{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6737264673311185,
  "eval_steps": 500,
  "global_step": 4867,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013842746400885937,
      "grad_norm": 1.760809302330017,
      "learning_rate": 3.1095890410958907e-06,
      "loss": 0.8443,
      "step": 10
    },
    {
      "epoch": 0.0027685492801771874,
      "grad_norm": 1.084220051765442,
      "learning_rate": 4.3424657534246575e-06,
      "loss": 0.7778,
      "step": 20
    },
    {
      "epoch": 0.004152823920265781,
      "grad_norm": 0.7984484434127808,
      "learning_rate": 5.575342465753425e-06,
      "loss": 0.7318,
      "step": 30
    },
    {
      "epoch": 0.005537098560354375,
      "grad_norm": 0.7084514498710632,
      "learning_rate": 6.808219178082193e-06,
      "loss": 0.7495,
      "step": 40
    },
    {
      "epoch": 0.0069213732004429675,
      "grad_norm": 0.860415518283844,
      "learning_rate": 8.04109589041096e-06,
      "loss": 0.711,
      "step": 50
    },
    {
      "epoch": 0.008305647840531562,
      "grad_norm": 0.7939516305923462,
      "learning_rate": 9.273972602739727e-06,
      "loss": 0.7147,
      "step": 60
    },
    {
      "epoch": 0.009689922480620155,
      "grad_norm": 3.751399517059326,
      "learning_rate": 1.0506849315068496e-05,
      "loss": 0.6994,
      "step": 70
    },
    {
      "epoch": 0.01107419712070875,
      "grad_norm": 0.7517818212509155,
      "learning_rate": 1.1739726027397262e-05,
      "loss": 0.71,
      "step": 80
    },
    {
      "epoch": 0.012458471760797342,
      "grad_norm": 0.7785382866859436,
      "learning_rate": 1.2972602739726028e-05,
      "loss": 0.7167,
      "step": 90
    },
    {
      "epoch": 0.013842746400885935,
      "grad_norm": 1.0320597887039185,
      "learning_rate": 1.4205479452054796e-05,
      "loss": 0.7018,
      "step": 100
    },
    {
      "epoch": 0.01522702104097453,
      "grad_norm": 0.8340734839439392,
      "learning_rate": 1.5438356164383563e-05,
      "loss": 0.7107,
      "step": 110
    },
    {
      "epoch": 0.016611295681063124,
      "grad_norm": 0.6760812401771545,
      "learning_rate": 1.6671232876712333e-05,
      "loss": 0.7112,
      "step": 120
    },
    {
      "epoch": 0.017995570321151715,
      "grad_norm": 0.7238306999206543,
      "learning_rate": 1.7904109589041096e-05,
      "loss": 0.7048,
      "step": 130
    },
    {
      "epoch": 0.01937984496124031,
      "grad_norm": 0.6886943578720093,
      "learning_rate": 1.9136986301369865e-05,
      "loss": 0.7171,
      "step": 140
    },
    {
      "epoch": 0.020764119601328904,
      "grad_norm": 0.7005371451377869,
      "learning_rate": 2e-05,
      "loss": 0.7078,
      "step": 150
    },
    {
      "epoch": 0.0221483942414175,
      "grad_norm": 0.6996603012084961,
      "learning_rate": 2e-05,
      "loss": 0.7155,
      "step": 160
    },
    {
      "epoch": 0.02353266888150609,
      "grad_norm": 0.6375390291213989,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 170
    },
    {
      "epoch": 0.024916943521594685,
      "grad_norm": 0.5814144015312195,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 180
    },
    {
      "epoch": 0.02630121816168328,
      "grad_norm": 0.5828060507774353,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 190
    },
    {
      "epoch": 0.02768549280177187,
      "grad_norm": 0.6017023921012878,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 200
    },
    {
      "epoch": 0.029069767441860465,
      "grad_norm": 0.6145961880683899,
      "learning_rate": 2e-05,
      "loss": 0.7014,
      "step": 210
    },
    {
      "epoch": 0.03045404208194906,
      "grad_norm": 0.5833798050880432,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 220
    },
    {
      "epoch": 0.03183831672203765,
      "grad_norm": 0.5166535973548889,
      "learning_rate": 2e-05,
      "loss": 0.7022,
      "step": 230
    },
    {
      "epoch": 0.03322259136212625,
      "grad_norm": 0.5050677061080933,
      "learning_rate": 2e-05,
      "loss": 0.6917,
      "step": 240
    },
    {
      "epoch": 0.03460686600221484,
      "grad_norm": 0.6089950203895569,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 250
    },
    {
      "epoch": 0.03599114064230343,
      "grad_norm": 0.5114278793334961,
      "learning_rate": 2e-05,
      "loss": 0.6877,
      "step": 260
    },
    {
      "epoch": 0.03737541528239203,
      "grad_norm": 0.5379406213760376,
      "learning_rate": 2e-05,
      "loss": 0.7092,
      "step": 270
    },
    {
      "epoch": 0.03875968992248062,
      "grad_norm": 0.5379545092582703,
      "learning_rate": 2e-05,
      "loss": 0.6992,
      "step": 280
    },
    {
      "epoch": 0.04014396456256921,
      "grad_norm": 0.5013362169265747,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 290
    },
    {
      "epoch": 0.04152823920265781,
      "grad_norm": 0.601866602897644,
      "learning_rate": 2e-05,
      "loss": 0.6834,
      "step": 300
    },
    {
      "epoch": 0.0429125138427464,
      "grad_norm": 0.5356491208076477,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 310
    },
    {
      "epoch": 0.044296788482835,
      "grad_norm": 0.48086288571357727,
      "learning_rate": 2e-05,
      "loss": 0.6776,
      "step": 320
    },
    {
      "epoch": 0.04568106312292359,
      "grad_norm": 0.4765521287918091,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 330
    },
    {
      "epoch": 0.04706533776301218,
      "grad_norm": 0.48504871129989624,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 340
    },
    {
      "epoch": 0.04844961240310078,
      "grad_norm": 0.6373012065887451,
      "learning_rate": 2e-05,
      "loss": 0.6813,
      "step": 350
    },
    {
      "epoch": 0.04983388704318937,
      "grad_norm": 0.47384321689605713,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 360
    },
    {
      "epoch": 0.05121816168327796,
      "grad_norm": 0.4874902367591858,
      "learning_rate": 2e-05,
      "loss": 0.6798,
      "step": 370
    },
    {
      "epoch": 0.05260243632336656,
      "grad_norm": 0.46476519107818604,
      "learning_rate": 2e-05,
      "loss": 0.6808,
      "step": 380
    },
    {
      "epoch": 0.05398671096345515,
      "grad_norm": 0.4813632369041443,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 390
    },
    {
      "epoch": 0.05537098560354374,
      "grad_norm": 0.4901427626609802,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 400
    },
    {
      "epoch": 0.05675526024363234,
      "grad_norm": 0.44426673650741577,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 410
    },
    {
      "epoch": 0.05813953488372093,
      "grad_norm": 0.4643983244895935,
      "learning_rate": 2e-05,
      "loss": 0.6779,
      "step": 420
    },
    {
      "epoch": 0.05952380952380952,
      "grad_norm": 0.4741644859313965,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 430
    },
    {
      "epoch": 0.06090808416389812,
      "grad_norm": 0.4387054443359375,
      "learning_rate": 2e-05,
      "loss": 0.6728,
      "step": 440
    },
    {
      "epoch": 0.06229235880398671,
      "grad_norm": 0.5127604007720947,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 450
    },
    {
      "epoch": 0.0636766334440753,
      "grad_norm": 0.4588662385940552,
      "learning_rate": 2e-05,
      "loss": 0.6945,
      "step": 460
    },
    {
      "epoch": 0.0650609080841639,
      "grad_norm": 0.47103187441825867,
      "learning_rate": 2e-05,
      "loss": 0.6935,
      "step": 470
    },
    {
      "epoch": 0.0664451827242525,
      "grad_norm": 0.4784078299999237,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 480
    },
    {
      "epoch": 0.06782945736434108,
      "grad_norm": 0.5959983468055725,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 490
    },
    {
      "epoch": 0.06921373200442968,
      "grad_norm": 0.4589248597621918,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 500
    },
    {
      "epoch": 0.07059800664451828,
      "grad_norm": 0.4595228433609009,
      "learning_rate": 2e-05,
      "loss": 0.6749,
      "step": 510
    },
    {
      "epoch": 0.07198228128460686,
      "grad_norm": 0.5400301218032837,
      "learning_rate": 2e-05,
      "loss": 0.6815,
      "step": 520
    },
    {
      "epoch": 0.07336655592469546,
      "grad_norm": 0.4362351894378662,
      "learning_rate": 2e-05,
      "loss": 0.6792,
      "step": 530
    },
    {
      "epoch": 0.07475083056478406,
      "grad_norm": 0.4854046404361725,
      "learning_rate": 2e-05,
      "loss": 0.6725,
      "step": 540
    },
    {
      "epoch": 0.07613510520487264,
      "grad_norm": 0.4978936016559601,
      "learning_rate": 2e-05,
      "loss": 0.6646,
      "step": 550
    },
    {
      "epoch": 0.07751937984496124,
      "grad_norm": 0.4682069420814514,
      "learning_rate": 2e-05,
      "loss": 0.6715,
      "step": 560
    },
    {
      "epoch": 0.07890365448504984,
      "grad_norm": 0.4608563482761383,
      "learning_rate": 2e-05,
      "loss": 0.6746,
      "step": 570
    },
    {
      "epoch": 0.08028792912513842,
      "grad_norm": 0.42644211649894714,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 580
    },
    {
      "epoch": 0.08167220376522702,
      "grad_norm": 0.4320621192455292,
      "learning_rate": 2e-05,
      "loss": 0.6616,
      "step": 590
    },
    {
      "epoch": 0.08305647840531562,
      "grad_norm": 0.4401155114173889,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 600
    },
    {
      "epoch": 0.0844407530454042,
      "grad_norm": 0.442085325717926,
