{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1092,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.027472527472527472,
      "grad_norm": 12.728795051574707,
      "learning_rate": 5e-06,
      "loss": 0.9765,
      "step": 10
    },
    {
      "epoch": 0.054945054945054944,
      "grad_norm": 2.706224203109741,
      "learning_rate": 5e-06,
      "loss": 0.839,
      "step": 20
    },
    {
      "epoch": 0.08241758241758242,
      "grad_norm": 1.059246301651001,
      "learning_rate": 5e-06,
      "loss": 0.7854,
      "step": 30
    },
    {
      "epoch": 0.10989010989010989,
      "grad_norm": 0.8609521389007568,
      "learning_rate": 5e-06,
      "loss": 0.7552,
      "step": 40
    },
    {
      "epoch": 0.13736263736263737,
      "grad_norm": 0.8304686546325684,
      "learning_rate": 5e-06,
      "loss": 0.7404,
      "step": 50
    },
    {
      "epoch": 0.16483516483516483,
      "grad_norm": 0.95517498254776,
      "learning_rate": 5e-06,
      "loss": 0.717,
      "step": 60
    },
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 0.8028889298439026,
      "learning_rate": 5e-06,
      "loss": 0.7069,
      "step": 70
    },
    {
      "epoch": 0.21978021978021978,
      "grad_norm": 1.287640929222107,
      "learning_rate": 5e-06,
      "loss": 0.6991,
      "step": 80
    },
    {
      "epoch": 0.24725274725274726,
      "grad_norm": 2.1982975006103516,
      "learning_rate": 5e-06,
      "loss": 0.6911,
      "step": 90
    },
    {
      "epoch": 0.27472527472527475,
      "grad_norm": 0.6827638149261475,
      "learning_rate": 5e-06,
      "loss": 0.6901,
      "step": 100
    },
    {
      "epoch": 0.3021978021978022,
      "grad_norm": 0.9066330194473267,
      "learning_rate": 5e-06,
      "loss": 0.6826,
      "step": 110
    },
    {
      "epoch": 0.32967032967032966,
      "grad_norm": 0.6553434133529663,
      "learning_rate": 5e-06,
      "loss": 0.6863,
      "step": 120
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.9135424494743347,
      "learning_rate": 5e-06,
      "loss": 0.6745,
      "step": 130
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 0.7348975539207458,
      "learning_rate": 5e-06,
      "loss": 0.6759,
      "step": 140
    },
    {
      "epoch": 0.41208791208791207,
      "grad_norm": 0.9071689248085022,
      "learning_rate": 5e-06,
      "loss": 0.6673,
      "step": 150
    },
    {
      "epoch": 0.43956043956043955,
      "grad_norm": 0.7530332207679749,
      "learning_rate": 5e-06,
      "loss": 0.6651,
      "step": 160
    },
    {
      "epoch": 0.46703296703296704,
      "grad_norm": 0.7991921305656433,
      "learning_rate": 5e-06,
      "loss": 0.6716,
      "step": 170
    },
    {
      "epoch": 0.4945054945054945,
      "grad_norm": 0.8702341318130493,
      "learning_rate": 5e-06,
      "loss": 0.6589,
      "step": 180
    },
    {
      "epoch": 0.521978021978022,
      "grad_norm": 0.5698891878128052,
      "learning_rate": 5e-06,
      "loss": 0.6618,
      "step": 190
    },
    {
      "epoch": 0.5494505494505495,
      "grad_norm": 0.5574747323989868,
      "learning_rate": 5e-06,
      "loss": 0.6629,
      "step": 200
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 0.6472902894020081,
      "learning_rate": 5e-06,
      "loss": 0.6582,
      "step": 210
    },
    {
      "epoch": 0.6043956043956044,
      "grad_norm": 0.5712385177612305,
      "learning_rate": 5e-06,
      "loss": 0.6551,
      "step": 220
    },
    {
      "epoch": 0.6318681318681318,
      "grad_norm": 0.492366224527359,
