{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9425684210526316,
  "eval_steps": 500,
  "global_step": 11193,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008421052631578948,
      "grad_norm": 1.6452618837356567,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 0.8848,
      "step": 10
    },
    {
      "epoch": 0.0016842105263157896,
      "grad_norm": 1.2745853662490845,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 0.8978,
      "step": 20
    },
    {
      "epoch": 0.0025263157894736842,
      "grad_norm": 1.1237419843673706,
      "learning_rate": 3.342134831460675e-06,
      "loss": 0.8238,
      "step": 30
    },
    {
      "epoch": 0.003368421052631579,
      "grad_norm": 1.066956877708435,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 0.7917,
      "step": 40
    },
    {
      "epoch": 0.004210526315789474,
      "grad_norm": 0.8702846765518188,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.7807,
      "step": 50
    },
    {
      "epoch": 0.0050526315789473685,
      "grad_norm": 0.7245010733604431,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.7296,
      "step": 60
    },
    {
      "epoch": 0.005894736842105263,
      "grad_norm": 0.7846991419792175,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.735,
      "step": 70
    },
    {
      "epoch": 0.006736842105263158,
      "grad_norm": 0.8695846796035767,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.7109,
      "step": 80
    },
    {
      "epoch": 0.007578947368421052,
      "grad_norm": 0.7376852035522461,
      "learning_rate": 1.005e-05,
      "loss": 0.7221,
      "step": 90
    },
    {
      "epoch": 0.008421052631578947,
      "grad_norm": 0.877524197101593,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.7048,
      "step": 100
    },
    {
      "epoch": 0.009263157894736843,
      "grad_norm": 0.7281895875930786,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.7126,
      "step": 110
    },
    {
      "epoch": 0.010105263157894737,
      "grad_norm": 0.8136023283004761,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.6877,
      "step": 120
    },
    {
      "epoch": 0.010947368421052631,
      "grad_norm": 0.8643974661827087,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.7093,
      "step": 130
    },
    {
      "epoch": 0.011789473684210527,
      "grad_norm": 0.787255585193634,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.6981,
      "step": 140
    },
    {
      "epoch": 0.01263157894736842,
      "grad_norm": 0.8877604603767395,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.7027,
      "step": 150
    },
    {
      "epoch": 0.013473684210526317,
      "grad_norm": 0.8588012456893921,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.6974,
      "step": 160
    },
    {
      "epoch": 0.01431578947368421,
      "grad_norm": 0.737384557723999,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.7082,
      "step": 170
    },
    {
      "epoch": 0.015157894736842105,
      "grad_norm": 0.8280077576637268,
      "learning_rate": 2e-05,
      "loss": 0.7238,
      "step": 180
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.8020002841949463,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 190
    },
    {
      "epoch": 0.016842105263157894,
      "grad_norm": 0.7412464618682861,
      "learning_rate": 2e-05,
      "loss": 0.683,
      "step": 200
    },
    {
      "epoch": 0.01768421052631579,
      "grad_norm": 0.6804697513580322,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 210
    },
    {
      "epoch": 0.018526315789473686,
      "grad_norm": 0.6289499402046204,
      "learning_rate": 2e-05,
      "loss": 0.6937,
      "step": 220
    },
    {
      "epoch": 0.01936842105263158,
      "grad_norm": 0.7443152070045471,
      "learning_rate": 2e-05,
      "loss": 0.6954,
      "step": 230
    },
    {
      "epoch": 0.020210526315789474,
      "grad_norm": 0.5574337244033813,
      "learning_rate": 2e-05,
      "loss": 0.6995,
      "step": 240
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 0.5961909294128418,
      "learning_rate": 2e-05,
      "loss": 0.6741,
      "step": 250
    },
    {
      "epoch": 0.021894736842105262,
      "grad_norm": 0.623269259929657,
      "learning_rate": 2e-05,
      "loss": 0.6884,
      "step": 260
    },
    {
      "epoch": 0.02273684210526316,
      "grad_norm": 0.6100159287452698,
      "learning_rate": 2e-05,
      "loss": 0.6802,
      "step": 270
    },
    {
      "epoch": 0.023578947368421053,
      "grad_norm": 0.5976904034614563,
      "learning_rate": 2e-05,
      "loss": 0.6818,
      "step": 280
    },
    {
      "epoch": 0.024421052631578948,
      "grad_norm": 0.5483630895614624,
      "learning_rate": 2e-05,
      "loss": 0.6959,
      "step": 290
    },
    {
      "epoch": 0.02526315789473684,
      "grad_norm": 0.5604531764984131,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 300
    },
    {
      "epoch": 0.026105263157894736,
      "grad_norm": 0.7242655158042908,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 310
    },
    {
      "epoch": 0.026947368421052633,
      "grad_norm": 0.6082198619842529,
      "learning_rate": 2e-05,
      "loss": 0.7195,
      "step": 320
    },
    {
      "epoch": 0.027789473684210527,
      "grad_norm": 0.6409816145896912,
      "learning_rate": 2e-05,
      "loss": 0.7018,
      "step": 330
    },
    {
      "epoch": 0.02863157894736842,
      "grad_norm": 0.5886685848236084,
      "learning_rate": 2e-05,
      "loss": 0.6722,
      "step": 340
    },
    {
      "epoch": 0.029473684210526315,
      "grad_norm": 0.5524492859840393,
      "learning_rate": 2e-05,
      "loss": 0.6978,
      "step": 350
    },
    {
      "epoch": 0.03031578947368421,
      "grad_norm": 0.5257862210273743,
      "learning_rate": 2e-05,
      "loss": 0.6724,
      "step": 360
    },
    {
      "epoch": 0.031157894736842107,
      "grad_norm": 0.4827383756637573,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 370
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.511551558971405,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 380
    },
    {
      "epoch": 0.032842105263157895,
      "grad_norm": 0.495875746011734,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 390
    },
    {
      "epoch": 0.03368421052631579,
      "grad_norm": 0.5335220098495483,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 400
    },
    {
      "epoch": 0.03452631578947368,
      "grad_norm": 0.49408501386642456,
      "learning_rate": 2e-05,
      "loss": 0.6896,
      "step": 410
    },
    {
      "epoch": 0.03536842105263158,
      "grad_norm": 0.5984386801719666,
      "learning_rate": 2e-05,
      "loss": 0.6874,
      "step": 420
    },
    {
      "epoch": 0.03621052631578947,
      "grad_norm": 0.5150274634361267,
      "learning_rate": 2e-05,
      "loss": 0.6681,
      "step": 430
    },
    {
      "epoch": 0.03705263157894737,
      "grad_norm": 0.5099483132362366,
      "learning_rate": 2e-05,
      "loss": 0.7014,
      "step": 440
    },
    {
      "epoch": 0.037894736842105266,
      "grad_norm": 0.5528599619865417,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 450
    },
    {
      "epoch": 0.03873684210526316,
      "grad_norm": 0.5252863168716431,
      "learning_rate": 2e-05,
      "loss": 0.6827,
      "step": 460
    },
    {
      "epoch": 0.039578947368421054,
      "grad_norm": 0.5065030455589294,
      "learning_rate": 2e-05,
      "loss": 0.6915,
      "step": 470
    },
    {
      "epoch": 0.04042105263157895,
      "grad_norm": 0.5258328318595886,
      "learning_rate": 2e-05,
      "loss": 0.6889,
      "step": 480
    },
    {
      "epoch": 0.04126315789473684,
      "grad_norm": 0.5116169452667236,
      "learning_rate": 2e-05,
      "loss": 0.68,
      "step": 490
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 0.48490849137306213,
      "learning_rate": 2e-05,
      "loss": 0.6715,
      "step": 500
    },
    {
      "epoch": 0.04294736842105263,
      "grad_norm": 0.5084254145622253,
      "learning_rate": 2e-05,
      "loss": 0.6712,
      "step": 510
    },
    {
      "epoch": 0.043789473684210524,
      "grad_norm": 0.49414074420928955,
      "learning_rate": 2e-05,
      "loss": 0.6802,
      "step": 520
    },
    {
      "epoch": 0.04463157894736842,
      "grad_norm": 0.46595826745033264,
      "learning_rate": 2e-05,
      "loss": 0.6946,
      "step": 530
    },
    {
      "epoch": 0.04547368421052632,
      "grad_norm": 0.5120992064476013,
      "learning_rate": 2e-05,
      "loss": 0.6693,
      "step": 540
    },
    {
      "epoch": 0.04631578947368421,
      "grad_norm": 0.5484461784362793,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 550
    },
    {
      "epoch": 0.04715789473684211,
      "grad_norm": 0.5065701603889465,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 560
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.4997612237930298,
      "learning_rate": 2e-05,
      "loss": 0.6826,
      "step": 570
    },
    {
      "epoch": 0.048842105263157895,
      "grad_norm": 0.5135060548782349,
      "learning_rate": 2e-05,
      "loss": 0.6654,
      "step": 580
    },
    {
      "epoch": 0.04968421052631579,
      "grad_norm": 0.49073326587677,
      "learning_rate": 2e-05,
      "loss": 0.6721,
      "step": 590
    },
    {
      "epoch": 0.05052631578947368,
      "grad_norm": 0.4621580243110657,
      "learning_rate": 2e-05,
      "loss": 0.7012,
      "step": 600
    },
    {
      "epoch": 0.05136842105263158,
      "grad_norm": 0.5011141300201416,
      "learning_rate": 2e-05,
      "loss": 0.6714,
      "step": 610
    },
    {
      "epoch": 0.05221052631578947,
      "grad_norm": 0.46268361806869507,
      "learning_rate": 2e-05,
      "loss": 0.6728,
      "step": 620
    },
    {
      "epoch": 0.053052631578947365,
      "grad_norm": 0.4844551086425781,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 630
    },
    {
      "epoch": 0.053894736842105266,
      "grad_norm": 0.4990077018737793,
      "learning_rate": 2e-05,
      "loss": 0.6796,
      "step": 640
    },
    {
      "epoch": 0.05473684210526316,
      "grad_norm": 0.49000492691993713,
