{
  "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": 6.242694070966219,
      "learning_rate": 5e-06,
      "loss": 1.0388,
      "step": 10
    },
    {
      "epoch": 0.072992700729927,
      "grad_norm": 6.140265839791405,
      "learning_rate": 5e-06,
      "loss": 0.9087,
      "step": 20
    },
    {
      "epoch": 0.10948905109489052,
      "grad_norm": 2.0729558067197855,
      "learning_rate": 5e-06,
      "loss": 0.8718,
      "step": 30
    },
    {
      "epoch": 0.145985401459854,
      "grad_norm": 2.0048530569345027,
      "learning_rate": 5e-06,
      "loss": 0.8466,
      "step": 40
    },
    {
      "epoch": 0.18248175182481752,
      "grad_norm": 1.2903801782452229,
      "learning_rate": 5e-06,
      "loss": 0.8272,
      "step": 50
    },
    {
      "epoch": 0.21897810218978103,
      "grad_norm": 1.9320500006452086,
      "learning_rate": 5e-06,
      "loss": 0.8176,
      "step": 60
    },
    {
      "epoch": 0.25547445255474455,
      "grad_norm": 1.3130529470395709,
      "learning_rate": 5e-06,
      "loss": 0.8039,
      "step": 70
    },
    {
      "epoch": 0.291970802919708,
      "grad_norm": 1.1106618130287023,
      "learning_rate": 5e-06,
      "loss": 0.7935,
      "step": 80
    },
    {
      "epoch": 0.3284671532846715,
      "grad_norm": 0.9675468601929832,
      "learning_rate": 5e-06,
      "loss": 0.78,
      "step": 90
    },
    {
      "epoch": 0.36496350364963503,
      "grad_norm": 0.6972886370921517,
      "learning_rate": 5e-06,
      "loss": 0.7753,
      "step": 100
    },
    {
      "epoch": 0.40145985401459855,
      "grad_norm": 0.6872017378102949,
      "learning_rate": 5e-06,
      "loss": 0.7722,
      "step": 110
    },
    {
      "epoch": 0.43795620437956206,
      "grad_norm": 0.6670734728993309,
      "learning_rate": 5e-06,
      "loss": 0.7673,
      "step": 120
    },
    {
      "epoch": 0.4744525547445255,
      "grad_norm": 0.6673276704616633,
      "learning_rate": 5e-06,
      "loss": 0.7643,
      "step": 130
    },
    {
      "epoch": 0.5109489051094891,
      "grad_norm": 0.7068821433326216,
      "learning_rate": 5e-06,
      "loss": 0.7637,
      "step": 140
    },
    {
      "epoch": 0.5474452554744526,
      "grad_norm": 0.8071055608747135,
      "learning_rate": 5e-06,
      "loss": 0.7604,
      "step": 150
    },
    {
      "epoch": 0.583941605839416,
      "grad_norm": 0.6868966364491296,
      "learning_rate": 5e-06,
      "loss": 0.7522,
      "step": 160
    },
    {
      "epoch": 0.6204379562043796,
      "grad_norm": 0.9124104678824304,
      "learning_rate": 5e-06,
      "loss": 0.7547,
      "step": 170
    },
    {
      "epoch": 0.656934306569343,
      "grad_norm": 0.6792590544713151,
      "learning_rate": 5e-06,
      "loss": 0.7562,
      "step": 180
    },
    {
      "epoch": 0.6934306569343066,
      "grad_norm": 0.617023036902581,
      "learning_rate": 5e-06,
      "loss": 0.756,
      "step": 190
    },
    {
      "epoch": 0.7299270072992701,
      "grad_norm": 0.6319144841610138,
      "learning_rate": 5e-06,
      "loss": 0.751,
      "step": 200
    },
    {
      "epoch": 0.7664233576642335,
      "grad_norm": 0.6320543677861779,
      "learning_rate": 5e-06,
      "loss": 0.748,
      "step": 210
    },
    {
      "epoch": 0.8029197080291971,
      "grad_norm": 0.6210695741046822,
      "learning_rate": 5e-06,
      "loss": 0.7489,