      "learning_rate": 5e-06,
      "loss": 0.6485,
      "step": 230
    },
    {
      "epoch": 0.6593406593406593,
      "grad_norm": 0.7048478126525879,
      "learning_rate": 5e-06,
      "loss": 0.6484,
      "step": 240
    },
    {
      "epoch": 0.6868131868131868,
      "grad_norm": 0.537118136882782,
      "learning_rate": 5e-06,
      "loss": 0.6488,
      "step": 250
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.49864494800567627,
      "learning_rate": 5e-06,
      "loss": 0.6463,
      "step": 260
    },
    {
      "epoch": 0.7417582417582418,
      "grad_norm": 0.7438676953315735,
      "learning_rate": 5e-06,
      "loss": 0.6498,
      "step": 270
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.5052060484886169,
      "learning_rate": 5e-06,
      "loss": 0.6464,
      "step": 280
    },
    {
      "epoch": 0.7967032967032966,
      "grad_norm": 0.548953115940094,
      "learning_rate": 5e-06,
      "loss": 0.6441,
      "step": 290
    },
    {
      "epoch": 0.8241758241758241,
      "grad_norm": 0.6289977431297302,
      "learning_rate": 5e-06,
      "loss": 0.6415,
      "step": 300
    },
    {
      "epoch": 0.8516483516483516,
      "grad_norm": 0.5599405765533447,
      "learning_rate": 5e-06,
      "loss": 0.6419,
      "step": 310
    },
    {
      "epoch": 0.8791208791208791,
      "grad_norm": 0.6185189485549927,
      "learning_rate": 5e-06,
      "loss": 0.6371,
      "step": 320
    },
    {
      "epoch": 0.9065934065934066,
      "grad_norm": 0.5025250911712646,
      "learning_rate": 5e-06,
      "loss": 0.6388,
      "step": 330
    },
    {
      "epoch": 0.9340659340659341,
      "grad_norm": 0.596026599407196,
      "learning_rate": 5e-06,
      "loss": 0.6371,
      "step": 340
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 0.5150969624519348,
      "learning_rate": 5e-06,
      "loss": 0.6323,
      "step": 350
    },
    {
      "epoch": 0.989010989010989,
      "grad_norm": 0.5902632474899292,
      "learning_rate": 5e-06,
      "loss": 0.6304,
      "step": 360
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6319476366043091,
      "eval_runtime": 33.2599,
      "eval_samples_per_second": 294.589,
      "eval_steps_per_second": 1.173,
      "step": 364
    },
    {
      "epoch": 1.0164835164835164,
      "grad_norm": 0.684147298336029,
      "learning_rate": 5e-06,
      "loss": 0.6119,
      "step": 370
    },
    {
      "epoch": 1.043956043956044,
      "grad_norm": 0.6040191054344177,
      "learning_rate": 5e-06,
      "loss": 0.5939,
      "step": 380
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 0.6056487560272217,
      "learning_rate": 5e-06,
      "loss": 0.5955,
      "step": 390
    },
    {
      "epoch": 1.098901098901099,
      "grad_norm": 0.6380589008331299,
      "learning_rate": 5e-06,
      "loss": 0.5917,
      "step": 400
    },
    {
      "epoch": 1.1263736263736264,
      "grad_norm": 0.734009861946106,
      "learning_rate": 5e-06,
      "loss": 0.5912,
      "step": 410
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 0.5856278538703918,
      "learning_rate": 5e-06,
      "loss": 0.5887,
      "step": 420
    },
    {
      "epoch": 1.1813186813186813,
      "grad_norm": 0.5818033218383789,
      "learning_rate": 5e-06,
      "loss": 0.5959,
      "step": 430
    },
    {
      "epoch": 1.2087912087912087,
      "grad_norm": 0.5466902852058411,
      "learning_rate": 5e-06,
      "loss": 0.5927,
      "step": 440
    },
    {
      "epoch": 1.2362637362637363,
      "grad_norm": 0.5269283056259155,
