{
  "best_global_step": 3000,
  "best_metric": 3.9151711463928223,
  "best_model_checkpoint": "/kaggle/working/checkpoints/checkpoint-3000",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 3279,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04577706569008926,
      "grad_norm": 0.950545608997345,
      "learning_rate": 2.474747474747475e-05,
      "loss": 26.02827392578125,
      "step": 50
    },
    {
      "epoch": 0.09155413138017852,
      "grad_norm": 0.9253817200660706,
      "learning_rate": 5e-05,
      "loss": 25.6423583984375,
      "step": 100
    },
    {
      "epoch": 0.1373311970702678,
      "grad_norm": 1.8487942218780518,
      "learning_rate": 4.9213836477987426e-05,
      "loss": 25.38603759765625,
      "step": 150
    },
    {
      "epoch": 0.18310826276035705,
      "grad_norm": 0.9318061470985413,
      "learning_rate": 4.842767295597484e-05,
      "loss": 25.09637451171875,
      "step": 200
    },
    {
      "epoch": 0.22888532845044632,
      "grad_norm": 1.2386919260025024,
      "learning_rate": 4.7641509433962266e-05,
      "loss": 24.62987548828125,
      "step": 250
    },
    {
      "epoch": 0.2746623941405356,
      "grad_norm": 0.9762535691261292,
      "learning_rate": 4.685534591194969e-05,
      "loss": 24.32124755859375,
      "step": 300
    },
    {
      "epoch": 0.3204394598306249,
      "grad_norm": 1.3880475759506226,
      "learning_rate": 4.606918238993711e-05,
      "loss": 24.26835205078125,
      "step": 350
    },
    {
      "epoch": 0.3662165255207141,
      "grad_norm": 0.7988409399986267,
      "learning_rate": 4.528301886792453e-05,
      "loss": 23.8698388671875,
      "step": 400
    },
    {
      "epoch": 0.41199359121080337,
      "grad_norm": 0.7432121634483337,
      "learning_rate": 4.449685534591195e-05,
      "loss": 23.6873876953125,
      "step": 450
    },
    {
      "epoch": 0.45777065690089264,
      "grad_norm": 2.9906067848205566,
      "learning_rate": 4.3710691823899376e-05,
      "loss": 23.65133056640625,
      "step": 500
    },
    {
      "epoch": 0.45777065690089264,
      "eval_loss": 4.262399196624756,
      "eval_runtime": 11.9465,
      "eval_samples_per_second": 59.097,
      "eval_steps_per_second": 7.45,
      "step": 500
    },
    {
      "epoch": 0.5035477225909819,
      "grad_norm": 0.7991481423377991,
      "learning_rate": 4.292452830188679e-05,
      "loss": 23.3265283203125,
      "step": 550
    },
    {
      "epoch": 0.5493247882810712,
      "grad_norm": 0.8507863283157349,
      "learning_rate": 4.213836477987422e-05,
      "loss": 23.256435546875,
      "step": 600
    },
    {
      "epoch": 0.5951018539711604,
      "grad_norm": 0.8129443526268005,
      "learning_rate": 4.135220125786164e-05,
      "loss": 23.2051806640625,
      "step": 650
    },
    {
      "epoch": 0.6408789196612498,
      "grad_norm": 0.7767907977104187,
      "learning_rate": 4.0566037735849064e-05,
      "loss": 22.90304443359375,
      "step": 700
    },
    {
      "epoch": 0.686655985351339,
      "grad_norm": 0.8058496117591858,
      "learning_rate": 3.977987421383648e-05,
      "loss": 22.94605224609375,
      "step": 750
    },
    {
      "epoch": 0.7324330510414282,
      "grad_norm": 0.7904766798019409,
      "learning_rate": 3.8993710691823904e-05,
      "loss": 22.76631591796875,
      "step": 800
    },
    {
      "epoch": 0.7782101167315175,
      "grad_norm": 0.7378940582275391,