      "learning_rate": 2e-05,
      "loss": 0.6845,
      "step": 650
    },
    {
      "epoch": 0.055578947368421054,
      "grad_norm": 0.4686015248298645,
      "learning_rate": 2e-05,
      "loss": 0.6876,
      "step": 660
    },
    {
      "epoch": 0.05642105263157895,
      "grad_norm": 0.5337815284729004,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 670
    },
    {
      "epoch": 0.05726315789473684,
      "grad_norm": 0.49088242650032043,
      "learning_rate": 2e-05,
      "loss": 0.6586,
      "step": 680
    },
    {
      "epoch": 0.058105263157894736,
      "grad_norm": 0.46112194657325745,
      "learning_rate": 2e-05,
      "loss": 0.6704,
      "step": 690
    },
    {
      "epoch": 0.05894736842105263,
      "grad_norm": 0.5169646739959717,
      "learning_rate": 2e-05,
      "loss": 0.6587,
      "step": 700
    },
    {
      "epoch": 0.059789473684210524,
      "grad_norm": 0.4750337600708008,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 710
    },
    {
      "epoch": 0.06063157894736842,
      "grad_norm": 0.47679170966148376,
      "learning_rate": 2e-05,
      "loss": 0.6688,
      "step": 720
    },
    {
      "epoch": 0.06147368421052631,
      "grad_norm": 0.4437170922756195,
      "learning_rate": 2e-05,
      "loss": 0.6641,
      "step": 730
    },
    {
      "epoch": 0.06231578947368421,
      "grad_norm": 0.48904526233673096,
      "learning_rate": 2e-05,
      "loss": 0.6901,
      "step": 740
    },
    {
      "epoch": 0.06315789473684211,
      "grad_norm": 0.47720208764076233,
      "learning_rate": 2e-05,
      "loss": 0.6612,
      "step": 750
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.45092350244522095,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 760
    },
    {
      "epoch": 0.0648421052631579,
      "grad_norm": 0.5482639670372009,
      "learning_rate": 2e-05,
      "loss": 0.6644,
      "step": 770
    },
    {
      "epoch": 0.06568421052631579,
      "grad_norm": 0.47330716252326965,
      "learning_rate": 2e-05,
      "loss": 0.6659,
      "step": 780
    },
    {
      "epoch": 0.06652631578947368,
      "grad_norm": 0.4877111613750458,
      "learning_rate": 2e-05,
      "loss": 0.682,
      "step": 790
    },
    {
      "epoch": 0.06736842105263158,
      "grad_norm": 0.4415607154369354,
      "learning_rate": 2e-05,
      "loss": 0.6642,
      "step": 800
    },
    {
      "epoch": 0.06821052631578947,
      "grad_norm": 0.5016757249832153,
      "learning_rate": 2e-05,
      "loss": 0.6929,
      "step": 810
    },
    {
      "epoch": 0.06905263157894737,
      "grad_norm": 0.44279953837394714,
      "learning_rate": 2e-05,
      "loss": 0.665,
      "step": 820
    },
    {
      "epoch": 0.06989473684210526,
      "grad_norm": 0.46184229850769043,
      "learning_rate": 2e-05,
      "loss": 0.6689,
      "step": 830
    },
    {
      "epoch": 0.07073684210526315,
      "grad_norm": 0.4588353633880615,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 840
    },
    {
      "epoch": 0.07157894736842105,
      "grad_norm": 0.46397942304611206,
      "learning_rate": 2e-05,
      "loss": 0.6678,
      "step": 850
    },
    {
      "epoch": 0.07242105263157894,
      "grad_norm": 0.4994259178638458,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 860
    },
    {
      "epoch": 0.07326315789473684,
      "grad_norm": 0.43922510743141174,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 870
    },
    {
      "epoch": 0.07410526315789474,
      "grad_norm": 0.4572885036468506,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 880
    },
    {
      "epoch": 0.07494736842105264,
      "grad_norm": 0.5367586612701416,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 890
    },
    {
      "epoch": 0.07578947368421053,
      "grad_norm": 0.4777562916278839,
      "learning_rate": 2e-05,
      "loss": 0.6713,
      "step": 900
    },
    {
      "epoch": 0.07663157894736843,
      "grad_norm": 0.47622188925743103,
      "learning_rate": 2e-05,
      "loss": 0.6596,
      "step": 910
    },
    {
      "epoch": 0.07747368421052632,
      "grad_norm": 0.46543118357658386,
      "learning_rate": 2e-05,
      "loss": 0.6734,
      "step": 920
    },
    {
      "epoch": 0.07831578947368421,
      "grad_norm": 0.460198312997818,
      "learning_rate": 2e-05,
      "loss": 0.6605,
      "step": 930
    },
    {
      "epoch": 0.07915789473684211,
      "grad_norm": 0.5260334014892578,
      "learning_rate": 2e-05,
      "loss": 0.6579,
      "step": 940
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5725151896476746,
      "learning_rate": 2e-05,
      "loss": 0.6633,
      "step": 950
    },
    {
      "epoch": 0.0808421052631579,
      "grad_norm": 0.4703020751476288,
      "learning_rate": 2e-05,
      "loss": 0.6642,
      "step": 960
    },
    {
      "epoch": 0.08168421052631579,
      "grad_norm": 0.46530941128730774,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 970
    },
    {
      "epoch": 0.08252631578947368,
      "grad_norm": 0.486493855714798,
      "learning_rate": 2e-05,
      "loss": 0.6488,
      "step": 980
    },
    {
      "epoch": 0.08336842105263158,
      "grad_norm": 0.4559321999549866,
      "learning_rate": 2e-05,
      "loss": 0.6543,
      "step": 990
    },
    {
      "epoch": 0.08421052631578947,
      "grad_norm": 0.4368566870689392,
      "learning_rate": 2e-05,
      "loss": 0.6695,
      "step": 1000
    },
    {
      "epoch": 0.08505263157894737,
      "grad_norm": 0.5327503681182861,
      "learning_rate": 2e-05,
      "loss": 0.6673,
      "step": 1010
    },
    {
      "epoch": 0.08589473684210526,
      "grad_norm": 0.4721207618713379,
      "learning_rate": 2e-05,
      "loss": 0.661,
      "step": 1020
    },
    {
      "epoch": 0.08673684210526315,
      "grad_norm": 0.4939779043197632,
      "learning_rate": 2e-05,
      "loss": 0.661,
      "step": 1030
    },
    {
      "epoch": 0.08757894736842105,
      "grad_norm": 0.5211036205291748,
      "learning_rate": 2e-05,
      "loss": 0.675,
      "step": 1040
    },
    {
      "epoch": 0.08842105263157894,
      "grad_norm": 0.46270859241485596,
      "learning_rate": 2e-05,
      "loss": 0.6489,
      "step": 1050
    },
    {
      "epoch": 0.08926315789473684,
      "grad_norm": 0.47459232807159424,
      "learning_rate": 2e-05,
      "loss": 0.6745,
      "step": 1060
    },
    {
      "epoch": 0.09010526315789473,
      "grad_norm": 0.4822026491165161,
      "learning_rate": 2e-05,
      "loss": 0.6526,
      "step": 1070
    },
    {
      "epoch": 0.09094736842105264,
      "grad_norm": 0.4457903802394867,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 1080
    },
    {
      "epoch": 0.09178947368421053,
      "grad_norm": 0.47300124168395996,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 1090
    },
    {
      "epoch": 0.09263157894736843,
      "grad_norm": 0.4323350191116333,
      "learning_rate": 2e-05,
      "loss": 0.6504,
      "step": 1100
    },
    {
      "epoch": 0.09347368421052632,
      "grad_norm": 0.4604896903038025,
      "learning_rate": 2e-05,
      "loss": 0.6919,
      "step": 1110
    },
    {
      "epoch": 0.09431578947368421,
      "grad_norm": 0.5188210010528564,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 1120
    },
    {
      "epoch": 0.09515789473684211,
      "grad_norm": 0.45725199580192566,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 1130
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.5177894234657288,
      "learning_rate": 2e-05,
      "loss": 0.6717,
      "step": 1140
    },
    {
      "epoch": 0.0968421052631579,
      "grad_norm": 0.4111453592777252,
      "learning_rate": 2e-05,
      "loss": 0.6502,
      "step": 1150
    },
    {
      "epoch": 0.09768421052631579,
      "grad_norm": 0.4603477716445923,
      "learning_rate": 2e-05,
      "loss": 0.6706,
      "step": 1160
    },
    {
      "epoch": 0.09852631578947368,
      "grad_norm": 0.46829044818878174,
      "learning_rate": 2e-05,
      "loss": 0.6777,
      "step": 1170
    },
    {
      "epoch": 0.09936842105263158,
      "grad_norm": 0.4708096981048584,
      "learning_rate": 2e-05,
      "loss": 0.6644,
      "step": 1180
    },
    {
      "epoch": 0.10021052631578947,
      "grad_norm": 0.549617350101471,
      "learning_rate": 2e-05,
      "loss": 0.6651,
      "step": 1190
    },
    {
      "epoch": 0.10105263157894737,
      "grad_norm": 0.4585419297218323,
      "learning_rate": 2e-05,
      "loss": 0.6312,
      "step": 1200
    },
    {
      "epoch": 0.10189473684210526,
      "grad_norm": 0.44110795855522156,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 1210
    },
    {
      "epoch": 0.10273684210526315,
      "grad_norm": 0.43249937891960144,
      "learning_rate": 2e-05,
      "loss": 0.6506,
      "step": 1220
    },
    {
      "epoch": 0.10357894736842105,
      "grad_norm": 1.2289555072784424,
      "learning_rate": 2e-05,
      "loss": 0.6548,
      "step": 1230
    },
    {
      "epoch": 0.10442105263157894,
      "grad_norm": 0.446972131729126,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 1240
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 0.4171697199344635,
      "learning_rate": 2e-05,
      "loss": 0.6726,
      "step": 1250
    },
    {
      "epoch": 0.10610526315789473,
      "grad_norm": 0.5522831678390503,
      "learning_rate": 2e-05,
      "loss": 0.6818,
      "step": 1260
    },
    {
      "epoch": 0.10694736842105264,
      "grad_norm": 0.4587099850177765,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 1270
    },
    {
      "epoch": 0.10778947368421053,
      "grad_norm": 0.458698034286499,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 1280
    },
    {
      "epoch": 0.10863157894736843,
      "grad_norm": 0.4405956566333771,
      "learning_rate": 2e-05,
      "loss": 0.6711,
      "step": 1290
    },
    {
      "epoch": 0.10947368421052632,
      "grad_norm": 0.41542065143585205,
      "learning_rate": 2e-05,
      "loss": 0.6635,
      "step": 1300
    },
    {
      "epoch": 0.11031578947368421,
      "grad_norm": 0.45384955406188965,
      "learning_rate": 2e-05,
      "loss": 0.6456,