      "learning_rate": 2e-05,
      "loss": 0.6716,
      "step": 610
    },
    {
      "epoch": 0.0858250276854928,
      "grad_norm": 0.45256638526916504,
      "learning_rate": 2e-05,
      "loss": 0.6819,
      "step": 620
    },
    {
      "epoch": 0.0872093023255814,
      "grad_norm": 0.4555166959762573,
      "learning_rate": 2e-05,
      "loss": 0.6592,
      "step": 630
    },
    {
      "epoch": 0.08859357696567,
      "grad_norm": 0.45368900895118713,
      "learning_rate": 2e-05,
      "loss": 0.6717,
      "step": 640
    },
    {
      "epoch": 0.08997785160575858,
      "grad_norm": 0.4519157111644745,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 650
    },
    {
      "epoch": 0.09136212624584718,
      "grad_norm": 0.462415486574173,
      "learning_rate": 2e-05,
      "loss": 0.6709,
      "step": 660
    },
    {
      "epoch": 0.09274640088593578,
      "grad_norm": 0.501905083656311,
      "learning_rate": 2e-05,
      "loss": 0.6708,
      "step": 670
    },
    {
      "epoch": 0.09413067552602436,
      "grad_norm": 0.4555647075176239,
      "learning_rate": 2e-05,
      "loss": 0.6727,
      "step": 680
    },
    {
      "epoch": 0.09551495016611296,
      "grad_norm": 0.4773111045360565,
      "learning_rate": 2e-05,
      "loss": 0.6683,
      "step": 690
    },
    {
      "epoch": 0.09689922480620156,
      "grad_norm": 0.4588943421840668,
      "learning_rate": 2e-05,
      "loss": 0.6831,
      "step": 700
    },
    {
      "epoch": 0.09828349944629014,
      "grad_norm": 0.4584910571575165,
      "learning_rate": 2e-05,
      "loss": 0.6691,
      "step": 710
    },
    {
      "epoch": 0.09966777408637874,
      "grad_norm": 0.44037702679634094,
      "learning_rate": 2e-05,
      "loss": 0.6666,
      "step": 720
    },
    {
      "epoch": 0.10105204872646734,
      "grad_norm": 0.44435733556747437,
      "learning_rate": 2e-05,
      "loss": 0.6664,
      "step": 730
    },
    {
      "epoch": 0.10243632336655592,
      "grad_norm": 0.43205612897872925,
      "learning_rate": 2e-05,
      "loss": 0.6657,
      "step": 740
    },
    {
      "epoch": 0.10382059800664452,
      "grad_norm": 0.4339076280593872,
      "learning_rate": 2e-05,
      "loss": 0.6667,
      "step": 750
    },
    {
      "epoch": 0.10520487264673312,
      "grad_norm": 0.7289295196533203,
      "learning_rate": 2e-05,
      "loss": 0.6622,
      "step": 760
    },
    {
      "epoch": 0.1065891472868217,
      "grad_norm": 0.46540015935897827,
      "learning_rate": 2e-05,
      "loss": 0.6643,
      "step": 770
    },
    {
      "epoch": 0.1079734219269103,
      "grad_norm": 0.43192070722579956,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 780
    },
    {
      "epoch": 0.1093576965669989,
      "grad_norm": 0.4702797532081604,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 790
    },
    {
      "epoch": 0.11074197120708748,
      "grad_norm": 0.4620274603366852,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 800
    },
    {
      "epoch": 0.11212624584717608,
      "grad_norm": 0.4694966971874237,
      "learning_rate": 2e-05,
      "loss": 0.6596,
      "step": 810
    },
    {
      "epoch": 0.11351052048726468,
      "grad_norm": 0.5121780633926392,
      "learning_rate": 2e-05,
      "loss": 0.672,
      "step": 820
    },
    {
      "epoch": 0.11489479512735326,
      "grad_norm": 0.4460819959640503,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 830
    },
    {
      "epoch": 0.11627906976744186,
      "grad_norm": 0.4590769112110138,
      "learning_rate": 2e-05,
      "loss": 0.6636,
      "step": 840
    },
    {
      "epoch": 0.11766334440753046,
      "grad_norm": 0.4578828811645508,
      "learning_rate": 2e-05,
      "loss": 0.6563,
      "step": 850
    },
    {
      "epoch": 0.11904761904761904,
      "grad_norm": 0.4304739832878113,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 860
    },
    {
      "epoch": 0.12043189368770764,
      "grad_norm": 0.496647447347641,
      "learning_rate": 2e-05,
      "loss": 0.6537,
      "step": 870
    },
    {
      "epoch": 0.12181616832779624,
      "grad_norm": 0.46330809593200684,
      "learning_rate": 2e-05,
      "loss": 0.6556,
      "step": 880
    },
    {
      "epoch": 0.12320044296788484,
      "grad_norm": 0.4382369816303253,
      "learning_rate": 2e-05,
      "loss": 0.6794,
      "step": 890
    },
    {
      "epoch": 0.12458471760797342,
      "grad_norm": 0.48029306530952454,
      "learning_rate": 2e-05,
      "loss": 0.6684,
      "step": 900
    },
    {
      "epoch": 0.12596899224806202,
      "grad_norm": 0.4352469742298126,
      "learning_rate": 2e-05,
      "loss": 0.6405,
      "step": 910
    },
    {
      "epoch": 0.1273532668881506,
      "grad_norm": 0.44442424178123474,
      "learning_rate": 2e-05,
      "loss": 0.6775,
      "step": 920
    },
    {
      "epoch": 0.1287375415282392,
      "grad_norm": 0.4895966649055481,
      "learning_rate": 2e-05,
      "loss": 0.6671,
      "step": 930
    },
    {
      "epoch": 0.1301218161683278,
      "grad_norm": 0.46154484152793884,
      "learning_rate": 2e-05,
      "loss": 0.6613,
      "step": 940
    },
    {
      "epoch": 0.13150609080841638,
      "grad_norm": 0.47581154108047485,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 950
    },
    {
      "epoch": 0.132890365448505,
      "grad_norm": 0.49340739846229553,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 960
    },
    {
      "epoch": 0.13427464008859358,
      "grad_norm": 0.6075978875160217,
      "learning_rate": 2e-05,
      "loss": 0.6661,
      "step": 970
    },
    {
      "epoch": 0.13565891472868216,
      "grad_norm": 0.46497467160224915,
      "learning_rate": 2e-05,
      "loss": 0.6804,
      "step": 980
    },
    {
      "epoch": 0.13704318936877077,
      "grad_norm": 0.43133673071861267,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 990
    },
    {
      "epoch": 0.13842746400885936,
      "grad_norm": 0.4831154942512512,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 1000
    },
    {
      "epoch": 0.13981173864894794,
      "grad_norm": 0.4346819519996643,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 1010
    },
    {
      "epoch": 0.14119601328903655,
      "grad_norm": 0.43662431836128235,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 1020
    },
    {
      "epoch": 0.14258028792912514,
      "grad_norm": 0.49456480145454407,
      "learning_rate": 2e-05,
      "loss": 0.6526,
      "step": 1030
    },
    {
      "epoch": 0.14396456256921372,
      "grad_norm": 0.5407099723815918,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 1040
    },
    {
      "epoch": 0.14534883720930233,
      "grad_norm": 0.4813271462917328,
      "learning_rate": 2e-05,
      "loss": 0.6492,
      "step": 1050
    },
    {
      "epoch": 0.14673311184939092,
      "grad_norm": 0.4658135771751404,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 1060
    },
    {
      "epoch": 0.1481173864894795,
      "grad_norm": 0.4237080514431,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 1070
    },
    {
      "epoch": 0.14950166112956811,
      "grad_norm": 0.459164559841156,
      "learning_rate": 2e-05,
      "loss": 0.6643,
      "step": 1080
    },
    {
      "epoch": 0.1508859357696567,
      "grad_norm": 0.4798552691936493,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 1090
    },
    {
      "epoch": 0.15227021040974528,
      "grad_norm": 0.462601900100708,
      "learning_rate": 2e-05,
      "loss": 0.6386,
      "step": 1100
    },
    {
      "epoch": 0.1536544850498339,
      "grad_norm": 0.42169931530952454,
      "learning_rate": 2e-05,
      "loss": 0.6352,
      "step": 1110
    },
    {
      "epoch": 0.15503875968992248,
      "grad_norm": 0.443325012922287,
      "learning_rate": 2e-05,
      "loss": 0.6426,
      "step": 1120
    },
    {
      "epoch": 0.15642303433001106,
      "grad_norm": 0.44760993123054504,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 1130
    },
    {
      "epoch": 0.15780730897009967,
      "grad_norm": 0.43030163645744324,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 1140
    },
    {
      "epoch": 0.15919158361018826,
      "grad_norm": 0.4333432912826538,
      "learning_rate": 2e-05,
      "loss": 0.6455,
      "step": 1150
    },
    {
      "epoch": 0.16057585825027684,
      "grad_norm": 0.44685250520706177,
      "learning_rate": 2e-05,
      "loss": 0.6562,
      "step": 1160
    },
    {
      "epoch": 0.16196013289036545,
      "grad_norm": 0.5148652195930481,
      "learning_rate": 2e-05,
      "loss": 0.6522,
      "step": 1170
    },
    {
      "epoch": 0.16334440753045404,
      "grad_norm": 0.4822603762149811,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 1180
    },
    {
      "epoch": 0.16472868217054262,
      "grad_norm": 0.4302799105644226,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 1190
    },
    {
      "epoch": 0.16611295681063123,
      "grad_norm": 0.49725010991096497,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 1200
    },
    {
      "epoch": 0.16749723145071982,
      "grad_norm": 0.4514305293560028,
      "learning_rate": 2e-05,
      "loss": 0.6523,
      "step": 1210
    },
    {
      "epoch": 0.1688815060908084,
      "grad_norm": 0.4627952575683594,
      "learning_rate": 2e-05,
      "loss": 0.6384,
      "step": 1220
    },
    {
      "epoch": 0.17026578073089702,
      "grad_norm": 0.42806243896484375,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 1230
    },
    {
      "epoch": 0.1716500553709856,
