{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 972,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.030864197530864196,
      "grad_norm": 5.446850666927686,
      "learning_rate": 5e-06,
      "loss": 0.9199,
      "step": 10
    },
    {
      "epoch": 0.06172839506172839,
      "grad_norm": 2.663126142630872,
      "learning_rate": 5e-06,
      "loss": 0.821,
      "step": 20
    },
    {
      "epoch": 0.09259259259259259,
      "grad_norm": 1.1642696107304236,
      "learning_rate": 5e-06,
      "loss": 0.7706,
      "step": 30
    },
    {
      "epoch": 0.12345679012345678,
      "grad_norm": 4.280110588222716,
      "learning_rate": 5e-06,
      "loss": 0.7596,
      "step": 40
    },
    {
      "epoch": 0.15432098765432098,
      "grad_norm": 4.412787627263132,
      "learning_rate": 5e-06,
      "loss": 0.746,
      "step": 50
    },
    {
      "epoch": 0.18518518518518517,
      "grad_norm": 1.5711972402992538,
      "learning_rate": 5e-06,
      "loss": 0.7267,
      "step": 60
    },
    {
      "epoch": 0.21604938271604937,
      "grad_norm": 0.9451756673135364,
      "learning_rate": 5e-06,
      "loss": 0.7186,
      "step": 70
    },
    {
      "epoch": 0.24691358024691357,
      "grad_norm": 0.9020860720269491,
      "learning_rate": 5e-06,
      "loss": 0.7099,
      "step": 80
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 0.7223983788610128,
      "learning_rate": 5e-06,
      "loss": 0.7001,
      "step": 90
    },
    {
      "epoch": 0.30864197530864196,
      "grad_norm": 0.8632286972903407,
      "learning_rate": 5e-06,
      "loss": 0.6965,
      "step": 100
    },
    {
      "epoch": 0.3395061728395062,
      "grad_norm": 0.6600509272889379,
      "learning_rate": 5e-06,
      "loss": 0.6878,
      "step": 110
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.8445466577407515,
      "learning_rate": 5e-06,
      "loss": 0.685,
      "step": 120
    },
    {
      "epoch": 0.4012345679012346,
      "grad_norm": 0.7606146778204088,
      "learning_rate": 5e-06,
      "loss": 0.6826,
      "step": 130
    },
    {
      "epoch": 0.43209876543209874,
      "grad_norm": 0.6093510086614801,
      "learning_rate": 5e-06,
      "loss": 0.6828,
      "step": 140
    },
    {
      "epoch": 0.46296296296296297,
      "grad_norm": 0.7249517817925787,
      "learning_rate": 5e-06,
      "loss": 0.6808,
      "step": 150
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 0.559508936091321,
      "learning_rate": 5e-06,
      "loss": 0.6806,
      "step": 160
    },
    {
      "epoch": 0.5246913580246914,
      "grad_norm": 0.6485112723957187,
      "learning_rate": 5e-06,
      "loss": 0.6755,
      "step": 170
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.5734053780909245,
      "learning_rate": 5e-06,
      "loss": 0.6676,
      "step": 180
    },
    {
      "epoch": 0.5864197530864198,
      "grad_norm": 0.7330429838481911,
      "learning_rate": 5e-06,
      "loss": 0.6695,
      "step": 190
    },
    {
      "epoch": 0.6172839506172839,
      "grad_norm": 0.7031639101048125,
      "learning_rate": 5e-06,
      "loss": 0.669,
      "step": 200
    },
    {
      "epoch": 0.6481481481481481,
      "grad_norm": 0.5618292117245931,
      "learning_rate": 5e-06,
      "loss": 0.6759,
      "step": 210
    },
    {
      "epoch": 0.6790123456790124,
      "grad_norm": 0.5384920681614226,
      "learning_rate": 5e-06,
      "loss": 0.6626,
      "step": 220
    },
    {
      "epoch": 0.7098765432098766,
      "grad_norm": 0.8695843119793333,
      "learning_rate": 5e-06,
      "loss": 0.6613,
      "step": 230
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 0.8530303198305609,
      "learning_rate": 5e-06,
      "loss": 0.6622,
      "step": 240
    },
    {
      "epoch": 0.7716049382716049,
      "grad_norm": 0.7246183829809985,
      "learning_rate": 5e-06,
      "loss": 0.6585,
      "step": 250
    },
    {
      "epoch": 0.8024691358024691,
      "grad_norm": 0.6467615123805251,