      "step": 1310
    },
    {
      "epoch": 0.11115789473684211,
      "grad_norm": 0.4372265636920929,
      "learning_rate": 2e-05,
      "loss": 0.6656,
      "step": 1320
    },
    {
      "epoch": 0.112,
      "grad_norm": 0.4737173318862915,
      "learning_rate": 2e-05,
      "loss": 0.6733,
      "step": 1330
    },
    {
      "epoch": 0.1128421052631579,
      "grad_norm": 0.45308926701545715,
      "learning_rate": 2e-05,
      "loss": 0.6657,
      "step": 1340
    },
    {
      "epoch": 0.11368421052631579,
      "grad_norm": 0.5928835868835449,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 1350
    },
    {
      "epoch": 0.11452631578947368,
      "grad_norm": 0.4407281279563904,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 1360
    },
    {
      "epoch": 0.11536842105263158,
      "grad_norm": 0.4661693871021271,
      "learning_rate": 2e-05,
      "loss": 0.6484,
      "step": 1370
    },
    {
      "epoch": 0.11621052631578947,
      "grad_norm": 0.42691245675086975,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 1380
    },
    {
      "epoch": 0.11705263157894737,
      "grad_norm": 0.47203555703163147,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 1390
    },
    {
      "epoch": 0.11789473684210526,
      "grad_norm": 0.5509248375892639,
      "learning_rate": 2e-05,
      "loss": 0.6564,
      "step": 1400
    },
    {
      "epoch": 0.11873684210526315,
      "grad_norm": 0.4351705312728882,
      "learning_rate": 2e-05,
      "loss": 0.6395,
      "step": 1410
    },
    {
      "epoch": 0.11957894736842105,
      "grad_norm": 0.47608712315559387,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 1420
    },
    {
      "epoch": 0.12042105263157894,
      "grad_norm": 0.4292226731777191,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 1430
    },
    {
      "epoch": 0.12126315789473684,
      "grad_norm": 0.45191818475723267,
      "learning_rate": 2e-05,
      "loss": 0.6883,
      "step": 1440
    },
    {
      "epoch": 0.12210526315789473,
      "grad_norm": 0.4589603543281555,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 1450
    },
    {
      "epoch": 0.12294736842105262,
      "grad_norm": 0.43541625142097473,
      "learning_rate": 2e-05,
      "loss": 0.6486,
      "step": 1460
    },
    {
      "epoch": 0.12378947368421053,
      "grad_norm": 0.445057213306427,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 1470
    },
    {
      "epoch": 0.12463157894736843,
      "grad_norm": 0.46918973326683044,
      "learning_rate": 2e-05,
      "loss": 0.6665,
      "step": 1480
    },
    {
      "epoch": 0.12547368421052632,
      "grad_norm": 0.43344002962112427,
      "learning_rate": 2e-05,
      "loss": 0.6417,
      "step": 1490
    },
    {
      "epoch": 0.12631578947368421,
      "grad_norm": 0.4777977168560028,
      "learning_rate": 2e-05,
      "loss": 0.6682,
      "step": 1500
    },
    {
      "epoch": 0.1271578947368421,
      "grad_norm": 0.5021777153015137,
      "learning_rate": 2e-05,
      "loss": 0.6712,
      "step": 1510
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.4618164300918579,
      "learning_rate": 2e-05,
      "loss": 0.6564,
      "step": 1520
    },
    {
      "epoch": 0.1288421052631579,
      "grad_norm": 0.40703055262565613,
      "learning_rate": 2e-05,
      "loss": 0.6763,
      "step": 1530
    },
    {
      "epoch": 0.1296842105263158,
      "grad_norm": 0.45353758335113525,
      "learning_rate": 2e-05,
      "loss": 0.652,
      "step": 1540
    },
    {
      "epoch": 0.13052631578947368,
      "grad_norm": 0.4652615189552307,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 1550
    },
    {
      "epoch": 0.13136842105263158,
      "grad_norm": 0.4542207419872284,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 1560
    },
    {
      "epoch": 0.13221052631578947,
      "grad_norm": 0.4470004141330719,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 1570
    },
    {
      "epoch": 0.13305263157894737,
      "grad_norm": 0.40870246291160583,
      "learning_rate": 2e-05,
      "loss": 0.6644,
      "step": 1580
    },
    {
      "epoch": 0.13389473684210526,
      "grad_norm": 0.42540448904037476,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 1590
    },
    {
      "epoch": 0.13473684210526315,
      "grad_norm": 0.4614720046520233,
      "learning_rate": 2e-05,
      "loss": 0.6607,
      "step": 1600
    },
    {
      "epoch": 0.13557894736842105,
      "grad_norm": 0.4467135965824127,
      "learning_rate": 2e-05,
      "loss": 0.6814,
      "step": 1610
    },
    {
      "epoch": 0.13642105263157894,
      "grad_norm": 0.4206336438655853,
      "learning_rate": 2e-05,
      "loss": 0.6639,
      "step": 1620
    },
    {
      "epoch": 0.13726315789473684,
      "grad_norm": 0.48927173018455505,
      "learning_rate": 2e-05,
      "loss": 0.6688,
      "step": 1630
    },
    {
      "epoch": 0.13810526315789473,
      "grad_norm": 0.45843756198883057,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 1640
    },
    {
      "epoch": 0.13894736842105262,
      "grad_norm": 0.449156254529953,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 1650
    },
    {
      "epoch": 0.13978947368421052,
      "grad_norm": 0.4332134425640106,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 1660
    },
    {
      "epoch": 0.1406315789473684,
      "grad_norm": 0.42238250374794006,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 1670
    },
    {
      "epoch": 0.1414736842105263,
      "grad_norm": 0.4564463198184967,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 1680
    },
    {
      "epoch": 0.1423157894736842,
      "grad_norm": 0.44483888149261475,
      "learning_rate": 2e-05,
      "loss": 0.6517,
      "step": 1690
    },
    {
      "epoch": 0.1431578947368421,
      "grad_norm": 0.5297313332557678,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 1700
    },
    {
      "epoch": 0.144,
      "grad_norm": 0.40007227659225464,
      "learning_rate": 2e-05,
      "loss": 0.6309,
      "step": 1710
    },
    {
      "epoch": 0.14484210526315788,
      "grad_norm": 0.4941370487213135,
      "learning_rate": 2e-05,
      "loss": 0.6561,
      "step": 1720
    },
    {
      "epoch": 0.14568421052631578,
      "grad_norm": 0.43409863114356995,
      "learning_rate": 2e-05,
      "loss": 0.6369,
      "step": 1730
    },
    {
      "epoch": 0.14652631578947367,
      "grad_norm": 0.44366204738616943,
      "learning_rate": 2e-05,
      "loss": 0.6416,
      "step": 1740
    },
    {
      "epoch": 0.14736842105263157,
      "grad_norm": 0.4011799395084381,
      "learning_rate": 2e-05,
      "loss": 0.6538,
      "step": 1750
    },
    {
      "epoch": 0.1482105263157895,
      "grad_norm": 0.40802836418151855,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 1760
    },
    {
      "epoch": 0.14905263157894738,
      "grad_norm": 0.4368784427642822,
      "learning_rate": 2e-05,
      "loss": 0.67,
      "step": 1770
    },
    {
      "epoch": 0.14989473684210528,
      "grad_norm": 0.4641876816749573,
      "learning_rate": 2e-05,
      "loss": 0.65,
      "step": 1780
    },
    {
      "epoch": 0.15073684210526317,
      "grad_norm": 0.4642079174518585,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 1790
    },
    {
      "epoch": 0.15157894736842106,
      "grad_norm": 0.4088793396949768,
      "learning_rate": 2e-05,
      "loss": 0.6439,
      "step": 1800
    },
    {
      "epoch": 0.15242105263157896,
      "grad_norm": 0.7351284027099609,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 1810
    },
    {
      "epoch": 0.15326315789473685,
      "grad_norm": 0.43712669610977173,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 1820
    },
    {
      "epoch": 0.15410526315789475,
      "grad_norm": 0.4419451951980591,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 1830
    },
    {
      "epoch": 0.15494736842105264,
      "grad_norm": 0.49290475249290466,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 1840
    },
    {
      "epoch": 0.15578947368421053,
      "grad_norm": 0.4438256323337555,
      "learning_rate": 2e-05,
      "loss": 0.6812,
      "step": 1850
    },
    {
      "epoch": 0.15663157894736843,
      "grad_norm": 0.40685737133026123,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 1860
    },
    {
      "epoch": 0.15747368421052632,
      "grad_norm": 0.49131378531455994,
      "learning_rate": 2e-05,
      "loss": 0.6722,
      "step": 1870
    },
    {
      "epoch": 0.15831578947368422,
      "grad_norm": 0.4030827283859253,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 1880
    },
    {
      "epoch": 0.1591578947368421,
      "grad_norm": 1.2659680843353271,
      "learning_rate": 2e-05,
      "loss": 0.6377,
      "step": 1890
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4396963119506836,
      "learning_rate": 2e-05,
      "loss": 0.6758,
      "step": 1900
    },
    {
      "epoch": 0.1608421052631579,
      "grad_norm": 0.42015787959098816,
      "learning_rate": 2e-05,
      "loss": 0.6465,
      "step": 1910
    },
    {
      "epoch": 0.1616842105263158,
      "grad_norm": 0.3945922553539276,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 1920
    },
    {
      "epoch": 0.16252631578947369,
      "grad_norm": 0.4661765396595001,
      "learning_rate": 2e-05,
      "loss": 0.654,
      "step": 1930
    },
    {
      "epoch": 0.16336842105263158,
      "grad_norm": 0.42668765783309937,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 1940
    },
    {
      "epoch": 0.16421052631578947,
      "grad_norm": 0.42763540148735046,
      "learning_rate": 2e-05,
      "loss": 0.672,
      "step": 1950
    },
    {
      "epoch": 0.16505263157894737,
      "grad_norm": 0.449554443359375,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 1960
    },
    {
      "epoch": 0.16589473684210526,
      "grad_norm": 0.3989042341709137,
      "learning_rate": 2e-05,
      "loss": 0.6293,
      "step": 1970
    },
    {
      "epoch": 0.16673684210526316,
      "grad_norm": 0.3974854648113251,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 1980
    },
    {
