{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 822,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0364963503649635,
      "grad_norm": 6.935588591840994,
      "learning_rate": 5e-06,
      "loss": 1.0327,
      "step": 10
    },
    {
      "epoch": 0.072992700729927,
      "grad_norm": 2.670201985668935,
      "learning_rate": 5e-06,
      "loss": 0.9089,
      "step": 20
    },
    {
      "epoch": 0.10948905109489052,
      "grad_norm": 6.488843905015272,
      "learning_rate": 5e-06,
      "loss": 0.8647,
      "step": 30
    },
    {
      "epoch": 0.145985401459854,
      "grad_norm": 1.750141549981801,
      "learning_rate": 5e-06,
      "loss": 0.8455,
      "step": 40
    },
    {
      "epoch": 0.18248175182481752,
      "grad_norm": 1.7426579902019592,
      "learning_rate": 5e-06,
      "loss": 0.8261,
      "step": 50
    },
    {
      "epoch": 0.21897810218978103,
      "grad_norm": 1.5150538320225337,
      "learning_rate": 5e-06,
      "loss": 0.8215,
      "step": 60
    },
    {
      "epoch": 0.25547445255474455,
      "grad_norm": 1.34752537394588,
      "learning_rate": 5e-06,
      "loss": 0.806,
      "step": 70
    },
    {
      "epoch": 0.291970802919708,
      "grad_norm": 0.8501658634743531,
      "learning_rate": 5e-06,
      "loss": 0.7944,
      "step": 80
    },
    {
      "epoch": 0.3284671532846715,
      "grad_norm": 1.1940581870093743,
      "learning_rate": 5e-06,
      "loss": 0.7843,
      "step": 90
    },
    {
      "epoch": 0.36496350364963503,
      "grad_norm": 1.5996734385655718,
      "learning_rate": 5e-06,
      "loss": 0.7782,
      "step": 100
    },
    {
      "epoch": 0.40145985401459855,
      "grad_norm": 1.0860044183217865,
      "learning_rate": 5e-06,
      "loss": 0.7704,
      "step": 110
    },
    {
      "epoch": 0.43795620437956206,
      "grad_norm": 0.6827596601808271,
      "learning_rate": 5e-06,
      "loss": 0.7721,
      "step": 120
    },
    {
      "epoch": 0.4744525547445255,
      "grad_norm": 0.8446630134792682,
      "learning_rate": 5e-06,
      "loss": 0.7694,
      "step": 130
    },
    {
      "epoch": 0.5109489051094891,
      "grad_norm": 0.6527357476166321,
      "learning_rate": 5e-06,
      "loss": 0.7607,
      "step": 140
    },
    {
      "epoch": 0.5474452554744526,
      "grad_norm": 0.8473812345074847,
      "learning_rate": 5e-06,
      "loss": 0.7654,
      "step": 150
    },
    {
      "epoch": 0.583941605839416,
      "grad_norm": 0.680871697828666,
      "learning_rate": 5e-06,
      "loss": 0.7569,
      "step": 160
    },
    {
      "epoch": 0.6204379562043796,
      "grad_norm": 0.7106980775523508,
      "learning_rate": 5e-06,
      "loss": 0.7558,
      "step": 170
    },
    {
      "epoch": 0.656934306569343,
      "grad_norm": 0.6257154252846203,
      "learning_rate": 5e-06,
      "loss": 0.7555,
      "step": 180
    },
    {
      "epoch": 0.6934306569343066,
      "grad_norm": 0.8083982567358466,
      "learning_rate": 5e-06,
      "loss": 0.7514,
      "step": 190
    },
    {
      "epoch": 0.7299270072992701,
      "grad_norm": 0.8444839075154996,
      "learning_rate": 5e-06,
      "loss": 0.7498,
      "step": 200
    },
    {
      "epoch": 0.7664233576642335,
      "grad_norm": 0.7002950498691304,
      "learning_rate": 5e-06,
      "loss": 0.747,
      "step": 210
    },
    {
      "epoch": 0.8029197080291971,
      "grad_norm": 0.8747715499460583,
      "learning_rate": 5e-06,
