{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 8072,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004956015363647627,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 4.2404,
      "step": 1
    },
    {
      "epoch": 0.04956015363647627,
      "grad_norm": 3.30698561668396,
      "learning_rate": 4.9435973720094217e-05,
      "loss": 1.484,
      "step": 100
    },
    {
      "epoch": 0.09912030727295254,
      "grad_norm": 2.357815980911255,
      "learning_rate": 4.8816164621296644e-05,
      "loss": 1.1277,
      "step": 200
    },
    {
      "epoch": 0.14868046090942882,
      "grad_norm": 2.1948750019073486,
      "learning_rate": 4.819635552249907e-05,
      "loss": 1.0774,
      "step": 300
    },
    {
      "epoch": 0.19824061454590508,
      "grad_norm": 2.270810842514038,
      "learning_rate": 4.75765464237015e-05,
      "loss": 1.0326,
      "step": 400
    },
    {
      "epoch": 0.24780076818238136,
      "grad_norm": 2.18212628364563,
      "learning_rate": 4.6956737324903934e-05,
      "loss": 1.0031,
      "step": 500
    },
    {
      "epoch": 0.24780076818238136,
      "eval_loss": 1.9913930892944336,
      "eval_runtime": 368.7305,
      "eval_samples_per_second": 6.571,
      "eval_steps_per_second": 3.287,
      "step": 500
    },
    {
      "epoch": 0.29736092181885765,
      "grad_norm": 1.9237951040267944,
      "learning_rate": 4.633692822610636e-05,
      "loss": 0.9989,
      "step": 600
    },
    {
      "epoch": 0.3469210754553339,
      "grad_norm": 2.1836133003234863,
      "learning_rate": 4.571711912730879e-05,
      "loss": 0.9794,
      "step": 700
    },
    {
      "epoch": 0.39648122909181016,
      "grad_norm": 1.7673603296279907,
      "learning_rate": 4.509731002851122e-05,
      "loss": 0.959,
      "step": 800
    },
    {
      "epoch": 0.4460413827282865,
      "grad_norm": 1.9361021518707275,
      "learning_rate": 4.447750092971365e-05,
      "loss": 0.9235,
      "step": 900
    },
    {
      "epoch": 0.49560153636476273,
      "grad_norm": 1.8775012493133545,
      "learning_rate": 4.385769183091608e-05,
      "loss": 0.9235,
      "step": 1000
    },
    {
      "epoch": 0.49560153636476273,
      "eval_loss": 1.903882384300232,
      "eval_runtime": 361.7269,
      "eval_samples_per_second": 6.698,
      "eval_steps_per_second": 3.351,
      "step": 1000
    },
    {
      "epoch": 0.545161690001239,
      "grad_norm": 1.8856130838394165,
      "learning_rate": 4.3237882732118514e-05,
      "loss": 0.9073,
      "step": 1100
    },
    {
      "epoch": 0.5947218436377153,
      "grad_norm": 1.8579490184783936,
      "learning_rate": 4.2618073633320935e-05,
      "loss": 0.8897,
      "step": 1200
    },
    {
      "epoch": 0.6442819972741916,
      "grad_norm": 1.821420431137085,
      "learning_rate": 4.199826453452337e-05,
      "loss": 0.8924,
      "step": 1300
    },
    {
      "epoch": 0.6938421509106678,
      "grad_norm": 1.6502270698547363,
      "learning_rate": 4.13784554357258e-05,
      "loss": 0.8813,
      "step": 1400
    },
    {
      "epoch": 0.7434023045471441,
      "grad_norm": 2.301623821258545,
      "learning_rate": 4.075864633692823e-05,
      "loss": 0.8831,
      "step": 1500
    },
    {
      "epoch": 0.7434023045471441,
      "eval_loss": 1.8592430353164673,
      "eval_runtime": 356.2419,
      "eval_samples_per_second": 6.802,
      "eval_steps_per_second": 3.402,
      "step": 1500
    },
    {
      "epoch": 0.7929624581836203,
      "grad_norm": 1.7900115251541138,
