{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.48084192439862544,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000859106529209622,
      "grad_norm": 10.069700241088867,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 2.3607,
      "step": 10
    },
    {
      "epoch": 0.001718213058419244,
      "grad_norm": 2.7327144145965576,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 2.1048,
      "step": 20
    },
    {
      "epoch": 0.002577319587628866,
      "grad_norm": 3.3271842002868652,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.7854,
      "step": 30
    },
    {
      "epoch": 0.003436426116838488,
      "grad_norm": 1.4632568359375,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.4144,
      "step": 40
    },
    {
      "epoch": 0.00429553264604811,
      "grad_norm": 0.96810382604599,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.2037,
      "step": 50
    },
    {
      "epoch": 0.005154639175257732,
      "grad_norm": 0.7168840169906616,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.0357,
      "step": 60
    },
    {
      "epoch": 0.006013745704467354,
      "grad_norm": 0.6445292830467224,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.9969,
      "step": 70
    },
    {
      "epoch": 0.006872852233676976,
      "grad_norm": 0.6907113194465637,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0041,
      "step": 80
    },
    {
      "epoch": 0.007731958762886598,
      "grad_norm": 0.6093459725379944,
      "learning_rate": 1.005e-05,
      "loss": 0.9553,
      "step": 90
    },
    {
      "epoch": 0.00859106529209622,
      "grad_norm": 0.6630317568778992,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.9573,
      "step": 100
    },
    {
      "epoch": 0.009450171821305841,
      "grad_norm": 0.6552422642707825,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.9811,
      "step": 110
    },
    {
      "epoch": 0.010309278350515464,
      "grad_norm": 0.673576831817627,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.9248,
      "step": 120
    },
    {
      "epoch": 0.011168384879725086,
      "grad_norm": 0.6397550106048584,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.9155,
      "step": 130
    },
    {
      "epoch": 0.012027491408934709,
      "grad_norm": 0.6486415863037109,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.9121,
      "step": 140
    },
    {
      "epoch": 0.01288659793814433,
      "grad_norm": 0.5658463835716248,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.9371,
      "step": 150
    },
    {
      "epoch": 0.013745704467353952,
      "grad_norm": 0.5926650762557983,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.8875,
      "step": 160
    },
    {
      "epoch": 0.014604810996563574,
      "grad_norm": 0.576126217842102,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.9006,
      "step": 170
    },
    {
      "epoch": 0.015463917525773196,
      "grad_norm": 0.5573123097419739,
      "learning_rate": 2e-05,
      "loss": 0.9152,
      "step": 180
    },
    {
      "epoch": 0.01632302405498282,
      "grad_norm": 0.5773940086364746,
      "learning_rate": 2e-05,
      "loss": 0.8955,
      "step": 190
    },
    {
      "epoch": 0.01718213058419244,
      "grad_norm": 0.5635377764701843,
      "learning_rate": 2e-05,
      "loss": 0.8897,
      "step": 200
    },
    {
      "epoch": 0.01804123711340206,
      "grad_norm": 0.5612330436706543,
      "learning_rate": 2e-05,
      "loss": 0.8959,
      "step": 210
    },
    {
      "epoch": 0.018900343642611683,
      "grad_norm": 0.6228774785995483,
      "learning_rate": 2e-05,
      "loss": 0.8906,
      "step": 220
    },
    {
      "epoch": 0.019759450171821305,
      "grad_norm": 0.5534242391586304,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 230
    },
    {
      "epoch": 0.020618556701030927,
      "grad_norm": 0.5617905855178833,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 240
    },
    {
      "epoch": 0.02147766323024055,
      "grad_norm": 0.5597807765007019,
      "learning_rate": 2e-05,
      "loss": 0.8846,
      "step": 250
    },
    {
      "epoch": 0.022336769759450172,
      "grad_norm": 0.5115612149238586,
      "learning_rate": 2e-05,
      "loss": 0.8671,
      "step": 260
    },
    {
      "epoch": 0.023195876288659795,
      "grad_norm": 0.5317944288253784,
      "learning_rate": 2e-05,
      "loss": 0.8915,
      "step": 270
    },
    {
      "epoch": 0.024054982817869417,
      "grad_norm": 0.5044589638710022,
      "learning_rate": 2e-05,
      "loss": 0.8656,
      "step": 280
    },
    {
      "epoch": 0.02491408934707904,
      "grad_norm": 0.5581314563751221,
      "learning_rate": 2e-05,
      "loss": 0.8764,
      "step": 290
    },
    {
      "epoch": 0.02577319587628866,
      "grad_norm": 0.5628443956375122,
      "learning_rate": 2e-05,
      "loss": 0.8941,
      "step": 300
    },
    {
      "epoch": 0.02663230240549828,
      "grad_norm": 0.560197114944458,
      "learning_rate": 2e-05,
      "loss": 0.8491,
      "step": 310
    },
    {
      "epoch": 0.027491408934707903,
      "grad_norm": 0.524337649345398,
      "learning_rate": 2e-05,
      "loss": 0.8857,
      "step": 320
    },
    {
      "epoch": 0.028350515463917526,
      "grad_norm": 0.683029055595398,
      "learning_rate": 2e-05,
      "loss": 0.8597,
      "step": 330
    },
    {
      "epoch": 0.029209621993127148,
      "grad_norm": 0.5435221195220947,
      "learning_rate": 2e-05,
      "loss": 0.8483,
      "step": 340
    },
    {
      "epoch": 0.03006872852233677,
      "grad_norm": 0.5640478134155273,
      "learning_rate": 2e-05,
      "loss": 0.8865,
      "step": 350
    },
    {
      "epoch": 0.030927835051546393,
      "grad_norm": 0.4870714545249939,
      "learning_rate": 2e-05,
      "loss": 0.8616,
      "step": 360
    },
    {
      "epoch": 0.03178694158075601,
      "grad_norm": 0.4941020607948303,
      "learning_rate": 2e-05,
      "loss": 0.8745,
      "step": 370
    },
    {
      "epoch": 0.03264604810996564,
      "grad_norm": 0.4704708755016327,
      "learning_rate": 2e-05,
      "loss": 0.8337,
      "step": 380
    },
    {
      "epoch": 0.03350515463917526,
      "grad_norm": 0.5146850347518921,
      "learning_rate": 2e-05,
      "loss": 0.8392,
      "step": 390
    },
    {
      "epoch": 0.03436426116838488,
      "grad_norm": 0.5625430345535278,
      "learning_rate": 2e-05,
      "loss": 0.8679,
      "step": 400
    },
    {
      "epoch": 0.0352233676975945,
      "grad_norm": 0.5005638003349304,
      "learning_rate": 2e-05,
      "loss": 0.8686,
      "step": 410
    },
    {
      "epoch": 0.03608247422680412,
      "grad_norm": 0.5048443078994751,
      "learning_rate": 2e-05,
      "loss": 0.8587,
      "step": 420
    },
    {
      "epoch": 0.036941580756013746,
      "grad_norm": 0.5067149996757507,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 430
    },
    {
      "epoch": 0.037800687285223365,
      "grad_norm": 0.5074286460876465,
      "learning_rate": 2e-05,
      "loss": 0.8589,
      "step": 440
    },
    {
      "epoch": 0.03865979381443299,
      "grad_norm": 0.4815230965614319,
      "learning_rate": 2e-05,
      "loss": 0.8735,
      "step": 450
    },
    {
      "epoch": 0.03951890034364261,
      "grad_norm": 0.4682285785675049,
      "learning_rate": 2e-05,
      "loss": 0.8988,
      "step": 460
    },
    {
      "epoch": 0.040378006872852236,
      "grad_norm": 0.4762243628501892,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 470
    },
    {
      "epoch": 0.041237113402061855,
      "grad_norm": 0.48740312457084656,
      "learning_rate": 2e-05,
      "loss": 0.819,
      "step": 480
    },
    {
      "epoch": 0.04209621993127148,
      "grad_norm": 0.4676920473575592,
      "learning_rate": 2e-05,
      "loss": 0.8169,
      "step": 490
    },
    {
      "epoch": 0.0429553264604811,
      "grad_norm": 0.4818402826786041,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 500
    },
    {
      "epoch": 0.04381443298969072,
      "grad_norm": 0.536251425743103,
      "learning_rate": 2e-05,
      "loss": 0.8857,
      "step": 510
    },
    {
      "epoch": 0.044673539518900345,
      "grad_norm": 0.46843159198760986,
      "learning_rate": 2e-05,
      "loss": 0.8162,
      "step": 520
    },
    {
      "epoch": 0.04553264604810996,
      "grad_norm": 0.47590792179107666,
      "learning_rate": 2e-05,
      "loss": 0.8749,
      "step": 530
    },
    {
      "epoch": 0.04639175257731959,
      "grad_norm": 0.48866006731987,
      "learning_rate": 2e-05,
      "loss": 0.8761,
      "step": 540
    },
    {
      "epoch": 0.04725085910652921,
      "grad_norm": 0.529727041721344,
      "learning_rate": 2e-05,
      "loss": 0.8591,
      "step": 550
    },
    {
      "epoch": 0.048109965635738834,
      "grad_norm": 0.46046891808509827,
      "learning_rate": 2e-05,
      "loss": 0.84,
      "step": 560
    },
    {
      "epoch": 0.04896907216494845,
      "grad_norm": 0.503052830696106,
      "learning_rate": 2e-05,
      "loss": 0.841,
      "step": 570
    },
    {
      "epoch": 0.04982817869415808,
      "grad_norm": 0.47345227003097534,
      "learning_rate": 2e-05,
      "loss": 0.8206,
      "step": 580
    },
    {
      "epoch": 0.0506872852233677,
      "grad_norm": 0.4749337136745453,
      "learning_rate": 2e-05,
      "loss": 0.8191,
      "step": 590
    },
    {
      "epoch": 0.05154639175257732,
      "grad_norm": 0.45893123745918274,
      "learning_rate": 2e-05,
      "loss": 0.8512,
      "step": 600
    },
    {
      "epoch": 0.05240549828178694,
      "grad_norm": 0.4918938875198364,
      "learning_rate": 2e-05,
      "loss": 0.8324,
      "step": 610
    },
    {
      "epoch": 0.05326460481099656,
      "grad_norm": 0.5130265951156616,
      "learning_rate": 2e-05,
      "loss": 0.8261,
      "step": 620
    },
    {
      "epoch": 0.05412371134020619,
      "grad_norm": 0.48524969816207886,
      "learning_rate": 2e-05,
      "loss": 0.8402,
      "step": 630
    },
    {
      "epoch": 0.054982817869415807,
      "grad_norm": 0.4443276822566986,
      "learning_rate": 2e-05,
      "loss": 0.778,
      "step": 640
    },
    {
      "epoch": 0.05584192439862543,
      "grad_norm": 0.48050761222839355,
      "learning_rate": 2e-05,
      "loss": 0.8328,
      "step": 650
    },
    {
      "epoch": 0.05670103092783505,
      "grad_norm": 0.468749463558197,
      "learning_rate": 2e-05,
      "loss": 0.8126,
      "step": 660
    },
    {
      "epoch": 0.05756013745704467,
      "grad_norm": 0.47499385476112366,
      "learning_rate": 2e-05,
      "loss": 0.8567,
      "step": 670
    },
    {
      "epoch": 0.058419243986254296,
      "grad_norm": 0.47868019342422485,
      "learning_rate": 2e-05,
      "loss": 0.8247,
      "step": 680
    },
    {
      "epoch": 0.059278350515463915,
      "grad_norm": 0.4504624009132385,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 690