      "learning_rate": 5e-06,
      "loss": 0.5928,
      "step": 450
    },
    {
      "epoch": 1.2637362637362637,
      "grad_norm": 0.5852891802787781,
      "learning_rate": 5e-06,
      "loss": 0.5888,
      "step": 460
    },
    {
      "epoch": 1.2912087912087913,
      "grad_norm": 0.5838508605957031,
      "learning_rate": 5e-06,
      "loss": 0.5868,
      "step": 470
    },
    {
      "epoch": 1.3186813186813187,
      "grad_norm": 0.5828148722648621,
      "learning_rate": 5e-06,
      "loss": 0.5881,
      "step": 480
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 0.5466414093971252,
      "learning_rate": 5e-06,
      "loss": 0.5844,
      "step": 490
    },
    {
      "epoch": 1.3736263736263736,
      "grad_norm": 0.6251242160797119,
      "learning_rate": 5e-06,
      "loss": 0.5867,
      "step": 500
    },
    {
      "epoch": 1.401098901098901,
      "grad_norm": 0.6761026382446289,
      "learning_rate": 5e-06,
      "loss": 0.5878,
      "step": 510
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.8003695011138916,
      "learning_rate": 5e-06,
      "loss": 0.5858,
      "step": 520
    },
    {
      "epoch": 1.456043956043956,
      "grad_norm": 0.6401771903038025,
      "learning_rate": 5e-06,
      "loss": 0.5859,
      "step": 530
    },
    {
      "epoch": 1.4835164835164836,
      "grad_norm": 0.722094714641571,
      "learning_rate": 5e-06,
      "loss": 0.5885,
      "step": 540
    },
    {
      "epoch": 1.510989010989011,
      "grad_norm": 0.6249004006385803,
      "learning_rate": 5e-06,
      "loss": 0.5846,
      "step": 550
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.5867748856544495,
      "learning_rate": 5e-06,
      "loss": 0.5841,
      "step": 560
    },
    {
      "epoch": 1.565934065934066,
      "grad_norm": 0.6063013672828674,
      "learning_rate": 5e-06,
      "loss": 0.5854,
      "step": 570
    },
    {
      "epoch": 1.5934065934065935,
      "grad_norm": 0.538633406162262,
      "learning_rate": 5e-06,
      "loss": 0.5812,
      "step": 580
    },
    {
      "epoch": 1.620879120879121,
      "grad_norm": 0.6249384880065918,
      "learning_rate": 5e-06,
      "loss": 0.5802,
      "step": 590
    },
    {
      "epoch": 1.6483516483516483,
      "grad_norm": 0.5973653197288513,
      "learning_rate": 5e-06,
      "loss": 0.5848,
      "step": 600
    },
    {
      "epoch": 1.6758241758241759,
      "grad_norm": 0.5950356125831604,
      "learning_rate": 5e-06,
      "loss": 0.5807,
      "step": 610
    },
    {
      "epoch": 1.7032967032967035,
      "grad_norm": 0.5238327980041504,
      "learning_rate": 5e-06,
      "loss": 0.5753,
      "step": 620
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 0.5083211660385132,
      "learning_rate": 5e-06,
      "loss": 0.5801,
      "step": 630
    },
    {
      "epoch": 1.7582417582417582,
      "grad_norm": 0.8316366076469421,
      "learning_rate": 5e-06,
      "loss": 0.5745,
      "step": 640
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 0.5875478386878967,
      "learning_rate": 5e-06,
      "loss": 0.58,
      "step": 650
    },
    {
      "epoch": 1.8131868131868132,
      "grad_norm": 0.5625612735748291,
      "learning_rate": 5e-06,
      "loss": 0.5721,
      "step": 660
    },
    {
      "epoch": 1.8406593406593408,
      "grad_norm": 0.5859466791152954,
      "learning_rate": 5e-06,
      "loss": 0.5784,
      "step": 670
    },
    {
      "epoch": 1.8681318681318682,
      "grad_norm": 0.5293060541152954,
      "learning_rate": 5e-06,
      "loss": 0.575,