      "learning_rate": 4.013883723813065e-05,
      "loss": 0.8828,
      "step": 1600
    },
    {
      "epoch": 0.8425226118200967,
      "grad_norm": 1.7043579816818237,
      "learning_rate": 3.951902813933309e-05,
      "loss": 0.8722,
      "step": 1700
    },
    {
      "epoch": 0.892082765456573,
      "grad_norm": 1.7436531782150269,
      "learning_rate": 3.8899219040535514e-05,
      "loss": 0.8537,
      "step": 1800
    },
    {
      "epoch": 0.9416429190930492,
      "grad_norm": 1.583708643913269,
      "learning_rate": 3.827940994173795e-05,
      "loss": 0.8487,
      "step": 1900
    },
    {
      "epoch": 0.9912030727295255,
      "grad_norm": 1.6228455305099487,
      "learning_rate": 3.7659600842940377e-05,
      "loss": 0.836,
      "step": 2000
    },
    {
      "epoch": 0.9912030727295255,
      "eval_loss": 1.8353278636932373,
      "eval_runtime": 357.9669,
      "eval_samples_per_second": 6.769,
      "eval_steps_per_second": 3.386,
      "step": 2000
    },
    {
      "epoch": 1.0406393259819104,
      "grad_norm": 1.728725552558899,
      "learning_rate": 3.7039791744142804e-05,
      "loss": 0.75,
      "step": 2100
    },
    {
      "epoch": 1.090199479618387,
      "grad_norm": 1.7833404541015625,
      "learning_rate": 3.641998264534524e-05,
      "loss": 0.7392,
      "step": 2200
    },
    {
      "epoch": 1.1397596332548632,
      "grad_norm": 1.4817216396331787,
      "learning_rate": 3.5800173546547666e-05,
      "loss": 0.7349,
      "step": 2300
    },
    {
      "epoch": 1.1893197868913394,
      "grad_norm": 1.362112045288086,
      "learning_rate": 3.5180364447750094e-05,
      "loss": 0.7279,
      "step": 2400
    },
    {
      "epoch": 1.2388799405278157,
      "grad_norm": 1.6990805864334106,
      "learning_rate": 3.456055534895252e-05,
      "loss": 0.7303,
      "step": 2500
    },
    {
      "epoch": 1.2388799405278157,
      "eval_loss": 1.9066349267959595,
      "eval_runtime": 361.0756,
      "eval_samples_per_second": 6.711,
      "eval_steps_per_second": 3.357,
      "step": 2500
    },
    {
      "epoch": 1.288440094164292,
      "grad_norm": 1.5519583225250244,
      "learning_rate": 3.3940746250154956e-05,
      "loss": 0.739,
      "step": 2600
    },
    {
      "epoch": 1.3380002478007682,
      "grad_norm": 1.9304168224334717,
      "learning_rate": 3.3320937151357384e-05,
      "loss": 0.7384,
      "step": 2700
    },
    {
      "epoch": 1.3875604014372445,
      "grad_norm": 2.0171093940734863,
      "learning_rate": 3.270112805255982e-05,
      "loss": 0.7314,
      "step": 2800
    },
    {
      "epoch": 1.4371205550737207,
      "grad_norm": 1.5190871953964233,
      "learning_rate": 3.208131895376224e-05,
      "loss": 0.723,
      "step": 2900
    },
    {
      "epoch": 1.486680708710197,
      "grad_norm": 1.839789867401123,
      "learning_rate": 3.1461509854964674e-05,
      "loss": 0.7402,
      "step": 3000
    },
    {
      "epoch": 1.486680708710197,
      "eval_loss": 1.939505696296692,
      "eval_runtime": 357.1918,
      "eval_samples_per_second": 6.783,
      "eval_steps_per_second": 3.393,
      "step": 3000
    },
    {
      "epoch": 1.5362408623466732,
      "grad_norm": 1.6048545837402344,
      "learning_rate": 3.08417007561671e-05,
      "loss": 0.722,
      "step": 3100
    },
    {
      "epoch": 1.5858010159831495,
      "grad_norm": 1.897549033164978,
      "learning_rate": 3.0221891657369533e-05,
      "loss": 0.7244,
      "step": 3200
    },
    {