      "loss": 0.7455,
      "step": 220
    },
    {
      "epoch": 0.8394160583941606,
      "grad_norm": 0.8840789126221499,
      "learning_rate": 5e-06,
      "loss": 0.7437,
      "step": 230
    },
    {
      "epoch": 0.8759124087591241,
      "grad_norm": 0.6784150124578601,
      "learning_rate": 5e-06,
      "loss": 0.7421,
      "step": 240
    },
    {
      "epoch": 0.9124087591240876,
      "grad_norm": 0.6074960130500583,
      "learning_rate": 5e-06,
      "loss": 0.7423,
      "step": 250
    },
    {
      "epoch": 0.948905109489051,
      "grad_norm": 0.6838717489885743,
      "learning_rate": 5e-06,
      "loss": 0.7427,
      "step": 260
    },
    {
      "epoch": 0.9854014598540146,
      "grad_norm": 0.6526592018350256,
      "learning_rate": 5e-06,
      "loss": 0.7416,
      "step": 270
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7413060665130615,
      "eval_runtime": 26.3056,
      "eval_samples_per_second": 279.864,
      "eval_steps_per_second": 1.102,
      "step": 274
    },
    {
      "epoch": 1.0218978102189782,
      "grad_norm": 0.8154022012318396,
      "learning_rate": 5e-06,
      "loss": 0.7114,
      "step": 280
    },
    {
      "epoch": 1.0583941605839415,
      "grad_norm": 0.8045050782757963,
      "learning_rate": 5e-06,
      "loss": 0.6861,
      "step": 290
    },
    {
      "epoch": 1.094890510948905,
      "grad_norm": 0.6684516327458488,
      "learning_rate": 5e-06,
      "loss": 0.6847,
      "step": 300
    },
    {
      "epoch": 1.1313868613138687,
      "grad_norm": 0.6133382471958105,
      "learning_rate": 5e-06,
      "loss": 0.6888,
      "step": 310
    },
    {
      "epoch": 1.167883211678832,
      "grad_norm": 0.6070090310932738,
      "learning_rate": 5e-06,
      "loss": 0.6864,
      "step": 320
    },
    {
      "epoch": 1.2043795620437956,
      "grad_norm": 0.6691556220694154,
      "learning_rate": 5e-06,
      "loss": 0.6858,
      "step": 330
    },
    {
      "epoch": 1.2408759124087592,
      "grad_norm": 0.7214448473176965,
      "learning_rate": 5e-06,
      "loss": 0.6888,
      "step": 340
    },
    {
      "epoch": 1.2773722627737225,
      "grad_norm": 0.6313497992938364,
      "learning_rate": 5e-06,
      "loss": 0.6895,
      "step": 350
    },
    {
      "epoch": 1.313868613138686,
      "grad_norm": 0.6011076115864562,
      "learning_rate": 5e-06,
      "loss": 0.6894,
      "step": 360
    },
    {
      "epoch": 1.3503649635036497,
      "grad_norm": 0.6417225834212155,
      "learning_rate": 5e-06,
      "loss": 0.6842,
      "step": 370
    },
    {
      "epoch": 1.3868613138686132,
      "grad_norm": 0.8658988992278465,
      "learning_rate": 5e-06,
      "loss": 0.6865,
      "step": 380
    },
    {
      "epoch": 1.4233576642335766,
      "grad_norm": 1.1272922956683207,
      "learning_rate": 5e-06,
      "loss": 0.6866,
      "step": 390
    },
    {
      "epoch": 1.4598540145985401,
      "grad_norm": 0.5914462231869502,
      "learning_rate": 5e-06,
      "loss": 0.6865,
      "step": 400
    },
    {
      "epoch": 1.4963503649635037,
      "grad_norm": 0.5668202854702565,
      "learning_rate": 5e-06,
      "loss": 0.6846,
      "step": 410
    },
    {
      "epoch": 1.5328467153284673,
      "grad_norm": 0.6635467334798604,
      "learning_rate": 5e-06,
      "loss": 0.6921,
      "step": 420
    },
    {
      "epoch": 1.5693430656934306,
      "grad_norm": 0.7643912156465277,
      "learning_rate": 5e-06,
