{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2901,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05174197999310107,
      "grad_norm": 20.0,
      "learning_rate": 3.356164383561644e-06,
      "loss": 2.9015182495117187,
      "step": 50
    },
    {
      "epoch": 0.10348395998620213,
      "grad_norm": 10.8125,
      "learning_rate": 6.78082191780822e-06,
      "loss": 1.9724142456054687,
      "step": 100
    },
    {
      "epoch": 0.1552259399793032,
      "grad_norm": 12.875,
      "learning_rate": 9.999970742404346e-06,
      "loss": 1.6598092651367187,
      "step": 150
    },
    {
      "epoch": 0.20696791997240427,
      "grad_norm": 8.5,
      "learning_rate": 9.990871149623236e-06,
      "loss": 1.5588894653320313,
      "step": 200
    },
    {
      "epoch": 0.25870989996550536,
      "grad_norm": 8.25,
      "learning_rate": 9.965551392499958e-06,
      "loss": 1.5084381103515625,
      "step": 250
    },
    {
      "epoch": 0.3104518799586064,
      "grad_norm": 8.4375,
      "learning_rate": 9.924093759398476e-06,
      "loss": 1.4763961791992188,
      "step": 300
    },
    {
      "epoch": 0.3621938599517075,
      "grad_norm": 7.5,
      "learning_rate": 9.86663298624003e-06,
      "loss": 1.45826416015625,
      "step": 350
    },
    {
      "epoch": 0.41393583994480854,
      "grad_norm": 6.90625,
      "learning_rate": 9.793355818615905e-06,
      "loss": 1.4410775756835938,
      "step": 400
    },
    {
      "epoch": 0.46567781993790963,
      "grad_norm": 7.09375,
      "learning_rate": 9.7045004048705e-06,
      "loss": 1.4302684020996095,
      "step": 450
    },
    {
      "epoch": 0.5174197999310107,
      "grad_norm": 6.46875,
      "learning_rate": 9.600355522127174e-06,
      "loss": 1.4159922790527344,
      "step": 500
    },
    {
      "epoch": 0.5174197999310107,
      "eval_loss": 1.3360999822616577,
      "eval_runtime": 48.5888,
      "eval_samples_per_second": 51.452,
      "eval_steps_per_second": 0.556,
      "step": 500
    },
    {
      "epoch": 0.5691617799241118,
      "grad_norm": 6.46875,
      "learning_rate": 9.481259637772251e-06,
      "loss": 1.408507080078125,
      "step": 550
    },
    {
      "epoch": 0.6209037599172128,
      "grad_norm": 8.0,
      "learning_rate": 9.347599809447331e-06,
      "loss": 1.406485595703125,
      "step": 600
    },
    {
      "epoch": 0.672645739910314,
      "grad_norm": 6.59375,
      "learning_rate": 9.199810427124904e-06,
      "loss": 1.3980792236328126,
      "step": 650
    },
    {
      "epoch": 0.724387719903415,
      "grad_norm": 6.1875,
      "learning_rate": 9.03837180135546e-06,
      "loss": 1.3900822448730468,
      "step": 700
    },
    {
      "epoch": 0.776129699896516,
      "grad_norm": 6.375,
      "learning_rate": 8.86380860227426e-06,
      "loss": 1.377584228515625,
      "step": 750
    },
    {
      "epoch": 0.8278716798896171,
      "grad_norm": 6.46875,
      "learning_rate": 8.676688154440947e-06,
      "loss": 1.3831480407714845,
      "step": 800
    },
    {
      "epoch": 0.8796136598827182,
      "grad_norm": 6.84375,
      "learning_rate": 8.477618593053693e-06,
      "loss": 1.3737605285644532,
      "step": 850
    },
    {
      "epoch": 0.9313556398758193,
      "grad_norm": 6.5625,
      "learning_rate": 8.26724688753015e-06,
      "loss": 1.3781863403320314,
      "step": 900
    },
    {
      "epoch": 0.9830976198689203,
      "grad_norm": 6.0,
      "learning_rate": 8.04625673887846e-06,
      "loss": 1.3639559936523438,
      "step": 950
    },
    {
      "epoch": 1.0341497067954466,
      "grad_norm": 5.84375,
      "learning_rate": 7.815366357691863e-06,
      "loss": 1.3568319702148437,
      "step": 1000
    },
    {
      "epoch": 1.0341497067954466,
