_rate": 1.8603351955307266e-05,
      "loss": 0.1116,
      "step": 2500
    },
    {
      "epoch": 4.19,
      "grad_norm": 0.10890375822782516,
      "learning_rate": 1.8324022346368716e-05,
      "loss": 0.0992,
      "step": 3000
    },
    {
      "epoch": 4.89,
      "grad_norm": 4.843805313110352,
      "learning_rate": 1.804469273743017e-05,
      "loss": 0.0723,
      "step": 3500
    },
    {
      "epoch": 5.59,
      "grad_norm": 0.6422103643417358,
      "learning_rate": 1.776536312849162e-05,
      "loss": 0.0543,
      "step": 4000
    },
    {
      "epoch": 6.28,
      "grad_norm": 0.04145967587828636,
      "learning_rate": 1.7486033519553075e-05,
      "loss": 0.0479,
      "step": 4500
    },
    {
      "epoch": 6.98,
      "grad_norm": 0.18585649132728577,
      "learning_rate": 1.7206703910614527e-05,
      "loss": 0.0472,
      "step": 5000
    },
    {
      "epoch": 7.68,
      "grad_norm": 1.1734449863433838,
      "learning_rate": 1.6927374301675977e-05,
      "loss": 0.037,
      "step": 5500
    },
    {
      "epoch": 8.38,
      "grad_norm": 5.980381965637207,
      "learning_rate": 1.664804469273743e-05,
      "loss": 0.0352,
      "step": 6000
    },
    {
      "epoch": 9.08,
      "grad_norm": 0.02630673162639141,
      "learning_rate": 1.6368715083798886e-05,
      "loss": 0.031,
      "step": 6500
    },
    {
      "epoch": 9.78,
      "grad_norm": 0.34635427594184875,
      "learning_rate": 1.6089385474860336e-05,
      "loss": 0.0249,
      "step": 7000
    },
    {
      "epoch": 10.47,
      "grad_norm": 0.038617122918367386,
      "learning_rate": 1.581005586592179e-05,
      "loss": 0.0236,
      "step": 7500
    },
    {
      "epoch": 11.17,
      "grad_norm": 0.011197003535926342,
      "learning_rate": 1.553072625698324e-05,
      "loss": 0.024,
      "step": 8000
    },
    {
      "epoch": 11.87,
      "grad_norm": 0.04000287503004074,
      "learning_rate": 1.5251396648044694e-05,
      "loss": 0.0257,
      "step": 8500
    },
    {
      "epoch": 12.57,
      "grad_norm": 0.01364846620708704,
      "learning_rate": 1.4972067039106146e-05,
      "loss": 0.0185,
      "step": 9000
    },
    {
      "epoch": 13.27,
      "grad_norm": 0.030796006321907043,
      "learning_rate": 1.4692737430167599e-05,
      "loss": 0.0193,
      "step": 9500
    },
    {
      "epoch": 13.97,
      "grad_norm": 0.008376175537705421,
      "learning_rate": 1.4413407821229052e-05,
      "loss": 0.0177,
      "step": 10000
    },
    {
      "epoch": 14.66,
      "grad_norm": 0.004392937291413546,
      "learning_rate": 1.4134078212290503e-05,
      "loss": 0.0127,
      "step": 10500
    },
    {
      "epoch": 15.36,
      "grad_norm": 0.0029910472221672535,
      "learning_rate": 1.3854748603351957e-05,
      "loss": 0.014,
      "step": 11000
    },
    {
      "epoch": 16.06,
      "grad_norm": 0.0029884399846196175,
      "learning_rate": 1.357541899441341e-05,
      "loss": 0.0156,
      "step": 11500
    },
    {
      "epoch": 16.76,
      "grad_norm": 7.065054416656494,
      "learning_rate": 1.3296089385474861e-05,
      "loss": 0.0132,
      "step": 12000
    },
    {
      "epoch": 17.46,
      "grad_norm": 0.002008031355217099,
      "learning_rate": 1.3016759776536314e-05,
      "loss": 0.0105,
      "step": 12500
    },
    {
      "epoch": 18.16,
      "grad_norm": 0.004853041376918554,
      "learning_rate": 1.2737430167597766e-05,
      "loss": 0.0091,
      "step": 13000
    },
    {
      "epoch": 18.85,
      "grad_norm": 0.009278771467506886,