      "loss": 0.6875,
      "step": 430
    },
    {
      "epoch": 1.6058394160583942,
      "grad_norm": 0.6238580970561005,
      "learning_rate": 5e-06,
      "loss": 0.6863,
      "step": 440
    },
    {
      "epoch": 1.6423357664233578,
      "grad_norm": 0.6822599541559131,
      "learning_rate": 5e-06,
      "loss": 0.6846,
      "step": 450
    },
    {
      "epoch": 1.6788321167883211,
      "grad_norm": 0.6619644198552357,
      "learning_rate": 5e-06,
      "loss": 0.6852,
      "step": 460
    },
    {
      "epoch": 1.7153284671532847,
      "grad_norm": 0.5553682134217856,
      "learning_rate": 5e-06,
      "loss": 0.6903,
      "step": 470
    },
    {
      "epoch": 1.7518248175182483,
      "grad_norm": 0.68779697897221,
      "learning_rate": 5e-06,
      "loss": 0.6857,
      "step": 480
    },
    {
      "epoch": 1.7883211678832116,
      "grad_norm": 0.6635201246353984,
      "learning_rate": 5e-06,
      "loss": 0.6875,
      "step": 490
    },
    {
      "epoch": 1.8248175182481752,
      "grad_norm": 0.6121605469304142,
      "learning_rate": 5e-06,
      "loss": 0.6868,
      "step": 500
    },
    {
      "epoch": 1.8613138686131387,
      "grad_norm": 0.619764705154293,
      "learning_rate": 5e-06,
      "loss": 0.6823,
      "step": 510
    },
    {
      "epoch": 1.897810218978102,
      "grad_norm": 0.5832433162588191,
      "learning_rate": 5e-06,
      "loss": 0.681,
      "step": 520
    },
    {
      "epoch": 1.9343065693430657,
      "grad_norm": 0.7973902031324107,
      "learning_rate": 5e-06,
      "loss": 0.6801,
      "step": 530
    },
    {
      "epoch": 1.9708029197080292,
      "grad_norm": 0.6524573193193096,
      "learning_rate": 5e-06,
      "loss": 0.6856,
      "step": 540
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7306408882141113,
      "eval_runtime": 29.0253,
      "eval_samples_per_second": 253.641,
      "eval_steps_per_second": 0.999,
      "step": 548
    },
    {
      "epoch": 2.0072992700729926,
      "grad_norm": 1.0397746521566786,
      "learning_rate": 5e-06,
      "loss": 0.6708,
      "step": 550
    },
    {
      "epoch": 2.0437956204379564,
      "grad_norm": 0.7448998852970502,
      "learning_rate": 5e-06,
      "loss": 0.6358,
      "step": 560
    },
    {
      "epoch": 2.0802919708029197,
      "grad_norm": 0.821170183708594,
      "learning_rate": 5e-06,
      "loss": 0.6306,
      "step": 570
    },
    {
      "epoch": 2.116788321167883,
      "grad_norm": 0.8192091109666185,
      "learning_rate": 5e-06,
      "loss": 0.6289,
      "step": 580
    },
    {
      "epoch": 2.153284671532847,
      "grad_norm": 0.6531060593153705,
      "learning_rate": 5e-06,
      "loss": 0.6303,
      "step": 590
    },
    {
      "epoch": 2.18978102189781,
      "grad_norm": 0.6207596231842625,
      "learning_rate": 5e-06,
      "loss": 0.6291,
      "step": 600
    },
    {
      "epoch": 2.2262773722627736,
      "grad_norm": 0.6299880389445941,
      "learning_rate": 5e-06,
      "loss": 0.6266,
      "step": 610
    },
    {
      "epoch": 2.2627737226277373,
      "grad_norm": 0.7577058406502929,
      "learning_rate": 5e-06,
      "loss": 0.6316,
      "step": 620
    },
    {
      "epoch": 2.2992700729927007,
      "grad_norm": 0.7972966242533703,
      "learning_rate": 5e-06,
      "loss": 0.6324,
      "step": 630
    },
    {
      "epoch": 2.335766423357664,
      "grad_norm": 0.6748643973886247,
      "learning_rate": 5e-06,