      "eval_loss": 1.2891921997070312,
      "eval_runtime": 36.787,
      "eval_samples_per_second": 67.959,
      "eval_steps_per_second": 0.734,
      "step": 1000
    },
    {
      "epoch": 1.0858916867885477,
      "grad_norm": 5.84375,
      "learning_rate": 7.575326129988268e-06,
      "loss": 1.3579391479492187,
      "step": 1050
    },
    {
      "epoch": 1.137633666781649,
      "grad_norm": 6.15625,
      "learning_rate": 7.326916178480796e-06,
      "loss": 1.353500213623047,
      "step": 1100
    },
    {
      "epoch": 1.1893756467747498,
      "grad_norm": 5.625,
      "learning_rate": 7.070943827205048e-06,
      "loss": 1.3528330993652344,
      "step": 1150
    },
    {
      "epoch": 1.241117626767851,
      "grad_norm": 5.96875,
      "learning_rate": 6.808240977742987e-06,
      "loss": 1.3461532592773438,
      "step": 1200
    },
    {
      "epoch": 1.2928596067609521,
      "grad_norm": 5.28125,
      "learning_rate": 6.539661405570615e-06,
      "loss": 1.3481027221679687,
      "step": 1250
    },
    {
      "epoch": 1.344601586754053,
      "grad_norm": 6.09375,
      "learning_rate": 6.266077985316224e-06,
      "loss": 1.348444061279297,
      "step": 1300
    },
    {
      "epoch": 1.3963435667471542,
      "grad_norm": 6.1875,
      "learning_rate": 5.988379853947046e-06,
      "loss": 1.3519085693359374,
      "step": 1350
    },
    {
      "epoch": 1.4480855467402551,
      "grad_norm": 5.375,
      "learning_rate": 5.7074695211038385e-06,
      "loss": 1.3478456115722657,
      "step": 1400
    },
    {
      "epoch": 1.4998275267333563,
      "grad_norm": 5.84375,
      "learning_rate": 5.424259935974736e-06,
      "loss": 1.3487643432617187,
      "step": 1450
    },
    {
      "epoch": 1.5515695067264574,
      "grad_norm": 6.09375,
      "learning_rate": 5.139671520240915e-06,
      "loss": 1.3462490844726562,
      "step": 1500
    },
    {
      "epoch": 1.5515695067264574,
      "eval_loss": 1.2737045288085938,
      "eval_runtime": 35.6507,
      "eval_samples_per_second": 70.125,
      "eval_steps_per_second": 0.757,
      "step": 1500
    },
    {
      "epoch": 1.6033114867195586,
      "grad_norm": 10.125,
      "learning_rate": 4.854629176736911e-06,
      "loss": 1.34809326171875,
      "step": 1550
    },
    {
      "epoch": 1.6550534667126595,
      "grad_norm": 5.84375,
      "learning_rate": 4.570059283547328e-06,
      "loss": 1.343378448486328,
      "step": 1600
    },
    {
      "epoch": 1.7067954467057604,
      "grad_norm": 6.0625,
      "learning_rate": 4.286886683309053e-06,
      "loss": 1.341541748046875,
      "step": 1650
    },
    {
      "epoch": 1.7585374266988616,
      "grad_norm": 5.40625,
      "learning_rate": 4.0060316775036025e-06,
      "loss": 1.3389480590820313,
      "step": 1700
    },
    {
      "epoch": 1.8102794066919627,
      "grad_norm": 5.75,
      "learning_rate": 3.728407035508089e-06,
      "loss": 1.3394900512695314,
      "step": 1750
    },
    {
      "epoch": 1.862021386685064,
      "grad_norm": 5.8125,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 1.3372491455078126,
      "step": 1800
    },
    {
      "epoch": 1.913763366678165,
      "grad_norm": 5.71875,
      "learning_rate": 3.1864444952335823e-06,
      "loss": 1.3394888305664063,
      "step": 1850
    },
    {
      "epoch": 1.965505346671266,
      "grad_norm": 5.59375,
      "learning_rate": 2.9238679570873217e-06,
      "loss": 1.34095458984375,
      "step": 1900
    },
    {
      "epoch": 2.0165574335977925,
      "grad_norm": 6.0625,
      "learning_rate": 2.66803877865493e-06,
      "loss": 1.3422109985351562,
      "step": 1950
    },
    {
      "epoch": 2.068299413590893,
      "grad_norm": 5.65625,
      "learning_rate": 2.4197883962114164e-06,