      "epoch": 0.16757894736842105,
      "grad_norm": 0.4603552222251892,
      "learning_rate": 2e-05,
      "loss": 0.6417,
      "step": 1990
    },
    {
      "epoch": 0.16842105263157894,
      "grad_norm": 0.42975977063179016,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 2000
    },
    {
      "epoch": 0.16926315789473684,
      "grad_norm": 0.4494156241416931,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 2010
    },
    {
      "epoch": 0.17010526315789473,
      "grad_norm": 0.43503236770629883,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 2020
    },
    {
      "epoch": 0.17094736842105263,
      "grad_norm": 0.4536204934120178,
      "learning_rate": 2e-05,
      "loss": 0.6413,
      "step": 2030
    },
    {
      "epoch": 0.17178947368421052,
      "grad_norm": 0.44027554988861084,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 2040
    },
    {
      "epoch": 0.1726315789473684,
      "grad_norm": 0.43327200412750244,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 2050
    },
    {
      "epoch": 0.1734736842105263,
      "grad_norm": 0.45290684700012207,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 2060
    },
    {
      "epoch": 0.1743157894736842,
      "grad_norm": 0.4675568640232086,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 2070
    },
    {
      "epoch": 0.1751578947368421,
      "grad_norm": 0.4425695836544037,
      "learning_rate": 2e-05,
      "loss": 0.6317,
      "step": 2080
    },
    {
      "epoch": 0.176,
      "grad_norm": 0.42776045203208923,
      "learning_rate": 2e-05,
      "loss": 0.6872,
      "step": 2090
    },
    {
      "epoch": 0.17684210526315788,
      "grad_norm": 0.4114616811275482,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 2100
    },
    {
      "epoch": 0.17768421052631578,
      "grad_norm": 0.41503867506980896,
      "learning_rate": 2e-05,
      "loss": 0.6504,
      "step": 2110
    },
    {
      "epoch": 0.17852631578947367,
      "grad_norm": 0.4861103594303131,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 2120
    },
    {
      "epoch": 0.17936842105263157,
      "grad_norm": 0.512718141078949,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 2130
    },
    {
      "epoch": 0.18021052631578946,
      "grad_norm": 0.45127350091934204,
      "learning_rate": 2e-05,
      "loss": 0.6752,
      "step": 2140
    },
    {
      "epoch": 0.18105263157894738,
      "grad_norm": 0.3995113670825958,
      "learning_rate": 2e-05,
      "loss": 0.6498,
      "step": 2150
    },
    {
      "epoch": 0.18189473684210528,
      "grad_norm": 0.39956554770469666,
      "learning_rate": 2e-05,
      "loss": 0.6284,
      "step": 2160
    },
    {
      "epoch": 0.18273684210526317,
      "grad_norm": 0.5028399229049683,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 2170
    },
    {
      "epoch": 0.18357894736842106,
      "grad_norm": 0.4250260591506958,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 2180
    },
    {
      "epoch": 0.18442105263157896,
      "grad_norm": 0.4396851658821106,
      "learning_rate": 2e-05,
      "loss": 0.6561,
      "step": 2190
    },
    {
      "epoch": 0.18526315789473685,
      "grad_norm": 0.49678802490234375,
      "learning_rate": 2e-05,
      "loss": 0.6744,
      "step": 2200
    },
    {
      "epoch": 0.18610526315789475,
      "grad_norm": 0.49938711524009705,
      "learning_rate": 2e-05,
      "loss": 0.6266,
      "step": 2210
    },
    {
      "epoch": 0.18694736842105264,
      "grad_norm": 0.4113972783088684,
      "learning_rate": 2e-05,
      "loss": 0.6611,
      "step": 2220
    },
    {
      "epoch": 0.18778947368421053,
      "grad_norm": 0.4604727029800415,
      "learning_rate": 2e-05,
      "loss": 0.6476,
      "step": 2230
    },
    {
      "epoch": 0.18863157894736843,
      "grad_norm": 0.39633166790008545,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 2240
    },
    {
      "epoch": 0.18947368421052632,
      "grad_norm": 0.39922624826431274,
      "learning_rate": 2e-05,
      "loss": 0.6544,
      "step": 2250
    },
    {
      "epoch": 0.19031578947368422,
      "grad_norm": 0.4407571852207184,
      "learning_rate": 2e-05,
      "loss": 0.6725,
      "step": 2260
    },
    {
      "epoch": 0.1911578947368421,
      "grad_norm": 0.4127040505409241,
      "learning_rate": 2e-05,
      "loss": 0.6622,
      "step": 2270
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.42205244302749634,
      "learning_rate": 2e-05,
      "loss": 0.6445,
      "step": 2280
    },
    {
      "epoch": 0.1928421052631579,
      "grad_norm": 0.4369538724422455,
      "learning_rate": 2e-05,
      "loss": 0.6544,
      "step": 2290
    },
    {
      "epoch": 0.1936842105263158,
      "grad_norm": 0.46700286865234375,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 2300
    },
    {
      "epoch": 0.19452631578947369,
      "grad_norm": 1.3719345331192017,
      "learning_rate": 2e-05,
      "loss": 0.662,
      "step": 2310
    },
    {
      "epoch": 0.19536842105263158,
      "grad_norm": 0.8232142925262451,
      "learning_rate": 2e-05,
      "loss": 0.6373,
      "step": 2320
    },
    {
      "epoch": 0.19621052631578947,
      "grad_norm": 0.45501574873924255,
      "learning_rate": 2e-05,
      "loss": 0.663,
      "step": 2330
    },
    {
      "epoch": 0.19705263157894737,
      "grad_norm": 0.38035255670547485,
      "learning_rate": 2e-05,
      "loss": 0.6413,
      "step": 2340
    },
    {
      "epoch": 0.19789473684210526,
      "grad_norm": 0.42564475536346436,
      "learning_rate": 2e-05,
      "loss": 0.6566,
      "step": 2350
    },
    {
      "epoch": 0.19873684210526316,
      "grad_norm": 0.42836716771125793,
      "learning_rate": 2e-05,
      "loss": 0.6443,
      "step": 2360
    },
    {
      "epoch": 0.19957894736842105,
      "grad_norm": 0.4486235976219177,
      "learning_rate": 2e-05,
      "loss": 0.6445,
      "step": 2370
    },
    {
      "epoch": 0.20042105263157894,
      "grad_norm": 0.43508684635162354,
      "learning_rate": 2e-05,
      "loss": 0.6485,
      "step": 2380
    },
    {
      "epoch": 0.20126315789473684,
      "grad_norm": 0.43823474645614624,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 2390
    },
    {
      "epoch": 0.20210526315789473,
      "grad_norm": 0.4307081699371338,
      "learning_rate": 2e-05,
      "loss": 0.6383,
      "step": 2400
    },
    {
      "epoch": 0.20294736842105263,
      "grad_norm": 0.45380184054374695,
      "learning_rate": 2e-05,
      "loss": 0.6461,
      "step": 2410
    },
    {
      "epoch": 0.20378947368421052,
      "grad_norm": 0.4245603084564209,
      "learning_rate": 2e-05,
      "loss": 0.6485,
      "step": 2420
    },
    {
      "epoch": 0.20463157894736841,
      "grad_norm": 0.42921018600463867,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 2430
    },
    {
      "epoch": 0.2054736842105263,
      "grad_norm": 0.40832093358039856,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 2440
    },
    {
      "epoch": 0.2063157894736842,
      "grad_norm": 0.40788644552230835,
      "learning_rate": 2e-05,
      "loss": 0.6509,
      "step": 2450
    },
    {
      "epoch": 0.2071578947368421,
      "grad_norm": 0.3845847547054291,
      "learning_rate": 2e-05,
      "loss": 0.6618,
      "step": 2460
    },
    {
      "epoch": 0.208,
      "grad_norm": 0.4283493161201477,
      "learning_rate": 2e-05,
      "loss": 0.6459,
      "step": 2470
    },
    {
      "epoch": 0.20884210526315788,
      "grad_norm": 0.4894337058067322,
      "learning_rate": 2e-05,
      "loss": 0.6477,
      "step": 2480
    },
    {
      "epoch": 0.20968421052631578,
      "grad_norm": 0.5109031200408936,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 2490
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 0.44362136721611023,
      "learning_rate": 2e-05,
      "loss": 0.63,
      "step": 2500
    },
    {
      "epoch": 0.21136842105263157,
      "grad_norm": 0.40784358978271484,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 2510
    },
    {
      "epoch": 0.21221052631578946,
      "grad_norm": 0.37488508224487305,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 2520
    },
    {
      "epoch": 0.21305263157894735,
      "grad_norm": 0.40795764327049255,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 2530
    },
    {
      "epoch": 0.21389473684210528,
      "grad_norm": 0.4658086895942688,
      "learning_rate": 2e-05,
      "loss": 0.6756,
      "step": 2540
    },
    {
      "epoch": 0.21473684210526317,
      "grad_norm": 0.42823630571365356,
      "learning_rate": 2e-05,
      "loss": 0.6718,
      "step": 2550
    },
    {
      "epoch": 0.21557894736842106,
      "grad_norm": 0.4981653094291687,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 2560
    },
    {
      "epoch": 0.21642105263157896,
      "grad_norm": 0.4345998764038086,
      "learning_rate": 2e-05,
      "loss": 0.6336,
      "step": 2570
    },
    {
      "epoch": 0.21726315789473685,
      "grad_norm": 0.39910784363746643,
      "learning_rate": 2e-05,
      "loss": 0.6257,
      "step": 2580
    },
    {
      "epoch": 0.21810526315789475,
      "grad_norm": 0.3906060755252838,
      "learning_rate": 2e-05,
      "loss": 0.633,
      "step": 2590
    },
    {
      "epoch": 0.21894736842105264,
      "grad_norm": 0.40513890981674194,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 2600
    },
    {
      "epoch": 0.21978947368421053,
      "grad_norm": 0.40323129296302795,
      "learning_rate": 2e-05,
      "loss": 0.6448,
      "step": 2610
    },
    {
      "epoch": 0.22063157894736843,
      "grad_norm": 0.4804961085319519,
      "learning_rate": 2e-05,
      "loss": 0.6521,
      "step": 2620
    },
    {
      "epoch": 0.22147368421052632,
      "grad_norm": 0.4444946348667145,
      "learning_rate": 2e-05,
      "loss": 0.6628,
      "step": 2630
    },
    {
      "epoch": 0.22231578947368422,
      "grad_norm": 0.5287010073661804,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 2640
    },
    {
      "epoch": 0.2231578947368421,
      "grad_norm": 0.41825786232948303,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 2650