      "learning_rate": 3.820754716981133e-05,
      "loss": 22.6701904296875,
      "step": 850
    },
    {
      "epoch": 0.8239871824216067,
      "grad_norm": 0.9154200553894043,
      "learning_rate": 3.7421383647798744e-05,
      "loss": 22.53183349609375,
      "step": 900
    },
    {
      "epoch": 0.8697642481116961,
      "grad_norm": 0.8045285940170288,
      "learning_rate": 3.663522012578616e-05,
      "loss": 22.55562744140625,
      "step": 950
    },
    {
      "epoch": 0.9155413138017853,
      "grad_norm": 1.2049347162246704,
      "learning_rate": 3.5849056603773584e-05,
      "loss": 22.30632568359375,
      "step": 1000
    },
    {
      "epoch": 0.9155413138017853,
      "eval_loss": 4.1011881828308105,
      "eval_runtime": 11.9002,
      "eval_samples_per_second": 59.327,
      "eval_steps_per_second": 7.479,
      "step": 1000
    },
    {
      "epoch": 0.9613183794918746,
      "grad_norm": 0.8331048488616943,
      "learning_rate": 3.506289308176101e-05,
      "loss": 22.24485107421875,
      "step": 1050
    },
    {
      "epoch": 1.0064087891966125,
      "grad_norm": 0.7863403558731079,
      "learning_rate": 3.4276729559748424e-05,
      "loss": 21.805654296875,
      "step": 1100
    },
    {
      "epoch": 1.0521858548867018,
      "grad_norm": 0.8146964907646179,
      "learning_rate": 3.349056603773585e-05,
      "loss": 22.14404052734375,
      "step": 1150
    },
    {
      "epoch": 1.0979629205767911,
      "grad_norm": 0.7487536072731018,
      "learning_rate": 3.270440251572327e-05,
      "loss": 22.0405126953125,
      "step": 1200
    },
    {
      "epoch": 1.1437399862668802,
      "grad_norm": 0.7023004293441772,
      "learning_rate": 3.1918238993710694e-05,
      "loss": 21.87947021484375,
      "step": 1250
    },
    {
      "epoch": 1.1895170519569696,
      "grad_norm": 0.8285568356513977,
      "learning_rate": 3.113207547169811e-05,
      "loss": 21.8118017578125,
      "step": 1300
    },
    {
      "epoch": 1.2352941176470589,
      "grad_norm": 0.8095338940620422,
      "learning_rate": 3.0345911949685535e-05,
      "loss": 21.712080078125,
      "step": 1350
    },
    {
      "epoch": 1.281071183337148,
      "grad_norm": 0.7827941179275513,
      "learning_rate": 2.9559748427672958e-05,
      "loss": 21.71359375,
      "step": 1400
    },
    {
      "epoch": 1.3268482490272373,
      "grad_norm": 0.7628238201141357,
      "learning_rate": 2.8773584905660378e-05,
      "loss": 21.643154296875,
      "step": 1450
    },
    {
      "epoch": 1.3726253147173266,
      "grad_norm": 0.802753746509552,
      "learning_rate": 2.7987421383647798e-05,
      "loss": 21.62091796875,
      "step": 1500
    },
    {
      "epoch": 1.3726253147173266,
      "eval_loss": 4.01209020614624,
      "eval_runtime": 11.9226,
      "eval_samples_per_second": 59.215,
      "eval_steps_per_second": 7.465,
      "step": 1500
    },
    {
      "epoch": 1.418402380407416,
      "grad_norm": 0.7911498546600342,
      "learning_rate": 2.720125786163522e-05,
      "loss": 21.520439453125,
      "step": 1550
    },
    {
      "epoch": 1.4641794460975053,
      "grad_norm": 0.7967684864997864,
      "learning_rate": 2.641509433962264e-05,
      "loss": 21.51121337890625,
      "step": 1600
    },
    {
      "epoch": 1.5099565117875944,
      "grad_norm": 0.8296108841896057,
      "learning_rate": 2.5628930817610065e-05,
      "loss": 21.4302001953125,
      "step": 1650
    },