      "loss": 1.3344174194335938,
      "step": 2000
    },
    {
      "epoch": 2.068299413590893,
      "eval_loss": 1.269867181777954,
      "eval_runtime": 37.0763,
      "eval_samples_per_second": 67.429,
      "eval_steps_per_second": 0.728,
      "step": 2000
    },
    {
      "epoch": 2.1200413935839943,
      "grad_norm": 5.84375,
      "learning_rate": 2.1799236151982623e-06,
      "loss": 1.3353939819335938,
      "step": 2050
    },
    {
      "epoch": 2.1717833735770955,
      "grad_norm": 5.5,
      "learning_rate": 1.9492239881327103e-06,
      "loss": 1.3307119750976562,
      "step": 2100
    },
    {
      "epoch": 2.2235253535701966,
      "grad_norm": 6.65625,
      "learning_rate": 1.7284392810881568e-06,
      "loss": 1.3394575500488282,
      "step": 2150
    },
    {
      "epoch": 2.275267333563298,
      "grad_norm": 6.59375,
      "learning_rate": 1.5182870369794838e-06,
      "loss": 1.3347157287597655,
      "step": 2200
    },
    {
      "epoch": 2.327009313556399,
      "grad_norm": 5.25,
      "learning_rate": 1.3194502435725787e-06,
      "loss": 1.3337785339355468,
      "step": 2250
    },
    {
      "epoch": 2.3787512935494997,
      "grad_norm": 5.78125,
      "learning_rate": 1.1325751137969381e-06,
      "loss": 1.3284304809570313,
      "step": 2300
    },
    {
      "epoch": 2.430493273542601,
      "grad_norm": 6.09375,
      "learning_rate": 9.582689855752515e-07,
      "loss": 1.3337535095214843,
      "step": 2350
    },
    {
      "epoch": 2.482235253535702,
      "grad_norm": 5.875,
      "learning_rate": 7.970983479954631e-07,
      "loss": 1.3316661071777345,
      "step": 2400
    },
    {
      "epoch": 2.533977233528803,
      "grad_norm": 5.6875,
      "learning_rate": 6.495870002401822e-07,
      "loss": 1.3348265075683594,
      "step": 2450
    },
    {
      "epoch": 2.5857192135219043,
      "grad_norm": 5.59375,
      "learning_rate": 5.162143492568511e-07,
      "loss": 1.3390269470214844,
      "step": 2500
    },
    {
      "epoch": 2.5857192135219043,
      "eval_loss": 1.269131064414978,
      "eval_runtime": 36.297,
      "eval_samples_per_second": 68.876,
      "eval_steps_per_second": 0.744,
      "step": 2500
    },
    {
      "epoch": 2.637461193515005,
      "grad_norm": 5.8125,
      "learning_rate": 3.9741385170120273e-07,
      "loss": 1.3350772094726562,
      "step": 2550
    },
    {
      "epoch": 2.689203173508106,
      "grad_norm": 5.75,
      "learning_rate": 2.935716052176163e-07,
      "loss": 1.336797637939453,
      "step": 2600
    },
    {
      "epoch": 2.7409451535012073,
      "grad_norm": 5.84375,
      "learning_rate": 2.050250936346937e-07,
      "loss": 1.336610107421875,
      "step": 2650
    },
    {
      "epoch": 2.7926871334943084,
      "grad_norm": 5.5625,
      "learning_rate": 1.3206209015409532e-07,
      "loss": 1.3384811401367187,
      "step": 2700
    },
    {
      "epoch": 2.8444291134874096,
      "grad_norm": 6.03125,
      "learning_rate": 7.491972209725807e-08,
      "loss": 1.3393882751464843,
      "step": 2750
    },
    {
      "epoch": 2.8961710934805103,
      "grad_norm": 7.21875,
      "learning_rate": 3.3783700249533925e-08,
      "loss": 1.3347865295410157,
      "step": 2800
    },
    {
      "epoch": 2.947913073473612,
      "grad_norm": 5.71875,
      "learning_rate": 8.787715306344258e-09,
      "loss": 1.342543487548828,
      "step": 2850
    },
    {
      "epoch": 2.9996550534667126,
      "grad_norm": 6.15625,
      "learning_rate": 1.3003382891563221e-11,
      "loss": 1.3331295776367187,
      "step": 2900
    }
  ],
  "logging_steps": 50,
  "max_steps": 2901,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.833997897629958e+18,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}