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.39499178528785706,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 2660
    },
    {
      "epoch": 0.2248421052631579,
      "grad_norm": 0.3989652991294861,
      "learning_rate": 2e-05,
      "loss": 0.6354,
      "step": 2670
    },
    {
      "epoch": 0.2256842105263158,
      "grad_norm": 0.411063015460968,
      "learning_rate": 2e-05,
      "loss": 0.6465,
      "step": 2680
    },
    {
      "epoch": 0.2265263157894737,
      "grad_norm": 0.4129523038864136,
      "learning_rate": 2e-05,
      "loss": 0.6675,
      "step": 2690
    },
    {
      "epoch": 0.22736842105263158,
      "grad_norm": 0.43651139736175537,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 2700
    },
    {
      "epoch": 0.22821052631578947,
      "grad_norm": 0.40520063042640686,
      "learning_rate": 2e-05,
      "loss": 0.6567,
      "step": 2710
    },
    {
      "epoch": 0.22905263157894737,
      "grad_norm": 0.4163479506969452,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 2720
    },
    {
      "epoch": 0.22989473684210526,
      "grad_norm": 0.4244038164615631,
      "learning_rate": 2e-05,
      "loss": 0.6448,
      "step": 2730
    },
    {
      "epoch": 0.23073684210526316,
      "grad_norm": 0.40219080448150635,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 2740
    },
    {
      "epoch": 0.23157894736842105,
      "grad_norm": 0.4042263627052307,
      "learning_rate": 2e-05,
      "loss": 0.6508,
      "step": 2750
    },
    {
      "epoch": 0.23242105263157894,
      "grad_norm": 0.41499000787734985,
      "learning_rate": 2e-05,
      "loss": 0.6791,
      "step": 2760
    },
    {
      "epoch": 0.23326315789473684,
      "grad_norm": 0.39045268297195435,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 2770
    },
    {
      "epoch": 0.23410526315789473,
      "grad_norm": 0.41404953598976135,
      "learning_rate": 2e-05,
      "loss": 0.659,
      "step": 2780
    },
    {
      "epoch": 0.23494736842105263,
      "grad_norm": 0.4036545753479004,
      "learning_rate": 2e-05,
      "loss": 0.6339,
      "step": 2790
    },
    {
      "epoch": 0.23578947368421052,
      "grad_norm": 0.4176270663738251,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 2800
    },
    {
      "epoch": 0.23663157894736842,
      "grad_norm": 0.4281472861766815,
      "learning_rate": 2e-05,
      "loss": 0.6316,
      "step": 2810
    },
    {
      "epoch": 0.2374736842105263,
      "grad_norm": 0.39686036109924316,
      "learning_rate": 2e-05,
      "loss": 0.6486,
      "step": 2820
    },
    {
      "epoch": 0.2383157894736842,
      "grad_norm": 1.6661343574523926,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 2830
    },
    {
      "epoch": 0.2391578947368421,
      "grad_norm": 0.41678765416145325,
      "learning_rate": 2e-05,
      "loss": 0.6561,
      "step": 2840
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.392876535654068,
      "learning_rate": 2e-05,
      "loss": 0.631,
      "step": 2850
    },
    {
      "epoch": 0.24084210526315789,
      "grad_norm": 0.3897719979286194,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 2860
    },
    {
      "epoch": 0.24168421052631578,
      "grad_norm": 0.4358420968055725,
      "learning_rate": 2e-05,
      "loss": 0.6566,
      "step": 2870
    },
    {
      "epoch": 0.24252631578947367,
      "grad_norm": 0.39645540714263916,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 2880
    },
    {
      "epoch": 0.24336842105263157,
      "grad_norm": 0.4347623288631439,
      "learning_rate": 2e-05,
      "loss": 0.6526,
      "step": 2890
    },
    {
      "epoch": 0.24421052631578946,
      "grad_norm": 0.4221624732017517,
      "learning_rate": 2e-05,
      "loss": 0.6608,
      "step": 2900
    },
    {
      "epoch": 0.24505263157894736,
      "grad_norm": 0.41466405987739563,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 2910
    },
    {
      "epoch": 0.24589473684210525,
      "grad_norm": 0.4393434226512909,
      "learning_rate": 2e-05,
      "loss": 0.6584,
      "step": 2920
    },
    {
      "epoch": 0.24673684210526317,
      "grad_norm": 0.38062044978141785,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 2930
    },
    {
      "epoch": 0.24757894736842107,
      "grad_norm": 0.4067496359348297,
      "learning_rate": 2e-05,
      "loss": 0.65,
      "step": 2940
    },
    {
      "epoch": 0.24842105263157896,
      "grad_norm": 0.36574116349220276,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 2950
    },
    {
      "epoch": 0.24926315789473685,
      "grad_norm": 0.47488290071487427,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 2960
    },
    {
      "epoch": 0.2501052631578947,
      "grad_norm": 0.3768024742603302,
      "learning_rate": 2e-05,
      "loss": 0.6425,
      "step": 2970
    },
    {
      "epoch": 0.25094736842105264,
      "grad_norm": 0.47155463695526123,
      "learning_rate": 2e-05,
      "loss": 0.6514,
      "step": 2980
    },
    {
      "epoch": 0.2517894736842105,
      "grad_norm": 0.4444313645362854,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 2990
    },
    {
      "epoch": 0.25263157894736843,
      "grad_norm": 0.41606080532073975,
      "learning_rate": 2e-05,
      "loss": 0.6342,
      "step": 3000
    },
    {
      "epoch": 0.2534736842105263,
      "grad_norm": 0.39055007696151733,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 3010
    },
    {
      "epoch": 0.2543157894736842,
      "grad_norm": 0.4286848306655884,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 3020
    },
    {
      "epoch": 0.2551578947368421,
      "grad_norm": 0.43060189485549927,
      "learning_rate": 2e-05,
      "loss": 0.6462,
      "step": 3030
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.6774581670761108,
      "learning_rate": 2e-05,
      "loss": 0.6484,
      "step": 3040
    },
    {
      "epoch": 0.25684210526315787,
      "grad_norm": 0.38739949464797974,
      "learning_rate": 2e-05,
      "loss": 0.6397,
      "step": 3050
    },
    {
      "epoch": 0.2576842105263158,
      "grad_norm": 0.5650710463523865,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 3060
    },
    {
      "epoch": 0.25852631578947366,
      "grad_norm": 0.4666733741760254,
      "learning_rate": 2e-05,
      "loss": 0.6396,
      "step": 3070
    },
    {
      "epoch": 0.2593684210526316,
      "grad_norm": 0.44759392738342285,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 3080
    },
    {
      "epoch": 0.26021052631578945,
      "grad_norm": 0.4109085500240326,
      "learning_rate": 2e-05,
      "loss": 0.6399,
      "step": 3090
    },
    {
      "epoch": 0.26105263157894737,
      "grad_norm": 0.6701503992080688,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 3100
    },
    {
      "epoch": 0.26189473684210524,
      "grad_norm": 0.3906865119934082,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 3110
    },
    {
      "epoch": 0.26273684210526316,
      "grad_norm": 0.5981286764144897,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 3120
    },
    {
      "epoch": 0.2635789473684211,
      "grad_norm": 0.3991749882698059,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 3130
    },
    {
      "epoch": 0.26442105263157895,
      "grad_norm": 0.41412943601608276,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 3140
    },
    {
      "epoch": 0.26526315789473687,
      "grad_norm": 0.406644344329834,
      "learning_rate": 2e-05,
      "loss": 0.6267,
      "step": 3150
    },
    {
      "epoch": 0.26610526315789473,
      "grad_norm": 0.39435264468193054,
      "learning_rate": 2e-05,
      "loss": 0.6367,
      "step": 3160
    },
    {
      "epoch": 0.26694736842105266,
      "grad_norm": 0.442579984664917,
      "learning_rate": 2e-05,
      "loss": 0.6252,
      "step": 3170
    },
    {
      "epoch": 0.2677894736842105,
      "grad_norm": 0.4149363040924072,
      "learning_rate": 2e-05,
      "loss": 0.6281,
      "step": 3180
    },
    {
      "epoch": 0.26863157894736844,
      "grad_norm": 0.4123855531215668,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 3190
    },
    {
      "epoch": 0.2694736842105263,
      "grad_norm": 0.3868705630302429,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 3200
    },
    {
      "epoch": 0.27031578947368423,
      "grad_norm": 0.49756893515586853,
      "learning_rate": 2e-05,
      "loss": 0.6611,
      "step": 3210
    },
    {
      "epoch": 0.2711578947368421,
      "grad_norm": 0.40090250968933105,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 3220
    },
    {
      "epoch": 0.272,
      "grad_norm": 0.47787225246429443,
      "learning_rate": 2e-05,
      "loss": 0.6351,
      "step": 3230
    },
    {
      "epoch": 0.2728421052631579,
      "grad_norm": 0.4524623155593872,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 3240
    },
    {
      "epoch": 0.2736842105263158,
      "grad_norm": 0.41923800110816956,
      "learning_rate": 2e-05,
      "loss": 0.66,
      "step": 3250
    },
    {
      "epoch": 0.2745263157894737,
      "grad_norm": 0.4373698830604553,
      "learning_rate": 2e-05,
      "loss": 0.6301,
      "step": 3260
    },
    {
      "epoch": 0.2753684210526316,
      "grad_norm": 0.4100465178489685,
      "learning_rate": 2e-05,
      "loss": 0.6325,
      "step": 3270
    },
    {
      "epoch": 0.27621052631578946,
      "grad_norm": 0.4169442057609558,
      "learning_rate": 2e-05,
      "loss": 0.6522,
      "step": 3280
    },
    {
      "epoch": 0.2770526315789474,
      "grad_norm": 0.40628781914711,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 3290
    },
    {
      "epoch": 0.27789473684210525,
      "grad_norm": 0.3915088474750519,
      "learning_rate": 2e-05,
      "loss": 0.644,
      "step": 3300
    },
    {
      "epoch": 0.27873684210526317,
      "grad_norm": 0.5214042067527771,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 3310
    },
    {
      "epoch": 0.27957894736842104,
      "grad_norm": 0.4097200930118561,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 3320
    },
    {
      "epoch": 0.28042105263157896,