      "loss": 0.6276,
      "step": 640
    },
    {
      "epoch": 2.372262773722628,
      "grad_norm": 0.6136288389697012,
      "learning_rate": 5e-06,
      "loss": 0.6376,
      "step": 650
    },
    {
      "epoch": 2.408759124087591,
      "grad_norm": 0.6659808481432584,
      "learning_rate": 5e-06,
      "loss": 0.6287,
      "step": 660
    },
    {
      "epoch": 2.445255474452555,
      "grad_norm": 0.6729354487925674,
      "learning_rate": 5e-06,
      "loss": 0.6355,
      "step": 670
    },
    {
      "epoch": 2.4817518248175183,
      "grad_norm": 0.5975205704498788,
      "learning_rate": 5e-06,
      "loss": 0.6361,
      "step": 680
    },
    {
      "epoch": 2.5182481751824817,
      "grad_norm": 0.6572927598737819,
      "learning_rate": 5e-06,
      "loss": 0.6377,
      "step": 690
    },
    {
      "epoch": 2.554744525547445,
      "grad_norm": 0.5847912296345943,
      "learning_rate": 5e-06,
      "loss": 0.6348,
      "step": 700
    },
    {
      "epoch": 2.591240875912409,
      "grad_norm": 0.5992831189054557,
      "learning_rate": 5e-06,
      "loss": 0.6409,
      "step": 710
    },
    {
      "epoch": 2.627737226277372,
      "grad_norm": 0.6534494990778235,
      "learning_rate": 5e-06,
      "loss": 0.6295,
      "step": 720
    },
    {
      "epoch": 2.664233576642336,
      "grad_norm": 0.7980552337710113,
      "learning_rate": 5e-06,
      "loss": 0.6353,
      "step": 730
    },
    {
      "epoch": 2.7007299270072993,
      "grad_norm": 0.5906002533417896,
      "learning_rate": 5e-06,
      "loss": 0.6363,
      "step": 740
    },
    {
      "epoch": 2.7372262773722627,
      "grad_norm": 0.6317871178225047,
      "learning_rate": 5e-06,
      "loss": 0.6349,
      "step": 750
    },
    {
      "epoch": 2.7737226277372264,
      "grad_norm": 0.5701689110159228,
      "learning_rate": 5e-06,
      "loss": 0.6407,
      "step": 760
    },
    {
      "epoch": 2.81021897810219,
      "grad_norm": 0.7789184105520879,
      "learning_rate": 5e-06,
      "loss": 0.631,
      "step": 770
    },
    {
      "epoch": 2.846715328467153,
      "grad_norm": 0.7395144400572116,
      "learning_rate": 5e-06,
      "loss": 0.6298,
      "step": 780
    },
    {
      "epoch": 2.883211678832117,
      "grad_norm": 0.7833250133924173,
      "learning_rate": 5e-06,
      "loss": 0.6325,
      "step": 790
    },
    {
      "epoch": 2.9197080291970803,
      "grad_norm": 0.750312659575106,
      "learning_rate": 5e-06,
      "loss": 0.6425,
      "step": 800
    },
    {
      "epoch": 2.9562043795620436,
      "grad_norm": 0.6284417569809958,
      "learning_rate": 5e-06,
      "loss": 0.6323,
      "step": 810
    },
    {
      "epoch": 2.9927007299270074,
      "grad_norm": 0.5631024071115539,
      "learning_rate": 5e-06,
      "loss": 0.6383,
      "step": 820
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7370563745498657,
      "eval_runtime": 26.6576,
      "eval_samples_per_second": 276.169,
      "eval_steps_per_second": 1.088,
      "step": 822
    },
    {
      "epoch": 3.0,
      "step": 822,
      "total_flos": 1376880615751680.0,
      "train_loss": 0.7024027848765798,
      "train_runtime": 5290.9875,
      "train_samples_per_second": 79.308,
      "train_steps_per_second": 0.155
    }
  ],
  "logging_steps": 10,
  "max_steps": 822,
  "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": 1376880615751680.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}