      "grad_norm": 0.43487489223480225,
      "learning_rate": 2e-05,
      "loss": 0.648,
      "step": 1240
    },
    {
      "epoch": 0.17303433001107418,
      "grad_norm": 0.4223099648952484,
      "learning_rate": 2e-05,
      "loss": 0.6417,
      "step": 1250
    },
    {
      "epoch": 0.1744186046511628,
      "grad_norm": 0.46509769558906555,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 1260
    },
    {
      "epoch": 0.17580287929125138,
      "grad_norm": 0.44957759976387024,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 1270
    },
    {
      "epoch": 0.17718715393134,
      "grad_norm": 0.4843831956386566,
      "learning_rate": 2e-05,
      "loss": 0.6369,
      "step": 1280
    },
    {
      "epoch": 0.17857142857142858,
      "grad_norm": 0.42932626605033875,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 1290
    },
    {
      "epoch": 0.17995570321151716,
      "grad_norm": 0.43993765115737915,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 1300
    },
    {
      "epoch": 0.18133997785160577,
      "grad_norm": 0.4289640188217163,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 1310
    },
    {
      "epoch": 0.18272425249169436,
      "grad_norm": 0.5552659034729004,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 1320
    },
    {
      "epoch": 0.18410852713178294,
      "grad_norm": 0.4631248414516449,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 1330
    },
    {
      "epoch": 0.18549280177187155,
      "grad_norm": 0.47540920972824097,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 1340
    },
    {
      "epoch": 0.18687707641196014,
      "grad_norm": 0.4299192726612091,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 1350
    },
    {
      "epoch": 0.18826135105204872,
      "grad_norm": 0.4769332706928253,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 1360
    },
    {
      "epoch": 0.18964562569213733,
      "grad_norm": 0.4631207287311554,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 1370
    },
    {
      "epoch": 0.19102990033222592,
      "grad_norm": 0.43387413024902344,
      "learning_rate": 2e-05,
      "loss": 0.6144,
      "step": 1380
    },
    {
      "epoch": 0.1924141749723145,
      "grad_norm": 0.4758026599884033,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 1390
    },
    {
      "epoch": 0.1937984496124031,
      "grad_norm": 0.43073710799217224,
      "learning_rate": 2e-05,
      "loss": 0.6293,
      "step": 1400
    },
    {
      "epoch": 0.1951827242524917,
      "grad_norm": 0.5552448630332947,
      "learning_rate": 2e-05,
      "loss": 0.6298,
      "step": 1410
    },
    {
      "epoch": 0.19656699889258028,
      "grad_norm": 0.44213637709617615,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 1420
    },
    {
      "epoch": 0.1979512735326689,
      "grad_norm": 0.49454012513160706,
      "learning_rate": 2e-05,
      "loss": 0.6275,
      "step": 1430
    },
    {
      "epoch": 0.19933554817275748,
      "grad_norm": 0.4356294274330139,
      "learning_rate": 2e-05,
      "loss": 0.6324,
      "step": 1440
    },
    {
      "epoch": 0.20071982281284606,
      "grad_norm": 0.4898640215396881,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 1450
    },
    {
      "epoch": 0.20210409745293467,
      "grad_norm": 0.4469102919101715,
      "learning_rate": 2e-05,
      "loss": 0.6375,
      "step": 1460
    },
    {
      "epoch": 0.20348837209302326,
      "grad_norm": 0.4744914472103119,
      "learning_rate": 2e-05,
      "loss": 0.6237,
      "step": 1470
    },
    {
      "epoch": 0.20487264673311184,
      "grad_norm": 0.4621232748031616,
      "learning_rate": 2e-05,
      "loss": 0.6383,
      "step": 1480
    },
    {
      "epoch": 0.20625692137320045,
      "grad_norm": 0.4394593834877014,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 1490
    },
    {
      "epoch": 0.20764119601328904,
      "grad_norm": 0.4693845808506012,
      "learning_rate": 2e-05,
      "loss": 0.6398,
      "step": 1500
    },
    {
      "epoch": 0.20902547065337762,
      "grad_norm": 0.44942423701286316,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 1510
    },
    {
      "epoch": 0.21040974529346623,
      "grad_norm": 0.4683709740638733,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 1520
    },
    {
      "epoch": 0.21179401993355482,
      "grad_norm": 0.4374249279499054,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 1530
    },
    {
      "epoch": 0.2131782945736434,
      "grad_norm": 0.4350873827934265,
      "learning_rate": 2e-05,
      "loss": 0.6076,
      "step": 1540
    },
    {
      "epoch": 0.214562569213732,
      "grad_norm": 0.49674034118652344,
      "learning_rate": 2e-05,
      "loss": 0.6112,
      "step": 1550
    },
    {
      "epoch": 0.2159468438538206,
      "grad_norm": 0.44921791553497314,
      "learning_rate": 2e-05,
      "loss": 0.6396,
      "step": 1560
    },
    {
      "epoch": 0.21733111849390918,
      "grad_norm": 0.4455544054508209,
      "learning_rate": 2e-05,
      "loss": 0.6101,
      "step": 1570
    },
    {
      "epoch": 0.2187153931339978,
      "grad_norm": 0.4863693416118622,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 1580
    },
    {
      "epoch": 0.22009966777408638,
      "grad_norm": 0.4984947443008423,
      "learning_rate": 2e-05,
      "loss": 0.6202,
      "step": 1590
    },
    {
      "epoch": 0.22148394241417496,
      "grad_norm": 0.4522366523742676,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 1600
    },
    {
      "epoch": 0.22286821705426357,
      "grad_norm": 0.45198485255241394,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 1610
    },
    {
      "epoch": 0.22425249169435216,
      "grad_norm": 0.4920777678489685,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 1620
    },
    {
      "epoch": 0.22563676633444074,
      "grad_norm": 0.43350744247436523,
      "learning_rate": 2e-05,
      "loss": 0.6261,
      "step": 1630
    },
    {
      "epoch": 0.22702104097452935,
      "grad_norm": 0.664797842502594,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 1640
    },
    {
      "epoch": 0.22840531561461794,
      "grad_norm": 0.48066094517707825,
      "learning_rate": 2e-05,
      "loss": 0.6344,
      "step": 1650
    },
    {
      "epoch": 0.22978959025470652,
      "grad_norm": 0.4406963586807251,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 1660
    },
    {
      "epoch": 0.23117386489479513,
      "grad_norm": 0.4168227016925812,
      "learning_rate": 2e-05,
      "loss": 0.6161,
      "step": 1670
    },
    {
      "epoch": 0.23255813953488372,
      "grad_norm": 0.42777279019355774,
      "learning_rate": 2e-05,
      "loss": 0.6184,
      "step": 1680
    },
    {
      "epoch": 0.2339424141749723,
      "grad_norm": 0.43606796860694885,
      "learning_rate": 2e-05,
      "loss": 0.6152,
      "step": 1690
    },
    {
      "epoch": 0.2353266888150609,
      "grad_norm": 0.4511287808418274,
      "learning_rate": 2e-05,
      "loss": 0.6258,
      "step": 1700
    },
    {
      "epoch": 0.2367109634551495,
      "grad_norm": 0.4567727744579315,
      "learning_rate": 2e-05,
      "loss": 0.6199,
      "step": 1710
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 0.44054827094078064,
      "learning_rate": 2e-05,
      "loss": 0.6141,
      "step": 1720
    },
    {
      "epoch": 0.2394795127353267,
      "grad_norm": 0.43460410833358765,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 1730
    },
    {
      "epoch": 0.24086378737541528,
      "grad_norm": 0.4277775287628174,
      "learning_rate": 2e-05,
      "loss": 0.6191,
      "step": 1740
    },
    {
      "epoch": 0.24224806201550386,
      "grad_norm": 0.4418438673019409,
      "learning_rate": 2e-05,
      "loss": 0.6254,
      "step": 1750
    },
    {
      "epoch": 0.24363233665559247,
      "grad_norm": 0.453975647687912,
      "learning_rate": 2e-05,
      "loss": 0.6068,
      "step": 1760
    },
    {
      "epoch": 0.24501661129568106,
      "grad_norm": 0.45285025238990784,
      "learning_rate": 2e-05,
      "loss": 0.6282,
      "step": 1770
    },
    {
      "epoch": 0.24640088593576967,
      "grad_norm": 0.4464224576950073,
      "learning_rate": 2e-05,
      "loss": 0.6174,
      "step": 1780
    },
    {
      "epoch": 0.24778516057585825,
      "grad_norm": 0.4236616790294647,
      "learning_rate": 2e-05,
      "loss": 0.6229,
      "step": 1790
    },
    {
      "epoch": 0.24916943521594684,
      "grad_norm": 0.4533737897872925,
      "learning_rate": 2e-05,
      "loss": 0.6224,
      "step": 1800
    },
    {
      "epoch": 0.2505537098560354,
      "grad_norm": 0.44441816210746765,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 1810
    },
    {
      "epoch": 0.25193798449612403,
      "grad_norm": 0.4282686114311218,
      "learning_rate": 2e-05,
      "loss": 0.6078,
      "step": 1820
    },
    {
      "epoch": 0.25332225913621265,
      "grad_norm": 0.43642640113830566,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 1830
    },
    {
      "epoch": 0.2547065337763012,
      "grad_norm": 0.5253880620002747,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 1840
    },
    {
      "epoch": 0.2560908084163898,
      "grad_norm": 0.45619305968284607,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 1850
    },
    {
      "epoch": 0.2574750830564784,
      "grad_norm": 0.5161237716674805,
      "learning_rate": 2e-05,
      "loss": 0.6272,
      "step": 1860
    },
    {
      "epoch": 0.258859357696567,
      "grad_norm": 0.465443879365921,