      "epoch": 1.635361169619626,
      "grad_norm": 1.4337180852890015,
      "learning_rate": 2.960208255857196e-05,
      "loss": 0.719,
      "step": 3300
    },
    {
      "epoch": 1.684921323256102,
      "grad_norm": 1.6136891841888428,
      "learning_rate": 2.8982273459774388e-05,
      "loss": 0.7163,
      "step": 3400
    },
    {
      "epoch": 1.7344814768925785,
      "grad_norm": 1.6104755401611328,
      "learning_rate": 2.836246436097682e-05,
      "loss": 0.7196,
      "step": 3500
    },
    {
      "epoch": 1.7344814768925785,
      "eval_loss": 1.9235131740570068,
      "eval_runtime": 356.6603,
      "eval_samples_per_second": 6.794,
      "eval_steps_per_second": 3.398,
      "step": 3500
    },
    {
      "epoch": 1.7840416305290545,
      "grad_norm": 1.4487009048461914,
      "learning_rate": 2.774265526217925e-05,
      "loss": 0.7232,
      "step": 3600
    },
    {
      "epoch": 1.833601784165531,
      "grad_norm": 1.6850144863128662,
      "learning_rate": 2.712284616338168e-05,
      "loss": 0.7134,
      "step": 3700
    },
    {
      "epoch": 1.883161937802007,
      "grad_norm": 1.427085518836975,
      "learning_rate": 2.650303706458411e-05,
      "loss": 0.7205,
      "step": 3800
    },
    {
      "epoch": 1.9327220914384835,
      "grad_norm": 1.330480694770813,
      "learning_rate": 2.588322796578654e-05,
      "loss": 0.7072,
      "step": 3900
    },
    {
      "epoch": 1.9822822450749598,
      "grad_norm": 1.7128593921661377,
      "learning_rate": 2.526341886698897e-05,
      "loss": 0.709,
      "step": 4000
    },
    {
      "epoch": 1.9822822450749598,
      "eval_loss": 1.8779014348983765,
      "eval_runtime": 360.7502,
      "eval_samples_per_second": 6.717,
      "eval_steps_per_second": 3.36,
      "step": 4000
    },
    {
      "epoch": 2.031718498327345,
      "grad_norm": 1.6081246137619019,
      "learning_rate": 2.46436097681914e-05,
      "loss": 0.6167,
      "step": 4100
    },
    {
      "epoch": 2.081278651963821,
      "grad_norm": 2.1736507415771484,
      "learning_rate": 2.402380066939383e-05,
      "loss": 0.562,
      "step": 4200
    },
    {
      "epoch": 2.1308388056002974,
      "grad_norm": 1.9378325939178467,
      "learning_rate": 2.3403991570596257e-05,
      "loss": 0.5701,
      "step": 4300
    },
    {
      "epoch": 2.180398959236774,
      "grad_norm": 1.6395761966705322,
      "learning_rate": 2.278418247179869e-05,
      "loss": 0.5634,
      "step": 4400
    },
    {
      "epoch": 2.22995911287325,
      "grad_norm": 1.547757863998413,
      "learning_rate": 2.2164373373001116e-05,
      "loss": 0.5664,
      "step": 4500
    },
    {
      "epoch": 2.22995911287325,
      "eval_loss": 2.0903470516204834,
      "eval_runtime": 360.0829,
      "eval_samples_per_second": 6.729,
      "eval_steps_per_second": 3.366,
      "step": 4500
    },
    {
      "epoch": 2.2795192665097264,
      "grad_norm": 2.287740468978882,
      "learning_rate": 2.1544564274203547e-05,
      "loss": 0.5761,
      "step": 4600
    },
    {
      "epoch": 2.3290794201462024,
      "grad_norm": 1.8061494827270508,
      "learning_rate": 2.092475517540598e-05,
      "loss": 0.5847,
      "step": 4700
    },
    {
      "epoch": 2.378639573782679,
      "grad_norm": 1.8823641538619995,
      "learning_rate": 2.0304946076608406e-05,
      "loss": 0.5769,
      "step": 4800
    },
    {
      "epoch": 2.428199727419155,
      "grad_norm": 1.9263031482696533,
      "learning_rate": 1.9685136977810837e-05,
      "loss": 0.5767,