      "learning_rate": 5e-06,
      "loss": 0.6663,
      "step": 260
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.6051023696593527,
      "learning_rate": 5e-06,
      "loss": 0.6578,
      "step": 270
    },
    {
      "epoch": 0.8641975308641975,
      "grad_norm": 0.5325735601636417,
      "learning_rate": 5e-06,
      "loss": 0.6628,
      "step": 280
    },
    {
      "epoch": 0.8950617283950617,
      "grad_norm": 0.5624534504885511,
      "learning_rate": 5e-06,
      "loss": 0.6551,
      "step": 290
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 0.5105392081865389,
      "learning_rate": 5e-06,
      "loss": 0.6585,
      "step": 300
    },
    {
      "epoch": 0.9567901234567902,
      "grad_norm": 0.7882251080908214,
      "learning_rate": 5e-06,
      "loss": 0.6578,
      "step": 310
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 0.7218610961797266,
      "learning_rate": 5e-06,
      "loss": 0.6542,
      "step": 320
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6550179123878479,
      "eval_runtime": 31.8484,
      "eval_samples_per_second": 273.641,
      "eval_steps_per_second": 1.099,
      "step": 324
    },
    {
      "epoch": 1.0185185185185186,
      "grad_norm": 0.7148119037457841,
      "learning_rate": 5e-06,
      "loss": 0.624,
      "step": 330
    },
    {
      "epoch": 1.0493827160493827,
      "grad_norm": 0.8285481607039845,
      "learning_rate": 5e-06,
      "loss": 0.6138,
      "step": 340
    },
    {
      "epoch": 1.0802469135802468,
      "grad_norm": 0.5405908730469038,
      "learning_rate": 5e-06,
      "loss": 0.6154,
      "step": 350
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.6837013751545337,
      "learning_rate": 5e-06,
      "loss": 0.613,
      "step": 360
    },
    {
      "epoch": 1.1419753086419753,
      "grad_norm": 0.8086734597067436,
      "learning_rate": 5e-06,
      "loss": 0.6051,
      "step": 370
    },
    {
      "epoch": 1.1728395061728394,
      "grad_norm": 0.6168727254851533,
      "learning_rate": 5e-06,
      "loss": 0.6047,
      "step": 380
    },
    {
      "epoch": 1.2037037037037037,
      "grad_norm": 0.6222564529681911,
      "learning_rate": 5e-06,
      "loss": 0.6135,
      "step": 390
    },
    {
      "epoch": 1.2345679012345678,
      "grad_norm": 0.6502199448782471,
      "learning_rate": 5e-06,
      "loss": 0.6092,
      "step": 400
    },
    {
      "epoch": 1.2654320987654322,
      "grad_norm": 0.642957942043697,
      "learning_rate": 5e-06,
      "loss": 0.6151,
      "step": 410
    },
    {
      "epoch": 1.2962962962962963,
      "grad_norm": 0.6973190504371732,
      "learning_rate": 5e-06,
      "loss": 0.6087,
      "step": 420
    },
    {
      "epoch": 1.3271604938271606,
      "grad_norm": 0.5681720826825847,
      "learning_rate": 5e-06,
      "loss": 0.6058,
      "step": 430
    },
    {
      "epoch": 1.3580246913580247,
      "grad_norm": 0.5560347945291988,
      "learning_rate": 5e-06,
      "loss": 0.6152,
      "step": 440
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.5202152575422632,
      "learning_rate": 5e-06,
      "loss": 0.6112,
      "step": 450
    },
    {
      "epoch": 1.4197530864197532,
      "grad_norm": 0.6652453548193578,
      "learning_rate": 5e-06,
      "loss": 0.6084,
      "step": 460
    },
    {
      "epoch": 1.4506172839506173,
      "grad_norm": 0.5959726615804369,
      "learning_rate": 5e-06,
      "loss": 0.611,
      "step": 470
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 0.5753042580268376,
      "learning_rate": 5e-06,
      "loss": 0.6099,
      "step": 480
    },
    {
      "epoch": 1.5123456790123457,
      "grad_norm": 0.5657589973831328,
      "learning_rate": 5e-06,
      "loss": 0.6158,
      "step": 490
    },
    {
      "epoch": 1.5432098765432098,
      "grad_norm": 0.48962281372551314,
      "learning_rate": 5e-06,
      "loss": 0.612,
      "step": 500
    },
    {
      "epoch": 1.574074074074074,