      "step": 220
    },
    {
      "epoch": 0.8394160583941606,
      "grad_norm": 0.7516050498994822,
      "learning_rate": 5e-06,
      "loss": 0.7453,
      "step": 230
    },
    {
      "epoch": 0.8759124087591241,
      "grad_norm": 0.5948362298346876,
      "learning_rate": 5e-06,
      "loss": 0.7444,
      "step": 240
    },
    {
      "epoch": 0.9124087591240876,
      "grad_norm": 0.7329567143609019,
      "learning_rate": 5e-06,
      "loss": 0.7444,
      "step": 250
    },
    {
      "epoch": 0.948905109489051,
      "grad_norm": 0.5724746440812548,
      "learning_rate": 5e-06,
      "loss": 0.7372,
      "step": 260
    },
    {
      "epoch": 0.9854014598540146,
      "grad_norm": 0.5915513551849372,
      "learning_rate": 5e-06,
      "loss": 0.7414,
      "step": 270
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.737393319606781,
      "eval_runtime": 26.5119,
      "eval_samples_per_second": 278.403,
      "eval_steps_per_second": 1.094,
      "step": 274
    },
    {
      "epoch": 1.0218978102189782,
      "grad_norm": 1.1407224725998888,
      "learning_rate": 5e-06,
      "loss": 0.707,
      "step": 280
    },
    {
      "epoch": 1.0583941605839415,
      "grad_norm": 0.8137371947663772,
      "learning_rate": 5e-06,
      "loss": 0.6965,
      "step": 290
    },
    {
      "epoch": 1.094890510948905,
      "grad_norm": 0.7624656836561821,
      "learning_rate": 5e-06,
      "loss": 0.6929,
      "step": 300
    },
    {
      "epoch": 1.1313868613138687,
      "grad_norm": 0.6731270507290879,
      "learning_rate": 5e-06,
      "loss": 0.6907,
      "step": 310
    },
    {
      "epoch": 1.167883211678832,
      "grad_norm": 0.6477624974220811,
      "learning_rate": 5e-06,
      "loss": 0.6883,
      "step": 320
    },
    {
      "epoch": 1.2043795620437956,
      "grad_norm": 0.6806120359954909,
      "learning_rate": 5e-06,
      "loss": 0.6896,
      "step": 330
    },
    {
      "epoch": 1.2408759124087592,
      "grad_norm": 0.6534094167466027,
      "learning_rate": 5e-06,
      "loss": 0.6834,
      "step": 340
    },
    {
      "epoch": 1.2773722627737225,
      "grad_norm": 0.7053154028357804,
      "learning_rate": 5e-06,
      "loss": 0.6938,
      "step": 350
    },
    {
      "epoch": 1.313868613138686,
      "grad_norm": 0.5843975789107667,
      "learning_rate": 5e-06,
      "loss": 0.6905,
      "step": 360
    },
    {
      "epoch": 1.3503649635036497,
      "grad_norm": 0.6607671591107314,
      "learning_rate": 5e-06,
      "loss": 0.6887,
      "step": 370
    },
    {
      "epoch": 1.3868613138686132,
      "grad_norm": 0.6948472684652132,
      "learning_rate": 5e-06,
      "loss": 0.6863,
      "step": 380
    },
    {
      "epoch": 1.4233576642335766,
      "grad_norm": 0.6575033962434685,
      "learning_rate": 5e-06,
      "loss": 0.6886,
      "step": 390
    },
    {
      "epoch": 1.4598540145985401,
      "grad_norm": 0.5872317321934506,
      "learning_rate": 5e-06,
      "loss": 0.6868,
      "step": 400
    },
    {
      "epoch": 1.4963503649635037,
      "grad_norm": 0.5826458330462528,
      "learning_rate": 5e-06,
      "loss": 0.6895,
      "step": 410
    },
    {
      "epoch": 1.5328467153284673,
      "grad_norm": 0.6464190966828731,
      "learning_rate": 5e-06,
      "loss": 0.6836,
      "step": 420
    },
    {
      "epoch": 1.5693430656934306,
      "grad_norm": 0.7159795940614451,
