{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.025,
      "grad_norm": 14.020469665527344,
      "learning_rate": 5e-06,
      "loss": 0.8991,
      "step": 10
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.1231062412261963,
      "learning_rate": 5e-06,
      "loss": 0.7786,
      "step": 20
    },
    {
      "epoch": 0.075,
      "grad_norm": 0.8827582597732544,
      "learning_rate": 5e-06,
      "loss": 0.7194,
      "step": 30
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8402972221374512,
      "learning_rate": 5e-06,
      "loss": 0.6925,
      "step": 40
    },
    {
      "epoch": 0.125,
      "grad_norm": 0.8711609840393066,
      "learning_rate": 5e-06,
      "loss": 0.6664,
      "step": 50
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.020943284034729,
      "learning_rate": 5e-06,
      "loss": 0.6468,
      "step": 60
    },
    {
      "epoch": 0.175,
      "grad_norm": 0.6615628600120544,
      "learning_rate": 5e-06,
      "loss": 0.6438,
      "step": 70
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6699038743972778,
      "learning_rate": 5e-06,
      "loss": 0.6357,
      "step": 80
    },
    {
      "epoch": 0.225,
      "grad_norm": 0.6830587387084961,
      "learning_rate": 5e-06,
      "loss": 0.6206,
      "step": 90
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5887433886528015,
      "learning_rate": 5e-06,
      "loss": 0.613,
      "step": 100
    },
    {
      "epoch": 0.275,
      "grad_norm": 0.491720587015152,
      "learning_rate": 5e-06,
      "loss": 0.6036,
      "step": 110
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6611770391464233,
      "learning_rate": 5e-06,
      "loss": 0.6014,
      "step": 120
    },
    {
      "epoch": 0.325,
      "grad_norm": 0.6511964797973633,
      "learning_rate": 5e-06,
      "loss": 0.6006,
      "step": 130
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5848156213760376,
      "learning_rate": 5e-06,
      "loss": 0.6091,
      "step": 140
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.6775133609771729,
      "learning_rate": 5e-06,
      "loss": 0.6004,
      "step": 150
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5474533438682556,
      "learning_rate": 5e-06,
      "loss": 0.6068,
      "step": 160
    },
    {
      "epoch": 0.425,
      "grad_norm": 0.4722808301448822,
      "learning_rate": 5e-06,
      "loss": 0.5916,
      "step": 170
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5218152403831482,
      "learning_rate": 5e-06,
      "loss": 0.5972,
      "step": 180
    },
    {
      "epoch": 0.475,
      "grad_norm": 0.703032374382019,
      "learning_rate": 5e-06,
      "loss": 0.5896,
      "step": 190
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6441683769226074,
      "learning_rate": 5e-06,
      "loss": 0.592,
      "step": 200
    },
    {
      "epoch": 0.525,
      "grad_norm": 0.5189586281776428,
      "learning_rate": 5e-06,
      "loss": 0.5887,
      "step": 210
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5127268433570862,
      "learning_rate": 5e-06,
      "loss": 0.5825,
      "step": 220
    },
    {
      "epoch": 0.575,
      "grad_norm": 0.5413115620613098,
      "learning_rate": 5e-06,
      "loss": 0.5804,
      "step": 230
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.48631197214126587,
      "learning_rate": 5e-06,
      "loss": 0.5836,
      "step": 240
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.554779052734375,
      "learning_rate": 5e-06,
      "loss": 0.5831,
      "step": 250
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5633807182312012,
      "learning_rate": 5e-06,
      "loss": 0.5815,
      "step": 260
    },
    {
      "epoch": 0.675,
      "grad_norm": 0.5503019690513611,
      "learning_rate": 5e-06,
      "loss": 0.5771,
      "step": 270
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4752262830734253,
      "learning_rate": 5e-06,
      "loss": 0.5778,
      "step": 280
    },
    {
      "epoch": 0.725,
      "grad_norm": 0.6272807121276855,
      "learning_rate": 5e-06,
      "loss": 0.5763,
      "step": 290
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6500713229179382,
      "learning_rate": 5e-06,
      "loss": 0.573,
      "step": 300
    },
    {
      "epoch": 0.775,
      "grad_norm": 0.5538522005081177,
      "learning_rate": 5e-06,
      "loss": 0.575,
      "step": 310
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.674211323261261,
      "learning_rate": 5e-06,
      "loss": 0.5775,