      "step": 4900
    },
    {
      "epoch": 2.4777598810556314,
      "grad_norm": 1.8592733144760132,
      "learning_rate": 1.9065327879013265e-05,
      "loss": 0.5676,
      "step": 5000
    },
    {
      "epoch": 2.4777598810556314,
      "eval_loss": 2.1367952823638916,
      "eval_runtime": 358.8557,
      "eval_samples_per_second": 6.752,
      "eval_steps_per_second": 3.377,
      "step": 5000
    },
    {
      "epoch": 2.5273200346921074,
      "grad_norm": 1.823480248451233,
      "learning_rate": 1.8445518780215696e-05,
      "loss": 0.5719,
      "step": 5100
    },
    {
      "epoch": 2.576880188328584,
      "grad_norm": 1.9600774049758911,
      "learning_rate": 1.7825709681418124e-05,
      "loss": 0.5679,
      "step": 5200
    },
    {
      "epoch": 2.62644034196506,
      "grad_norm": 1.8784770965576172,
      "learning_rate": 1.7205900582620555e-05,
      "loss": 0.5743,
      "step": 5300
    },
    {
      "epoch": 2.6760004956015364,
      "grad_norm": 1.8293484449386597,
      "learning_rate": 1.6586091483822982e-05,
      "loss": 0.5593,
      "step": 5400
    },
    {
      "epoch": 2.7255606492380124,
      "grad_norm": 2.11364483833313,
      "learning_rate": 1.5966282385025413e-05,
      "loss": 0.5734,
      "step": 5500
    },
    {
      "epoch": 2.7255606492380124,
      "eval_loss": 2.2089123725891113,
      "eval_runtime": 357.9895,
      "eval_samples_per_second": 6.768,
      "eval_steps_per_second": 3.386,
      "step": 5500
    },
    {
      "epoch": 2.775120802874489,
      "grad_norm": 1.7178990840911865,
      "learning_rate": 1.534647328622784e-05,
      "loss": 0.5628,
      "step": 5600
    },
    {
      "epoch": 2.8246809565109654,
      "grad_norm": 2.3309977054595947,
      "learning_rate": 1.4726664187430272e-05,
      "loss": 0.5703,
      "step": 5700
    },
    {
      "epoch": 2.8742411101474414,
      "grad_norm": 1.5812301635742188,
      "learning_rate": 1.41068550886327e-05,
      "loss": 0.5684,
      "step": 5800
    },
    {
      "epoch": 2.9238012637839175,
      "grad_norm": 1.6736018657684326,
      "learning_rate": 1.3487045989835131e-05,
      "loss": 0.5672,
      "step": 5900
    },
    {
      "epoch": 2.973361417420394,
      "grad_norm": 1.706903338432312,
      "learning_rate": 1.2867236891037562e-05,
      "loss": 0.5599,
      "step": 6000
    },
    {
      "epoch": 2.973361417420394,
      "eval_loss": 2.1762378215789795,
      "eval_runtime": 358.2691,
      "eval_samples_per_second": 6.763,
      "eval_steps_per_second": 3.383,
      "step": 6000
    },
    {
      "epoch": 3.0227976706727793,
      "grad_norm": 2.0913047790527344,
      "learning_rate": 1.224742779223999e-05,
      "loss": 0.4929,
      "step": 6100
    },
    {
      "epoch": 3.0723578243092553,
      "grad_norm": 1.9073973894119263,
      "learning_rate": 1.162761869344242e-05,
      "loss": 0.3932,
      "step": 6200
    },
    {
      "epoch": 3.121917977945732,
      "grad_norm": 2.258511781692505,
      "learning_rate": 1.100780959464485e-05,
      "loss": 0.3935,
      "step": 6300
    },
    {
      "epoch": 3.171478131582208,
      "grad_norm": 2.307436227798462,
      "learning_rate": 1.038800049584728e-05,
      "loss": 0.3901,
      "step": 6400
    },
    {
      "epoch": 3.2210382852186843,
      "grad_norm": 1.9356776475906372,
      "learning_rate": 9.768191397049709e-06,
      "loss": 0.397,
      "step": 6500
    },
    {
      "epoch": 3.2210382852186843,
      "eval_loss": 2.597865343093872,