      "learning_rate": 5e-06,
      "loss": 0.6834,
      "step": 430
    },
    {
      "epoch": 1.6058394160583942,
      "grad_norm": 0.74415061125301,
      "learning_rate": 5e-06,
      "loss": 0.6922,
      "step": 440
    },
    {
      "epoch": 1.6423357664233578,
      "grad_norm": 0.5766064803057743,
      "learning_rate": 5e-06,
      "loss": 0.6848,
      "step": 450
    },
    {
      "epoch": 1.6788321167883211,
      "grad_norm": 0.6927585803335585,
      "learning_rate": 5e-06,
      "loss": 0.6839,
      "step": 460
    },
    {
      "epoch": 1.7153284671532847,
      "grad_norm": 0.551695429707459,
      "learning_rate": 5e-06,
      "loss": 0.6859,
      "step": 470
    },
    {
      "epoch": 1.7518248175182483,
      "grad_norm": 0.6002867275771926,
      "learning_rate": 5e-06,
      "loss": 0.6872,
      "step": 480
    },
    {
      "epoch": 1.7883211678832116,
      "grad_norm": 0.5786825602682476,
      "learning_rate": 5e-06,
      "loss": 0.6882,
      "step": 490
    },
    {
      "epoch": 1.8248175182481752,
      "grad_norm": 0.6235940806026485,
      "learning_rate": 5e-06,
      "loss": 0.6869,
      "step": 500
    },
    {
      "epoch": 1.8613138686131387,
      "grad_norm": 0.6421720075253999,
      "learning_rate": 5e-06,
      "loss": 0.6869,
      "step": 510
    },
    {
      "epoch": 1.897810218978102,
      "grad_norm": 0.6399996565023783,
      "learning_rate": 5e-06,
      "loss": 0.683,
      "step": 520
    },
    {
      "epoch": 1.9343065693430657,
      "grad_norm": 0.6345252795641959,
      "learning_rate": 5e-06,
      "loss": 0.683,
      "step": 530
    },
    {
      "epoch": 1.9708029197080292,
      "grad_norm": 0.6174798987748756,
      "learning_rate": 5e-06,
      "loss": 0.6788,
      "step": 540
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.724261999130249,
      "eval_runtime": 29.7165,
      "eval_samples_per_second": 248.38,
      "eval_steps_per_second": 0.976,
      "step": 548
    },
    {
      "epoch": 2.0072992700729926,
      "grad_norm": 0.9503885013468557,
      "learning_rate": 5e-06,
      "loss": 0.6733,
      "step": 550
    },
    {
      "epoch": 2.0437956204379564,
      "grad_norm": 0.7079333930728279,
      "learning_rate": 5e-06,
      "loss": 0.6294,
      "step": 560
    },
    {
      "epoch": 2.0802919708029197,
      "grad_norm": 0.8055321987953024,
      "learning_rate": 5e-06,
      "loss": 0.6282,
      "step": 570
    },
    {
      "epoch": 2.116788321167883,
      "grad_norm": 0.7013752206943129,
      "learning_rate": 5e-06,
      "loss": 0.6346,
      "step": 580
    },
    {
      "epoch": 2.153284671532847,
      "grad_norm": 0.795586932183297,
      "learning_rate": 5e-06,
      "loss": 0.6309,
      "step": 590
    },
    {
      "epoch": 2.18978102189781,
      "grad_norm": 0.6072532710663935,
      "learning_rate": 5e-06,
      "loss": 0.6288,
      "step": 600
    },
    {
      "epoch": 2.2262773722627736,
      "grad_norm": 0.7308694689768187,
      "learning_rate": 5e-06,
      "loss": 0.6335,
      "step": 610
    },
    {
      "epoch": 2.2627737226277373,
      "grad_norm": 0.8888891048172273,
      "learning_rate": 5e-06,
      "loss": 0.6347,
      "step": 620
    },
    {
      "epoch": 2.2992700729927007,
      "grad_norm": 0.5694467380633018,
      "learning_rate": 5e-06,
      "loss": 0.6358,
      "step": 630
    },
    {