      "step": 320
    },
    {
      "epoch": 0.825,
      "grad_norm": 0.5151064991950989,
      "learning_rate": 5e-06,
      "loss": 0.5695,
      "step": 330
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.45805805921554565,
      "learning_rate": 5e-06,
      "loss": 0.5658,
      "step": 340
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.907016932964325,
      "learning_rate": 5e-06,
      "loss": 0.5682,
      "step": 350
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.4508662521839142,
      "learning_rate": 5e-06,
      "loss": 0.5691,
      "step": 360
    },
    {
      "epoch": 0.925,
      "grad_norm": 0.5657577514648438,
      "learning_rate": 5e-06,
      "loss": 0.5661,
      "step": 370
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.4820933938026428,
      "learning_rate": 5e-06,
      "loss": 0.5677,
      "step": 380
    },
    {
      "epoch": 0.975,
      "grad_norm": 0.4836273491382599,
      "learning_rate": 5e-06,
      "loss": 0.5654,
      "step": 390
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4653589427471161,
      "learning_rate": 5e-06,
      "loss": 0.5566,
      "step": 400
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5628939270973206,
      "eval_runtime": 36.7288,
      "eval_samples_per_second": 292.985,
      "eval_steps_per_second": 1.171,
      "step": 400
    },
    {
      "epoch": 1.025,
      "grad_norm": 0.5971388816833496,
      "learning_rate": 5e-06,
      "loss": 0.526,
      "step": 410
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.6474968791007996,
      "learning_rate": 5e-06,
      "loss": 0.5298,
      "step": 420
    },
    {
      "epoch": 1.075,
      "grad_norm": 0.7014959454536438,
      "learning_rate": 5e-06,
      "loss": 0.5299,
      "step": 430
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.7876601219177246,
      "learning_rate": 5e-06,
      "loss": 0.5273,
      "step": 440
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.5993332862854004,
      "learning_rate": 5e-06,
      "loss": 0.5274,
      "step": 450
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.476075142621994,
      "learning_rate": 5e-06,
      "loss": 0.5241,
      "step": 460
    },
    {
      "epoch": 1.175,
      "grad_norm": 0.509283721446991,
      "learning_rate": 5e-06,
      "loss": 0.5204,
      "step": 470
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5607204437255859,
      "learning_rate": 5e-06,
      "loss": 0.524,
      "step": 480
    },
    {
      "epoch": 1.225,
      "grad_norm": 0.6008129715919495,
      "learning_rate": 5e-06,
      "loss": 0.5247,
      "step": 490
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.5496598482131958,
      "learning_rate": 5e-06,
      "loss": 0.5224,
      "step": 500
    },
    {
      "epoch": 1.275,
      "grad_norm": 0.46281903982162476,
      "learning_rate": 5e-06,
      "loss": 0.5223,
      "step": 510
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.5407223105430603,
      "learning_rate": 5e-06,
      "loss": 0.5233,
      "step": 520
    },
    {
      "epoch": 1.325,
      "grad_norm": 0.6812359690666199,
      "learning_rate": 5e-06,
      "loss": 0.5253,
      "step": 530
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.5700986385345459,
      "learning_rate": 5e-06,
      "loss": 0.5198,
      "step": 540
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.7843871712684631,
      "learning_rate": 5e-06,
      "loss": 0.5199,
      "step": 550
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.5454580187797546,
      "learning_rate": 5e-06,
      "loss": 0.5206,
      "step": 560
    },
    {
      "epoch": 1.425,
      "grad_norm": 0.5220310091972351,
      "learning_rate": 5e-06,
      "loss": 0.5213,
      "step": 570
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.5269823670387268,
      "learning_rate": 5e-06,
      "loss": 0.5222,
      "step": 580
    },
    {
      "epoch": 1.475,
      "grad_norm": 0.5376739501953125,
      "learning_rate": 5e-06,
      "loss": 0.5194,
      "step": 590
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.58550626039505,
      "learning_rate": 5e-06,
      "loss": 0.5238,
      "step": 600
    },
    {
      "epoch": 1.525,
      "grad_norm": 0.5343301296234131,
      "learning_rate": 5e-06,
      "loss": 0.515,
      "step": 610
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.6829107999801636,
      "learning_rate": 5e-06,
      "loss": 0.519,
      "step": 620
    },
    {
      "epoch": 1.575,