      "learning_rate": 2e-05,
      "loss": 0.6094,
      "step": 1870
    },
    {
      "epoch": 0.2602436323366556,
      "grad_norm": 0.5024373531341553,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 1880
    },
    {
      "epoch": 0.2616279069767442,
      "grad_norm": 0.46063631772994995,
      "learning_rate": 2e-05,
      "loss": 0.612,
      "step": 1890
    },
    {
      "epoch": 0.26301218161683276,
      "grad_norm": 0.4636746346950531,
      "learning_rate": 2e-05,
      "loss": 0.6178,
      "step": 1900
    },
    {
      "epoch": 0.2643964562569214,
      "grad_norm": 0.46138960123062134,
      "learning_rate": 2e-05,
      "loss": 0.5964,
      "step": 1910
    },
    {
      "epoch": 0.26578073089701,
      "grad_norm": 0.4579954445362091,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 1920
    },
    {
      "epoch": 0.26716500553709854,
      "grad_norm": 0.469111829996109,
      "learning_rate": 2e-05,
      "loss": 0.6114,
      "step": 1930
    },
    {
      "epoch": 0.26854928017718716,
      "grad_norm": 0.45288708806037903,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 1940
    },
    {
      "epoch": 0.26993355481727577,
      "grad_norm": 0.4392133951187134,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 1950
    },
    {
      "epoch": 0.2713178294573643,
      "grad_norm": 0.4534960091114044,
      "learning_rate": 2e-05,
      "loss": 0.6201,
      "step": 1960
    },
    {
      "epoch": 0.27270210409745294,
      "grad_norm": 0.5160900950431824,
      "learning_rate": 2e-05,
      "loss": 0.6325,
      "step": 1970
    },
    {
      "epoch": 0.27408637873754155,
      "grad_norm": 0.43736329674720764,
      "learning_rate": 2e-05,
      "loss": 0.6097,
      "step": 1980
    },
    {
      "epoch": 0.2754706533776301,
      "grad_norm": 0.48259100317955017,
      "learning_rate": 2e-05,
      "loss": 0.6128,
      "step": 1990
    },
    {
      "epoch": 0.2768549280177187,
      "grad_norm": 0.4431242048740387,
      "learning_rate": 2e-05,
      "loss": 0.6073,
      "step": 2000
    },
    {
      "epoch": 0.2782392026578073,
      "grad_norm": 0.4775521755218506,
      "learning_rate": 2e-05,
      "loss": 0.6157,
      "step": 2010
    },
    {
      "epoch": 0.2796234772978959,
      "grad_norm": 0.4758862853050232,
      "learning_rate": 2e-05,
      "loss": 0.6046,
      "step": 2020
    },
    {
      "epoch": 0.2810077519379845,
      "grad_norm": 0.48884904384613037,
      "learning_rate": 2e-05,
      "loss": 0.6014,
      "step": 2030
    },
    {
      "epoch": 0.2823920265780731,
      "grad_norm": 0.4610230326652527,
      "learning_rate": 2e-05,
      "loss": 0.6073,
      "step": 2040
    },
    {
      "epoch": 0.28377630121816166,
      "grad_norm": 0.4403124451637268,
      "learning_rate": 2e-05,
      "loss": 0.5874,
      "step": 2050
    },
    {
      "epoch": 0.2851605758582503,
      "grad_norm": 0.4460209906101227,
      "learning_rate": 2e-05,
      "loss": 0.5903,
      "step": 2060
    },
    {
      "epoch": 0.2865448504983389,
      "grad_norm": 0.447582870721817,
      "learning_rate": 2e-05,
      "loss": 0.6094,
      "step": 2070
    },
    {
      "epoch": 0.28792912513842744,
      "grad_norm": 0.47697415947914124,
      "learning_rate": 2e-05,
      "loss": 0.6105,
      "step": 2080
    },
    {
      "epoch": 0.28931339977851606,
      "grad_norm": 0.4408358633518219,
      "learning_rate": 2e-05,
      "loss": 0.6027,
      "step": 2090
    },
    {
      "epoch": 0.29069767441860467,
      "grad_norm": 0.44841769337654114,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 2100
    },
    {
      "epoch": 0.2920819490586932,
      "grad_norm": 0.45582085847854614,
      "learning_rate": 2e-05,
      "loss": 0.5907,
      "step": 2110
    },
    {
      "epoch": 0.29346622369878184,
      "grad_norm": 0.45418334007263184,
      "learning_rate": 2e-05,
      "loss": 0.6227,
      "step": 2120
    },
    {
      "epoch": 0.29485049833887045,
      "grad_norm": 0.4470612406730652,
      "learning_rate": 2e-05,
      "loss": 0.598,
      "step": 2130
    },
    {
      "epoch": 0.296234772978959,
      "grad_norm": 0.4700588285923004,
      "learning_rate": 2e-05,
      "loss": 0.5871,
      "step": 2140
    },
    {
      "epoch": 0.2976190476190476,
      "grad_norm": 0.5149217844009399,
      "learning_rate": 2e-05,
      "loss": 0.6015,
      "step": 2150
    },
    {
      "epoch": 0.29900332225913623,
      "grad_norm": 0.4543517827987671,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 2160
    },
    {
      "epoch": 0.3003875968992248,
      "grad_norm": 0.4750279486179352,
      "learning_rate": 2e-05,
      "loss": 0.611,
      "step": 2170
    },
    {
      "epoch": 0.3017718715393134,
      "grad_norm": 0.46759718656539917,
      "learning_rate": 2e-05,
      "loss": 0.6138,
      "step": 2180
    },
    {
      "epoch": 0.303156146179402,
      "grad_norm": 0.4290469288825989,
      "learning_rate": 2e-05,
      "loss": 0.5935,
      "step": 2190
    },
    {
      "epoch": 0.30454042081949056,
      "grad_norm": 0.4428347647190094,
      "learning_rate": 2e-05,
      "loss": 0.5926,
      "step": 2200
    },
    {
      "epoch": 0.3059246954595792,
      "grad_norm": 0.42132094502449036,
      "learning_rate": 2e-05,
      "loss": 0.5767,
      "step": 2210
    },
    {
      "epoch": 0.3073089700996678,
      "grad_norm": 0.4626559913158417,
      "learning_rate": 2e-05,
      "loss": 0.6038,
      "step": 2220
    },
    {
      "epoch": 0.30869324473975635,
      "grad_norm": 0.4289563000202179,
      "learning_rate": 2e-05,
      "loss": 0.5854,
      "step": 2230
    },
    {
      "epoch": 0.31007751937984496,
      "grad_norm": 0.4292846918106079,
      "learning_rate": 2e-05,
      "loss": 0.5782,
      "step": 2240
    },
    {
      "epoch": 0.31146179401993357,
      "grad_norm": 0.507136881351471,
      "learning_rate": 2e-05,
      "loss": 0.6047,
      "step": 2250
    },
    {
      "epoch": 0.3128460686600221,
      "grad_norm": 0.4628688097000122,
      "learning_rate": 2e-05,
      "loss": 0.617,
      "step": 2260
    },
    {
      "epoch": 0.31423034330011074,
      "grad_norm": 0.44342857599258423,
      "learning_rate": 2e-05,
      "loss": 0.5925,
      "step": 2270
    },
    {
      "epoch": 0.31561461794019935,
      "grad_norm": 0.47981342673301697,
      "learning_rate": 2e-05,
      "loss": 0.6012,
      "step": 2280
    },
    {
      "epoch": 0.3169988925802879,
      "grad_norm": 0.4353543817996979,
      "learning_rate": 2e-05,
      "loss": 0.5976,
      "step": 2290
    },
    {
      "epoch": 0.3183831672203765,
      "grad_norm": 0.45344865322113037,
      "learning_rate": 2e-05,
      "loss": 0.6178,
      "step": 2300
    },
    {
      "epoch": 0.31976744186046513,
      "grad_norm": 0.4914526641368866,
      "learning_rate": 2e-05,
      "loss": 0.5883,
      "step": 2310
    },
    {
      "epoch": 0.3211517165005537,
      "grad_norm": 0.5003400444984436,
      "learning_rate": 2e-05,
      "loss": 0.5878,
      "step": 2320
    },
    {
      "epoch": 0.3225359911406423,
      "grad_norm": 0.4585626721382141,
      "learning_rate": 2e-05,
      "loss": 0.6045,
      "step": 2330
    },
    {
      "epoch": 0.3239202657807309,
      "grad_norm": 0.4700723886489868,
      "learning_rate": 2e-05,
      "loss": 0.5991,
      "step": 2340
    },
    {
      "epoch": 0.32530454042081947,
      "grad_norm": 0.42575058341026306,
      "learning_rate": 2e-05,
      "loss": 0.5882,
      "step": 2350
    },
    {
      "epoch": 0.3266888150609081,
      "grad_norm": 0.445823073387146,
      "learning_rate": 2e-05,
      "loss": 0.602,
      "step": 2360
    },
    {
      "epoch": 0.3280730897009967,
      "grad_norm": 0.5308166742324829,
      "learning_rate": 2e-05,
      "loss": 0.5991,
      "step": 2370
    },
    {
      "epoch": 0.32945736434108525,
      "grad_norm": 0.45579978823661804,
      "learning_rate": 2e-05,
      "loss": 0.5849,
      "step": 2380
    },
    {
      "epoch": 0.33084163898117386,
      "grad_norm": 0.6233319044113159,
      "learning_rate": 2e-05,
      "loss": 0.5986,
      "step": 2390
    },
    {
      "epoch": 0.33222591362126247,
      "grad_norm": 0.4533543288707733,
      "learning_rate": 2e-05,
      "loss": 0.6032,
      "step": 2400
    },
    {
      "epoch": 0.333610188261351,
      "grad_norm": 0.48076629638671875,
      "learning_rate": 2e-05,
      "loss": 0.5865,
      "step": 2410
    },
    {
      "epoch": 0.33499446290143964,
      "grad_norm": 0.5166177749633789,
      "learning_rate": 2e-05,
      "loss": 0.5781,
      "step": 2420
    },
    {
      "epoch": 0.33637873754152825,
      "grad_norm": 0.5142418146133423,
      "learning_rate": 2e-05,
      "loss": 0.6013,
      "step": 2430
    },
    {
      "epoch": 0.3377630121816168,
      "grad_norm": 0.5224171876907349,
      "learning_rate": 2e-05,
      "loss": 0.5877,
      "step": 2440
    },
    {
      "epoch": 0.3391472868217054,
      "grad_norm": 0.4636785686016083,
      "learning_rate": 2e-05,
      "loss": 0.6055,
      "step": 2450
    },
    {
      "epoch": 0.34053156146179403,
      "grad_norm": 0.4638904929161072,
      "learning_rate": 2e-05,
      "loss": 0.5898,
      "step": 2460
    },
    {
      "epoch": 0.3419158361018826,
      "grad_norm": 0.535005509853363,
      "learning_rate": 2e-05,
      "loss": 0.5949,
      "step": 2470
    },
    {
      "epoch": 0.3433001107419712,
      "grad_norm": 0.4827979803085327,
      "learning_rate": 2e-05,
      "loss": 0.5995,
      "step": 2480