    },
    {
      "epoch": 0.06013745704467354,
      "grad_norm": 0.481806218624115,
      "learning_rate": 2e-05,
      "loss": 0.8208,
      "step": 700
    },
    {
      "epoch": 0.06099656357388316,
      "grad_norm": 2.282531261444092,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 710
    },
    {
      "epoch": 0.061855670103092786,
      "grad_norm": 0.48688116669654846,
      "learning_rate": 2e-05,
      "loss": 0.8329,
      "step": 720
    },
    {
      "epoch": 0.0627147766323024,
      "grad_norm": 0.5048409104347229,
      "learning_rate": 2e-05,
      "loss": 0.8492,
      "step": 730
    },
    {
      "epoch": 0.06357388316151202,
      "grad_norm": 0.493664413690567,
      "learning_rate": 2e-05,
      "loss": 0.8449,
      "step": 740
    },
    {
      "epoch": 0.06443298969072164,
      "grad_norm": 0.4665951728820801,
      "learning_rate": 2e-05,
      "loss": 0.8156,
      "step": 750
    },
    {
      "epoch": 0.06529209621993128,
      "grad_norm": 0.4632743000984192,
      "learning_rate": 2e-05,
      "loss": 0.8592,
      "step": 760
    },
    {
      "epoch": 0.0661512027491409,
      "grad_norm": 0.5417582392692566,
      "learning_rate": 2e-05,
      "loss": 0.8185,
      "step": 770
    },
    {
      "epoch": 0.06701030927835051,
      "grad_norm": 0.5666975378990173,
      "learning_rate": 2e-05,
      "loss": 0.8282,
      "step": 780
    },
    {
      "epoch": 0.06786941580756013,
      "grad_norm": 0.4715534746646881,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 790
    },
    {
      "epoch": 0.06872852233676977,
      "grad_norm": 0.4496767818927765,
      "learning_rate": 2e-05,
      "loss": 0.8049,
      "step": 800
    },
    {
      "epoch": 0.06958762886597938,
      "grad_norm": 0.47709542512893677,
      "learning_rate": 2e-05,
      "loss": 0.84,
      "step": 810
    },
    {
      "epoch": 0.070446735395189,
      "grad_norm": 0.49228349328041077,
      "learning_rate": 2e-05,
      "loss": 0.8107,
      "step": 820
    },
    {
      "epoch": 0.07130584192439862,
      "grad_norm": 0.4816240072250366,
      "learning_rate": 2e-05,
      "loss": 0.8094,
      "step": 830
    },
    {
      "epoch": 0.07216494845360824,
      "grad_norm": 1.0014556646347046,
      "learning_rate": 2e-05,
      "loss": 0.8432,
      "step": 840
    },
    {
      "epoch": 0.07302405498281787,
      "grad_norm": 0.4686058461666107,
      "learning_rate": 2e-05,
      "loss": 0.8189,
      "step": 850
    },
    {
      "epoch": 0.07388316151202749,
      "grad_norm": 0.4596993923187256,
      "learning_rate": 2e-05,
      "loss": 0.8065,
      "step": 860
    },
    {
      "epoch": 0.07474226804123711,
      "grad_norm": 0.48179084062576294,
      "learning_rate": 2e-05,
      "loss": 0.829,
      "step": 870
    },
    {
      "epoch": 0.07560137457044673,
      "grad_norm": 0.5536651611328125,
      "learning_rate": 2e-05,
      "loss": 0.8229,
      "step": 880
    },
    {
      "epoch": 0.07646048109965636,
      "grad_norm": 0.49012085795402527,
      "learning_rate": 2e-05,
      "loss": 0.8365,
      "step": 890
    },
    {
      "epoch": 0.07731958762886598,
      "grad_norm": 0.46614527702331543,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 900
    },
    {
      "epoch": 0.0781786941580756,
      "grad_norm": 0.49526447057724,
      "learning_rate": 2e-05,
      "loss": 0.823,
      "step": 910
    },
    {
      "epoch": 0.07903780068728522,
      "grad_norm": 0.4854036271572113,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 920
    },
    {
      "epoch": 0.07989690721649484,
      "grad_norm": 0.46326735615730286,
      "learning_rate": 2e-05,
      "loss": 0.8298,
      "step": 930
    },
    {
      "epoch": 0.08075601374570447,
      "grad_norm": 0.48059317469596863,
      "learning_rate": 2e-05,
      "loss": 0.8074,
      "step": 940
    },
    {
      "epoch": 0.08161512027491409,
      "grad_norm": 0.4574795365333557,
      "learning_rate": 2e-05,
      "loss": 0.8097,
      "step": 950
    },
    {
      "epoch": 0.08247422680412371,
      "grad_norm": 0.44573479890823364,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 960
    },
    {
      "epoch": 0.08333333333333333,
      "grad_norm": 0.4629650115966797,
      "learning_rate": 2e-05,
      "loss": 0.798,
      "step": 970
    },
    {
      "epoch": 0.08419243986254296,
      "grad_norm": 0.47595855593681335,
      "learning_rate": 2e-05,
      "loss": 0.8105,
      "step": 980
    },
    {
      "epoch": 0.08505154639175258,
      "grad_norm": 0.4622196555137634,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 990
    },
    {
      "epoch": 0.0859106529209622,
      "grad_norm": 0.4406432807445526,
      "learning_rate": 2e-05,
      "loss": 0.849,
      "step": 1000
    },
    {
      "epoch": 0.08676975945017182,
      "grad_norm": 0.47430476546287537,
      "learning_rate": 2e-05,
      "loss": 0.8334,
      "step": 1010
    },
    {
      "epoch": 0.08762886597938144,
      "grad_norm": 0.4646911323070526,
      "learning_rate": 2e-05,
      "loss": 0.8302,
      "step": 1020
    },
    {
      "epoch": 0.08848797250859107,
      "grad_norm": 0.503621518611908,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 1030
    },
    {
      "epoch": 0.08934707903780069,
      "grad_norm": 0.42737218737602234,
      "learning_rate": 2e-05,
      "loss": 0.7868,
      "step": 1040
    },
    {
      "epoch": 0.09020618556701031,
      "grad_norm": 0.4449130892753601,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 1050
    },
    {
      "epoch": 0.09106529209621993,
      "grad_norm": 0.4552324414253235,
      "learning_rate": 2e-05,
      "loss": 0.8151,
      "step": 1060
    },
    {
      "epoch": 0.09192439862542956,
      "grad_norm": 0.4285755455493927,
      "learning_rate": 2e-05,
      "loss": 0.7987,
      "step": 1070
    },
    {
      "epoch": 0.09278350515463918,
      "grad_norm": 0.44612982869148254,
      "learning_rate": 2e-05,
      "loss": 0.8091,
      "step": 1080
    },
    {
      "epoch": 0.0936426116838488,
      "grad_norm": 0.41484394669532776,
      "learning_rate": 2e-05,
      "loss": 0.802,
      "step": 1090
    },
    {
      "epoch": 0.09450171821305842,
      "grad_norm": 0.43857651948928833,
      "learning_rate": 2e-05,
      "loss": 0.8051,
      "step": 1100
    },
    {
      "epoch": 0.09536082474226804,
      "grad_norm": 0.4621300995349884,
      "learning_rate": 2e-05,
      "loss": 0.8114,
      "step": 1110
    },
    {
      "epoch": 0.09621993127147767,
      "grad_norm": 0.46743395924568176,
      "learning_rate": 2e-05,
      "loss": 0.791,
      "step": 1120
    },
    {
      "epoch": 0.09707903780068729,
      "grad_norm": 0.4864065945148468,
      "learning_rate": 2e-05,
      "loss": 0.8253,
      "step": 1130
    },
    {
      "epoch": 0.0979381443298969,
      "grad_norm": 0.4746995270252228,
      "learning_rate": 2e-05,
      "loss": 0.8027,
      "step": 1140
    },
    {
      "epoch": 0.09879725085910653,
      "grad_norm": 0.43570324778556824,
      "learning_rate": 2e-05,
      "loss": 0.7926,
      "step": 1150
    },
    {
      "epoch": 0.09965635738831616,
      "grad_norm": 0.6272392868995667,
      "learning_rate": 2e-05,
      "loss": 0.7873,
      "step": 1160
    },
    {
      "epoch": 0.10051546391752578,
      "grad_norm": 0.4445926547050476,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 1170
    },
    {
      "epoch": 0.1013745704467354,
      "grad_norm": 0.44602736830711365,
      "learning_rate": 2e-05,
      "loss": 0.8186,
      "step": 1180
    },
    {
      "epoch": 0.10223367697594501,
      "grad_norm": 0.49426504969596863,
      "learning_rate": 2e-05,
      "loss": 0.8112,
      "step": 1190
    },
    {
      "epoch": 0.10309278350515463,
      "grad_norm": 0.4448431730270386,
      "learning_rate": 2e-05,
      "loss": 0.8012,
      "step": 1200
    },
    {
      "epoch": 0.10395189003436427,
      "grad_norm": 0.49004390835762024,
      "learning_rate": 2e-05,
      "loss": 0.8141,
      "step": 1210
    },
    {
      "epoch": 0.10481099656357389,
      "grad_norm": 0.48352155089378357,
      "learning_rate": 2e-05,
      "loss": 0.8125,
      "step": 1220
    },
    {
      "epoch": 0.1056701030927835,
      "grad_norm": 0.43718385696411133,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 1230
    },
    {
      "epoch": 0.10652920962199312,
      "grad_norm": 0.46456244587898254,
      "learning_rate": 2e-05,
      "loss": 0.8023,
      "step": 1240
    },
    {
      "epoch": 0.10738831615120274,
      "grad_norm": 0.4470098316669464,
      "learning_rate": 2e-05,
      "loss": 0.7836,
      "step": 1250
    },
    {
      "epoch": 0.10824742268041238,
      "grad_norm": 0.44037750363349915,
      "learning_rate": 2e-05,
      "loss": 0.8215,
      "step": 1260
    },
    {
      "epoch": 0.109106529209622,
      "grad_norm": 0.4431336224079132,
      "learning_rate": 2e-05,
      "loss": 0.8164,
      "step": 1270
    },
    {
      "epoch": 0.10996563573883161,
      "grad_norm": 0.4316923916339874,
      "learning_rate": 2e-05,
      "loss": 0.824,
      "step": 1280
    },
    {
      "epoch": 0.11082474226804123,
      "grad_norm": 0.4511774778366089,
      "learning_rate": 2e-05,
      "loss": 0.8211,
      "step": 1290
    },
    {
      "epoch": 0.11168384879725086,
      "grad_norm": 0.41615837812423706,
      "learning_rate": 2e-05,
      "loss": 0.7879,
      "step": 1300
    },
    {
      "epoch": 0.11254295532646048,
      "grad_norm": 0.42174649238586426,
      "learning_rate": 2e-05,
      "loss": 0.7917,
      "step": 1310
    },
    {
      "epoch": 0.1134020618556701,
      "grad_norm": 0.5034571886062622,
      "learning_rate": 2e-05,
      "loss": 0.8056,
      "step": 1320
    },
    {
      "epoch": 0.11426116838487972,
      "grad_norm": 0.4601093530654907,
      "learning_rate": 2e-05,
      "loss": 0.819,
      "step": 1330
    },
    {
      "epoch": 0.11512027491408934,
      "grad_norm": 0.4226849377155304,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 1340
    },
    {
      "epoch": 0.11597938144329897,
      "grad_norm": 0.4490329921245575,
      "learning_rate": 2e-05,
      "loss": 0.8005,
      "step": 1350
    },
    {
      "epoch": 0.11683848797250859,
      "grad_norm": 0.43054598569869995,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 1360
    },
    {
      "epoch": 0.11769759450171821,
      "grad_norm": 0.4534497559070587,
      "learning_rate": 2e-05,
      "loss": 0.7901,
      "step": 1370
    },
    {
      "epoch": 0.11855670103092783,
      "grad_norm": 0.4433537423610687,
      "learning_rate": 2e-05,
      "loss": 0.8133,
      "step": 1380
    },
    {
      "epoch": 0.11941580756013746,
      "grad_norm": 0.4343333840370178,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 1390
    },
    {
      "epoch": 0.12027491408934708,
      "grad_norm": 0.42904624342918396,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 1400