      "grad_norm": 0.7822884917259216,
      "learning_rate": 5e-06,
      "loss": 0.5103,
      "step": 630
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5208476781845093,
      "learning_rate": 5e-06,
      "loss": 0.515,
      "step": 640
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.6014963984489441,
      "learning_rate": 5e-06,
      "loss": 0.5162,
      "step": 650
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.5044267177581787,
      "learning_rate": 5e-06,
      "loss": 0.5073,
      "step": 660
    },
    {
      "epoch": 1.675,
      "grad_norm": 0.6184217929840088,
      "learning_rate": 5e-06,
      "loss": 0.5114,
      "step": 670
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.5359528064727783,
      "learning_rate": 5e-06,
      "loss": 0.5074,
      "step": 680
    },
    {
      "epoch": 1.725,
      "grad_norm": 0.5592532753944397,
      "learning_rate": 5e-06,
      "loss": 0.5129,
      "step": 690
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6100260615348816,
      "learning_rate": 5e-06,
      "loss": 0.5147,
      "step": 700
    },
    {
      "epoch": 1.775,
      "grad_norm": 0.542510986328125,
      "learning_rate": 5e-06,
      "loss": 0.5091,
      "step": 710
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.5745417475700378,
      "learning_rate": 5e-06,
      "loss": 0.5064,
      "step": 720
    },
    {
      "epoch": 1.825,
      "grad_norm": 0.48998454213142395,
      "learning_rate": 5e-06,
      "loss": 0.5114,
      "step": 730
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.48566123843193054,
      "learning_rate": 5e-06,
      "loss": 0.5136,
      "step": 740
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.5040503740310669,
      "learning_rate": 5e-06,
      "loss": 0.5139,
      "step": 750
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.5267658233642578,
      "learning_rate": 5e-06,
      "loss": 0.5099,
      "step": 760
    },
    {
      "epoch": 1.925,
      "grad_norm": 0.6203451156616211,
      "learning_rate": 5e-06,
      "loss": 0.5062,
      "step": 770
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.5038536787033081,
      "learning_rate": 5e-06,
      "loss": 0.5062,
      "step": 780
    },
    {
      "epoch": 1.975,
      "grad_norm": 0.571147620677948,
      "learning_rate": 5e-06,
      "loss": 0.5072,
      "step": 790
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5538639426231384,
      "learning_rate": 5e-06,
      "loss": 0.5106,
      "step": 800
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5380541682243347,
      "eval_runtime": 36.1734,
      "eval_samples_per_second": 297.484,
      "eval_steps_per_second": 1.189,
      "step": 800
    },
    {
      "epoch": 2.025,
      "grad_norm": 0.5824851989746094,
      "learning_rate": 5e-06,
      "loss": 0.4697,
      "step": 810
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.5304216742515564,
      "learning_rate": 5e-06,
      "loss": 0.4702,
      "step": 820
    },
    {
      "epoch": 2.075,
      "grad_norm": 0.6173574924468994,
      "learning_rate": 5e-06,
      "loss": 0.4704,
      "step": 830
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.5461722016334534,
      "learning_rate": 5e-06,
      "loss": 0.4737,
      "step": 840
    },
    {
      "epoch": 2.125,
      "grad_norm": 0.5816064476966858,
      "learning_rate": 5e-06,
      "loss": 0.472,
      "step": 850
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.5272941589355469,
      "learning_rate": 5e-06,
      "loss": 0.47,
      "step": 860
    },
    {
      "epoch": 2.175,
      "grad_norm": 0.6730207800865173,
      "learning_rate": 5e-06,
      "loss": 0.4747,
      "step": 870
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.5066570043563843,
      "learning_rate": 5e-06,
      "loss": 0.4721,
      "step": 880
    },
    {
      "epoch": 2.225,
      "grad_norm": 0.5733837485313416,
      "learning_rate": 5e-06,
      "loss": 0.47,
      "step": 890
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.5919420719146729,
      "learning_rate": 5e-06,
      "loss": 0.4697,
      "step": 900
    },
    {
      "epoch": 2.275,
      "grad_norm": 0.600871741771698,
      "learning_rate": 5e-06,
      "loss": 0.473,
      "step": 910
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.5913262367248535,
      "learning_rate": 5e-06,
      "loss": 0.4734,
      "step": 920
    },
    {
      "epoch": 2.325,
      "grad_norm": 0.6537840962409973,
      "learning_rate": 5e-06,