      "learning_rate": 1.2458100558659219e-05,
      "loss": 0.0144,
      "step": 13500
    },
    {
      "epoch": 19.55,
      "grad_norm": 0.002231143182143569,
      "learning_rate": 1.2178770949720671e-05,
      "loss": 0.0099,
      "step": 14000
    },
    {
      "epoch": 20.25,
      "grad_norm": 0.003844016930088401,
      "learning_rate": 1.1899441340782123e-05,
      "loss": 0.0132,
      "step": 14500
    },
    {
      "epoch": 20.95,
      "grad_norm": 0.0021888066548854113,
      "learning_rate": 1.1620111731843577e-05,
      "loss": 0.0126,
      "step": 15000
    },
    {
      "epoch": 21.65,
      "grad_norm": 16.925048828125,
      "learning_rate": 1.134078212290503e-05,
      "loss": 0.0077,
      "step": 15500
    },
    {
      "epoch": 22.35,
      "grad_norm": 0.0013325487961992621,
      "learning_rate": 1.1061452513966481e-05,
      "loss": 0.0108,
      "step": 16000
    },
    {
      "epoch": 23.04,
      "grad_norm": 0.0013474252773448825,
      "learning_rate": 1.0782122905027934e-05,
      "loss": 0.0078,
      "step": 16500
    },
    {
      "epoch": 23.74,
      "grad_norm": 0.0014491812326014042,
      "learning_rate": 1.0502793296089386e-05,
      "loss": 0.0048,
      "step": 17000
    },
    {
      "epoch": 24.44,
      "grad_norm": 0.0027435130905359983,
      "learning_rate": 1.0223463687150838e-05,
      "loss": 0.0072,
      "step": 17500
    },
    {
      "epoch": 25.14,
      "grad_norm": 0.06291136890649796,
      "learning_rate": 9.944134078212291e-06,
      "loss": 0.008,
      "step": 18000
    },
    {
      "epoch": 25.84,
      "grad_norm": 0.021899724379181862,
      "learning_rate": 9.664804469273744e-06,
      "loss": 0.0087,
      "step": 18500
    },
    {
      "epoch": 26.54,
      "grad_norm": 15.27237319946289,
      "learning_rate": 9.385474860335197e-06,
      "loss": 0.0052,
      "step": 19000
    },
    {
      "epoch": 27.23,
      "grad_norm": 0.009569833986461163,
      "learning_rate": 9.106145251396648e-06,
      "loss": 0.0073,
      "step": 19500
    },
    {
      "epoch": 27.93,
      "grad_norm": 0.0003999493783339858,
      "learning_rate": 8.826815642458101e-06,
      "loss": 0.0048,
      "step": 20000
    },
    {
      "epoch": 28.63,
      "grad_norm": 0.0033733926247805357,
      "learning_rate": 8.547486033519554e-06,
      "loss": 0.0056,
      "step": 20500
    },
    {
      "epoch": 29.33,
      "grad_norm": 0.04343196749687195,
      "learning_rate": 8.268156424581007e-06,
      "loss": 0.0056,
      "step": 21000
    },
    {
      "epoch": 30.03,
      "grad_norm": 0.0003582398348953575,
      "learning_rate": 7.988826815642458e-06,
      "loss": 0.0031,
      "step": 21500
    },
    {
      "epoch": 30.73,
      "grad_norm": 0.00021692673908546567,
      "learning_rate": 7.709497206703911e-06,
      "loss": 0.0035,
      "step": 22000
    },
    {
      "epoch": 31.42,
      "grad_norm": 0.00029874147730879486,
      "learning_rate": 7.430167597765364e-06,
      "loss": 0.0026,
      "step": 22500
    },
    {
      "epoch": 32.12,
      "grad_norm": 0.0002098192780977115,
      "learning_rate": 7.150837988826816e-06,
      "loss": 0.0048,
      "step": 23000
    },
    {
      "epoch": 32.82,
      "grad_norm": 0.010424572043120861,
      "learning_rate": 6.871508379888268e-06,
      "loss": 0.0054,
      "step": 23500
    },
    {
      "epoch": 33.52,
      "grad_norm": 0.1450434923171997,
      "learning_rate": 6.592178770949721e-06,
      "loss": 0.0031,
      "step": 24000
    },
    {
      "epoch": 34.22,
      "grad_norm": 0.0003349222242832184,