    },
    {
      "epoch": 0.1211340206185567,
      "grad_norm": 0.4655359983444214,
      "learning_rate": 2e-05,
      "loss": 0.8211,
      "step": 1410
    },
    {
      "epoch": 0.12199312714776632,
      "grad_norm": 0.4095928966999054,
      "learning_rate": 2e-05,
      "loss": 0.812,
      "step": 1420
    },
    {
      "epoch": 0.12285223367697594,
      "grad_norm": 0.41238921880722046,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 1430
    },
    {
      "epoch": 0.12371134020618557,
      "grad_norm": 0.4852188229560852,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 1440
    },
    {
      "epoch": 0.12457044673539519,
      "grad_norm": 0.44924116134643555,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 1450
    },
    {
      "epoch": 0.1254295532646048,
      "grad_norm": 0.4346265494823456,
      "learning_rate": 2e-05,
      "loss": 0.7992,
      "step": 1460
    },
    {
      "epoch": 0.12628865979381443,
      "grad_norm": 0.4454062581062317,
      "learning_rate": 2e-05,
      "loss": 0.797,
      "step": 1470
    },
    {
      "epoch": 0.12714776632302405,
      "grad_norm": 0.4472755193710327,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 1480
    },
    {
      "epoch": 0.12800687285223367,
      "grad_norm": 0.43495413661003113,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 1490
    },
    {
      "epoch": 0.12886597938144329,
      "grad_norm": 0.42517977952957153,
      "learning_rate": 2e-05,
      "loss": 0.8023,
      "step": 1500
    },
    {
      "epoch": 0.12972508591065293,
      "grad_norm": 0.5048529505729675,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 1510
    },
    {
      "epoch": 0.13058419243986255,
      "grad_norm": 0.41410166025161743,
      "learning_rate": 2e-05,
      "loss": 0.7798,
      "step": 1520
    },
    {
      "epoch": 0.13144329896907217,
      "grad_norm": 0.4294983744621277,
      "learning_rate": 2e-05,
      "loss": 0.7878,
      "step": 1530
    },
    {
      "epoch": 0.1323024054982818,
      "grad_norm": 0.42941445112228394,
      "learning_rate": 2e-05,
      "loss": 0.7884,
      "step": 1540
    },
    {
      "epoch": 0.1331615120274914,
      "grad_norm": 0.4035552740097046,
      "learning_rate": 2e-05,
      "loss": 0.792,
      "step": 1550
    },
    {
      "epoch": 0.13402061855670103,
      "grad_norm": 0.4652338922023773,
      "learning_rate": 2e-05,
      "loss": 0.777,
      "step": 1560
    },
    {
      "epoch": 0.13487972508591065,
      "grad_norm": 0.4364311993122101,
      "learning_rate": 2e-05,
      "loss": 0.7723,
      "step": 1570
    },
    {
      "epoch": 0.13573883161512026,
      "grad_norm": 0.4207146465778351,
      "learning_rate": 2e-05,
      "loss": 0.7612,
      "step": 1580
    },
    {
      "epoch": 0.13659793814432988,
      "grad_norm": 0.4272591769695282,
      "learning_rate": 2e-05,
      "loss": 0.8061,
      "step": 1590
    },
    {
      "epoch": 0.13745704467353953,
      "grad_norm": 0.4509749710559845,
      "learning_rate": 2e-05,
      "loss": 0.7926,
      "step": 1600
    },
    {
      "epoch": 0.13831615120274915,
      "grad_norm": 0.4478810429573059,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 1610
    },
    {
      "epoch": 0.13917525773195877,
      "grad_norm": 0.4207334518432617,
      "learning_rate": 2e-05,
      "loss": 0.7678,
      "step": 1620
    },
    {
      "epoch": 0.1400343642611684,
      "grad_norm": 0.4360485374927521,
      "learning_rate": 2e-05,
      "loss": 0.7909,
      "step": 1630
    },
    {
      "epoch": 0.140893470790378,
      "grad_norm": 0.4345305562019348,
      "learning_rate": 2e-05,
      "loss": 0.8139,
      "step": 1640
    },
    {
      "epoch": 0.14175257731958762,
      "grad_norm": 0.43382421135902405,
      "learning_rate": 2e-05,
      "loss": 0.797,
      "step": 1650
    },
    {
      "epoch": 0.14261168384879724,
      "grad_norm": 0.42108669877052307,
      "learning_rate": 2e-05,
      "loss": 0.804,
      "step": 1660
    },
    {
      "epoch": 0.14347079037800686,
      "grad_norm": 0.45418545603752136,
      "learning_rate": 2e-05,
      "loss": 0.7852,
      "step": 1670
    },
    {
      "epoch": 0.14432989690721648,
      "grad_norm": 0.4910484552383423,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 1680
    },
    {
      "epoch": 0.14518900343642613,
      "grad_norm": 0.4287989139556885,
      "learning_rate": 2e-05,
      "loss": 0.8165,
      "step": 1690
    },
    {
      "epoch": 0.14604810996563575,
      "grad_norm": 0.5834421515464783,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 1700
    },
    {
      "epoch": 0.14690721649484537,
      "grad_norm": 0.5063323378562927,
      "learning_rate": 2e-05,
      "loss": 0.7734,
      "step": 1710
    },
    {
      "epoch": 0.14776632302405499,
      "grad_norm": 0.4238131046295166,
      "learning_rate": 2e-05,
      "loss": 0.8028,
      "step": 1720
    },
    {
      "epoch": 0.1486254295532646,
      "grad_norm": 0.38663214445114136,
      "learning_rate": 2e-05,
      "loss": 0.7988,
      "step": 1730
    },
    {
      "epoch": 0.14948453608247422,
      "grad_norm": 0.44796040654182434,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 1740
    },
    {
      "epoch": 0.15034364261168384,
      "grad_norm": 0.41491448879241943,
      "learning_rate": 2e-05,
      "loss": 0.7628,
      "step": 1750
    },
    {
      "epoch": 0.15120274914089346,
      "grad_norm": 0.4768643379211426,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 1760
    },
    {
      "epoch": 0.15206185567010308,
      "grad_norm": 0.42717429995536804,
      "learning_rate": 2e-05,
      "loss": 0.7955,
      "step": 1770
    },
    {
      "epoch": 0.15292096219931273,
      "grad_norm": 0.43356406688690186,
      "learning_rate": 2e-05,
      "loss": 0.7826,
      "step": 1780
    },
    {
      "epoch": 0.15378006872852235,
      "grad_norm": 0.45616310834884644,
      "learning_rate": 2e-05,
      "loss": 0.7821,
      "step": 1790
    },
    {
      "epoch": 0.15463917525773196,
      "grad_norm": 0.4642743468284607,
      "learning_rate": 2e-05,
      "loss": 0.7832,
      "step": 1800
    },
    {
      "epoch": 0.15549828178694158,
      "grad_norm": 0.42061248421669006,
      "learning_rate": 2e-05,
      "loss": 0.7829,
      "step": 1810
    },
    {
      "epoch": 0.1563573883161512,
      "grad_norm": 0.473254919052124,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 1820
    },
    {
      "epoch": 0.15721649484536082,
      "grad_norm": 0.4449467957019806,
      "learning_rate": 2e-05,
      "loss": 0.7904,
      "step": 1830
    },
    {
      "epoch": 0.15807560137457044,
      "grad_norm": 0.4272187054157257,
      "learning_rate": 2e-05,
      "loss": 0.7939,
      "step": 1840
    },
    {
      "epoch": 0.15893470790378006,
      "grad_norm": 0.4838572144508362,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 1850
    },
    {
      "epoch": 0.15979381443298968,
      "grad_norm": 0.44314128160476685,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 1860
    },
    {
      "epoch": 0.16065292096219932,
      "grad_norm": 0.44182509183883667,
      "learning_rate": 2e-05,
      "loss": 0.7908,
      "step": 1870
    },
    {
      "epoch": 0.16151202749140894,
      "grad_norm": 0.4147513508796692,
      "learning_rate": 2e-05,
      "loss": 0.7883,
      "step": 1880
    },
    {
      "epoch": 0.16237113402061856,
      "grad_norm": 0.44913968443870544,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 1890
    },
    {
      "epoch": 0.16323024054982818,
      "grad_norm": 0.43962720036506653,
      "learning_rate": 2e-05,
      "loss": 0.7889,
      "step": 1900
    },
    {
      "epoch": 0.1640893470790378,
      "grad_norm": 0.41255542635917664,
      "learning_rate": 2e-05,
      "loss": 0.7715,
      "step": 1910
    },
    {
      "epoch": 0.16494845360824742,
      "grad_norm": 0.4223017692565918,
      "learning_rate": 2e-05,
      "loss": 0.7746,
      "step": 1920
    },
    {
      "epoch": 0.16580756013745704,
      "grad_norm": 0.4158386290073395,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 1930
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 0.42182308435440063,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 1940
    },
    {
      "epoch": 0.16752577319587628,
      "grad_norm": 0.43559759855270386,
      "learning_rate": 2e-05,
      "loss": 0.7756,
      "step": 1950
    },
    {
      "epoch": 0.16838487972508592,
      "grad_norm": 0.4135313332080841,
      "learning_rate": 2e-05,
      "loss": 0.7793,
      "step": 1960
    },
    {
      "epoch": 0.16924398625429554,
      "grad_norm": 0.4393354058265686,
      "learning_rate": 2e-05,
      "loss": 0.7774,
      "step": 1970
    },
    {
      "epoch": 0.17010309278350516,
      "grad_norm": 0.40443214774131775,
      "learning_rate": 2e-05,
      "loss": 0.7891,
      "step": 1980
    },
    {
      "epoch": 0.17096219931271478,
      "grad_norm": 0.44918447732925415,
      "learning_rate": 2e-05,
      "loss": 0.7923,
      "step": 1990
    },
    {
      "epoch": 0.1718213058419244,
      "grad_norm": 0.4612440764904022,
      "learning_rate": 2e-05,
      "loss": 0.7896,
      "step": 2000
    },
    {
      "epoch": 0.17268041237113402,
      "grad_norm": 0.43252137303352356,
      "learning_rate": 2e-05,
      "loss": 0.7628,
      "step": 2010
    },
    {
      "epoch": 0.17353951890034364,
      "grad_norm": 0.4581339657306671,
      "learning_rate": 2e-05,
      "loss": 0.8109,
      "step": 2020
    },
    {
      "epoch": 0.17439862542955326,
      "grad_norm": 0.4455418884754181,
      "learning_rate": 2e-05,
      "loss": 0.7978,
      "step": 2030
    },
    {
      "epoch": 0.17525773195876287,
      "grad_norm": 0.4451400637626648,
      "learning_rate": 2e-05,
      "loss": 0.7832,
      "step": 2040
    },
    {
      "epoch": 0.17611683848797252,
      "grad_norm": 0.4293883144855499,
      "learning_rate": 2e-05,
      "loss": 0.8197,
      "step": 2050
    },
    {
      "epoch": 0.17697594501718214,
      "grad_norm": 0.41408681869506836,
      "learning_rate": 2e-05,
      "loss": 0.7809,
      "step": 2060
    },
    {
      "epoch": 0.17783505154639176,
      "grad_norm": 0.4040951132774353,
      "learning_rate": 2e-05,
      "loss": 0.794,
      "step": 2070
    },
    {
      "epoch": 0.17869415807560138,
      "grad_norm": 0.8779594898223877,
      "learning_rate": 2e-05,
      "loss": 0.7654,
      "step": 2080
    },
    {
      "epoch": 0.179553264604811,
      "grad_norm": 0.45652055740356445,
      "learning_rate": 2e-05,
      "loss": 0.7794,
      "step": 2090
    },
    {
      "epoch": 0.18041237113402062,
      "grad_norm": 0.42162561416625977,
      "learning_rate": 2e-05,
      "loss": 0.7856,
      "step": 2100
    },
    {
      "epoch": 0.18127147766323023,
      "grad_norm": 0.4086996614933014,
      "learning_rate": 2e-05,