    {
      "epoch": 1.5557335774776837,
      "grad_norm": 0.799700140953064,
      "learning_rate": 2.4842767295597485e-05,
      "loss": 21.27000244140625,
      "step": 1700
    },
    {
      "epoch": 1.601510643167773,
      "grad_norm": 0.7474468946456909,
      "learning_rate": 2.405660377358491e-05,
      "loss": 21.2945166015625,
      "step": 1750
    },
    {
      "epoch": 1.6472877088578621,
      "grad_norm": 0.8560371994972229,
      "learning_rate": 2.327044025157233e-05,
      "loss": 21.26280517578125,
      "step": 1800
    },
    {
      "epoch": 1.6930647745479515,
      "grad_norm": 0.7146944999694824,
      "learning_rate": 2.248427672955975e-05,
      "loss": 21.30601806640625,
      "step": 1850
    },
    {
      "epoch": 1.7388418402380408,
      "grad_norm": 0.6856210827827454,
      "learning_rate": 2.1698113207547172e-05,
      "loss": 21.30221435546875,
      "step": 1900
    },
    {
      "epoch": 1.7846189059281299,
      "grad_norm": 0.6851561069488525,
      "learning_rate": 2.0911949685534592e-05,
      "loss": 21.15216552734375,
      "step": 1950
    },
    {
      "epoch": 1.8303959716182194,
      "grad_norm": 0.8206948637962341,
      "learning_rate": 2.0125786163522016e-05,
      "loss": 21.1641748046875,
      "step": 2000
    },
    {
      "epoch": 1.8303959716182194,
      "eval_loss": 3.951535940170288,
      "eval_runtime": 11.8555,
      "eval_samples_per_second": 59.55,
      "eval_steps_per_second": 7.507,
      "step": 2000
    },
    {
      "epoch": 1.8761730373083085,
      "grad_norm": 0.7400506734848022,
      "learning_rate": 1.9339622641509436e-05,
      "loss": 21.05539306640625,
      "step": 2050
    },
    {
      "epoch": 1.9219501029983979,
      "grad_norm": 0.6370204091072083,
      "learning_rate": 1.8553459119496856e-05,
      "loss": 21.11379638671875,
      "step": 2100
    },
    {
      "epoch": 1.9677271686884872,
      "grad_norm": 0.8683918118476868,
      "learning_rate": 1.7767295597484276e-05,
      "loss": 21.1146435546875,
      "step": 2150
    },
    {
      "epoch": 2.012817578393225,
      "grad_norm": 0.7333459258079529,
      "learning_rate": 1.69811320754717e-05,
      "loss": 20.737890625,
      "step": 2200
    },
    {
      "epoch": 2.058594644083314,
      "grad_norm": 0.680478572845459,
      "learning_rate": 1.619496855345912e-05,
      "loss": 21.101630859375,
      "step": 2250
    },
    {
      "epoch": 2.1043717097734036,
      "grad_norm": 0.8762068152427673,
      "learning_rate": 1.540880503144654e-05,
      "loss": 21.0653857421875,
      "step": 2300
    },
    {
      "epoch": 2.1501487754634927,
      "grad_norm": 0.8605155348777771,
      "learning_rate": 1.4622641509433963e-05,
      "loss": 21.01951171875,
      "step": 2350
    },
    {
      "epoch": 2.1959258411535822,
      "grad_norm": 0.8293413519859314,
      "learning_rate": 1.3836477987421385e-05,
      "loss": 20.94828369140625,
      "step": 2400
    },
    {
      "epoch": 2.2417029068436714,
      "grad_norm": 0.6899017095565796,
      "learning_rate": 1.3050314465408805e-05,
      "loss": 20.9380712890625,
      "step": 2450
    },
    {
      "epoch": 2.2874799725337605,
      "grad_norm": 0.7276245951652527,
      "learning_rate": 1.2264150943396227e-05,
      "loss": 21.033203125,
      "step": 2500
    },
    {
      "epoch": 2.2874799725337605,
      "eval_loss": 3.9277429580688477,
      "eval_runtime": 11.8976,