      "step": 680
    },
    {
      "epoch": 1.8956043956043955,
      "grad_norm": 0.516545832157135,
      "learning_rate": 5e-06,
      "loss": 0.5771,
      "step": 690
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.5076315999031067,
      "learning_rate": 5e-06,
      "loss": 0.5759,
      "step": 700
    },
    {
      "epoch": 1.9505494505494505,
      "grad_norm": 0.46912071108818054,
      "learning_rate": 5e-06,
      "loss": 0.5738,
      "step": 710
    },
    {
      "epoch": 1.978021978021978,
      "grad_norm": 0.5850493907928467,
      "learning_rate": 5e-06,
      "loss": 0.5776,
      "step": 720
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6049710512161255,
      "eval_runtime": 33.2692,
      "eval_samples_per_second": 294.506,
      "eval_steps_per_second": 1.172,
      "step": 728
    },
    {
      "epoch": 2.0054945054945055,
      "grad_norm": 0.8460399508476257,
      "learning_rate": 5e-06,
      "loss": 0.5647,
      "step": 730
    },
    {
      "epoch": 2.032967032967033,
      "grad_norm": 0.5859711170196533,
      "learning_rate": 5e-06,
      "loss": 0.5343,
      "step": 740
    },
    {
      "epoch": 2.0604395604395602,
      "grad_norm": 0.5478792190551758,
      "learning_rate": 5e-06,
      "loss": 0.5346,
      "step": 750
    },
    {
      "epoch": 2.087912087912088,
      "grad_norm": 0.5353866815567017,
      "learning_rate": 5e-06,
      "loss": 0.5386,
      "step": 760
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 0.589070737361908,
      "learning_rate": 5e-06,
      "loss": 0.5364,
      "step": 770
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.7154620289802551,
      "learning_rate": 5e-06,
      "loss": 0.5357,
      "step": 780
    },
    {
      "epoch": 2.17032967032967,
      "grad_norm": 0.5796002149581909,
      "learning_rate": 5e-06,
      "loss": 0.532,
      "step": 790
    },
    {
      "epoch": 2.197802197802198,
      "grad_norm": 0.5556730031967163,
      "learning_rate": 5e-06,
      "loss": 0.5389,
      "step": 800
    },
    {
      "epoch": 2.2252747252747254,
      "grad_norm": 0.5793793201446533,
      "learning_rate": 5e-06,
      "loss": 0.5323,
      "step": 810
    },
    {
      "epoch": 2.2527472527472527,
      "grad_norm": 0.6171457767486572,
      "learning_rate": 5e-06,
      "loss": 0.5325,
      "step": 820
    },
    {
      "epoch": 2.28021978021978,
      "grad_norm": 0.563108503818512,
      "learning_rate": 5e-06,
      "loss": 0.5305,
      "step": 830
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.6023669242858887,
      "learning_rate": 5e-06,
      "loss": 0.5336,
      "step": 840
    },
    {
      "epoch": 2.3351648351648353,
      "grad_norm": 0.6838552355766296,
      "learning_rate": 5e-06,
      "loss": 0.5299,
      "step": 850
    },
    {
      "epoch": 2.3626373626373627,
      "grad_norm": 0.5465499758720398,
      "learning_rate": 5e-06,
      "loss": 0.5352,
      "step": 860
    },
    {
      "epoch": 2.39010989010989,
      "grad_norm": 0.6073447465896606,
      "learning_rate": 5e-06,
      "loss": 0.534,
      "step": 870
    },
    {
      "epoch": 2.4175824175824174,
      "grad_norm": 0.5182047486305237,
      "learning_rate": 5e-06,
      "loss": 0.5317,
      "step": 880
    },
    {
      "epoch": 2.4450549450549453,
      "grad_norm": 0.543487548828125,
      "learning_rate": 5e-06,
      "loss": 0.5381,
      "step": 890
    },
    {
      "epoch": 2.4725274725274726,
      "grad_norm": 0.5238776803016663,
      "learning_rate": 5e-06,
      "loss": 0.5381,