      "eval_runtime": 361.0516,
      "eval_samples_per_second": 6.711,
      "eval_steps_per_second": 3.357,
      "step": 6500
    },
    {
      "epoch": 3.2705984388551603,
      "grad_norm": 2.4313840866088867,
      "learning_rate": 9.14838229825214e-06,
      "loss": 0.3951,
      "step": 6600
    },
    {
      "epoch": 3.320158592491637,
      "grad_norm": 2.332550048828125,
      "learning_rate": 8.52857319945457e-06,
      "loss": 0.3919,
      "step": 6700
    },
    {
      "epoch": 3.369718746128113,
      "grad_norm": 2.176802396774292,
      "learning_rate": 7.908764100656999e-06,
      "loss": 0.399,
      "step": 6800
    },
    {
      "epoch": 3.4192788997645893,
      "grad_norm": 2.2205862998962402,
      "learning_rate": 7.288955001859428e-06,
      "loss": 0.3926,
      "step": 6900
    },
    {
      "epoch": 3.4688390534010654,
      "grad_norm": 2.149296760559082,
      "learning_rate": 6.6691459030618575e-06,
      "loss": 0.3841,
      "step": 7000
    },
    {
      "epoch": 3.4688390534010654,
      "eval_loss": 2.627171754837036,
      "eval_runtime": 359.3677,
      "eval_samples_per_second": 6.742,
      "eval_steps_per_second": 3.373,
      "step": 7000
    },
    {
      "epoch": 3.518399207037542,
      "grad_norm": 2.374778985977173,
      "learning_rate": 6.049336804264287e-06,
      "loss": 0.3836,
      "step": 7100
    },
    {
      "epoch": 3.567959360674018,
      "grad_norm": 2.034752607345581,
      "learning_rate": 5.429527705466716e-06,
      "loss": 0.3945,
      "step": 7200
    },
    {
      "epoch": 3.6175195143104943,
      "grad_norm": 2.6508090496063232,
      "learning_rate": 4.809718606669146e-06,
      "loss": 0.3958,
      "step": 7300
    },
    {
      "epoch": 3.667079667946971,
      "grad_norm": 2.4304332733154297,
      "learning_rate": 4.189909507871576e-06,
      "loss": 0.3853,
      "step": 7400
    },
    {
      "epoch": 3.716639821583447,
      "grad_norm": 2.457920551300049,
      "learning_rate": 3.570100409074005e-06,
      "loss": 0.3815,
      "step": 7500
    },
    {
      "epoch": 3.716639821583447,
      "eval_loss": 2.6776695251464844,
      "eval_runtime": 358.5716,
      "eval_samples_per_second": 6.757,
      "eval_steps_per_second": 3.38,
      "step": 7500
    },
    {
      "epoch": 3.7661999752199233,
      "grad_norm": 2.602248191833496,
      "learning_rate": 2.950291310276435e-06,
      "loss": 0.3863,
      "step": 7600
    },
    {
      "epoch": 3.8157601288563994,
      "grad_norm": 2.3076493740081787,
      "learning_rate": 2.330482211478865e-06,
      "loss": 0.3882,
      "step": 7700
    },
    {
      "epoch": 3.865320282492876,
      "grad_norm": 2.247474193572998,
      "learning_rate": 1.7106731126812943e-06,
      "loss": 0.384,
      "step": 7800
    },
    {
      "epoch": 3.914880436129352,
      "grad_norm": 2.400883913040161,
      "learning_rate": 1.0908640138837239e-06,
      "loss": 0.377,
      "step": 7900
    },
    {
      "epoch": 3.9644405897658284,
      "grad_norm": 2.1551015377044678,
      "learning_rate": 4.710549150861535e-07,
      "loss": 0.3834,
      "step": 8000
    },
    {
      "epoch": 3.9644405897658284,
      "eval_loss": 2.66070818901062,
      "eval_runtime": 356.8712,
      "eval_samples_per_second": 6.79,
      "eval_steps_per_second": 3.396,
      "step": 8000
    }
  ],
  "logging_steps": 100,
  "max_steps": 8072,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.6417343971328e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}