      "epoch": 2.335766423357664,
      "grad_norm": 0.6996612017894321,
      "learning_rate": 5e-06,
      "loss": 0.6331,
      "step": 640
    },
    {
      "epoch": 2.372262773722628,
      "grad_norm": 0.6798927912381371,
      "learning_rate": 5e-06,
      "loss": 0.6355,
      "step": 650
    },
    {
      "epoch": 2.408759124087591,
      "grad_norm": 0.7832060679040136,
      "learning_rate": 5e-06,
      "loss": 0.6306,
      "step": 660
    },
    {
      "epoch": 2.445255474452555,
      "grad_norm": 0.8518259483264891,
      "learning_rate": 5e-06,
      "loss": 0.6329,
      "step": 670
    },
    {
      "epoch": 2.4817518248175183,
      "grad_norm": 0.8847565076314947,
      "learning_rate": 5e-06,
      "loss": 0.6355,
      "step": 680
    },
    {
      "epoch": 2.5182481751824817,
      "grad_norm": 0.8790006051896251,
      "learning_rate": 5e-06,
      "loss": 0.6341,
      "step": 690
    },
    {
      "epoch": 2.554744525547445,
      "grad_norm": 0.7205003318722193,
      "learning_rate": 5e-06,
      "loss": 0.6353,
      "step": 700
    },
    {
      "epoch": 2.591240875912409,
      "grad_norm": 0.6695065635591323,
      "learning_rate": 5e-06,
      "loss": 0.6336,
      "step": 710
    },
    {
      "epoch": 2.627737226277372,
      "grad_norm": 0.701791622519896,
      "learning_rate": 5e-06,
      "loss": 0.6392,
      "step": 720
    },
    {
      "epoch": 2.664233576642336,
      "grad_norm": 0.6510579026114841,
      "learning_rate": 5e-06,
      "loss": 0.6325,
      "step": 730
    },
    {
      "epoch": 2.7007299270072993,
      "grad_norm": 0.6391591776977092,
      "learning_rate": 5e-06,
      "loss": 0.6379,
      "step": 740
    },
    {
      "epoch": 2.7372262773722627,
      "grad_norm": 0.6379410196843575,
      "learning_rate": 5e-06,
      "loss": 0.6428,
      "step": 750
    },
    {
      "epoch": 2.7737226277372264,
      "grad_norm": 0.5663117162856989,
      "learning_rate": 5e-06,
      "loss": 0.6365,
      "step": 760
    },
    {
      "epoch": 2.81021897810219,
      "grad_norm": 0.5976599060822565,
      "learning_rate": 5e-06,
      "loss": 0.6337,
      "step": 770
    },
    {
      "epoch": 2.846715328467153,
      "grad_norm": 0.5764241288666703,
      "learning_rate": 5e-06,
      "loss": 0.6378,
      "step": 780
    },
    {
      "epoch": 2.883211678832117,
      "grad_norm": 0.5923449235752416,
      "learning_rate": 5e-06,
      "loss": 0.6414,
      "step": 790
    },
    {
      "epoch": 2.9197080291970803,
      "grad_norm": 0.7012989231730122,
      "learning_rate": 5e-06,
      "loss": 0.6389,
      "step": 800
    },
    {
      "epoch": 2.9562043795620436,
      "grad_norm": 0.669400873506838,
      "learning_rate": 5e-06,
      "loss": 0.6429,
      "step": 810
    },
    {
      "epoch": 2.9927007299270074,
      "grad_norm": 0.6342356482957713,
      "learning_rate": 5e-06,
      "loss": 0.6398,
      "step": 820
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7273038625717163,
      "eval_runtime": 26.4167,
      "eval_samples_per_second": 279.406,
      "eval_steps_per_second": 1.098,
      "step": 822
    },
    {
      "epoch": 3.0,
      "step": 822,
      "total_flos": 1376880615751680.0,
      "train_loss": 0.7032751636783572,
      "train_runtime": 5303.7541,
      "train_samples_per_second": 79.315,
      "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
}