      "loss": 0.7851,
      "step": 2110
    },
    {
      "epoch": 0.18213058419243985,
      "grad_norm": 0.4172867238521576,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 2120
    },
    {
      "epoch": 0.18298969072164947,
      "grad_norm": 0.4956143796443939,
      "learning_rate": 2e-05,
      "loss": 0.7823,
      "step": 2130
    },
    {
      "epoch": 0.18384879725085912,
      "grad_norm": 0.41764020919799805,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 2140
    },
    {
      "epoch": 0.18470790378006874,
      "grad_norm": 0.40405750274658203,
      "learning_rate": 2e-05,
      "loss": 0.8238,
      "step": 2150
    },
    {
      "epoch": 0.18556701030927836,
      "grad_norm": 0.4923059642314911,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 2160
    },
    {
      "epoch": 0.18642611683848798,
      "grad_norm": 0.44558045268058777,
      "learning_rate": 2e-05,
      "loss": 0.7791,
      "step": 2170
    },
    {
      "epoch": 0.1872852233676976,
      "grad_norm": 0.395918071269989,
      "learning_rate": 2e-05,
      "loss": 0.7709,
      "step": 2180
    },
    {
      "epoch": 0.18814432989690721,
      "grad_norm": 0.40383675694465637,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 2190
    },
    {
      "epoch": 0.18900343642611683,
      "grad_norm": 0.42041248083114624,
      "learning_rate": 2e-05,
      "loss": 0.7589,
      "step": 2200
    },
    {
      "epoch": 0.18986254295532645,
      "grad_norm": 0.406876802444458,
      "learning_rate": 2e-05,
      "loss": 0.7947,
      "step": 2210
    },
    {
      "epoch": 0.19072164948453607,
      "grad_norm": 0.4463612139225006,
      "learning_rate": 2e-05,
      "loss": 0.7608,
      "step": 2220
    },
    {
      "epoch": 0.19158075601374572,
      "grad_norm": 0.42114925384521484,
      "learning_rate": 2e-05,
      "loss": 0.8017,
      "step": 2230
    },
    {
      "epoch": 0.19243986254295534,
      "grad_norm": 0.5391575694084167,
      "learning_rate": 2e-05,
      "loss": 0.8105,
      "step": 2240
    },
    {
      "epoch": 0.19329896907216496,
      "grad_norm": 0.4293856620788574,
      "learning_rate": 2e-05,
      "loss": 0.803,
      "step": 2250
    },
    {
      "epoch": 0.19415807560137457,
      "grad_norm": 0.4132872521877289,
      "learning_rate": 2e-05,
      "loss": 0.773,
      "step": 2260
    },
    {
      "epoch": 0.1950171821305842,
      "grad_norm": 0.42714643478393555,
      "learning_rate": 2e-05,
      "loss": 0.7969,
      "step": 2270
    },
    {
      "epoch": 0.1958762886597938,
      "grad_norm": 0.416900634765625,
      "learning_rate": 2e-05,
      "loss": 0.7813,
      "step": 2280
    },
    {
      "epoch": 0.19673539518900343,
      "grad_norm": 0.40540578961372375,
      "learning_rate": 2e-05,
      "loss": 0.7846,
      "step": 2290
    },
    {
      "epoch": 0.19759450171821305,
      "grad_norm": 0.3905494511127472,
      "learning_rate": 2e-05,
      "loss": 0.7636,
      "step": 2300
    },
    {
      "epoch": 0.19845360824742267,
      "grad_norm": 0.4042344391345978,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 2310
    },
    {
      "epoch": 0.19931271477663232,
      "grad_norm": 0.41851627826690674,
      "learning_rate": 2e-05,
      "loss": 0.7524,
      "step": 2320
    },
    {
      "epoch": 0.20017182130584193,
      "grad_norm": 0.45878472924232483,
      "learning_rate": 2e-05,
      "loss": 0.7791,
      "step": 2330
    },
    {
      "epoch": 0.20103092783505155,
      "grad_norm": 0.3952105641365051,
      "learning_rate": 2e-05,
      "loss": 0.7652,
      "step": 2340
    },
    {
      "epoch": 0.20189003436426117,
      "grad_norm": 0.4032493829727173,
      "learning_rate": 2e-05,
      "loss": 0.7915,
      "step": 2350
    },
    {
      "epoch": 0.2027491408934708,
      "grad_norm": 0.4258342981338501,
      "learning_rate": 2e-05,
      "loss": 0.7769,
      "step": 2360
    },
    {
      "epoch": 0.2036082474226804,
      "grad_norm": 0.43611422181129456,
      "learning_rate": 2e-05,
      "loss": 0.8027,
      "step": 2370
    },
    {
      "epoch": 0.20446735395189003,
      "grad_norm": 0.41574737429618835,
      "learning_rate": 2e-05,
      "loss": 0.7815,
      "step": 2380
    },
    {
      "epoch": 0.20532646048109965,
      "grad_norm": 0.4091911017894745,
      "learning_rate": 2e-05,
      "loss": 0.7698,
      "step": 2390
    },
    {
      "epoch": 0.20618556701030927,
      "grad_norm": 0.4178768992424011,
      "learning_rate": 2e-05,
      "loss": 0.7955,
      "step": 2400
    },
    {
      "epoch": 0.20704467353951891,
      "grad_norm": 0.4205310046672821,
      "learning_rate": 2e-05,
      "loss": 0.754,
      "step": 2410
    },
    {
      "epoch": 0.20790378006872853,
      "grad_norm": 0.3947570323944092,
      "learning_rate": 2e-05,
      "loss": 0.7749,
      "step": 2420
    },
    {
      "epoch": 0.20876288659793815,
      "grad_norm": 0.425428181886673,
      "learning_rate": 2e-05,
      "loss": 0.7831,
      "step": 2430
    },
    {
      "epoch": 0.20962199312714777,
      "grad_norm": 0.4144151210784912,
      "learning_rate": 2e-05,
      "loss": 0.7924,
      "step": 2440
    },
    {
      "epoch": 0.2104810996563574,
      "grad_norm": 0.40834271907806396,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 2450
    },
    {
      "epoch": 0.211340206185567,
      "grad_norm": 0.570460855960846,
      "learning_rate": 2e-05,
      "loss": 0.7918,
      "step": 2460
    },
    {
      "epoch": 0.21219931271477663,
      "grad_norm": 0.41472381353378296,
      "learning_rate": 2e-05,
      "loss": 0.7746,
      "step": 2470
    },
    {
      "epoch": 0.21305841924398625,
      "grad_norm": 0.4356991648674011,
      "learning_rate": 2e-05,
      "loss": 0.793,
      "step": 2480
    },
    {
      "epoch": 0.21391752577319587,
      "grad_norm": 0.425836443901062,
      "learning_rate": 2e-05,
      "loss": 0.7826,
      "step": 2490
    },
    {
      "epoch": 0.21477663230240548,
      "grad_norm": 0.9032119512557983,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 2500
    },
    {
      "epoch": 0.21563573883161513,
      "grad_norm": 0.4030701518058777,
      "learning_rate": 2e-05,
      "loss": 0.7539,
      "step": 2510
    },
    {
      "epoch": 0.21649484536082475,
      "grad_norm": 0.4129731357097626,
      "learning_rate": 2e-05,
      "loss": 0.7652,
      "step": 2520
    },
    {
      "epoch": 0.21735395189003437,
      "grad_norm": 0.4277688264846802,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 2530
    },
    {
      "epoch": 0.218213058419244,
      "grad_norm": 0.40256595611572266,
      "learning_rate": 2e-05,
      "loss": 0.7498,
      "step": 2540
    },
    {
      "epoch": 0.2190721649484536,
      "grad_norm": 0.39388591051101685,
      "learning_rate": 2e-05,
      "loss": 0.7449,
      "step": 2550
    },
    {
      "epoch": 0.21993127147766323,
      "grad_norm": 0.41771769523620605,
      "learning_rate": 2e-05,
      "loss": 0.7658,
      "step": 2560
    },
    {
      "epoch": 0.22079037800687284,
      "grad_norm": 0.43381044268608093,
      "learning_rate": 2e-05,
      "loss": 0.7379,
      "step": 2570
    },
    {
      "epoch": 0.22164948453608246,
      "grad_norm": 0.4080195426940918,
      "learning_rate": 2e-05,
      "loss": 0.7623,
      "step": 2580
    },
    {
      "epoch": 0.22250859106529208,
      "grad_norm": 0.4249197840690613,
      "learning_rate": 2e-05,
      "loss": 0.7439,
      "step": 2590
    },
    {
      "epoch": 0.22336769759450173,
      "grad_norm": 0.4250815510749817,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 2600
    },
    {
      "epoch": 0.22422680412371135,
      "grad_norm": 0.4173683226108551,
      "learning_rate": 2e-05,
      "loss": 0.7402,
      "step": 2610
    },
    {
      "epoch": 0.22508591065292097,
      "grad_norm": 0.4079398810863495,
      "learning_rate": 2e-05,
      "loss": 0.7959,
      "step": 2620
    },
    {
      "epoch": 0.2259450171821306,
      "grad_norm": 0.4313492178916931,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 2630
    },
    {
      "epoch": 0.2268041237113402,
      "grad_norm": 0.3909526467323303,
      "learning_rate": 2e-05,
      "loss": 0.7483,
      "step": 2640
    },
    {
      "epoch": 0.22766323024054982,
      "grad_norm": 0.4261670410633087,
      "learning_rate": 2e-05,
      "loss": 0.7814,
      "step": 2650
    },
    {
      "epoch": 0.22852233676975944,
      "grad_norm": 0.4046727418899536,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 2660
    },
    {
      "epoch": 0.22938144329896906,
      "grad_norm": 0.4096175730228424,
      "learning_rate": 2e-05,
      "loss": 0.7568,
      "step": 2670
    },
    {
      "epoch": 0.23024054982817868,
      "grad_norm": 0.39225271344184875,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 2680
    },
    {
      "epoch": 0.23109965635738833,
      "grad_norm": 0.45596063137054443,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 2690
    },
    {
      "epoch": 0.23195876288659795,
      "grad_norm": 0.3877202272415161,
      "learning_rate": 2e-05,
      "loss": 0.7551,
      "step": 2700
    },
    {
      "epoch": 0.23281786941580757,
      "grad_norm": 0.4236013889312744,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 2710
    },
    {
      "epoch": 0.23367697594501718,
      "grad_norm": 0.3999956548213959,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 2720
    },
    {
      "epoch": 0.2345360824742268,
      "grad_norm": 0.47755834460258484,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 2730
    },
    {
      "epoch": 0.23539518900343642,
      "grad_norm": 0.4101397693157196,
      "learning_rate": 2e-05,
      "loss": 0.7715,
      "step": 2740
    },
    {
      "epoch": 0.23625429553264604,
      "grad_norm": 0.42583993077278137,
      "learning_rate": 2e-05,
      "loss": 0.7851,
      "step": 2750
    },
    {
      "epoch": 0.23711340206185566,
      "grad_norm": 0.4020129442214966,
      "learning_rate": 2e-05,
      "loss": 0.7394,
      "step": 2760
    },
    {
      "epoch": 0.23797250859106528,
      "grad_norm": 0.4051017761230469,
      "learning_rate": 2e-05,
      "loss": 0.7705,
      "step": 2770
    },
    {
      "epoch": 0.23883161512027493,
      "grad_norm": 0.4843144416809082,
      "learning_rate": 2e-05,
      "loss": 0.7836,
      "step": 2780
    },
    {
      "epoch": 0.23969072164948454,
      "grad_norm": 0.4086487591266632,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 2790
    },
    {
      "epoch": 0.24054982817869416,
      "grad_norm": 0.49289804697036743,
      "learning_rate": 2e-05,
      "loss": 0.8043,
      "step": 2800
    },
    {
      "epoch": 0.24140893470790378,
      "grad_norm": 0.4173983037471771,
      "learning_rate": 2e-05,
      "loss": 0.7655,
      "step": 2810
    },
    {
      "epoch": 0.2422680412371134,
      "grad_norm": 0.42129677534103394,