      "eval_samples_per_second": 59.34,
      "eval_steps_per_second": 7.48,
      "step": 2500
    },
    {
      "epoch": 2.33325703822385,
      "grad_norm": 0.6670855283737183,
      "learning_rate": 1.1477987421383648e-05,
      "loss": 20.87577880859375,
      "step": 2550
    },
    {
      "epoch": 2.379034103913939,
      "grad_norm": 0.7529842853546143,
      "learning_rate": 1.069182389937107e-05,
      "loss": 21.0073095703125,
      "step": 2600
    },
    {
      "epoch": 2.424811169604028,
      "grad_norm": 0.6587186455726624,
      "learning_rate": 9.905660377358492e-06,
      "loss": 20.84186279296875,
      "step": 2650
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.7700671553611755,
      "learning_rate": 9.119496855345912e-06,
      "loss": 20.9491015625,
      "step": 2700
    },
    {
      "epoch": 2.516365300984207,
      "grad_norm": 0.7723618745803833,
      "learning_rate": 8.333333333333334e-06,
      "loss": 20.86789306640625,
      "step": 2750
    },
    {
      "epoch": 2.562142366674296,
      "grad_norm": 0.7601158022880554,
      "learning_rate": 7.547169811320755e-06,
      "loss": 20.905009765625,
      "step": 2800
    },
    {
      "epoch": 2.6079194323643855,
      "grad_norm": 0.8978911638259888,
      "learning_rate": 6.761006289308176e-06,
      "loss": 20.7809033203125,
      "step": 2850
    },
    {
      "epoch": 2.6536964980544746,
      "grad_norm": 0.7408691048622131,
      "learning_rate": 5.974842767295598e-06,
      "loss": 20.84373779296875,
      "step": 2900
    },
    {
      "epoch": 2.699473563744564,
      "grad_norm": 0.7798228859901428,
      "learning_rate": 5.188679245283019e-06,
      "loss": 20.96936279296875,
      "step": 2950
    },
    {
      "epoch": 2.7452506294346533,
      "grad_norm": 0.7453646659851074,
      "learning_rate": 4.40251572327044e-06,
      "loss": 20.81538818359375,
      "step": 3000
    },
    {
      "epoch": 2.7452506294346533,
      "eval_loss": 3.9151711463928223,
      "eval_runtime": 11.9267,
      "eval_samples_per_second": 59.195,
      "eval_steps_per_second": 7.462,
      "step": 3000
    },
    {
      "epoch": 2.7910276951247424,
      "grad_norm": 0.7813104391098022,
      "learning_rate": 3.6163522012578618e-06,
      "loss": 20.85503662109375,
      "step": 3050
    },
    {
      "epoch": 2.836804760814832,
      "grad_norm": 0.7470929026603699,
      "learning_rate": 2.830188679245283e-06,
      "loss": 20.86689697265625,
      "step": 3100
    },
    {
      "epoch": 2.882581826504921,
      "grad_norm": 0.7827481031417847,
      "learning_rate": 2.0440251572327044e-06,
      "loss": 20.801142578125,
      "step": 3150
    },
    {
      "epoch": 2.9283588921950106,
      "grad_norm": 0.8394864201545715,
      "learning_rate": 1.257861635220126e-06,
      "loss": 20.861923828125,
      "step": 3200
    },
    {
      "epoch": 2.9741359578850997,
      "grad_norm": 0.7399877905845642,
      "learning_rate": 4.7169811320754717e-07,
      "loss": 20.7888671875,
      "step": 3250
    },
    {
      "epoch": 3.0,
      "step": 3279,
      "total_flos": 2.782532121447629e+16,
      "train_loss": 21.985386016929464,
      "train_runtime": 7952.9013,
      "train_samples_per_second": 26.364,
      "train_steps_per_second": 0.412
    }
  ],
  "logging_steps": 50,
  "max_steps": 3279,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2.782532121447629e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}