      "learning_rate": 6.312849162011173e-06,
      "loss": 0.0029,
      "step": 24500
    },
    {
      "epoch": 34.92,
      "grad_norm": 0.0007802957552485168,
      "learning_rate": 6.033519553072626e-06,
      "loss": 0.0023,
      "step": 25000
    },
    {
      "epoch": 35.61,
      "grad_norm": 0.0005215730634517968,
      "learning_rate": 5.754189944134078e-06,
      "loss": 0.0021,
      "step": 25500
    },
    {
      "epoch": 36.31,
      "grad_norm": 0.00011289090616628528,
      "learning_rate": 5.474860335195531e-06,
      "loss": 0.0026,
      "step": 26000
    },
    {
      "epoch": 37.01,
      "grad_norm": 0.013641629368066788,
      "learning_rate": 5.195530726256983e-06,
      "loss": 0.0032,
      "step": 26500
    },
    {
      "epoch": 37.71,
      "grad_norm": 0.0002486356534063816,
      "learning_rate": 4.916201117318436e-06,
      "loss": 0.002,
      "step": 27000
    },
    {
      "epoch": 38.41,
      "grad_norm": 9.347203013021499e-05,
      "learning_rate": 4.636871508379888e-06,
      "loss": 0.0012,
      "step": 27500
    },
    {
      "epoch": 39.11,
      "grad_norm": 0.00012943429464939982,
      "learning_rate": 4.357541899441341e-06,
      "loss": 0.0026,
      "step": 28000
    },
    {
      "epoch": 39.8,
      "grad_norm": 0.001636040979065001,
      "learning_rate": 4.078212290502794e-06,
      "loss": 0.0021,
      "step": 28500
    },
    {
      "epoch": 40.5,
      "grad_norm": 0.009456774219870567,
      "learning_rate": 3.798882681564246e-06,
      "loss": 0.0016,
      "step": 29000
    },
    {
      "epoch": 41.2,
      "grad_norm": 0.000818349071778357,
      "learning_rate": 3.5195530726256988e-06,
      "loss": 0.0011,
      "step": 29500
    },
    {
      "epoch": 41.9,
      "grad_norm": 6.146980012999848e-05,
      "learning_rate": 3.240223463687151e-06,
      "loss": 0.0016,
      "step": 30000
    },
    {
      "epoch": 42.6,
      "grad_norm": 4.060289211338386e-05,
      "learning_rate": 2.9608938547486037e-06,
      "loss": 0.0007,
      "step": 30500
    },
    {
      "epoch": 43.3,
      "grad_norm": 0.00016211264301091433,
      "learning_rate": 2.6815642458100562e-06,
      "loss": 0.0032,
      "step": 31000
    },
    {
      "epoch": 43.99,
      "grad_norm": 0.0002249513054266572,
      "learning_rate": 2.4022346368715087e-06,
      "loss": 0.0011,
      "step": 31500
    },
    {
      "epoch": 44.69,
      "grad_norm": 0.00012292143946979195,
      "learning_rate": 2.1229050279329612e-06,
      "loss": 0.0006,
      "step": 32000
    },
    {
      "epoch": 45.39,
      "grad_norm": 5.348429112927988e-05,
      "learning_rate": 1.8435754189944135e-06,
      "loss": 0.0007,
      "step": 32500
    },
    {
      "epoch": 46.09,
      "grad_norm": 5.2925995987607166e-05,
      "learning_rate": 1.564245810055866e-06,
      "loss": 0.0021,
      "step": 33000
    },
    {
      "epoch": 46.79,
      "grad_norm": 6.103613850427791e-05,
      "learning_rate": 1.2849162011173185e-06,
      "loss": 0.0009,
      "step": 33500
    },
    {
      "epoch": 47.49,
      "grad_norm": 0.0004408392414916307,
      "learning_rate": 1.005586592178771e-06,
      "loss": 0.0013,
      "step": 34000
    },
    {
      "epoch": 48.18,
      "grad_norm": 0.0011864238185808063,
      "learning_rate": 7.262569832402236e-07,
      "loss": 0.0013,
      "step": 34500
    }
  ],
  "logging_steps": 500,
  "max_steps": 35800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 2.970794017734924e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}