      "learning_rate": 2e-05,
      "loss": 0.7828,
      "step": 2820
    },
    {
      "epoch": 0.24312714776632302,
      "grad_norm": 0.3931340277194977,
      "learning_rate": 2e-05,
      "loss": 0.7744,
      "step": 2830
    },
    {
      "epoch": 0.24398625429553264,
      "grad_norm": 0.4406070113182068,
      "learning_rate": 2e-05,
      "loss": 0.7553,
      "step": 2840
    },
    {
      "epoch": 0.24484536082474226,
      "grad_norm": 0.42197829484939575,
      "learning_rate": 2e-05,
      "loss": 0.7825,
      "step": 2850
    },
    {
      "epoch": 0.24570446735395188,
      "grad_norm": 0.4124753773212433,
      "learning_rate": 2e-05,
      "loss": 0.7562,
      "step": 2860
    },
    {
      "epoch": 0.24656357388316152,
      "grad_norm": 0.4178576171398163,
      "learning_rate": 2e-05,
      "loss": 0.7618,
      "step": 2870
    },
    {
      "epoch": 0.24742268041237114,
      "grad_norm": 0.4552730321884155,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 2880
    },
    {
      "epoch": 0.24828178694158076,
      "grad_norm": 0.42891925573349,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 2890
    },
    {
      "epoch": 0.24914089347079038,
      "grad_norm": 0.38573965430259705,
      "learning_rate": 2e-05,
      "loss": 0.7537,
      "step": 2900
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4111515283584595,
      "learning_rate": 2e-05,
      "loss": 0.8234,
      "step": 2910
    },
    {
      "epoch": 0.2508591065292096,
      "grad_norm": 0.4234291911125183,
      "learning_rate": 2e-05,
      "loss": 0.7947,
      "step": 2920
    },
    {
      "epoch": 0.25171821305841924,
      "grad_norm": 0.4000537693500519,
      "learning_rate": 2e-05,
      "loss": 0.7626,
      "step": 2930
    },
    {
      "epoch": 0.25257731958762886,
      "grad_norm": 0.4151047170162201,
      "learning_rate": 2e-05,
      "loss": 0.773,
      "step": 2940
    },
    {
      "epoch": 0.2534364261168385,
      "grad_norm": 0.3972582221031189,
      "learning_rate": 2e-05,
      "loss": 0.7824,
      "step": 2950
    },
    {
      "epoch": 0.2542955326460481,
      "grad_norm": 0.42536866664886475,
      "learning_rate": 2e-05,
      "loss": 0.7875,
      "step": 2960
    },
    {
      "epoch": 0.2551546391752577,
      "grad_norm": 0.42411258816719055,
      "learning_rate": 2e-05,
      "loss": 0.7407,
      "step": 2970
    },
    {
      "epoch": 0.25601374570446733,
      "grad_norm": 0.4318678677082062,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 2980
    },
    {
      "epoch": 0.25687285223367695,
      "grad_norm": 0.41649696230888367,
      "learning_rate": 2e-05,
      "loss": 0.7668,
      "step": 2990
    },
    {
      "epoch": 0.25773195876288657,
      "grad_norm": 0.42106330394744873,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 3000
    },
    {
      "epoch": 0.25859106529209624,
      "grad_norm": 0.4055321216583252,
      "learning_rate": 2e-05,
      "loss": 0.7986,
      "step": 3010
    },
    {
      "epoch": 0.25945017182130586,
      "grad_norm": 0.41959694027900696,
      "learning_rate": 2e-05,
      "loss": 0.7908,
      "step": 3020
    },
    {
      "epoch": 0.2603092783505155,
      "grad_norm": 0.4296233654022217,
      "learning_rate": 2e-05,
      "loss": 0.7846,
      "step": 3030
    },
    {
      "epoch": 0.2611683848797251,
      "grad_norm": 0.41830459237098694,
      "learning_rate": 2e-05,
      "loss": 0.8027,
      "step": 3040
    },
    {
      "epoch": 0.2620274914089347,
      "grad_norm": 0.40822726488113403,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 3050
    },
    {
      "epoch": 0.26288659793814434,
      "grad_norm": 0.4108002483844757,
      "learning_rate": 2e-05,
      "loss": 0.7665,
      "step": 3060
    },
    {
      "epoch": 0.26374570446735396,
      "grad_norm": 0.389371782541275,
      "learning_rate": 2e-05,
      "loss": 0.7397,
      "step": 3070
    },
    {
      "epoch": 0.2646048109965636,
      "grad_norm": 0.42747220396995544,
      "learning_rate": 2e-05,
      "loss": 0.7874,
      "step": 3080
    },
    {
      "epoch": 0.2654639175257732,
      "grad_norm": 0.4048026502132416,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 3090
    },
    {
      "epoch": 0.2663230240549828,
      "grad_norm": 0.5471147298812866,
      "learning_rate": 2e-05,
      "loss": 0.7873,
      "step": 3100
    },
    {
      "epoch": 0.26718213058419243,
      "grad_norm": 0.38855239748954773,
      "learning_rate": 2e-05,
      "loss": 0.7339,
      "step": 3110
    },
    {
      "epoch": 0.26804123711340205,
      "grad_norm": 0.4075400233268738,
      "learning_rate": 2e-05,
      "loss": 0.7489,
      "step": 3120
    },
    {
      "epoch": 0.26890034364261167,
      "grad_norm": 0.4032638370990753,
      "learning_rate": 2e-05,
      "loss": 0.7778,
      "step": 3130
    },
    {
      "epoch": 0.2697594501718213,
      "grad_norm": 0.41255396604537964,
      "learning_rate": 2e-05,
      "loss": 0.7492,
      "step": 3140
    },
    {
      "epoch": 0.2706185567010309,
      "grad_norm": 0.3877742290496826,
      "learning_rate": 2e-05,
      "loss": 0.7523,
      "step": 3150
    },
    {
      "epoch": 0.27147766323024053,
      "grad_norm": 0.3896322548389435,
      "learning_rate": 2e-05,
      "loss": 0.7774,
      "step": 3160
    },
    {
      "epoch": 0.27233676975945015,
      "grad_norm": 0.40311911702156067,
      "learning_rate": 2e-05,
      "loss": 0.7681,
      "step": 3170
    },
    {
      "epoch": 0.27319587628865977,
      "grad_norm": 0.3949439227581024,
      "learning_rate": 2e-05,
      "loss": 0.7683,
      "step": 3180
    },
    {
      "epoch": 0.27405498281786944,
      "grad_norm": 0.4075079560279846,
      "learning_rate": 2e-05,
      "loss": 0.741,
      "step": 3190
    },
    {
      "epoch": 0.27491408934707906,
      "grad_norm": 0.4325898289680481,
      "learning_rate": 2e-05,
      "loss": 0.7927,
      "step": 3200
    },
    {
      "epoch": 0.2757731958762887,
      "grad_norm": 0.41164514422416687,
      "learning_rate": 2e-05,
      "loss": 0.7573,
      "step": 3210
    },
    {
      "epoch": 0.2766323024054983,
      "grad_norm": 0.41085177659988403,
      "learning_rate": 2e-05,
      "loss": 0.7886,
      "step": 3220
    },
    {
      "epoch": 0.2774914089347079,
      "grad_norm": 0.4105769097805023,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 3230
    },
    {
      "epoch": 0.27835051546391754,
      "grad_norm": 0.41889050602912903,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 3240
    },
    {
      "epoch": 0.27920962199312716,
      "grad_norm": 0.3920256197452545,
      "learning_rate": 2e-05,
      "loss": 0.747,
      "step": 3250
    },
    {
      "epoch": 0.2800687285223368,
      "grad_norm": 0.4325742721557617,
      "learning_rate": 2e-05,
      "loss": 0.7725,
      "step": 3260
    },
    {
      "epoch": 0.2809278350515464,
      "grad_norm": 0.42666274309158325,
      "learning_rate": 2e-05,
      "loss": 0.7847,
      "step": 3270
    },
    {
      "epoch": 0.281786941580756,
      "grad_norm": 0.39878711104393005,
      "learning_rate": 2e-05,
      "loss": 0.7614,
      "step": 3280
    },
    {
      "epoch": 0.28264604810996563,
      "grad_norm": 0.45834362506866455,
      "learning_rate": 2e-05,
      "loss": 0.7683,
      "step": 3290
    },
    {
      "epoch": 0.28350515463917525,
      "grad_norm": 0.39607352018356323,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 3300
    },
    {
      "epoch": 0.28436426116838487,
      "grad_norm": 0.3845934271812439,
      "learning_rate": 2e-05,
      "loss": 0.7696,
      "step": 3310
    },
    {
      "epoch": 0.2852233676975945,
      "grad_norm": 0.4206698536872864,
      "learning_rate": 2e-05,
      "loss": 0.768,
      "step": 3320
    },
    {
      "epoch": 0.2860824742268041,
      "grad_norm": 0.3820017874240875,
      "learning_rate": 2e-05,
      "loss": 0.7598,
      "step": 3330
    },
    {
      "epoch": 0.2869415807560137,
      "grad_norm": 0.3978671729564667,
      "learning_rate": 2e-05,
      "loss": 0.7423,
      "step": 3340
    },
    {
      "epoch": 0.28780068728522334,
      "grad_norm": 0.4384324550628662,
      "learning_rate": 2e-05,
      "loss": 0.7646,
      "step": 3350
    },
    {
      "epoch": 0.28865979381443296,
      "grad_norm": 0.4027288556098938,
      "learning_rate": 2e-05,
      "loss": 0.7585,
      "step": 3360
    },
    {
      "epoch": 0.28951890034364264,
      "grad_norm": 0.4325219690799713,
      "learning_rate": 2e-05,
      "loss": 0.7806,
      "step": 3370
    },
    {
      "epoch": 0.29037800687285226,
      "grad_norm": 0.4221478998661041,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 3380
    },
    {
      "epoch": 0.2912371134020619,
      "grad_norm": 0.44164422154426575,
      "learning_rate": 2e-05,
      "loss": 0.7693,
      "step": 3390
    },
    {
      "epoch": 0.2920962199312715,
      "grad_norm": 0.38554859161376953,
      "learning_rate": 2e-05,
      "loss": 0.763,
      "step": 3400
    },
    {
      "epoch": 0.2929553264604811,
      "grad_norm": 0.41126173734664917,
      "learning_rate": 2e-05,
      "loss": 0.75,
      "step": 3410
    },
    {
      "epoch": 0.29381443298969073,
      "grad_norm": 0.42341408133506775,
      "learning_rate": 2e-05,
      "loss": 0.7742,
      "step": 3420
    },
    {
      "epoch": 0.29467353951890035,
      "grad_norm": 0.45448067784309387,
      "learning_rate": 2e-05,
      "loss": 0.7734,
      "step": 3430
    },
    {
      "epoch": 0.29553264604810997,
      "grad_norm": 0.41353639960289,
      "learning_rate": 2e-05,
      "loss": 0.7904,
      "step": 3440
    },
    {
      "epoch": 0.2963917525773196,
      "grad_norm": 0.4858936667442322,
      "learning_rate": 2e-05,
      "loss": 0.7834,
      "step": 3450
    },
    {
      "epoch": 0.2972508591065292,
      "grad_norm": 0.4161330461502075,
      "learning_rate": 2e-05,
      "loss": 0.7774,
      "step": 3460
    },
    {
      "epoch": 0.2981099656357388,
      "grad_norm": 0.4159027338027954,
      "learning_rate": 2e-05,
      "loss": 0.7833,
      "step": 3470
    },
    {
      "epoch": 0.29896907216494845,
      "grad_norm": 0.4200475513935089,
      "learning_rate": 2e-05,
      "loss": 0.7544,
      "step": 3480
    },
    {
      "epoch": 0.29982817869415807,
      "grad_norm": 0.3868395984172821,
      "learning_rate": 2e-05,
      "loss": 0.7608,
      "step": 3490
    },
    {
      "epoch": 0.3006872852233677,
      "grad_norm": 0.44515305757522583,
      "learning_rate": 2e-05,
      "loss": 0.8271,
      "step": 3500
    },
    {
      "epoch": 0.3015463917525773,
      "grad_norm": 0.41017138957977295,
      "learning_rate": 2e-05,
      "loss": 0.7438,
      "step": 3510
    },
    {
      "epoch": 0.3024054982817869,
      "grad_norm": 0.3779100179672241,
      "learning_rate": 2e-05,
      "loss": 0.7563,
      "step": 3520