    },
    {
      "epoch": 0.3446843853820598,
      "grad_norm": 0.487187922000885,
      "learning_rate": 2e-05,
      "loss": 0.5991,
      "step": 2490
    },
    {
      "epoch": 0.34606866002214837,
      "grad_norm": 0.513708233833313,
      "learning_rate": 2e-05,
      "loss": 0.6189,
      "step": 2500
    },
    {
      "epoch": 0.347452934662237,
      "grad_norm": 0.4368520975112915,
      "learning_rate": 2e-05,
      "loss": 0.5885,
      "step": 2510
    },
    {
      "epoch": 0.3488372093023256,
      "grad_norm": 0.4899316430091858,
      "learning_rate": 2e-05,
      "loss": 0.5725,
      "step": 2520
    },
    {
      "epoch": 0.3502214839424142,
      "grad_norm": 0.48472902178764343,
      "learning_rate": 2e-05,
      "loss": 0.5796,
      "step": 2530
    },
    {
      "epoch": 0.35160575858250276,
      "grad_norm": 0.4706763029098511,
      "learning_rate": 2e-05,
      "loss": 0.5905,
      "step": 2540
    },
    {
      "epoch": 0.35299003322259137,
      "grad_norm": 0.45990684628486633,
      "learning_rate": 2e-05,
      "loss": 0.5512,
      "step": 2550
    },
    {
      "epoch": 0.35437430786268,
      "grad_norm": 0.44680994749069214,
      "learning_rate": 2e-05,
      "loss": 0.5854,
      "step": 2560
    },
    {
      "epoch": 0.35575858250276854,
      "grad_norm": 0.47117701172828674,
      "learning_rate": 2e-05,
      "loss": 0.5806,
      "step": 2570
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.47342702746391296,
      "learning_rate": 2e-05,
      "loss": 0.5817,
      "step": 2580
    },
    {
      "epoch": 0.35852713178294576,
      "grad_norm": 0.4331980049610138,
      "learning_rate": 2e-05,
      "loss": 0.6012,
      "step": 2590
    },
    {
      "epoch": 0.3599114064230343,
      "grad_norm": 0.48320332169532776,
      "learning_rate": 2e-05,
      "loss": 0.5814,
      "step": 2600
    },
    {
      "epoch": 0.36129568106312293,
      "grad_norm": 0.44478633999824524,
      "learning_rate": 2e-05,
      "loss": 0.5921,
      "step": 2610
    },
    {
      "epoch": 0.36267995570321154,
      "grad_norm": 0.45052891969680786,
      "learning_rate": 2e-05,
      "loss": 0.5965,
      "step": 2620
    },
    {
      "epoch": 0.3640642303433001,
      "grad_norm": 0.4961407482624054,
      "learning_rate": 2e-05,
      "loss": 0.5719,
      "step": 2630
    },
    {
      "epoch": 0.3654485049833887,
      "grad_norm": 0.44594651460647583,
      "learning_rate": 2e-05,
      "loss": 0.5916,
      "step": 2640
    },
    {
      "epoch": 0.3668327796234773,
      "grad_norm": 0.4535442292690277,
      "learning_rate": 2e-05,
      "loss": 0.5809,
      "step": 2650
    },
    {
      "epoch": 0.3682170542635659,
      "grad_norm": 0.4547760784626007,
      "learning_rate": 2e-05,
      "loss": 0.5779,
      "step": 2660
    },
    {
      "epoch": 0.3696013289036545,
      "grad_norm": 0.4994950294494629,
      "learning_rate": 2e-05,
      "loss": 0.5846,
      "step": 2670
    },
    {
      "epoch": 0.3709856035437431,
      "grad_norm": 0.4898880124092102,
      "learning_rate": 2e-05,
      "loss": 0.5666,
      "step": 2680
    },
    {
      "epoch": 0.37236987818383166,
      "grad_norm": 0.5061846375465393,
      "learning_rate": 2e-05,
      "loss": 0.5865,
      "step": 2690
    },
    {
      "epoch": 0.37375415282392027,
      "grad_norm": 0.4583248794078827,
      "learning_rate": 2e-05,
      "loss": 0.6017,
      "step": 2700
    },
    {
      "epoch": 0.3751384274640089,
      "grad_norm": 0.5035448670387268,
      "learning_rate": 2e-05,
      "loss": 0.5644,
      "step": 2710
    },
    {
      "epoch": 0.37652270210409744,
      "grad_norm": 0.4859085977077484,
      "learning_rate": 2e-05,
      "loss": 0.5872,
      "step": 2720
    },
    {
      "epoch": 0.37790697674418605,
      "grad_norm": 0.43833255767822266,
      "learning_rate": 2e-05,
      "loss": 0.5877,
      "step": 2730
    },
    {
      "epoch": 0.37929125138427466,
      "grad_norm": 0.4499039351940155,
      "learning_rate": 2e-05,
      "loss": 0.5856,
      "step": 2740
    },
    {
      "epoch": 0.3806755260243632,
      "grad_norm": 0.4616643488407135,
      "learning_rate": 2e-05,
      "loss": 0.5649,
      "step": 2750
    },
    {
      "epoch": 0.38205980066445183,
      "grad_norm": 0.4653758406639099,
      "learning_rate": 2e-05,
      "loss": 0.5971,
      "step": 2760
    },
    {
      "epoch": 0.38344407530454044,
      "grad_norm": 0.45399942994117737,
      "learning_rate": 2e-05,
      "loss": 0.5889,
      "step": 2770
    },
    {
      "epoch": 0.384828349944629,
      "grad_norm": 0.4801301658153534,
      "learning_rate": 2e-05,
      "loss": 0.5678,
      "step": 2780
    },
    {
      "epoch": 0.3862126245847176,
      "grad_norm": 0.4705391824245453,
      "learning_rate": 2e-05,
      "loss": 0.5811,
      "step": 2790
    },
    {
      "epoch": 0.3875968992248062,
      "grad_norm": 0.4649393856525421,
      "learning_rate": 2e-05,
      "loss": 0.5975,
      "step": 2800
    },
    {
      "epoch": 0.3889811738648948,
      "grad_norm": 0.46329256892204285,
      "learning_rate": 2e-05,
      "loss": 0.5754,
      "step": 2810
    },
    {
      "epoch": 0.3903654485049834,
      "grad_norm": 0.48445969820022583,
      "learning_rate": 2e-05,
      "loss": 0.578,
      "step": 2820
    },
    {
      "epoch": 0.391749723145072,
      "grad_norm": 0.5109854936599731,
      "learning_rate": 2e-05,
      "loss": 0.5662,
      "step": 2830
    },
    {
      "epoch": 0.39313399778516056,
      "grad_norm": 0.4359543025493622,
      "learning_rate": 2e-05,
      "loss": 0.5786,
      "step": 2840
    },
    {
      "epoch": 0.3945182724252492,
      "grad_norm": 0.46977749466896057,
      "learning_rate": 2e-05,
      "loss": 0.5938,
      "step": 2850
    },
    {
      "epoch": 0.3959025470653378,
      "grad_norm": 0.48194414377212524,
      "learning_rate": 2e-05,
      "loss": 0.5733,
      "step": 2860
    },
    {
      "epoch": 0.39728682170542634,
      "grad_norm": 0.49208399653434753,
      "learning_rate": 2e-05,
      "loss": 0.5886,
      "step": 2870
    },
    {
      "epoch": 0.39867109634551495,
      "grad_norm": 0.7230966687202454,
      "learning_rate": 2e-05,
      "loss": 0.5742,
      "step": 2880
    },
    {
      "epoch": 0.40005537098560356,
      "grad_norm": 0.4486839771270752,
      "learning_rate": 2e-05,
      "loss": 0.5734,
      "step": 2890
    },
    {
      "epoch": 0.4014396456256921,
      "grad_norm": 0.4740370512008667,
      "learning_rate": 2e-05,
      "loss": 0.5665,
      "step": 2900
    },
    {
      "epoch": 0.40282392026578073,
      "grad_norm": 0.4609792232513428,
      "learning_rate": 2e-05,
      "loss": 0.5838,
      "step": 2910
    },
    {
      "epoch": 0.40420819490586934,
      "grad_norm": 0.48856472969055176,
      "learning_rate": 2e-05,
      "loss": 0.5599,
      "step": 2920
    },
    {
      "epoch": 0.4055924695459579,
      "grad_norm": 0.6106773018836975,
      "learning_rate": 2e-05,
      "loss": 0.5784,
      "step": 2930
    },
    {
      "epoch": 0.4069767441860465,
      "grad_norm": 0.5273647904396057,
      "learning_rate": 2e-05,
      "loss": 0.5839,
      "step": 2940
    },
    {
      "epoch": 0.4083610188261351,
      "grad_norm": 0.46062320470809937,
      "learning_rate": 2e-05,
      "loss": 0.5701,
      "step": 2950
    },
    {
      "epoch": 0.4097452934662237,
      "grad_norm": 0.49606624245643616,
      "learning_rate": 2e-05,
      "loss": 0.5859,
      "step": 2960
    },
    {
      "epoch": 0.4111295681063123,
      "grad_norm": 0.4899214506149292,
      "learning_rate": 2e-05,
      "loss": 0.5774,
      "step": 2970
    },
    {
      "epoch": 0.4125138427464009,
      "grad_norm": 0.498847633600235,
      "learning_rate": 2e-05,
      "loss": 0.554,
      "step": 2980
    },
    {
      "epoch": 0.41389811738648946,
      "grad_norm": 0.5811676979064941,
      "learning_rate": 2e-05,
      "loss": 0.5845,
      "step": 2990
    },
    {
      "epoch": 0.4152823920265781,
      "grad_norm": 0.45967867970466614,
      "learning_rate": 2e-05,
      "loss": 0.5739,
      "step": 3000
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 0.4345003366470337,
      "learning_rate": 2e-05,
      "loss": 0.5604,
      "step": 3010
    },
    {
      "epoch": 0.41805094130675524,
      "grad_norm": 0.49184611439704895,
      "learning_rate": 2e-05,
      "loss": 0.5717,
      "step": 3020
    },
    {
      "epoch": 0.41943521594684385,
      "grad_norm": 0.44787752628326416,
      "learning_rate": 2e-05,
      "loss": 0.5556,
      "step": 3030
    },
    {
      "epoch": 0.42081949058693247,
      "grad_norm": 0.48062196373939514,
      "learning_rate": 2e-05,
      "loss": 0.5727,
      "step": 3040
    },
    {
      "epoch": 0.422203765227021,
      "grad_norm": 0.4502577483654022,
      "learning_rate": 2e-05,
      "loss": 0.5738,
      "step": 3050
    },
    {
      "epoch": 0.42358803986710963,
      "grad_norm": 0.5000416040420532,
      "learning_rate": 2e-05,
      "loss": 0.5646,
      "step": 3060
    },
    {
      "epoch": 0.42497231450719825,
      "grad_norm": 0.4944615662097931,
      "learning_rate": 2e-05,
      "loss": 0.5644,
      "step": 3070
    },
    {
      "epoch": 0.4263565891472868,
      "grad_norm": 0.4559347927570343,
      "learning_rate": 2e-05,
      "loss": 0.5744,
      "step": 3080
    },
    {
      "epoch": 0.4277408637873754,
      "grad_norm": 0.4439176619052887,
      "learning_rate": 2e-05,
      "loss": 0.5894,
      "step": 3090