      "step": 900
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5654677152633667,
      "learning_rate": 5e-06,
      "loss": 0.5316,
      "step": 910
    },
    {
      "epoch": 2.5274725274725274,
      "grad_norm": 0.6107837557792664,
      "learning_rate": 5e-06,
      "loss": 0.5368,
      "step": 920
    },
    {
      "epoch": 2.5549450549450547,
      "grad_norm": 0.5588423609733582,
      "learning_rate": 5e-06,
      "loss": 0.5402,
      "step": 930
    },
    {
      "epoch": 2.5824175824175826,
      "grad_norm": 0.5652673840522766,
      "learning_rate": 5e-06,
      "loss": 0.5382,
      "step": 940
    },
    {
      "epoch": 2.60989010989011,
      "grad_norm": 0.5125435590744019,
      "learning_rate": 5e-06,
      "loss": 0.539,
      "step": 950
    },
    {
      "epoch": 2.6373626373626373,
      "grad_norm": 0.5421128273010254,
      "learning_rate": 5e-06,
      "loss": 0.5335,
      "step": 960
    },
    {
      "epoch": 2.6648351648351647,
      "grad_norm": 0.5882076621055603,
      "learning_rate": 5e-06,
      "loss": 0.5341,
      "step": 970
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.6641466021537781,
      "learning_rate": 5e-06,
      "loss": 0.5301,
      "step": 980
    },
    {
      "epoch": 2.71978021978022,
      "grad_norm": 0.5694056153297424,
      "learning_rate": 5e-06,
      "loss": 0.537,
      "step": 990
    },
    {
      "epoch": 2.7472527472527473,
      "grad_norm": 0.48833751678466797,
      "learning_rate": 5e-06,
      "loss": 0.5333,
      "step": 1000
    },
    {
      "epoch": 2.7747252747252746,
      "grad_norm": 0.5480605959892273,
      "learning_rate": 5e-06,
      "loss": 0.5352,
      "step": 1010
    },
    {
      "epoch": 2.802197802197802,
      "grad_norm": 0.6873406767845154,
      "learning_rate": 5e-06,
      "loss": 0.5389,
      "step": 1020
    },
    {
      "epoch": 2.82967032967033,
      "grad_norm": 0.5708329081535339,
      "learning_rate": 5e-06,
      "loss": 0.5346,
      "step": 1030
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.5935104489326477,
      "learning_rate": 5e-06,
      "loss": 0.5331,
      "step": 1040
    },
    {
      "epoch": 2.8846153846153846,
      "grad_norm": 0.5802971124649048,
      "learning_rate": 5e-06,
      "loss": 0.53,
      "step": 1050
    },
    {
      "epoch": 2.912087912087912,
      "grad_norm": 0.5741780400276184,
      "learning_rate": 5e-06,
      "loss": 0.5291,
      "step": 1060
    },
    {
      "epoch": 2.9395604395604398,
      "grad_norm": 0.578640878200531,
      "learning_rate": 5e-06,
      "loss": 0.5326,
      "step": 1070
    },
    {
      "epoch": 2.967032967032967,
      "grad_norm": 0.5405354499816895,
      "learning_rate": 5e-06,
      "loss": 0.5314,
      "step": 1080
    },
    {
      "epoch": 2.9945054945054945,
      "grad_norm": 0.5508515238761902,
      "learning_rate": 5e-06,
      "loss": 0.5347,
      "step": 1090
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5989904403686523,
      "eval_runtime": 33.1874,
      "eval_samples_per_second": 295.232,
      "eval_steps_per_second": 1.175,
      "step": 1092
    },
    {
      "epoch": 3.0,
      "step": 1092,
      "total_flos": 5.156086030431106e+19,
      "train_loss": 0.6005454126731816,
      "train_runtime": 7773.4306,
      "train_samples_per_second": 71.841,
      "train_steps_per_second": 0.14
    }
  ],
  "logging_steps": 10,
  "max_steps": 1092,
  "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": 5.156086030431106e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}