      "grad_norm": 0.5867122795097997,
      "learning_rate": 5e-06,
      "loss": 0.6092,
      "step": 510
    },
    {
      "epoch": 1.6049382716049383,
      "grad_norm": 0.567887162579392,
      "learning_rate": 5e-06,
      "loss": 0.6049,
      "step": 520
    },
    {
      "epoch": 1.6358024691358026,
      "grad_norm": 0.5846370116192817,
      "learning_rate": 5e-06,
      "loss": 0.6102,
      "step": 530
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.7372794706158421,
      "learning_rate": 5e-06,
      "loss": 0.6121,
      "step": 540
    },
    {
      "epoch": 1.6975308641975309,
      "grad_norm": 0.5679832644741175,
      "learning_rate": 5e-06,
      "loss": 0.6039,
      "step": 550
    },
    {
      "epoch": 1.7283950617283952,
      "grad_norm": 0.4951982048137007,
      "learning_rate": 5e-06,
      "loss": 0.6097,
      "step": 560
    },
    {
      "epoch": 1.7592592592592593,
      "grad_norm": 0.5947044750562988,
      "learning_rate": 5e-06,
      "loss": 0.6108,
      "step": 570
    },
    {
      "epoch": 1.7901234567901234,
      "grad_norm": 0.5672737096963988,
      "learning_rate": 5e-06,
      "loss": 0.6095,
      "step": 580
    },
    {
      "epoch": 1.8209876543209877,
      "grad_norm": 0.5215432816515835,
      "learning_rate": 5e-06,
      "loss": 0.6096,
      "step": 590
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 0.5356125614039279,
      "learning_rate": 5e-06,
      "loss": 0.6037,
      "step": 600
    },
    {
      "epoch": 1.882716049382716,
      "grad_norm": 0.5290482585426443,
      "learning_rate": 5e-06,
      "loss": 0.6062,
      "step": 610
    },
    {
      "epoch": 1.9135802469135803,
      "grad_norm": 0.6241816332212873,
      "learning_rate": 5e-06,
      "loss": 0.6106,
      "step": 620
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 0.4959658493368521,
      "learning_rate": 5e-06,
      "loss": 0.6067,
      "step": 630
    },
    {
      "epoch": 1.9753086419753085,
      "grad_norm": 0.5282675010462219,
      "learning_rate": 5e-06,
      "loss": 0.6069,
      "step": 640
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6462317109107971,
      "eval_runtime": 32.1855,
      "eval_samples_per_second": 270.774,
      "eval_steps_per_second": 1.087,
      "step": 648
    },
    {
      "epoch": 2.006172839506173,
      "grad_norm": 0.7935581137695162,
      "learning_rate": 5e-06,
      "loss": 0.596,
      "step": 650
    },
    {
      "epoch": 2.037037037037037,
      "grad_norm": 0.6261816641543,
      "learning_rate": 5e-06,
      "loss": 0.5645,
      "step": 660
    },
    {
      "epoch": 2.067901234567901,
      "grad_norm": 0.5849160839297688,
      "learning_rate": 5e-06,
      "loss": 0.558,
      "step": 670
    },
    {
      "epoch": 2.0987654320987654,
      "grad_norm": 0.6105130235265663,
      "learning_rate": 5e-06,
      "loss": 0.5591,
      "step": 680
    },
    {
      "epoch": 2.1296296296296298,
      "grad_norm": 0.6033282280363377,
      "learning_rate": 5e-06,
      "loss": 0.5573,
      "step": 690
    },
    {
      "epoch": 2.1604938271604937,
      "grad_norm": 0.7739145588445309,
      "learning_rate": 5e-06,
      "loss": 0.5571,
      "step": 700
    },
    {
      "epoch": 2.191358024691358,
      "grad_norm": 0.626140801689634,
      "learning_rate": 5e-06,
      "loss": 0.56,
      "step": 710
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.5916889343710152,
      "learning_rate": 5e-06,
      "loss": 0.5571,
      "step": 720
    },
    {
      "epoch": 2.253086419753086,
      "grad_norm": 0.636584234517333,
      "learning_rate": 5e-06,
      "loss": 0.5638,
      "step": 730
    },
    {
      "epoch": 2.2839506172839505,
      "grad_norm": 0.5507349524263677,
      "learning_rate": 5e-06,
      "loss": 0.5724,
      "step": 740
    },
    {
      "epoch": 2.314814814814815,
      "grad_norm": 0.6732342797910674,
      "learning_rate": 5e-06,
      "loss": 0.567,
      "step": 750