    },
    {
      "epoch": 0.429125138427464,
      "grad_norm": 0.463576078414917,
      "learning_rate": 2e-05,
      "loss": 0.5674,
      "step": 3100
    },
    {
      "epoch": 0.4305094130675526,
      "grad_norm": 0.5356590747833252,
      "learning_rate": 2e-05,
      "loss": 0.5786,
      "step": 3110
    },
    {
      "epoch": 0.4318936877076412,
      "grad_norm": 0.5175563097000122,
      "learning_rate": 2e-05,
      "loss": 0.5535,
      "step": 3120
    },
    {
      "epoch": 0.4332779623477298,
      "grad_norm": 0.48170632123947144,
      "learning_rate": 2e-05,
      "loss": 0.5534,
      "step": 3130
    },
    {
      "epoch": 0.43466223698781836,
      "grad_norm": 0.5080723762512207,
      "learning_rate": 2e-05,
      "loss": 0.5581,
      "step": 3140
    },
    {
      "epoch": 0.436046511627907,
      "grad_norm": 0.6235473155975342,
      "learning_rate": 2e-05,
      "loss": 0.571,
      "step": 3150
    },
    {
      "epoch": 0.4374307862679956,
      "grad_norm": 0.4751434922218323,
      "learning_rate": 2e-05,
      "loss": 0.545,
      "step": 3160
    },
    {
      "epoch": 0.43881506090808414,
      "grad_norm": 0.4574141502380371,
      "learning_rate": 2e-05,
      "loss": 0.5701,
      "step": 3170
    },
    {
      "epoch": 0.44019933554817275,
      "grad_norm": 0.46059706807136536,
      "learning_rate": 2e-05,
      "loss": 0.5418,
      "step": 3180
    },
    {
      "epoch": 0.44158361018826137,
      "grad_norm": 0.4763917326927185,
      "learning_rate": 2e-05,
      "loss": 0.5756,
      "step": 3190
    },
    {
      "epoch": 0.4429678848283499,
      "grad_norm": 0.45666614174842834,
      "learning_rate": 2e-05,
      "loss": 0.5803,
      "step": 3200
    },
    {
      "epoch": 0.44435215946843853,
      "grad_norm": 0.5104175806045532,
      "learning_rate": 2e-05,
      "loss": 0.5612,
      "step": 3210
    },
    {
      "epoch": 0.44573643410852715,
      "grad_norm": 0.49704769253730774,
      "learning_rate": 2e-05,
      "loss": 0.5668,
      "step": 3220
    },
    {
      "epoch": 0.4471207087486157,
      "grad_norm": 0.4857429563999176,
      "learning_rate": 2e-05,
      "loss": 0.5731,
      "step": 3230
    },
    {
      "epoch": 0.4485049833887043,
      "grad_norm": 0.4693543314933777,
      "learning_rate": 2e-05,
      "loss": 0.5589,
      "step": 3240
    },
    {
      "epoch": 0.4498892580287929,
      "grad_norm": 0.473862886428833,
      "learning_rate": 2e-05,
      "loss": 0.5522,
      "step": 3250
    },
    {
      "epoch": 0.4512735326688815,
      "grad_norm": 0.4944894313812256,
      "learning_rate": 2e-05,
      "loss": 0.554,
      "step": 3260
    },
    {
      "epoch": 0.4526578073089701,
      "grad_norm": 0.47641077637672424,
      "learning_rate": 2e-05,
      "loss": 0.5618,
      "step": 3270
    },
    {
      "epoch": 0.4540420819490587,
      "grad_norm": 0.4841245412826538,
      "learning_rate": 2e-05,
      "loss": 0.5487,
      "step": 3280
    },
    {
      "epoch": 0.45542635658914726,
      "grad_norm": 0.4585040211677551,
      "learning_rate": 2e-05,
      "loss": 0.5577,
      "step": 3290
    },
    {
      "epoch": 0.4568106312292359,
      "grad_norm": 0.4938330352306366,
      "learning_rate": 2e-05,
      "loss": 0.5539,
      "step": 3300
    },
    {
      "epoch": 0.4581949058693245,
      "grad_norm": 0.49708372354507446,
      "learning_rate": 2e-05,
      "loss": 0.5715,
      "step": 3310
    },
    {
      "epoch": 0.45957918050941304,
      "grad_norm": 0.5134687423706055,
      "learning_rate": 2e-05,
      "loss": 0.5478,
      "step": 3320
    },
    {
      "epoch": 0.46096345514950166,
      "grad_norm": 0.4389688968658447,
      "learning_rate": 2e-05,
      "loss": 0.5509,
      "step": 3330
    },
    {
      "epoch": 0.46234772978959027,
      "grad_norm": 0.4843428134918213,
      "learning_rate": 2e-05,
      "loss": 0.5488,
      "step": 3340
    },
    {
      "epoch": 0.4637320044296788,
      "grad_norm": 0.4451432526111603,
      "learning_rate": 2e-05,
      "loss": 0.554,
      "step": 3350
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 0.47852882742881775,
      "learning_rate": 2e-05,
      "loss": 0.5773,
      "step": 3360
    },
    {
      "epoch": 0.46650055370985605,
      "grad_norm": 0.4663555324077606,
      "learning_rate": 2e-05,
      "loss": 0.5501,
      "step": 3370
    },
    {
      "epoch": 0.4678848283499446,
      "grad_norm": 0.49802884459495544,
      "learning_rate": 2e-05,
      "loss": 0.5659,
      "step": 3380
    },
    {
      "epoch": 0.4692691029900332,
      "grad_norm": 0.47362440824508667,
      "learning_rate": 2e-05,
      "loss": 0.5772,
      "step": 3390
    },
    {
      "epoch": 0.4706533776301218,
      "grad_norm": 0.45582494139671326,
      "learning_rate": 2e-05,
      "loss": 0.5482,
      "step": 3400
    },
    {
      "epoch": 0.4720376522702104,
      "grad_norm": 0.45218512415885925,
      "learning_rate": 2e-05,
      "loss": 0.5582,
      "step": 3410
    },
    {
      "epoch": 0.473421926910299,
      "grad_norm": 0.4743266701698303,
      "learning_rate": 2e-05,
      "loss": 0.5598,
      "step": 3420
    },
    {
      "epoch": 0.4748062015503876,
      "grad_norm": 0.4403548538684845,
      "learning_rate": 2e-05,
      "loss": 0.5621,
      "step": 3430
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.48124459385871887,
      "learning_rate": 2e-05,
      "loss": 0.5554,
      "step": 3440
    },
    {
      "epoch": 0.4775747508305648,
      "grad_norm": 0.48941442370414734,
      "learning_rate": 2e-05,
      "loss": 0.5551,
      "step": 3450
    },
    {
      "epoch": 0.4789590254706534,
      "grad_norm": 0.4785088300704956,
      "learning_rate": 2e-05,
      "loss": 0.5599,
      "step": 3460
    },
    {
      "epoch": 0.48034330011074194,
      "grad_norm": 0.4747934639453888,
      "learning_rate": 2e-05,
      "loss": 0.5538,
      "step": 3470
    },
    {
      "epoch": 0.48172757475083056,
      "grad_norm": 0.44214168190956116,
      "learning_rate": 2e-05,
      "loss": 0.5503,
      "step": 3480
    },
    {
      "epoch": 0.48311184939091917,
      "grad_norm": 0.48559412360191345,
      "learning_rate": 2e-05,
      "loss": 0.5299,
      "step": 3490
    },
    {
      "epoch": 0.4844961240310077,
      "grad_norm": 0.5481932163238525,
      "learning_rate": 2e-05,
      "loss": 0.5479,
      "step": 3500
    },
    {
      "epoch": 0.48588039867109634,
      "grad_norm": 0.44426339864730835,
      "learning_rate": 2e-05,
      "loss": 0.5447,
      "step": 3510
    },
    {
      "epoch": 0.48726467331118495,
      "grad_norm": 0.4596952199935913,
      "learning_rate": 2e-05,
      "loss": 0.5496,
      "step": 3520
    },
    {
      "epoch": 0.48864894795127356,
      "grad_norm": 0.504192590713501,
      "learning_rate": 2e-05,
      "loss": 0.5704,
      "step": 3530
    },
    {
      "epoch": 0.4900332225913621,
      "grad_norm": 0.5149878263473511,
      "learning_rate": 2e-05,
      "loss": 0.5742,
      "step": 3540
    },
    {
      "epoch": 0.49141749723145073,
      "grad_norm": 0.4706507921218872,
      "learning_rate": 2e-05,
      "loss": 0.5638,
      "step": 3550
    },
    {
      "epoch": 0.49280177187153934,
      "grad_norm": 0.4283595383167267,
      "learning_rate": 2e-05,
      "loss": 0.5222,
      "step": 3560
    },
    {
      "epoch": 0.4941860465116279,
      "grad_norm": 0.46484580636024475,
      "learning_rate": 2e-05,
      "loss": 0.5392,
      "step": 3570
    },
    {
      "epoch": 0.4955703211517165,
      "grad_norm": 0.4691891670227051,
      "learning_rate": 2e-05,
      "loss": 0.5667,
      "step": 3580
    },
    {
      "epoch": 0.4969545957918051,
      "grad_norm": 0.5468685030937195,
      "learning_rate": 2e-05,
      "loss": 0.5426,
      "step": 3590
    },
    {
      "epoch": 0.4983388704318937,
      "grad_norm": 0.4608685374259949,
      "learning_rate": 2e-05,
      "loss": 0.5453,
      "step": 3600
    },
    {
      "epoch": 0.4997231450719823,
      "grad_norm": 0.5088940262794495,
      "learning_rate": 2e-05,
      "loss": 0.5572,
      "step": 3610
    },
    {
      "epoch": 0.5011074197120708,
      "grad_norm": 0.45228078961372375,
      "learning_rate": 2e-05,
      "loss": 0.5465,
      "step": 3620
    },
    {
      "epoch": 0.5024916943521595,
      "grad_norm": 0.46081870794296265,
      "learning_rate": 2e-05,
      "loss": 0.5436,
      "step": 3630
    },
    {
      "epoch": 0.5038759689922481,
      "grad_norm": 0.460740327835083,
      "learning_rate": 2e-05,
      "loss": 0.5638,
      "step": 3640
    },
    {
      "epoch": 0.5052602436323367,
      "grad_norm": 0.5299831628799438,
      "learning_rate": 2e-05,
      "loss": 0.5479,
      "step": 3650
    },
    {
      "epoch": 0.5066445182724253,
      "grad_norm": 1.7227613925933838,
      "learning_rate": 2e-05,
      "loss": 0.5542,
      "step": 3660
    },
    {
      "epoch": 0.5080287929125138,
      "grad_norm": 0.5014874339103699,
      "learning_rate": 2e-05,
      "loss": 0.5428,
      "step": 3670
    },
    {
      "epoch": 0.5094130675526024,
      "grad_norm": 0.5587078928947449,
      "learning_rate": 2e-05,
      "loss": 0.5433,
      "step": 3680
    },
    {
      "epoch": 0.510797342192691,
      "grad_norm": 0.47746700048446655,
      "learning_rate": 2e-05,
      "loss": 0.544,
      "step": 3690
    },
    {
      "epoch": 0.5121816168327796,
      "grad_norm": 0.5514836311340332,
      "learning_rate": 2e-05,
      "loss": 0.5336,
      "step": 3700
    },
    {