      "loss": 0.4702,
      "step": 930
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.4861612915992737,
      "learning_rate": 5e-06,
      "loss": 0.4713,
      "step": 940
    },
    {
      "epoch": 2.375,
      "grad_norm": 0.5109017491340637,
      "learning_rate": 5e-06,
      "loss": 0.4714,
      "step": 950
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.560487687587738,
      "learning_rate": 5e-06,
      "loss": 0.4751,
      "step": 960
    },
    {
      "epoch": 2.425,
      "grad_norm": 0.618090033531189,
      "learning_rate": 5e-06,
      "loss": 0.4703,
      "step": 970
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.5797336101531982,
      "learning_rate": 5e-06,
      "loss": 0.4694,
      "step": 980
    },
    {
      "epoch": 2.475,
      "grad_norm": 0.7067740559577942,
      "learning_rate": 5e-06,
      "loss": 0.4695,
      "step": 990
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.7067424654960632,
      "learning_rate": 5e-06,
      "loss": 0.4772,
      "step": 1000
    },
    {
      "epoch": 2.525,
      "grad_norm": 0.5409600138664246,
      "learning_rate": 5e-06,
      "loss": 0.4696,
      "step": 1010
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.6049436926841736,
      "learning_rate": 5e-06,
      "loss": 0.4763,
      "step": 1020
    },
    {
      "epoch": 2.575,
      "grad_norm": 0.591397225856781,
      "learning_rate": 5e-06,
      "loss": 0.4757,
      "step": 1030
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.5943789482116699,
      "learning_rate": 5e-06,
      "loss": 0.475,
      "step": 1040
    },
    {
      "epoch": 2.625,
      "grad_norm": 0.5674725770950317,
      "learning_rate": 5e-06,
      "loss": 0.4732,
      "step": 1050
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.4904470443725586,
      "learning_rate": 5e-06,
      "loss": 0.4734,
      "step": 1060
    },
    {
      "epoch": 2.675,
      "grad_norm": 0.5225439667701721,
      "learning_rate": 5e-06,
      "loss": 0.4795,
      "step": 1070
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.6196999549865723,
      "learning_rate": 5e-06,
      "loss": 0.4751,
      "step": 1080
    },
    {
      "epoch": 2.725,
      "grad_norm": 0.7215800881385803,
      "learning_rate": 5e-06,
      "loss": 0.4696,
      "step": 1090
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.6502852439880371,
      "learning_rate": 5e-06,
      "loss": 0.474,
      "step": 1100
    },
    {
      "epoch": 2.775,
      "grad_norm": 0.512017011642456,
      "learning_rate": 5e-06,
      "loss": 0.4763,
      "step": 1110
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.5410541296005249,
      "learning_rate": 5e-06,
      "loss": 0.4705,
      "step": 1120
    },
    {
      "epoch": 2.825,
      "grad_norm": 0.5041213631629944,
      "learning_rate": 5e-06,
      "loss": 0.469,
      "step": 1130
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.5700607895851135,
      "learning_rate": 5e-06,
      "loss": 0.4749,
      "step": 1140
    },
    {
      "epoch": 2.875,
      "grad_norm": 0.4994681775569916,
      "learning_rate": 5e-06,
      "loss": 0.4711,
      "step": 1150
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.5657273530960083,
      "learning_rate": 5e-06,
      "loss": 0.4714,
      "step": 1160
    },
    {
      "epoch": 2.925,
      "grad_norm": 0.48754239082336426,
      "learning_rate": 5e-06,
      "loss": 0.4718,
      "step": 1170
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.5921581983566284,
      "learning_rate": 5e-06,
      "loss": 0.4729,
      "step": 1180
    },
    {
      "epoch": 2.975,
      "grad_norm": 0.5031780004501343,
      "learning_rate": 5e-06,
      "loss": 0.4747,
      "step": 1190
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6396584510803223,
      "learning_rate": 5e-06,
      "loss": 0.4685,
      "step": 1200
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.532584547996521,
      "eval_runtime": 36.8565,
      "eval_samples_per_second": 291.97,
      "eval_steps_per_second": 1.167,
      "step": 1200
    },
    {
      "epoch": 3.0,
      "step": 1200,
      "total_flos": 5.666028609022722e+19,
      "train_loss": 0.5332197288672129,
      "train_runtime": 8530.7812,
      "train_samples_per_second": 71.901,
      "train_steps_per_second": 0.141
    }
  ],
  "logging_steps": 10,
  "max_steps": 1200,
  "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": 5.666028609022722e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}