      "grad_norm": 0.45008161664009094,
      "learning_rate": 2e-05,
      "loss": 0.6607,
      "step": 3330
    },
    {
      "epoch": 0.2812631578947368,
      "grad_norm": 0.38959962129592896,
      "learning_rate": 2e-05,
      "loss": 0.625,
      "step": 3340
    },
    {
      "epoch": 0.28210526315789475,
      "grad_norm": 0.39687466621398926,
      "learning_rate": 2e-05,
      "loss": 0.6539,
      "step": 3350
    },
    {
      "epoch": 0.2829473684210526,
      "grad_norm": 0.4723809063434601,
      "learning_rate": 2e-05,
      "loss": 0.6299,
      "step": 3360
    },
    {
      "epoch": 0.28378947368421054,
      "grad_norm": 0.4434508979320526,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 3370
    },
    {
      "epoch": 0.2846315789473684,
      "grad_norm": 0.4188632071018219,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 3380
    },
    {
      "epoch": 0.2854736842105263,
      "grad_norm": 0.4126710295677185,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 3390
    },
    {
      "epoch": 0.2863157894736842,
      "grad_norm": 0.4203440248966217,
      "learning_rate": 2e-05,
      "loss": 0.6227,
      "step": 3400
    },
    {
      "epoch": 0.2871578947368421,
      "grad_norm": 0.469748854637146,
      "learning_rate": 2e-05,
      "loss": 0.6362,
      "step": 3410
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.4273054301738739,
      "learning_rate": 2e-05,
      "loss": 0.6672,
      "step": 3420
    },
    {
      "epoch": 0.2888421052631579,
      "grad_norm": 0.4482914209365845,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 3430
    },
    {
      "epoch": 0.28968421052631577,
      "grad_norm": 0.4641944169998169,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 3440
    },
    {
      "epoch": 0.2905263157894737,
      "grad_norm": 0.37492042779922485,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 3450
    },
    {
      "epoch": 0.29136842105263155,
      "grad_norm": 0.4156726002693176,
      "learning_rate": 2e-05,
      "loss": 0.6544,
      "step": 3460
    },
    {
      "epoch": 0.2922105263157895,
      "grad_norm": 0.3957166373729706,
      "learning_rate": 2e-05,
      "loss": 0.6511,
      "step": 3470
    },
    {
      "epoch": 0.29305263157894734,
      "grad_norm": 0.4040151834487915,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 3480
    },
    {
      "epoch": 0.29389473684210526,
      "grad_norm": 0.4258546531200409,
      "learning_rate": 2e-05,
      "loss": 0.6241,
      "step": 3490
    },
    {
      "epoch": 0.29473684210526313,
      "grad_norm": 0.44563421607017517,
      "learning_rate": 2e-05,
      "loss": 0.6333,
      "step": 3500
    },
    {
      "epoch": 0.29557894736842105,
      "grad_norm": 0.3894782066345215,
      "learning_rate": 2e-05,
      "loss": 0.6541,
      "step": 3510
    },
    {
      "epoch": 0.296421052631579,
      "grad_norm": 0.43910810351371765,
      "learning_rate": 2e-05,
      "loss": 0.6495,
      "step": 3520
    },
    {
      "epoch": 0.29726315789473684,
      "grad_norm": 0.4005981683731079,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 3530
    },
    {
      "epoch": 0.29810526315789476,
      "grad_norm": 0.4540840685367584,
      "learning_rate": 2e-05,
      "loss": 0.6302,
      "step": 3540
    },
    {
      "epoch": 0.29894736842105263,
      "grad_norm": 0.4289926588535309,
      "learning_rate": 2e-05,
      "loss": 0.6455,
      "step": 3550
    },
    {
      "epoch": 0.29978947368421055,
      "grad_norm": 0.4118243455886841,
      "learning_rate": 2e-05,
      "loss": 0.6546,
      "step": 3560
    },
    {
      "epoch": 0.3006315789473684,
      "grad_norm": 0.42285335063934326,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 3570
    },
    {
      "epoch": 0.30147368421052634,
      "grad_norm": 0.4720819294452667,
      "learning_rate": 2e-05,
      "loss": 0.6247,
      "step": 3580
    },
    {
      "epoch": 0.3023157894736842,
      "grad_norm": 0.4414178431034088,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 3590
    },
    {
      "epoch": 0.3031578947368421,
      "grad_norm": 1.913885235786438,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 3600
    },
    {
      "epoch": 0.304,
      "grad_norm": 0.3927784562110901,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 3610
    },
    {
      "epoch": 0.3048421052631579,
      "grad_norm": 0.5843016505241394,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 3620
    },
    {
      "epoch": 0.3056842105263158,
      "grad_norm": 0.40196493268013,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 3630
    },
    {
      "epoch": 0.3065263157894737,
      "grad_norm": 0.37512633204460144,
      "learning_rate": 2e-05,
      "loss": 0.6329,
      "step": 3640
    },
    {
      "epoch": 0.30736842105263157,
      "grad_norm": 0.4471430778503418,
      "learning_rate": 2e-05,
      "loss": 0.6157,
      "step": 3650
    },
    {
      "epoch": 0.3082105263157895,
      "grad_norm": 0.42406660318374634,
      "learning_rate": 2e-05,
      "loss": 0.628,
      "step": 3660
    },
    {
      "epoch": 0.30905263157894736,
      "grad_norm": 0.3814440071582794,
      "learning_rate": 2e-05,
      "loss": 0.6277,
      "step": 3670
    },
    {
      "epoch": 0.3098947368421053,
      "grad_norm": 0.42781972885131836,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 3680
    },
    {
      "epoch": 0.31073684210526314,
      "grad_norm": 0.4300000071525574,
      "learning_rate": 2e-05,
      "loss": 0.6233,
      "step": 3690
    },
    {
      "epoch": 0.31157894736842107,
      "grad_norm": 0.4076962172985077,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 3700
    },
    {
      "epoch": 0.31242105263157893,
      "grad_norm": 0.3952917456626892,
      "learning_rate": 2e-05,
      "loss": 0.6517,
      "step": 3710
    },
    {
      "epoch": 0.31326315789473685,
      "grad_norm": 0.44713258743286133,
      "learning_rate": 2e-05,
      "loss": 0.659,
      "step": 3720
    },
    {
      "epoch": 0.3141052631578947,
      "grad_norm": 0.43187597393989563,
      "learning_rate": 2e-05,
      "loss": 0.6614,
      "step": 3730
    },
    {
      "epoch": 0.31494736842105264,
      "grad_norm": 0.4329753816127777,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 3740
    },
    {
      "epoch": 0.3157894736842105,
      "grad_norm": 0.4955722391605377,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 3750
    },
    {
      "epoch": 0.31663157894736843,
      "grad_norm": 0.510044276714325,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 3760
    },
    {
      "epoch": 0.3174736842105263,
      "grad_norm": 0.40952301025390625,
      "learning_rate": 2e-05,
      "loss": 0.6325,
      "step": 3770
    },
    {
      "epoch": 0.3183157894736842,
      "grad_norm": 0.4995168149471283,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 3780
    },
    {
      "epoch": 0.3191578947368421,
      "grad_norm": 0.3866928219795227,
      "learning_rate": 2e-05,
      "loss": 0.6541,
      "step": 3790
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4026857614517212,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 3800
    },
    {
      "epoch": 0.3208421052631579,
      "grad_norm": 0.4273292124271393,
      "learning_rate": 2e-05,
      "loss": 0.6037,
      "step": 3810
    },
    {
      "epoch": 0.3216842105263158,
      "grad_norm": 0.3855188190937042,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 3820
    },
    {
      "epoch": 0.32252631578947366,
      "grad_norm": 0.3889329731464386,
      "learning_rate": 2e-05,
      "loss": 0.6381,
      "step": 3830
    },
    {
      "epoch": 0.3233684210526316,
      "grad_norm": 0.3755100965499878,
      "learning_rate": 2e-05,
      "loss": 0.6353,
      "step": 3840
    },
    {
      "epoch": 0.32421052631578945,
      "grad_norm": 0.41141578555107117,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 3850
    },
    {
      "epoch": 0.32505263157894737,
      "grad_norm": 0.6547090411186218,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 3860
    },
    {
      "epoch": 0.32589473684210524,
      "grad_norm": 0.5511941313743591,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 3870
    },
    {
      "epoch": 0.32673684210526316,
      "grad_norm": 0.3957040011882782,
      "learning_rate": 2e-05,
      "loss": 0.629,
      "step": 3880
    },
    {
      "epoch": 0.327578947368421,
      "grad_norm": 0.42204535007476807,
      "learning_rate": 2e-05,
      "loss": 0.6131,
      "step": 3890
    },
    {
      "epoch": 0.32842105263157895,
      "grad_norm": 0.44419601559638977,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 3900
    },
    {
      "epoch": 0.32926315789473687,
      "grad_norm": 0.4307175576686859,
      "learning_rate": 2e-05,
      "loss": 0.6285,
      "step": 3910
    },
    {
      "epoch": 0.33010526315789473,
      "grad_norm": 0.42792925238609314,
      "learning_rate": 2e-05,
      "loss": 0.6112,
      "step": 3920
    },
    {
      "epoch": 0.33094736842105266,
      "grad_norm": 0.43926912546157837,
      "learning_rate": 2e-05,
      "loss": 0.6384,
      "step": 3930
    },
    {
      "epoch": 0.3317894736842105,
      "grad_norm": 0.41949915885925293,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 3940
    },
    {
      "epoch": 0.33263157894736844,
      "grad_norm": 0.4190210998058319,
      "learning_rate": 2e-05,
      "loss": 0.6314,
      "step": 3950
    },
    {
      "epoch": 0.3334736842105263,
      "grad_norm": 0.43903157114982605,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 3960
    },
    {
      "epoch": 0.33431578947368423,
      "grad_norm": 0.46098485589027405,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 3970
    },
    {
      "epoch": 0.3351578947368421,
      "grad_norm": 0.6722385883331299,
      "learning_rate": 2e-05,
      "loss": 0.6348,
      "step": 3980
    },
    {
      "epoch": 0.336,
      "grad_norm": 0.4123629331588745,
      "learning_rate": 2e-05,
      "loss": 0.6666,
      "step": 3990
    },
    {
      "epoch": 0.3368421052631579,
      "grad_norm": 0.39911410212516785,
      "learning_rate": 2e-05,