      "epoch": 0.5135658914728682,
      "grad_norm": 0.4769381582736969,
      "learning_rate": 2e-05,
      "loss": 0.5469,
      "step": 3710
    },
    {
      "epoch": 0.5149501661129569,
      "grad_norm": 0.480053573846817,
      "learning_rate": 2e-05,
      "loss": 0.5343,
      "step": 3720
    },
    {
      "epoch": 0.5163344407530454,
      "grad_norm": 0.4887942969799042,
      "learning_rate": 2e-05,
      "loss": 0.5491,
      "step": 3730
    },
    {
      "epoch": 0.517718715393134,
      "grad_norm": 0.4910425543785095,
      "learning_rate": 2e-05,
      "loss": 0.5277,
      "step": 3740
    },
    {
      "epoch": 0.5191029900332226,
      "grad_norm": 0.4797256886959076,
      "learning_rate": 2e-05,
      "loss": 0.5396,
      "step": 3750
    },
    {
      "epoch": 0.5204872646733112,
      "grad_norm": 0.5521448850631714,
      "learning_rate": 2e-05,
      "loss": 0.5398,
      "step": 3760
    },
    {
      "epoch": 0.5218715393133998,
      "grad_norm": 0.5706764459609985,
      "learning_rate": 2e-05,
      "loss": 0.554,
      "step": 3770
    },
    {
      "epoch": 0.5232558139534884,
      "grad_norm": 0.5074754953384399,
      "learning_rate": 2e-05,
      "loss": 0.5439,
      "step": 3780
    },
    {
      "epoch": 0.5246400885935769,
      "grad_norm": 0.4842338263988495,
      "learning_rate": 2e-05,
      "loss": 0.5261,
      "step": 3790
    },
    {
      "epoch": 0.5260243632336655,
      "grad_norm": 0.46847695112228394,
      "learning_rate": 2e-05,
      "loss": 0.5464,
      "step": 3800
    },
    {
      "epoch": 0.5274086378737541,
      "grad_norm": 0.46739593148231506,
      "learning_rate": 2e-05,
      "loss": 0.5733,
      "step": 3810
    },
    {
      "epoch": 0.5287929125138427,
      "grad_norm": 0.5192055106163025,
      "learning_rate": 2e-05,
      "loss": 0.5424,
      "step": 3820
    },
    {
      "epoch": 0.5301771871539314,
      "grad_norm": 0.5271517634391785,
      "learning_rate": 2e-05,
      "loss": 0.5277,
      "step": 3830
    },
    {
      "epoch": 0.53156146179402,
      "grad_norm": 0.5242041945457458,
      "learning_rate": 2e-05,
      "loss": 0.525,
      "step": 3840
    },
    {
      "epoch": 0.5329457364341085,
      "grad_norm": 0.518164873123169,
      "learning_rate": 2e-05,
      "loss": 0.5182,
      "step": 3850
    },
    {
      "epoch": 0.5343300110741971,
      "grad_norm": 0.44636064767837524,
      "learning_rate": 2e-05,
      "loss": 0.5368,
      "step": 3860
    },
    {
      "epoch": 0.5357142857142857,
      "grad_norm": 0.48869842290878296,
      "learning_rate": 2e-05,
      "loss": 0.5519,
      "step": 3870
    },
    {
      "epoch": 0.5370985603543743,
      "grad_norm": 0.6531343460083008,
      "learning_rate": 2e-05,
      "loss": 0.5273,
      "step": 3880
    },
    {
      "epoch": 0.5384828349944629,
      "grad_norm": 0.47449296712875366,
      "learning_rate": 2e-05,
      "loss": 0.5304,
      "step": 3890
    },
    {
      "epoch": 0.5398671096345515,
      "grad_norm": 0.46483784914016724,
      "learning_rate": 2e-05,
      "loss": 0.5309,
      "step": 3900
    },
    {
      "epoch": 0.54125138427464,
      "grad_norm": 0.47348055243492126,
      "learning_rate": 2e-05,
      "loss": 0.5316,
      "step": 3910
    },
    {
      "epoch": 0.5426356589147286,
      "grad_norm": 0.5768376588821411,
      "learning_rate": 2e-05,
      "loss": 0.5129,
      "step": 3920
    },
    {
      "epoch": 0.5440199335548173,
      "grad_norm": 0.466141015291214,
      "learning_rate": 2e-05,
      "loss": 0.527,
      "step": 3930
    },
    {
      "epoch": 0.5454042081949059,
      "grad_norm": 0.48306912183761597,
      "learning_rate": 2e-05,
      "loss": 0.5402,
      "step": 3940
    },
    {
      "epoch": 0.5467884828349945,
      "grad_norm": 0.6273697018623352,
      "learning_rate": 2e-05,
      "loss": 0.5341,
      "step": 3950
    },
    {
      "epoch": 0.5481727574750831,
      "grad_norm": 0.47792720794677734,
      "learning_rate": 2e-05,
      "loss": 0.5425,
      "step": 3960
    },
    {
      "epoch": 0.5495570321151716,
      "grad_norm": 0.4689851999282837,
      "learning_rate": 2e-05,
      "loss": 0.5345,
      "step": 3970
    },
    {
      "epoch": 0.5509413067552602,
      "grad_norm": 0.49213728308677673,
      "learning_rate": 2e-05,
      "loss": 0.54,
      "step": 3980
    },
    {
      "epoch": 0.5523255813953488,
      "grad_norm": 0.5102055668830872,
      "learning_rate": 2e-05,
      "loss": 0.5454,
      "step": 3990
    },
    {
      "epoch": 0.5537098560354374,
      "grad_norm": 0.5814585089683533,
      "learning_rate": 2e-05,
      "loss": 0.5395,
      "step": 4000
    },
    {
      "epoch": 0.555094130675526,
      "grad_norm": 0.4902860224246979,
      "learning_rate": 2e-05,
      "loss": 0.5468,
      "step": 4010
    },
    {
      "epoch": 0.5564784053156147,
      "grad_norm": 0.5666091442108154,
      "learning_rate": 2e-05,
      "loss": 0.5216,
      "step": 4020
    },
    {
      "epoch": 0.5578626799557033,
      "grad_norm": 0.5402845144271851,
      "learning_rate": 2e-05,
      "loss": 0.5096,
      "step": 4030
    },
    {
      "epoch": 0.5592469545957918,
      "grad_norm": 0.45804664492607117,
      "learning_rate": 2e-05,
      "loss": 0.5188,
      "step": 4040
    },
    {
      "epoch": 0.5606312292358804,
      "grad_norm": 0.47594836354255676,
      "learning_rate": 2e-05,
      "loss": 0.5336,
      "step": 4050
    },
    {
      "epoch": 0.562015503875969,
      "grad_norm": 0.4569697678089142,
      "learning_rate": 2e-05,
      "loss": 0.5192,
      "step": 4060
    },
    {
      "epoch": 0.5633997785160576,
      "grad_norm": 0.47449183464050293,
      "learning_rate": 2e-05,
      "loss": 0.5505,
      "step": 4070
    },
    {
      "epoch": 0.5647840531561462,
      "grad_norm": 0.5270619988441467,
      "learning_rate": 2e-05,
      "loss": 0.5158,
      "step": 4080
    },
    {
      "epoch": 0.5661683277962348,
      "grad_norm": 0.5315555930137634,
      "learning_rate": 2e-05,
      "loss": 0.5233,
      "step": 4090
    },
    {
      "epoch": 0.5675526024363233,
      "grad_norm": 0.5249839425086975,
      "learning_rate": 2e-05,
      "loss": 0.5369,
      "step": 4100
    },
    {
      "epoch": 0.5689368770764119,
      "grad_norm": 0.48996680974960327,
      "learning_rate": 2e-05,
      "loss": 0.5066,
      "step": 4110
    },
    {
      "epoch": 0.5703211517165006,
      "grad_norm": 0.5099858045578003,
      "learning_rate": 2e-05,
      "loss": 0.5351,
      "step": 4120
    },
    {
      "epoch": 0.5717054263565892,
      "grad_norm": 0.4698827862739563,
      "learning_rate": 2e-05,
      "loss": 0.5206,
      "step": 4130
    },
    {
      "epoch": 0.5730897009966778,
      "grad_norm": 0.486258327960968,
      "learning_rate": 2e-05,
      "loss": 0.5271,
      "step": 4140
    },
    {
      "epoch": 0.5744739756367664,
      "grad_norm": 0.5019756555557251,
      "learning_rate": 2e-05,
      "loss": 0.5143,
      "step": 4150
    },
    {
      "epoch": 0.5758582502768549,
      "grad_norm": 0.5071268081665039,
      "learning_rate": 2e-05,
      "loss": 0.5161,
      "step": 4160
    },
    {
      "epoch": 0.5772425249169435,
      "grad_norm": 0.4707529842853546,
      "learning_rate": 2e-05,
      "loss": 0.5201,
      "step": 4170
    },
    {
      "epoch": 0.5786267995570321,
      "grad_norm": 0.5228965878486633,
      "learning_rate": 2e-05,
      "loss": 0.5157,
      "step": 4180
    },
    {
      "epoch": 0.5800110741971207,
      "grad_norm": 0.4988875091075897,
      "learning_rate": 2e-05,
      "loss": 0.5234,
      "step": 4190
    },
    {
      "epoch": 0.5813953488372093,
      "grad_norm": 0.4811486005783081,
      "learning_rate": 2e-05,
      "loss": 0.5456,
      "step": 4200
    },
    {
      "epoch": 0.582779623477298,
      "grad_norm": 0.5186240077018738,
      "learning_rate": 2e-05,
      "loss": 0.5412,
      "step": 4210
    },
    {
      "epoch": 0.5841638981173864,
      "grad_norm": 0.48254865407943726,
      "learning_rate": 2e-05,
      "loss": 0.5504,
      "step": 4220
    },
    {
      "epoch": 0.5855481727574751,
      "grad_norm": 0.4839778542518616,
      "learning_rate": 2e-05,
      "loss": 0.5087,
      "step": 4230
    },
    {
      "epoch": 0.5869324473975637,
      "grad_norm": 0.5022257566452026,
      "learning_rate": 2e-05,
      "loss": 0.5333,
      "step": 4240
    },
    {
      "epoch": 0.5883167220376523,
      "grad_norm": 0.5056033134460449,
      "learning_rate": 2e-05,
      "loss": 0.5305,
      "step": 4250
    },
    {
      "epoch": 0.5897009966777409,
      "grad_norm": 0.46993741393089294,
      "learning_rate": 2e-05,
      "loss": 0.5039,
      "step": 4260
    },
    {
      "epoch": 0.5910852713178295,
      "grad_norm": 0.5203388929367065,
      "learning_rate": 2e-05,
      "loss": 0.5347,
      "step": 4270
    },
    {
      "epoch": 0.592469545957918,
      "grad_norm": 0.5526265501976013,
      "learning_rate": 2e-05,
      "loss": 0.5099,
      "step": 4280
    },
    {
      "epoch": 0.5938538205980066,
      "grad_norm": 0.48970651626586914,
      "learning_rate": 2e-05,
      "loss": 0.5291,
      "step": 4290
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 0.5051175951957703,
      "learning_rate": 2e-05,
      "loss": 0.5173,
      "step": 4300
    },
    {
      "epoch": 0.5966223698781838,
      "grad_norm": 0.5191178321838379,
      "learning_rate": 2e-05,
      "loss": 0.5164,
      "step": 4310
    },
    {
      "epoch": 0.5980066445182725,