    },
    {
      "epoch": 0.30326460481099654,
      "grad_norm": 0.40496891736984253,
      "learning_rate": 2e-05,
      "loss": 0.781,
      "step": 3530
    },
    {
      "epoch": 0.30412371134020616,
      "grad_norm": 0.382946640253067,
      "learning_rate": 2e-05,
      "loss": 0.7728,
      "step": 3540
    },
    {
      "epoch": 0.30498281786941583,
      "grad_norm": 0.4126337468624115,
      "learning_rate": 2e-05,
      "loss": 0.7727,
      "step": 3550
    },
    {
      "epoch": 0.30584192439862545,
      "grad_norm": 0.38314563035964966,
      "learning_rate": 2e-05,
      "loss": 0.7638,
      "step": 3560
    },
    {
      "epoch": 0.30670103092783507,
      "grad_norm": 0.3995760679244995,
      "learning_rate": 2e-05,
      "loss": 0.795,
      "step": 3570
    },
    {
      "epoch": 0.3075601374570447,
      "grad_norm": 0.40734052658081055,
      "learning_rate": 2e-05,
      "loss": 0.7333,
      "step": 3580
    },
    {
      "epoch": 0.3084192439862543,
      "grad_norm": 0.41933971643447876,
      "learning_rate": 2e-05,
      "loss": 0.7641,
      "step": 3590
    },
    {
      "epoch": 0.30927835051546393,
      "grad_norm": 0.4051254093647003,
      "learning_rate": 2e-05,
      "loss": 0.7494,
      "step": 3600
    },
    {
      "epoch": 0.31013745704467355,
      "grad_norm": 0.40811339020729065,
      "learning_rate": 2e-05,
      "loss": 0.772,
      "step": 3610
    },
    {
      "epoch": 0.31099656357388317,
      "grad_norm": 0.3880476653575897,
      "learning_rate": 2e-05,
      "loss": 0.7377,
      "step": 3620
    },
    {
      "epoch": 0.3118556701030928,
      "grad_norm": 0.4227207601070404,
      "learning_rate": 2e-05,
      "loss": 0.7676,
      "step": 3630
    },
    {
      "epoch": 0.3127147766323024,
      "grad_norm": 0.36887118220329285,
      "learning_rate": 2e-05,
      "loss": 0.7683,
      "step": 3640
    },
    {
      "epoch": 0.313573883161512,
      "grad_norm": 0.3982653319835663,
      "learning_rate": 2e-05,
      "loss": 0.7394,
      "step": 3650
    },
    {
      "epoch": 0.31443298969072164,
      "grad_norm": 0.4001864492893219,
      "learning_rate": 2e-05,
      "loss": 0.7578,
      "step": 3660
    },
    {
      "epoch": 0.31529209621993126,
      "grad_norm": 0.39527979493141174,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 3670
    },
    {
      "epoch": 0.3161512027491409,
      "grad_norm": 0.4183419942855835,
      "learning_rate": 2e-05,
      "loss": 0.7808,
      "step": 3680
    },
    {
      "epoch": 0.3170103092783505,
      "grad_norm": 0.41624853014945984,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 3690
    },
    {
      "epoch": 0.3178694158075601,
      "grad_norm": 0.3790307939052582,
      "learning_rate": 2e-05,
      "loss": 0.7506,
      "step": 3700
    },
    {
      "epoch": 0.31872852233676974,
      "grad_norm": 0.3806716799736023,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 3710
    },
    {
      "epoch": 0.31958762886597936,
      "grad_norm": 0.39428409934043884,
      "learning_rate": 2e-05,
      "loss": 0.7638,
      "step": 3720
    },
    {
      "epoch": 0.32044673539518903,
      "grad_norm": 0.43839022517204285,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 3730
    },
    {
      "epoch": 0.32130584192439865,
      "grad_norm": 0.40026065707206726,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 3740
    },
    {
      "epoch": 0.32216494845360827,
      "grad_norm": 0.4190294146537781,
      "learning_rate": 2e-05,
      "loss": 0.7841,
      "step": 3750
    },
    {
      "epoch": 0.3230240549828179,
      "grad_norm": 0.38157919049263,
      "learning_rate": 2e-05,
      "loss": 0.7627,
      "step": 3760
    },
    {
      "epoch": 0.3238831615120275,
      "grad_norm": 0.408637672662735,
      "learning_rate": 2e-05,
      "loss": 0.7676,
      "step": 3770
    },
    {
      "epoch": 0.3247422680412371,
      "grad_norm": 0.4058610200881958,
      "learning_rate": 2e-05,
      "loss": 0.7519,
      "step": 3780
    },
    {
      "epoch": 0.32560137457044674,
      "grad_norm": 0.47251054644584656,
      "learning_rate": 2e-05,
      "loss": 0.7542,
      "step": 3790
    },
    {
      "epoch": 0.32646048109965636,
      "grad_norm": 0.4172389805316925,
      "learning_rate": 2e-05,
      "loss": 0.7818,
      "step": 3800
    },
    {
      "epoch": 0.327319587628866,
      "grad_norm": 0.4114549160003662,
      "learning_rate": 2e-05,
      "loss": 0.7731,
      "step": 3810
    },
    {
      "epoch": 0.3281786941580756,
      "grad_norm": 0.4147323966026306,
      "learning_rate": 2e-05,
      "loss": 0.8056,
      "step": 3820
    },
    {
      "epoch": 0.3290378006872852,
      "grad_norm": 0.43751680850982666,
      "learning_rate": 2e-05,
      "loss": 0.7631,
      "step": 3830
    },
    {
      "epoch": 0.32989690721649484,
      "grad_norm": 0.3997228741645813,
      "learning_rate": 2e-05,
      "loss": 0.7781,
      "step": 3840
    },
    {
      "epoch": 0.33075601374570446,
      "grad_norm": 0.39175930619239807,
      "learning_rate": 2e-05,
      "loss": 0.7651,
      "step": 3850
    },
    {
      "epoch": 0.3316151202749141,
      "grad_norm": 0.38528063893318176,
      "learning_rate": 2e-05,
      "loss": 0.7554,
      "step": 3860
    },
    {
      "epoch": 0.3324742268041237,
      "grad_norm": 0.38999444246292114,
      "learning_rate": 2e-05,
      "loss": 0.7723,
      "step": 3870
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.3951996862888336,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 3880
    },
    {
      "epoch": 0.33419243986254293,
      "grad_norm": 0.4000983238220215,
      "learning_rate": 2e-05,
      "loss": 0.7865,
      "step": 3890
    },
    {
      "epoch": 0.33505154639175255,
      "grad_norm": 0.47464287281036377,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 3900
    },
    {
      "epoch": 0.33591065292096217,
      "grad_norm": 0.4179074764251709,
      "learning_rate": 2e-05,
      "loss": 0.7352,
      "step": 3910
    },
    {
      "epoch": 0.33676975945017185,
      "grad_norm": 0.3948587477207184,
      "learning_rate": 2e-05,
      "loss": 0.7687,
      "step": 3920
    },
    {
      "epoch": 0.33762886597938147,
      "grad_norm": 0.4247511923313141,
      "learning_rate": 2e-05,
      "loss": 0.7564,
      "step": 3930
    },
    {
      "epoch": 0.3384879725085911,
      "grad_norm": 0.39262357354164124,
      "learning_rate": 2e-05,
      "loss": 0.7518,
      "step": 3940
    },
    {
      "epoch": 0.3393470790378007,
      "grad_norm": 0.40266096591949463,
      "learning_rate": 2e-05,
      "loss": 0.7515,
      "step": 3950
    },
    {
      "epoch": 0.3402061855670103,
      "grad_norm": 0.4001063108444214,
      "learning_rate": 2e-05,
      "loss": 0.7494,
      "step": 3960
    },
    {
      "epoch": 0.34106529209621994,
      "grad_norm": 0.3866184651851654,
      "learning_rate": 2e-05,
      "loss": 0.7547,
      "step": 3970
    },
    {
      "epoch": 0.34192439862542956,
      "grad_norm": 0.38115713000297546,
      "learning_rate": 2e-05,
      "loss": 0.7501,
      "step": 3980
    },
    {
      "epoch": 0.3427835051546392,
      "grad_norm": 0.4157053232192993,
      "learning_rate": 2e-05,
      "loss": 0.759,
      "step": 3990
    },
    {
      "epoch": 0.3436426116838488,
      "grad_norm": 0.40975072979927063,
      "learning_rate": 2e-05,
      "loss": 0.7584,
      "step": 4000
    },
    {
      "epoch": 0.3445017182130584,
      "grad_norm": 0.40007585287094116,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 4010
    },
    {
      "epoch": 0.34536082474226804,
      "grad_norm": 0.43180954456329346,
      "learning_rate": 2e-05,
      "loss": 0.7479,
      "step": 4020
    },
    {
      "epoch": 0.34621993127147765,
      "grad_norm": 0.42279163002967834,
      "learning_rate": 2e-05,
      "loss": 0.7318,
      "step": 4030
    },
    {
      "epoch": 0.3470790378006873,
      "grad_norm": 0.3827601969242096,
      "learning_rate": 2e-05,
      "loss": 0.7591,
      "step": 4040
    },
    {
      "epoch": 0.3479381443298969,
      "grad_norm": 0.4175439476966858,
      "learning_rate": 2e-05,
      "loss": 0.7446,
      "step": 4050
    },
    {
      "epoch": 0.3487972508591065,
      "grad_norm": 0.40465012192726135,
      "learning_rate": 2e-05,
      "loss": 0.7611,
      "step": 4060
    },
    {
      "epoch": 0.34965635738831613,
      "grad_norm": 0.4096953868865967,
      "learning_rate": 2e-05,
      "loss": 0.7598,
      "step": 4070
    },
    {
      "epoch": 0.35051546391752575,
      "grad_norm": 0.4131430983543396,
      "learning_rate": 2e-05,
      "loss": 0.7869,
      "step": 4080
    },
    {
      "epoch": 0.35137457044673537,
      "grad_norm": 0.4112975001335144,
      "learning_rate": 2e-05,
      "loss": 0.7799,
      "step": 4090
    },
    {
      "epoch": 0.35223367697594504,
      "grad_norm": 0.4253504276275635,
      "learning_rate": 2e-05,
      "loss": 0.7339,
      "step": 4100
    },
    {
      "epoch": 0.35309278350515466,
      "grad_norm": 0.4268980920314789,
      "learning_rate": 2e-05,
      "loss": 0.7537,
      "step": 4110
    },
    {
      "epoch": 0.3539518900343643,
      "grad_norm": 0.38124239444732666,
      "learning_rate": 2e-05,
      "loss": 0.7729,
      "step": 4120
    },
    {
      "epoch": 0.3548109965635739,
      "grad_norm": 0.37831252813339233,
      "learning_rate": 2e-05,
      "loss": 0.7397,
      "step": 4130
    },
    {
      "epoch": 0.3556701030927835,
      "grad_norm": 0.41880711913108826,
      "learning_rate": 2e-05,
      "loss": 0.7757,
      "step": 4140
    },
    {
      "epoch": 0.35652920962199314,
      "grad_norm": 0.43672290444374084,
      "learning_rate": 2e-05,
      "loss": 0.7425,
      "step": 4150
    },
    {
      "epoch": 0.35738831615120276,
      "grad_norm": 0.4058953523635864,
      "learning_rate": 2e-05,
      "loss": 0.7309,
      "step": 4160
    },
    {
      "epoch": 0.3582474226804124,
      "grad_norm": 0.39708343148231506,
      "learning_rate": 2e-05,
      "loss": 0.755,
      "step": 4170
    },
    {
      "epoch": 0.359106529209622,
      "grad_norm": 0.40715470910072327,
      "learning_rate": 2e-05,
      "loss": 0.7618,
      "step": 4180
    },
    {
      "epoch": 0.3599656357388316,
      "grad_norm": 0.37325555086135864,
      "learning_rate": 2e-05,
      "loss": 0.7381,
      "step": 4190
    },
    {
      "epoch": 0.36082474226804123,
      "grad_norm": 0.49437960982322693,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 4200
    },
    {
      "epoch": 0.36168384879725085,
      "grad_norm": 0.40334492921829224,
      "learning_rate": 2e-05,
      "loss": 0.7775,
      "step": 4210
    },
    {
      "epoch": 0.36254295532646047,
      "grad_norm": 0.42243167757987976,
      "learning_rate": 2e-05,
      "loss": 0.7548,
      "step": 4220
    },
    {
      "epoch": 0.3634020618556701,