      "loss": 0.6378,
      "step": 4000
    },
    {
      "epoch": 0.3376842105263158,
      "grad_norm": 0.4449623227119446,
      "learning_rate": 2e-05,
      "loss": 0.6136,
      "step": 4010
    },
    {
      "epoch": 0.3385263157894737,
      "grad_norm": 0.4165237545967102,
      "learning_rate": 2e-05,
      "loss": 0.6687,
      "step": 4020
    },
    {
      "epoch": 0.3393684210526316,
      "grad_norm": 0.41476067900657654,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 4030
    },
    {
      "epoch": 0.34021052631578946,
      "grad_norm": 0.42177554965019226,
      "learning_rate": 2e-05,
      "loss": 0.6223,
      "step": 4040
    },
    {
      "epoch": 0.3410526315789474,
      "grad_norm": 0.417418509721756,
      "learning_rate": 2e-05,
      "loss": 0.6317,
      "step": 4050
    },
    {
      "epoch": 0.34189473684210525,
      "grad_norm": 0.39303576946258545,
      "learning_rate": 2e-05,
      "loss": 0.6574,
      "step": 4060
    },
    {
      "epoch": 0.3427368421052632,
      "grad_norm": 0.41646191477775574,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 4070
    },
    {
      "epoch": 0.34357894736842104,
      "grad_norm": 0.4005908668041229,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 4080
    },
    {
      "epoch": 0.34442105263157896,
      "grad_norm": 0.6376783847808838,
      "learning_rate": 2e-05,
      "loss": 0.6309,
      "step": 4090
    },
    {
      "epoch": 0.3452631578947368,
      "grad_norm": 0.3953230679035187,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 4100
    },
    {
      "epoch": 0.34610526315789475,
      "grad_norm": 0.40265971422195435,
      "learning_rate": 2e-05,
      "loss": 0.6397,
      "step": 4110
    },
    {
      "epoch": 0.3469473684210526,
      "grad_norm": 0.4055187404155731,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 4120
    },
    {
      "epoch": 0.34778947368421054,
      "grad_norm": 0.4240468144416809,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 4130
    },
    {
      "epoch": 0.3486315789473684,
      "grad_norm": 0.42309707403182983,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 4140
    },
    {
      "epoch": 0.3494736842105263,
      "grad_norm": 0.42797455191612244,
      "learning_rate": 2e-05,
      "loss": 0.6294,
      "step": 4150
    },
    {
      "epoch": 0.3503157894736842,
      "grad_norm": 0.39371049404144287,
      "learning_rate": 2e-05,
      "loss": 0.6417,
      "step": 4160
    },
    {
      "epoch": 0.3511578947368421,
      "grad_norm": 0.4434272050857544,
      "learning_rate": 2e-05,
      "loss": 0.6404,
      "step": 4170
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.38733410835266113,
      "learning_rate": 2e-05,
      "loss": 0.6324,
      "step": 4180
    },
    {
      "epoch": 0.3528421052631579,
      "grad_norm": 0.37633654475212097,
      "learning_rate": 2e-05,
      "loss": 0.6292,
      "step": 4190
    },
    {
      "epoch": 0.35368421052631577,
      "grad_norm": 0.4404960870742798,
      "learning_rate": 2e-05,
      "loss": 0.6716,
      "step": 4200
    },
    {
      "epoch": 0.3545263157894737,
      "grad_norm": 0.4043072462081909,
      "learning_rate": 2e-05,
      "loss": 0.6221,
      "step": 4210
    },
    {
      "epoch": 0.35536842105263156,
      "grad_norm": 0.3863939642906189,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 4220
    },
    {
      "epoch": 0.3562105263157895,
      "grad_norm": 0.4015117883682251,
      "learning_rate": 2e-05,
      "loss": 0.6453,
      "step": 4230
    },
    {
      "epoch": 0.35705263157894734,
      "grad_norm": 0.39795029163360596,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 4240
    },
    {
      "epoch": 0.35789473684210527,
      "grad_norm": 0.3844669461250305,
      "learning_rate": 2e-05,
      "loss": 0.6361,
      "step": 4250
    },
    {
      "epoch": 0.35873684210526313,
      "grad_norm": 0.4071543514728546,
      "learning_rate": 2e-05,
      "loss": 0.647,
      "step": 4260
    },
    {
      "epoch": 0.35957894736842105,
      "grad_norm": 0.4376121759414673,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 4270
    },
    {
      "epoch": 0.3604210526315789,
      "grad_norm": 0.39380401372909546,
      "learning_rate": 2e-05,
      "loss": 0.6232,
      "step": 4280
    },
    {
      "epoch": 0.36126315789473684,
      "grad_norm": 0.39438238739967346,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 4290
    },
    {
      "epoch": 0.36210526315789476,
      "grad_norm": 0.3838787376880646,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 4300
    },
    {
      "epoch": 0.36294736842105263,
      "grad_norm": 0.4334008991718292,
      "learning_rate": 2e-05,
      "loss": 0.6381,
      "step": 4310
    },
    {
      "epoch": 0.36378947368421055,
      "grad_norm": 0.4696757197380066,
      "learning_rate": 2e-05,
      "loss": 0.6103,
      "step": 4320
    },
    {
      "epoch": 0.3646315789473684,
      "grad_norm": 0.47157326340675354,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 4330
    },
    {
      "epoch": 0.36547368421052634,
      "grad_norm": 0.39691030979156494,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 4340
    },
    {
      "epoch": 0.3663157894736842,
      "grad_norm": 0.3954697251319885,
      "learning_rate": 2e-05,
      "loss": 0.6397,
      "step": 4350
    },
    {
      "epoch": 0.3671578947368421,
      "grad_norm": 0.40735548734664917,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 4360
    },
    {
      "epoch": 0.368,
      "grad_norm": 0.3969827890396118,
      "learning_rate": 2e-05,
      "loss": 0.6341,
      "step": 4370
    },
    {
      "epoch": 0.3688421052631579,
      "grad_norm": 0.4401497542858124,
      "learning_rate": 2e-05,
      "loss": 0.639,
      "step": 4380
    },
    {
      "epoch": 0.3696842105263158,
      "grad_norm": 0.42447590827941895,
      "learning_rate": 2e-05,
      "loss": 0.5932,
      "step": 4390
    },
    {
      "epoch": 0.3705263157894737,
      "grad_norm": 0.4302249848842621,
      "learning_rate": 2e-05,
      "loss": 0.6603,
      "step": 4400
    },
    {
      "epoch": 0.37136842105263157,
      "grad_norm": 0.4111262261867523,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 4410
    },
    {
      "epoch": 0.3722105263157895,
      "grad_norm": 0.3921228349208832,
      "learning_rate": 2e-05,
      "loss": 0.6349,
      "step": 4420
    },
    {
      "epoch": 0.37305263157894736,
      "grad_norm": 0.41597214341163635,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 4430
    },
    {
      "epoch": 0.3738947368421053,
      "grad_norm": 0.39468562602996826,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 4440
    },
    {
      "epoch": 0.37473684210526315,
      "grad_norm": 0.4095419645309448,
      "learning_rate": 2e-05,
      "loss": 0.6402,
      "step": 4450
    },
    {
      "epoch": 0.37557894736842107,
      "grad_norm": 0.39670878648757935,
      "learning_rate": 2e-05,
      "loss": 0.6277,
      "step": 4460
    },
    {
      "epoch": 0.37642105263157893,
      "grad_norm": 0.38621458411216736,
      "learning_rate": 2e-05,
      "loss": 0.6156,
      "step": 4470
    },
    {
      "epoch": 0.37726315789473686,
      "grad_norm": 0.40281733870506287,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 4480
    },
    {
      "epoch": 0.3781052631578947,
      "grad_norm": 0.3874606490135193,
      "learning_rate": 2e-05,
      "loss": 0.6338,
      "step": 4490
    },
    {
      "epoch": 0.37894736842105264,
      "grad_norm": 0.49733874201774597,
      "learning_rate": 2e-05,
      "loss": 0.633,
      "step": 4500
    },
    {
      "epoch": 0.3797894736842105,
      "grad_norm": 0.3730744421482086,
      "learning_rate": 2e-05,
      "loss": 0.6439,
      "step": 4510
    },
    {
      "epoch": 0.38063157894736843,
      "grad_norm": 0.3587619960308075,
      "learning_rate": 2e-05,
      "loss": 0.6293,
      "step": 4520
    },
    {
      "epoch": 0.3814736842105263,
      "grad_norm": 0.4145994484424591,
      "learning_rate": 2e-05,
      "loss": 0.6222,
      "step": 4530
    },
    {
      "epoch": 0.3823157894736842,
      "grad_norm": 0.4197967052459717,
      "learning_rate": 2e-05,
      "loss": 0.6516,
      "step": 4540
    },
    {
      "epoch": 0.3831578947368421,
      "grad_norm": 0.39057019352912903,
      "learning_rate": 2e-05,
      "loss": 0.6265,
      "step": 4550
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.39443516731262207,
      "learning_rate": 2e-05,
      "loss": 0.6396,
      "step": 4560
    },
    {
      "epoch": 0.3848421052631579,
      "grad_norm": 0.40980201959609985,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 4570
    },
    {
      "epoch": 0.3856842105263158,
      "grad_norm": 0.4044901430606842,
      "learning_rate": 2e-05,
      "loss": 0.6049,
      "step": 4580
    },
    {
      "epoch": 0.38652631578947366,
      "grad_norm": 0.3937481939792633,
      "learning_rate": 2e-05,
      "loss": 0.6431,
      "step": 4590
    },
    {
      "epoch": 0.3873684210526316,
      "grad_norm": 0.4133276045322418,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 4600
    },
    {
      "epoch": 0.38821052631578945,
      "grad_norm": 0.37604787945747375,
      "learning_rate": 2e-05,
      "loss": 0.6254,
      "step": 4610
    },
    {
      "epoch": 0.38905263157894737,
      "grad_norm": 0.40275147557258606,
      "learning_rate": 2e-05,
      "loss": 0.6471,
      "step": 4620
    },
    {
      "epoch": 0.38989473684210524,
      "grad_norm": 0.43851664662361145,
      "learning_rate": 2e-05,
      "loss": 0.6562,
      "step": 4630
    },
    {
      "epoch": 0.39073684210526316,
      "grad_norm": 0.42869511246681213,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 4640
    },
    {
      "epoch": 0.391578947368421,
      "grad_norm": 0.40454980731010437,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 4650
    },
    {
      "epoch": 0.39242105263157895,
      "grad_norm": 0.4188549518585205,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 4660
    },
    {
      "epoch": 0.3932631578947368,
      "grad_norm": 0.38690853118896484,