      "grad_norm": 0.49129706621170044,
      "learning_rate": 2e-05,
      "loss": 0.5131,
      "step": 4320
    },
    {
      "epoch": 0.5993909191583611,
      "grad_norm": 0.519318699836731,
      "learning_rate": 2e-05,
      "loss": 0.5146,
      "step": 4330
    },
    {
      "epoch": 0.6007751937984496,
      "grad_norm": 0.48014533519744873,
      "learning_rate": 2e-05,
      "loss": 0.5283,
      "step": 4340
    },
    {
      "epoch": 0.6021594684385382,
      "grad_norm": 0.4873373806476593,
      "learning_rate": 2e-05,
      "loss": 0.5206,
      "step": 4350
    },
    {
      "epoch": 0.6035437430786268,
      "grad_norm": 0.5058743953704834,
      "learning_rate": 2e-05,
      "loss": 0.5289,
      "step": 4360
    },
    {
      "epoch": 0.6049280177187154,
      "grad_norm": 0.50196373462677,
      "learning_rate": 2e-05,
      "loss": 0.5294,
      "step": 4370
    },
    {
      "epoch": 0.606312292358804,
      "grad_norm": 0.49518486857414246,
      "learning_rate": 2e-05,
      "loss": 0.5194,
      "step": 4380
    },
    {
      "epoch": 0.6076965669988926,
      "grad_norm": 0.5034221410751343,
      "learning_rate": 1.9984835874829866e-05,
      "loss": 0.5118,
      "step": 4390
    },
    {
      "epoch": 0.6090808416389811,
      "grad_norm": 0.4924371540546417,
      "learning_rate": 1.993248227822551e-05,
      "loss": 0.504,
      "step": 4400
    },
    {
      "epoch": 0.6104651162790697,
      "grad_norm": 0.48116081953048706,
      "learning_rate": 1.984296973671751e-05,
      "loss": 0.5162,
      "step": 4410
    },
    {
      "epoch": 0.6118493909191584,
      "grad_norm": 0.5034663677215576,
      "learning_rate": 1.9716670620660985e-05,
      "loss": 0.5373,
      "step": 4420
    },
    {
      "epoch": 0.613233665559247,
      "grad_norm": 0.797784149646759,
      "learning_rate": 1.9554110331814234e-05,
      "loss": 0.5326,
      "step": 4430
    },
    {
      "epoch": 0.6146179401993356,
      "grad_norm": 0.4884178042411804,
      "learning_rate": 1.9355965117678084e-05,
      "loss": 0.5052,
      "step": 4440
    },
    {
      "epoch": 0.6160022148394242,
      "grad_norm": 0.48519402742385864,
      "learning_rate": 1.912305925832e-05,
      "loss": 0.5281,
      "step": 4450
    },
    {
      "epoch": 0.6173864894795127,
      "grad_norm": 0.5348500609397888,
      "learning_rate": 1.885636163738577e-05,
      "loss": 0.5193,
      "step": 4460
    },
    {
      "epoch": 0.6187707641196013,
      "grad_norm": 0.48519015312194824,
      "learning_rate": 1.8556981711563194e-05,
      "loss": 0.4953,
      "step": 4470
    },
    {
      "epoch": 0.6201550387596899,
      "grad_norm": 0.5260573625564575,
      "learning_rate": 1.822616489526479e-05,
      "loss": 0.5219,
      "step": 4480
    },
    {
      "epoch": 0.6215393133997785,
      "grad_norm": 0.5469380021095276,
      "learning_rate": 1.7865287379729137e-05,
      "loss": 0.509,
      "step": 4490
    },
    {
      "epoch": 0.6229235880398671,
      "grad_norm": 0.509768545627594,
      "learning_rate": 1.7475850408093257e-05,
      "loss": 0.5182,
      "step": 4500
    },
    {
      "epoch": 0.6243078626799557,
      "grad_norm": 0.48531392216682434,
      "learning_rate": 1.705947403025165e-05,
      "loss": 0.5041,
      "step": 4510
    },
    {
      "epoch": 0.6256921373200443,
      "grad_norm": 0.47997668385505676,
      "learning_rate": 1.6617890363481623e-05,
      "loss": 0.5265,
      "step": 4520
    },
    {
      "epoch": 0.6270764119601329,
      "grad_norm": 0.5013784170150757,
      "learning_rate": 1.6152936386870604e-05,
      "loss": 0.5109,
      "step": 4530
    },
    {
      "epoch": 0.6284606866002215,
      "grad_norm": 0.5075677037239075,
      "learning_rate": 1.5666546299520414e-05,
      "loss": 0.5175,
      "step": 4540
    },
    {
      "epoch": 0.6298449612403101,
      "grad_norm": 0.516105592250824,
      "learning_rate": 1.5160743474318298e-05,
      "loss": 0.5111,
      "step": 4550
    },
    {
      "epoch": 0.6312292358803987,
      "grad_norm": 0.516226589679718,
      "learning_rate": 1.463763204074678e-05,
      "loss": 0.495,
      "step": 4560
    },
    {
      "epoch": 0.6326135105204873,
      "grad_norm": 0.5707921385765076,
      "learning_rate": 1.4099388131747764e-05,
      "loss": 0.4963,
      "step": 4570
    },
    {
      "epoch": 0.6339977851605758,
      "grad_norm": 0.5220585465431213,
      "learning_rate": 1.3548250831053656e-05,
      "loss": 0.493,
      "step": 4580
    },
    {
      "epoch": 0.6353820598006644,
      "grad_norm": 0.5072370171546936,
      "learning_rate": 1.2986512858644518e-05,
      "loss": 0.5074,
      "step": 4590
    },
    {
      "epoch": 0.636766334440753,
      "grad_norm": 0.5144025087356567,
      "learning_rate": 1.2416511033079498e-05,
      "loss": 0.5034,
      "step": 4600
    },
    {
      "epoch": 0.6381506090808416,
      "grad_norm": 0.46612557768821716,
      "learning_rate": 1.1840616550379082e-05,
      "loss": 0.5094,
      "step": 4610
    },
    {
      "epoch": 0.6395348837209303,
      "grad_norm": 0.47627463936805725,
      "learning_rate": 1.1261225119897781e-05,
      "loss": 0.4917,
      "step": 4620
    },
    {
      "epoch": 0.6409191583610189,
      "grad_norm": 0.5091423392295837,
      "learning_rate": 1.068074699822187e-05,
      "loss": 0.5035,
      "step": 4630
    },
    {
      "epoch": 0.6423034330011074,
      "grad_norm": 0.4707886874675751,
      "learning_rate": 1.0101596962550937e-05,
      "loss": 0.507,
      "step": 4640
    },
    {
      "epoch": 0.643687707641196,
      "grad_norm": 0.5316179990768433,
      "learning_rate": 9.526184265273748e-06,
      "loss": 0.4922,
      "step": 4650
    },
    {
      "epoch": 0.6450719822812846,
      "grad_norm": 0.5667310357093811,
      "learning_rate": 8.956902611527237e-06,
      "loss": 0.4904,
      "step": 4660
    },
    {
      "epoch": 0.6464562569213732,
      "grad_norm": 0.5211129188537598,
      "learning_rate": 8.396120201431742e-06,
      "loss": 0.5113,
      "step": 4670
    },
    {
      "epoch": 0.6478405315614618,
      "grad_norm": 0.4864901900291443,
      "learning_rate": 7.846169878426593e-06,
      "loss": 0.5055,
      "step": 4680
    },
    {
      "epoch": 0.6492248062015504,
      "grad_norm": 0.4923262894153595,
      "learning_rate": 7.3093394246887686e-06,
      "loss": 0.4912,
      "step": 4690
    },
    {
      "epoch": 0.6506090808416389,
      "grad_norm": 0.5287879705429077,
      "learning_rate": 6.787862044005491e-06,
      "loss": 0.5153,
      "step": 4700
    },
    {
      "epoch": 0.6519933554817275,
      "grad_norm": 0.49825167655944824,
      "learning_rate": 6.2839070716918235e-06,
      "loss": 0.4906,
      "step": 4710
    },
    {
      "epoch": 0.6533776301218162,
      "grad_norm": 0.5226734280586243,
      "learning_rate": 5.799570950199793e-06,
      "loss": 0.4811,
      "step": 4720
    },
    {
      "epoch": 0.6547619047619048,
      "grad_norm": 0.5251850485801697,
      "learning_rate": 5.336868507960386e-06,
      "loss": 0.5159,
      "step": 4730
    },
    {
      "epoch": 0.6561461794019934,
      "grad_norm": 0.5130586624145508,
      "learning_rate": 4.897724577738157e-06,
      "loss": 0.5185,
      "step": 4740
    },
    {
      "epoch": 0.657530454042082,
      "grad_norm": 0.5192838311195374,
      "learning_rate": 4.483965989366025e-06,
      "loss": 0.4829,
      "step": 4750
    },
    {
      "epoch": 0.6589147286821705,
      "grad_norm": 0.49104416370391846,
      "learning_rate": 4.097313970170263e-06,
      "loss": 0.5119,
      "step": 4760
    },
    {
      "epoch": 0.6602990033222591,
      "grad_norm": 0.48275795578956604,
      "learning_rate": 3.73937698469977e-06,
      "loss": 0.4968,
      "step": 4770
    },
    {
      "epoch": 0.6616832779623477,
      "grad_norm": 0.5024860501289368,
      "learning_rate": 3.411644043546202e-06,
      "loss": 0.4932,
      "step": 4780
    },
    {
      "epoch": 0.6630675526024363,
      "grad_norm": 0.4804997146129608,
      "learning_rate": 3.1154785090901505e-06,
      "loss": 0.484,
      "step": 4790
    },
    {
      "epoch": 0.6644518272425249,
      "grad_norm": 0.48775431513786316,
      "learning_rate": 2.8521124239413213e-06,
      "loss": 0.4835,
      "step": 4800
    },
    {
      "epoch": 0.6658361018826136,
      "grad_norm": 0.5086537599563599,
      "learning_rate": 2.622641385666294e-06,
      "loss": 0.5068,
      "step": 4810
    },
    {
      "epoch": 0.667220376522702,
      "grad_norm": 0.5400722026824951,
      "learning_rate": 2.428019989124892e-06,
      "loss": 0.5151,
      "step": 4820
    },
    {
      "epoch": 0.6686046511627907,
      "grad_norm": 0.5105034708976746,
      "learning_rate": 2.269057855374944e-06,
      "loss": 0.4922,
      "step": 4830
    },
    {
      "epoch": 0.6699889258028793,
      "grad_norm": 0.4837234914302826,
      "learning_rate": 2.1464162636651277e-06,
      "loss": 0.4986,
      "step": 4840
    },
    {
      "epoch": 0.6713732004429679,
      "grad_norm": 0.4796830713748932,
      "learning_rate": 2.0606054005267244e-06,
      "loss": 0.4858,
      "step": 4850
    },
    {
      "epoch": 0.6727574750830565,
      "grad_norm": 0.4831264913082123,
      "learning_rate": 2.011982237408003e-06,
      "loss": 0.4885,
      "step": 4860
    }
  ],
  "logging_steps": 10,
  "max_steps": 4867,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2035,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1726419832950227e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}