    },
    {
      "epoch": 2.3456790123456788,
      "grad_norm": 0.5395015214035337,
      "learning_rate": 5e-06,
      "loss": 0.5569,
      "step": 760
    },
    {
      "epoch": 2.376543209876543,
      "grad_norm": 0.5992087208696053,
      "learning_rate": 5e-06,
      "loss": 0.5647,
      "step": 770
    },
    {
      "epoch": 2.4074074074074074,
      "grad_norm": 0.5102699543572234,
      "learning_rate": 5e-06,
      "loss": 0.5622,
      "step": 780
    },
    {
      "epoch": 2.4382716049382718,
      "grad_norm": 0.6025610989122395,
      "learning_rate": 5e-06,
      "loss": 0.5634,
      "step": 790
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 0.6231368998767735,
      "learning_rate": 5e-06,
      "loss": 0.5594,
      "step": 800
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5707370285498646,
      "learning_rate": 5e-06,
      "loss": 0.5685,
      "step": 810
    },
    {
      "epoch": 2.5308641975308643,
      "grad_norm": 0.5549089552658856,
      "learning_rate": 5e-06,
      "loss": 0.5629,
      "step": 820
    },
    {
      "epoch": 2.5617283950617287,
      "grad_norm": 0.542471712528822,
      "learning_rate": 5e-06,
      "loss": 0.5681,
      "step": 830
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 0.5566494869370542,
      "learning_rate": 5e-06,
      "loss": 0.5663,
      "step": 840
    },
    {
      "epoch": 2.623456790123457,
      "grad_norm": 0.5552433471790524,
      "learning_rate": 5e-06,
      "loss": 0.5635,
      "step": 850
    },
    {
      "epoch": 2.6543209876543212,
      "grad_norm": 0.5049248358370862,
      "learning_rate": 5e-06,
      "loss": 0.5654,
      "step": 860
    },
    {
      "epoch": 2.685185185185185,
      "grad_norm": 0.6071017582058569,
      "learning_rate": 5e-06,
      "loss": 0.5669,
      "step": 870
    },
    {
      "epoch": 2.7160493827160495,
      "grad_norm": 0.5625590442980464,
      "learning_rate": 5e-06,
      "loss": 0.5642,
      "step": 880
    },
    {
      "epoch": 2.746913580246914,
      "grad_norm": 0.5506548153090438,
      "learning_rate": 5e-06,
      "loss": 0.5638,
      "step": 890
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.6470830808425209,
      "learning_rate": 5e-06,
      "loss": 0.5629,
      "step": 900
    },
    {
      "epoch": 2.808641975308642,
      "grad_norm": 0.6202475264816772,
      "learning_rate": 5e-06,
      "loss": 0.566,
      "step": 910
    },
    {
      "epoch": 2.8395061728395063,
      "grad_norm": 0.6200010347166321,
      "learning_rate": 5e-06,
      "loss": 0.5672,
      "step": 920
    },
    {
      "epoch": 2.8703703703703702,
      "grad_norm": 0.5382597238916978,
      "learning_rate": 5e-06,
      "loss": 0.5674,
      "step": 930
    },
    {
      "epoch": 2.9012345679012346,
      "grad_norm": 0.6294871904049958,
      "learning_rate": 5e-06,
      "loss": 0.5714,
      "step": 940
    },
    {
      "epoch": 2.932098765432099,
      "grad_norm": 0.5413425077416242,
      "learning_rate": 5e-06,
      "loss": 0.5677,
      "step": 950
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.5307261656256946,
      "learning_rate": 5e-06,
      "loss": 0.5674,
      "step": 960
    },
    {
      "epoch": 2.993827160493827,
      "grad_norm": 0.519858507326187,
      "learning_rate": 5e-06,
      "loss": 0.5688,
      "step": 970
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6513996720314026,
      "eval_runtime": 31.4148,
      "eval_samples_per_second": 277.417,
      "eval_steps_per_second": 1.114,
      "step": 972
    },
    {
      "epoch": 3.0,
      "step": 972,
      "total_flos": 1628136202567680.0,
      "train_loss": 0.6230609294318368,
      "train_runtime": 6074.7502,
      "train_samples_per_second": 81.773,
      "train_steps_per_second": 0.16
    }
  ],
  "logging_steps": 10,
  "max_steps": 972,
  "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": 1628136202567680.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}