      "grad_norm": 0.4097069501876831,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 4230
    },
    {
      "epoch": 0.3642611683848797,
      "grad_norm": 0.40296828746795654,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 4240
    },
    {
      "epoch": 0.3651202749140893,
      "grad_norm": 0.4230409562587738,
      "learning_rate": 2e-05,
      "loss": 0.7513,
      "step": 4250
    },
    {
      "epoch": 0.36597938144329895,
      "grad_norm": 0.4135167896747589,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 4260
    },
    {
      "epoch": 0.36683848797250856,
      "grad_norm": 0.38695311546325684,
      "learning_rate": 2e-05,
      "loss": 0.7546,
      "step": 4270
    },
    {
      "epoch": 0.36769759450171824,
      "grad_norm": 0.4157143533229828,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 4280
    },
    {
      "epoch": 0.36855670103092786,
      "grad_norm": 0.4034111499786377,
      "learning_rate": 2e-05,
      "loss": 0.7428,
      "step": 4290
    },
    {
      "epoch": 0.3694158075601375,
      "grad_norm": 0.42762601375579834,
      "learning_rate": 2e-05,
      "loss": 0.7734,
      "step": 4300
    },
    {
      "epoch": 0.3702749140893471,
      "grad_norm": 0.38261547684669495,
      "learning_rate": 2e-05,
      "loss": 0.7536,
      "step": 4310
    },
    {
      "epoch": 0.3711340206185567,
      "grad_norm": 0.559786319732666,
      "learning_rate": 2e-05,
      "loss": 0.7356,
      "step": 4320
    },
    {
      "epoch": 0.37199312714776633,
      "grad_norm": 0.4105216860771179,
      "learning_rate": 2e-05,
      "loss": 0.7488,
      "step": 4330
    },
    {
      "epoch": 0.37285223367697595,
      "grad_norm": 0.4059801995754242,
      "learning_rate": 2e-05,
      "loss": 0.7527,
      "step": 4340
    },
    {
      "epoch": 0.37371134020618557,
      "grad_norm": 0.3986341953277588,
      "learning_rate": 2e-05,
      "loss": 0.7524,
      "step": 4350
    },
    {
      "epoch": 0.3745704467353952,
      "grad_norm": 0.38521715998649597,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 4360
    },
    {
      "epoch": 0.3754295532646048,
      "grad_norm": 0.40247276425361633,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 4370
    },
    {
      "epoch": 0.37628865979381443,
      "grad_norm": 0.4033831059932709,
      "learning_rate": 2e-05,
      "loss": 0.753,
      "step": 4380
    },
    {
      "epoch": 0.37714776632302405,
      "grad_norm": 0.3811618387699127,
      "learning_rate": 2e-05,
      "loss": 0.742,
      "step": 4390
    },
    {
      "epoch": 0.37800687285223367,
      "grad_norm": 0.38687312602996826,
      "learning_rate": 2e-05,
      "loss": 0.7357,
      "step": 4400
    },
    {
      "epoch": 0.3788659793814433,
      "grad_norm": 0.39860835671424866,
      "learning_rate": 2e-05,
      "loss": 0.771,
      "step": 4410
    },
    {
      "epoch": 0.3797250859106529,
      "grad_norm": 0.3995939791202545,
      "learning_rate": 2e-05,
      "loss": 0.7678,
      "step": 4420
    },
    {
      "epoch": 0.3805841924398625,
      "grad_norm": 0.390959769487381,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 4430
    },
    {
      "epoch": 0.38144329896907214,
      "grad_norm": 0.39426323771476746,
      "learning_rate": 2e-05,
      "loss": 0.7513,
      "step": 4440
    },
    {
      "epoch": 0.38230240549828176,
      "grad_norm": 0.3981815278530121,
      "learning_rate": 2e-05,
      "loss": 0.7519,
      "step": 4450
    },
    {
      "epoch": 0.38316151202749144,
      "grad_norm": 0.3980715870857239,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 4460
    },
    {
      "epoch": 0.38402061855670105,
      "grad_norm": 0.3965037167072296,
      "learning_rate": 2e-05,
      "loss": 0.7469,
      "step": 4470
    },
    {
      "epoch": 0.3848797250859107,
      "grad_norm": 0.377102792263031,
      "learning_rate": 2e-05,
      "loss": 0.7681,
      "step": 4480
    },
    {
      "epoch": 0.3857388316151203,
      "grad_norm": 0.40167808532714844,
      "learning_rate": 2e-05,
      "loss": 0.7422,
      "step": 4490
    },
    {
      "epoch": 0.3865979381443299,
      "grad_norm": 0.3795527517795563,
      "learning_rate": 2e-05,
      "loss": 0.7476,
      "step": 4500
    },
    {
      "epoch": 0.38745704467353953,
      "grad_norm": 0.39987045526504517,
      "learning_rate": 2e-05,
      "loss": 0.748,
      "step": 4510
    },
    {
      "epoch": 0.38831615120274915,
      "grad_norm": 0.41090428829193115,
      "learning_rate": 2e-05,
      "loss": 0.7552,
      "step": 4520
    },
    {
      "epoch": 0.38917525773195877,
      "grad_norm": 0.39889150857925415,
      "learning_rate": 2e-05,
      "loss": 0.7397,
      "step": 4530
    },
    {
      "epoch": 0.3900343642611684,
      "grad_norm": 0.394302636384964,
      "learning_rate": 2e-05,
      "loss": 0.7651,
      "step": 4540
    },
    {
      "epoch": 0.390893470790378,
      "grad_norm": 0.39259299635887146,
      "learning_rate": 2e-05,
      "loss": 0.7599,
      "step": 4550
    },
    {
      "epoch": 0.3917525773195876,
      "grad_norm": 0.4073551297187805,
      "learning_rate": 2e-05,
      "loss": 0.758,
      "step": 4560
    },
    {
      "epoch": 0.39261168384879724,
      "grad_norm": 0.4444946050643921,
      "learning_rate": 2e-05,
      "loss": 0.7486,
      "step": 4570
    },
    {
      "epoch": 0.39347079037800686,
      "grad_norm": 0.40218037366867065,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 4580
    },
    {
      "epoch": 0.3943298969072165,
      "grad_norm": 0.4302692115306854,
      "learning_rate": 2e-05,
      "loss": 0.7506,
      "step": 4590
    },
    {
      "epoch": 0.3951890034364261,
      "grad_norm": 0.41661903262138367,
      "learning_rate": 2e-05,
      "loss": 0.754,
      "step": 4600
    },
    {
      "epoch": 0.3960481099656357,
      "grad_norm": 0.40314728021621704,
      "learning_rate": 2e-05,
      "loss": 0.7241,
      "step": 4610
    },
    {
      "epoch": 0.39690721649484534,
      "grad_norm": 0.388095498085022,
      "learning_rate": 2e-05,
      "loss": 0.7596,
      "step": 4620
    },
    {
      "epoch": 0.39776632302405496,
      "grad_norm": 0.3835248053073883,
      "learning_rate": 2e-05,
      "loss": 0.7593,
      "step": 4630
    },
    {
      "epoch": 0.39862542955326463,
      "grad_norm": 0.38472631573677063,
      "learning_rate": 2e-05,
      "loss": 0.7353,
      "step": 4640
    },
    {
      "epoch": 0.39948453608247425,
      "grad_norm": 0.3875240087509155,
      "learning_rate": 2e-05,
      "loss": 0.786,
      "step": 4650
    },
    {
      "epoch": 0.40034364261168387,
      "grad_norm": 0.44080010056495667,
      "learning_rate": 2e-05,
      "loss": 0.7499,
      "step": 4660
    },
    {
      "epoch": 0.4012027491408935,
      "grad_norm": 0.41625672578811646,
      "learning_rate": 2e-05,
      "loss": 0.7563,
      "step": 4670
    },
    {
      "epoch": 0.4020618556701031,
      "grad_norm": 0.4100339412689209,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 4680
    },
    {
      "epoch": 0.4029209621993127,
      "grad_norm": 0.40571466088294983,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 4690
    },
    {
      "epoch": 0.40378006872852235,
      "grad_norm": 0.4143446087837219,
      "learning_rate": 2e-05,
      "loss": 0.7568,
      "step": 4700
    },
    {
      "epoch": 0.40463917525773196,
      "grad_norm": 0.44597920775413513,
      "learning_rate": 2e-05,
      "loss": 0.7653,
      "step": 4710
    },
    {
      "epoch": 0.4054982817869416,
      "grad_norm": 0.3949943780899048,
      "learning_rate": 2e-05,
      "loss": 0.73,
      "step": 4720
    },
    {
      "epoch": 0.4063573883161512,
      "grad_norm": 0.38090425729751587,
      "learning_rate": 2e-05,
      "loss": 0.7606,
      "step": 4730
    },
    {
      "epoch": 0.4072164948453608,
      "grad_norm": 0.4613902270793915,
      "learning_rate": 2e-05,
      "loss": 0.7535,
      "step": 4740
    },
    {
      "epoch": 0.40807560137457044,
      "grad_norm": 0.4230399429798126,
      "learning_rate": 2e-05,
      "loss": 0.7563,
      "step": 4750
    },
    {
      "epoch": 0.40893470790378006,
      "grad_norm": 0.38796481490135193,
      "learning_rate": 2e-05,
      "loss": 0.7077,
      "step": 4760
    },
    {
      "epoch": 0.4097938144329897,
      "grad_norm": 0.4022347629070282,
      "learning_rate": 2e-05,
      "loss": 0.7531,
      "step": 4770
    },
    {
      "epoch": 0.4106529209621993,
      "grad_norm": 0.4280019700527191,
      "learning_rate": 2e-05,
      "loss": 0.7438,
      "step": 4780
    },
    {
      "epoch": 0.4115120274914089,
      "grad_norm": 0.39707037806510925,
      "learning_rate": 2e-05,
      "loss": 0.7514,
      "step": 4790
    },
    {
      "epoch": 0.41237113402061853,
      "grad_norm": 0.39433586597442627,
      "learning_rate": 2e-05,
      "loss": 0.7213,
      "step": 4800
    },
    {
      "epoch": 0.41323024054982815,
      "grad_norm": 0.42996808886528015,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 4810
    },
    {
      "epoch": 0.41408934707903783,
      "grad_norm": 0.40540334582328796,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 4820
    },
    {
      "epoch": 0.41494845360824745,
      "grad_norm": 0.4186977744102478,
      "learning_rate": 2e-05,
      "loss": 0.7855,
      "step": 4830
    },
    {
      "epoch": 0.41580756013745707,
      "grad_norm": 0.4322900176048279,
      "learning_rate": 2e-05,
      "loss": 0.7456,
      "step": 4840
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 0.572149395942688,
      "learning_rate": 2e-05,
      "loss": 0.7579,
      "step": 4850
    },
    {
      "epoch": 0.4175257731958763,
      "grad_norm": 0.37685662508010864,
      "learning_rate": 2e-05,
      "loss": 0.7379,
      "step": 4860
    },
    {
      "epoch": 0.4183848797250859,
      "grad_norm": 0.46816933155059814,
      "learning_rate": 2e-05,
      "loss": 0.7356,
      "step": 4870
    },
    {
      "epoch": 0.41924398625429554,
      "grad_norm": 0.4115220308303833,
      "learning_rate": 2e-05,
      "loss": 0.7605,
      "step": 4880
    },
    {
      "epoch": 0.42010309278350516,
      "grad_norm": 0.40049082040786743,
      "learning_rate": 2e-05,
      "loss": 0.7376,
      "step": 4890
    },
    {
      "epoch": 0.4209621993127148,
      "grad_norm": 0.4084942936897278,
      "learning_rate": 2e-05,
      "loss": 0.787,
      "step": 4900
    },
    {
      "epoch": 0.4218213058419244,
      "grad_norm": 0.40400949120521545,
      "learning_rate": 2e-05,
      "loss": 0.7686,
      "step": 4910
    },
    {
      "epoch": 0.422680412371134,
      "grad_norm": 0.3788411021232605,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 4920
    },
    {
      "epoch": 0.42353951890034364,
      "grad_norm": 0.4002997577190399,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 4930
    },
    {
      "epoch": 0.42439862542955326,
      "grad_norm": 0.42359012365341187,
      "learning_rate": 2e-05,
      "loss": 0.7486,
      "step": 4940
    },
    {
      "epoch": 0.4252577319587629,
      "grad_norm": 0.4005122184753418,
      "learning_rate": 2e-05,
      "loss": 0.7424,
      "step": 4950
    },
    {
      "epoch": 0.4261168384879725,
      "grad_norm": 0.40040847659111023,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 4960
    },
    {
      "epoch": 0.4269759450171821,
      "grad_norm": 0.3970852196216583,
      "learning_rate": 2e-05,
      "loss": 0.7524,
      "step": 4970
    },
    {
      "epoch": 0.42783505154639173,
      "grad_norm": 0.4006688892841339,
      "learning_rate": 2e-05,
      "loss": 0.7419,
      "step": 4980
    },
    {
      "epoch": 0.42869415807560135,
      "grad_norm": 0.3915325999259949,
      "learning_rate": 2e-05,
      "loss": 0.7512,
      "step": 4990
    },
    {
      "epoch": 0.42955326460481097,
      "grad_norm": 0.43504029512405396,
      "learning_rate": 2e-05,
      "loss": 0.7499,
      "step": 5000
    },
    {
      "epoch": 0.43041237113402064,
      "grad_norm": 0.4006860852241516,
      "learning_rate": 2e-05,
      "loss": 0.7722,
      "step": 5010
    },
    {
      "epoch": 0.43127147766323026,
      "grad_norm": 0.42165443301200867,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 5020
    },
    {
      "epoch": 0.4321305841924399,
      "grad_norm": 0.39642852544784546,
      "learning_rate": 2e-05,
      "loss": 0.7544,
      "step": 5030
    },
    {
      "epoch": 0.4329896907216495,
      "grad_norm": 0.3671112656593323,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 5040
    },
    {
      "epoch": 0.4338487972508591,
      "grad_norm": 0.40421435236930847,
      "learning_rate": 2e-05,
      "loss": 0.7248,
      "step": 5050
    },
    {
      "epoch": 0.43470790378006874,
      "grad_norm": 0.3971313536167145,
      "learning_rate": 2e-05,
      "loss": 0.795,
      "step": 5060
    },
    {
      "epoch": 0.43556701030927836,
      "grad_norm": 0.407399445772171,
      "learning_rate": 2e-05,
      "loss": 0.7299,
      "step": 5070
    },
    {
      "epoch": 0.436426116838488,
      "grad_norm": 0.39452338218688965,
      "learning_rate": 2e-05,
      "loss": 0.7574,
      "step": 5080
    },
    {
      "epoch": 0.4372852233676976,
      "grad_norm": 0.4186619818210602,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.7799,
      "step": 5090
    },
    {
      "epoch": 0.4381443298969072,
      "grad_norm": 0.3862781524658203,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.7308,
      "step": 5100
    },
    {
      "epoch": 0.43900343642611683,
      "grad_norm": 0.3767087161540985,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.7321,
      "step": 5110
    },
    {
      "epoch": 0.43986254295532645,
      "grad_norm": 0.4062947630882263,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.7456,
      "step": 5120
    },
    {
      "epoch": 0.44072164948453607,
      "grad_norm": 0.4075610339641571,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.7509,
      "step": 5130
    },
    {
      "epoch": 0.4415807560137457,
      "grad_norm": 0.37007495760917664,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.7297,
      "step": 5140
    },
    {
      "epoch": 0.4424398625429553,
      "grad_norm": 0.43931183218955994,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.7577,
      "step": 5150
    },
    {
      "epoch": 0.44329896907216493,
      "grad_norm": 0.40736839175224304,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.7574,
      "step": 5160
    },
    {
      "epoch": 0.44415807560137455,
      "grad_norm": 0.3988030254840851,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.7436,
      "step": 5170
    },
    {
      "epoch": 0.44501718213058417,
      "grad_norm": 0.3912084400653839,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.7471,
      "step": 5180
    },
    {
      "epoch": 0.44587628865979384,
      "grad_norm": 0.42972642183303833,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.7465,
      "step": 5190
    },
    {
      "epoch": 0.44673539518900346,
      "grad_norm": 0.39708760380744934,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.7323,
      "step": 5200
    },
    {
      "epoch": 0.4475945017182131,
      "grad_norm": 0.3852071464061737,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.7224,
      "step": 5210
    },
    {
      "epoch": 0.4484536082474227,
      "grad_norm": 0.3795469403266907,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.7641,
      "step": 5220
    },
    {
      "epoch": 0.4493127147766323,
      "grad_norm": 0.4086882472038269,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.7327,
      "step": 5230
    },
    {
      "epoch": 0.45017182130584193,
      "grad_norm": 0.3924045264720917,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.7667,
      "step": 5240
    },
    {
      "epoch": 0.45103092783505155,
      "grad_norm": 0.39131811261177063,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.7531,
      "step": 5250
    },
    {
      "epoch": 0.4518900343642612,
      "grad_norm": 0.41451042890548706,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.7484,
      "step": 5260
    },
    {
      "epoch": 0.4527491408934708,
      "grad_norm": 0.3818811774253845,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.7043,
      "step": 5270
    },
    {
      "epoch": 0.4536082474226804,
      "grad_norm": 0.3967249095439911,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.7308,
      "step": 5280
    },
    {
      "epoch": 0.45446735395189003,
      "grad_norm": 0.380952924489975,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.7344,
      "step": 5290
    },
    {
      "epoch": 0.45532646048109965,
      "grad_norm": 0.37272271513938904,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.7463,
      "step": 5300
    },
    {
      "epoch": 0.45618556701030927,
      "grad_norm": 0.37142452597618103,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.7416,
      "step": 5310
    },
    {
      "epoch": 0.4570446735395189,
      "grad_norm": 0.36679983139038086,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.7411,
      "step": 5320
    },
    {
      "epoch": 0.4579037800687285,
      "grad_norm": 0.40842485427856445,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.7578,
      "step": 5330
    },
    {
      "epoch": 0.4587628865979381,
      "grad_norm": 0.38203635811805725,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.7369,
      "step": 5340
    },
    {
      "epoch": 0.45962199312714774,
      "grad_norm": 0.41601574420928955,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.7198,
      "step": 5350
    },
    {
      "epoch": 0.46048109965635736,
      "grad_norm": 0.36368340253829956,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.7484,
      "step": 5360
    },
    {
      "epoch": 0.46134020618556704,
      "grad_norm": 0.3763279616832733,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.7313,
      "step": 5370
    },
    {
      "epoch": 0.46219931271477666,
      "grad_norm": 0.3989553451538086,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.7271,
      "step": 5380
    },
    {
      "epoch": 0.4630584192439863,
      "grad_norm": 0.36512890458106995,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.7242,
      "step": 5390
    },
    {
      "epoch": 0.4639175257731959,
      "grad_norm": 0.3865336775779724,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.7304,
      "step": 5400
    },
    {
      "epoch": 0.4647766323024055,
      "grad_norm": 0.3916051983833313,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.7432,
      "step": 5410
    },
    {
      "epoch": 0.46563573883161513,
      "grad_norm": 0.3612634837627411,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.7238,
      "step": 5420
    },
    {
      "epoch": 0.46649484536082475,
      "grad_norm": 0.357876718044281,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.7263,
      "step": 5430
    },
    {
      "epoch": 0.46735395189003437,
      "grad_norm": 0.37259751558303833,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.7396,
      "step": 5440
    },
    {
      "epoch": 0.468213058419244,
      "grad_norm": 0.3941054046154022,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.7171,
      "step": 5450
    },
    {
      "epoch": 0.4690721649484536,
      "grad_norm": 0.4026050567626953,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.7376,
      "step": 5460
    },
    {
      "epoch": 0.4699312714776632,
      "grad_norm": 0.3972621560096741,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.7168,
      "step": 5470
    },
    {
      "epoch": 0.47079037800687284,
      "grad_norm": 0.3538913428783417,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.7454,
      "step": 5480
    },
    {
      "epoch": 0.47164948453608246,
      "grad_norm": 0.37729233503341675,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.7244,
      "step": 5490
    },
    {
      "epoch": 0.4725085910652921,
      "grad_norm": 0.36948126554489136,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.7151,
      "step": 5500
    },
    {
      "epoch": 0.4733676975945017,
      "grad_norm": 0.41187506914138794,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.7327,
      "step": 5510
    },
    {
      "epoch": 0.4742268041237113,
      "grad_norm": 0.3823601007461548,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.7593,
      "step": 5520
    },
    {
      "epoch": 0.47508591065292094,
      "grad_norm": 0.4354175925254822,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.7443,
      "step": 5530
    },
    {
      "epoch": 0.47594501718213056,
      "grad_norm": 0.9211698770523071,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.7271,
      "step": 5540
    },
    {
      "epoch": 0.47680412371134023,
      "grad_norm": 0.36963316798210144,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.752,
      "step": 5550
    },
    {
      "epoch": 0.47766323024054985,
      "grad_norm": 0.35199132561683655,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.7203,
      "step": 5560
    },
    {
      "epoch": 0.47852233676975947,
      "grad_norm": 0.3772638142108917,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.7101,
      "step": 5570
    },
    {
      "epoch": 0.4793814432989691,
      "grad_norm": 0.36131879687309265,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6895,
      "step": 5580
    },
    {
      "epoch": 0.4802405498281787,
      "grad_norm": 0.3838633596897125,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.721,
      "step": 5590
    },
    {
      "epoch": 0.48084192439862544,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.02480831771572509,
      "train_runtime": 1776.8274,
      "train_samples_per_second": 37.8,
      "train_steps_per_second": 3.15
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}