      "learning_rate": 2e-05,
      "loss": 0.6092,
      "step": 4670
    },
    {
      "epoch": 0.39410526315789474,
      "grad_norm": 0.41081416606903076,
      "learning_rate": 2e-05,
      "loss": 0.6332,
      "step": 4680
    },
    {
      "epoch": 0.39494736842105266,
      "grad_norm": 0.44359102845191956,
      "learning_rate": 2e-05,
      "loss": 0.6287,
      "step": 4690
    },
    {
      "epoch": 0.3957894736842105,
      "grad_norm": 0.40810829401016235,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 4700
    },
    {
      "epoch": 0.39663157894736845,
      "grad_norm": 0.3910294771194458,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 4710
    },
    {
      "epoch": 0.3974736842105263,
      "grad_norm": 0.386960506439209,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 4720
    },
    {
      "epoch": 0.39831578947368423,
      "grad_norm": 0.3930663764476776,
      "learning_rate": 2e-05,
      "loss": 0.6257,
      "step": 4730
    },
    {
      "epoch": 0.3991578947368421,
      "grad_norm": 0.40631163120269775,
      "learning_rate": 2e-05,
      "loss": 0.6203,
      "step": 4740
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3676679730415344,
      "learning_rate": 2e-05,
      "loss": 0.6199,
      "step": 4750
    },
    {
      "epoch": 0.4008421052631579,
      "grad_norm": 0.3718276619911194,
      "learning_rate": 2e-05,
      "loss": 0.6448,
      "step": 4760
    },
    {
      "epoch": 0.4016842105263158,
      "grad_norm": 0.3798637092113495,
      "learning_rate": 2e-05,
      "loss": 0.6487,
      "step": 4770
    },
    {
      "epoch": 0.4025263157894737,
      "grad_norm": 0.3911955654621124,
      "learning_rate": 2e-05,
      "loss": 0.6309,
      "step": 4780
    },
    {
      "epoch": 0.4033684210526316,
      "grad_norm": 0.40748897194862366,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 4790
    },
    {
      "epoch": 0.40421052631578946,
      "grad_norm": 0.4001414477825165,
      "learning_rate": 2e-05,
      "loss": 0.6258,
      "step": 4800
    },
    {
      "epoch": 0.4050526315789474,
      "grad_norm": 0.40826529264450073,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 4810
    },
    {
      "epoch": 0.40589473684210525,
      "grad_norm": 0.4198121428489685,
      "learning_rate": 2e-05,
      "loss": 0.6242,
      "step": 4820
    },
    {
      "epoch": 0.4067368421052632,
      "grad_norm": 0.4132035970687866,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 4830
    },
    {
      "epoch": 0.40757894736842104,
      "grad_norm": 0.4702431857585907,
      "learning_rate": 2e-05,
      "loss": 0.6418,
      "step": 4840
    },
    {
      "epoch": 0.40842105263157896,
      "grad_norm": 0.40195342898368835,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 4850
    },
    {
      "epoch": 0.40926315789473683,
      "grad_norm": 0.42084524035453796,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 4860
    },
    {
      "epoch": 0.41010526315789475,
      "grad_norm": 0.41681522130966187,
      "learning_rate": 2e-05,
      "loss": 0.6381,
      "step": 4870
    },
    {
      "epoch": 0.4109473684210526,
      "grad_norm": 0.41127869486808777,
      "learning_rate": 2e-05,
      "loss": 0.6307,
      "step": 4880
    },
    {
      "epoch": 0.41178947368421054,
      "grad_norm": 0.44477570056915283,
      "learning_rate": 2e-05,
      "loss": 0.6263,
      "step": 4890
    },
    {
      "epoch": 0.4126315789473684,
      "grad_norm": 0.4360620081424713,
      "learning_rate": 2e-05,
      "loss": 0.6148,
      "step": 4900
    },
    {
      "epoch": 0.4134736842105263,
      "grad_norm": 0.3812357783317566,
      "learning_rate": 2e-05,
      "loss": 0.6252,
      "step": 4910
    },
    {
      "epoch": 0.4143157894736842,
      "grad_norm": 0.4152975082397461,
      "learning_rate": 2e-05,
      "loss": 0.641,
      "step": 4920
    },
    {
      "epoch": 0.4151578947368421,
      "grad_norm": 0.40829282999038696,
      "learning_rate": 2e-05,
      "loss": 0.6378,
      "step": 4930
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.4044029116630554,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 4940
    },
    {
      "epoch": 0.4168421052631579,
      "grad_norm": 0.3890173137187958,
      "learning_rate": 2e-05,
      "loss": 0.639,
      "step": 4950
    },
    {
      "epoch": 0.41768421052631577,
      "grad_norm": 0.42901378870010376,
      "learning_rate": 2e-05,
      "loss": 0.6386,
      "step": 4960
    },
    {
      "epoch": 0.4185263157894737,
      "grad_norm": 0.42894312739372253,
      "learning_rate": 2e-05,
      "loss": 0.6414,
      "step": 4970
    },
    {
      "epoch": 0.41936842105263156,
      "grad_norm": 0.3889273405075073,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 4980
    },
    {
      "epoch": 0.4202105263157895,
      "grad_norm": 0.5117679238319397,
      "learning_rate": 2e-05,
      "loss": 0.6284,
      "step": 4990
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 0.5317552089691162,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 5000
    },
    {
      "epoch": 0.42189473684210527,
      "grad_norm": 0.39053019881248474,
      "learning_rate": 2e-05,
      "loss": 0.6034,
      "step": 5010
    },
    {
      "epoch": 0.42273684210526313,
      "grad_norm": 0.4187553822994232,
      "learning_rate": 2e-05,
      "loss": 0.6256,
      "step": 5020
    },
    {
      "epoch": 0.42357894736842105,
      "grad_norm": 0.3733656406402588,
      "learning_rate": 2e-05,
      "loss": 0.6347,
      "step": 5030
    },
    {
      "epoch": 0.4244210526315789,
      "grad_norm": 0.40386319160461426,
      "learning_rate": 2e-05,
      "loss": 0.6093,
      "step": 5040
    },
    {
      "epoch": 0.42526315789473684,
      "grad_norm": 0.373512864112854,
      "learning_rate": 2e-05,
      "loss": 0.6252,
      "step": 5050
    },
    {
      "epoch": 0.4261052631578947,
      "grad_norm": 0.42248404026031494,
      "learning_rate": 2e-05,
      "loss": 0.6427,
      "step": 5060
    },
    {
      "epoch": 0.42694736842105263,
      "grad_norm": 0.42513394355773926,
      "learning_rate": 2e-05,
      "loss": 0.6271,
      "step": 5070
    },
    {
      "epoch": 0.42778947368421055,
      "grad_norm": 0.4904797673225403,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 5080
    },
    {
      "epoch": 0.4286315789473684,
      "grad_norm": 0.3930179476737976,
      "learning_rate": 2e-05,
      "loss": 0.6033,
      "step": 5090
    },
    {
      "epoch": 0.42947368421052634,
      "grad_norm": 0.3941327929496765,
      "learning_rate": 2e-05,
      "loss": 0.6074,
      "step": 5100
    },
    {
      "epoch": 0.4303157894736842,
      "grad_norm": 0.4101496636867523,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 5110
    },
    {
      "epoch": 0.43115789473684213,
      "grad_norm": 0.4099400043487549,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 5120
    },
    {
      "epoch": 0.432,
      "grad_norm": 0.4460608959197998,
      "learning_rate": 2e-05,
      "loss": 0.6084,
      "step": 5130
    },
    {
      "epoch": 0.4328421052631579,
      "grad_norm": 0.45957818627357483,
      "learning_rate": 2e-05,
      "loss": 0.6271,
      "step": 5140
    },
    {
      "epoch": 0.4336842105263158,
      "grad_norm": 0.4305644631385803,
      "learning_rate": 2e-05,
      "loss": 0.6134,
      "step": 5150
    },
    {
      "epoch": 0.4345263157894737,
      "grad_norm": 0.39713993668556213,
      "learning_rate": 2e-05,
      "loss": 0.6344,
      "step": 5160
    },
    {
      "epoch": 0.43536842105263157,
      "grad_norm": 0.3930221199989319,
      "learning_rate": 2e-05,
      "loss": 0.6362,
      "step": 5170
    },
    {
      "epoch": 0.4362105263157895,
      "grad_norm": 0.4276049733161926,
      "learning_rate": 2e-05,
      "loss": 0.6389,
      "step": 5180
    },
    {
      "epoch": 0.43705263157894736,
      "grad_norm": 0.3759106695652008,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 5190
    },
    {
      "epoch": 0.4378947368421053,
      "grad_norm": 0.3932980000972748,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 5200
    },
    {
      "epoch": 0.43873684210526315,
      "grad_norm": 0.5513986349105835,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 5210
    },
    {
      "epoch": 0.43957894736842107,
      "grad_norm": 0.4309244453907013,
      "learning_rate": 2e-05,
      "loss": 0.6448,
      "step": 5220
    },
    {
      "epoch": 0.44042105263157894,
      "grad_norm": 0.3919553756713867,
      "learning_rate": 2e-05,
      "loss": 0.6233,
      "step": 5230
    },
    {
      "epoch": 0.44126315789473686,
      "grad_norm": 0.3774227201938629,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 5240
    },
    {
      "epoch": 0.4421052631578947,
      "grad_norm": 0.35983794927597046,
      "learning_rate": 2e-05,
      "loss": 0.6223,
      "step": 5250
    },
    {
      "epoch": 0.44294736842105265,
      "grad_norm": 0.4020618200302124,
      "learning_rate": 2e-05,
      "loss": 0.6597,
      "step": 5260
    },
    {
      "epoch": 0.4437894736842105,
      "grad_norm": 0.6328001022338867,
      "learning_rate": 2e-05,
      "loss": 0.6577,
      "step": 5270
    },
    {
      "epoch": 0.44463157894736843,
      "grad_norm": 0.4017433822154999,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 5280
    },
    {
      "epoch": 0.4454736842105263,
      "grad_norm": 0.37539350986480713,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 5290
    },
    {
      "epoch": 0.4463157894736842,
      "grad_norm": 0.41220203042030334,
      "learning_rate": 2e-05,
      "loss": 0.622,
      "step": 5300
    },
    {
      "epoch": 0.4471578947368421,
      "grad_norm": 0.4161873459815979,
      "learning_rate": 2e-05,
      "loss": 0.627,
      "step": 5310
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.42080461978912354,
      "learning_rate": 2e-05,
      "loss": 0.6454,
      "step": 5320
    },
    {
      "epoch": 0.4488421052631579,
      "grad_norm": 0.3736485242843628,
      "learning_rate": 2e-05,
      "loss": 0.6233,
      "step": 5330
    },
    {
      "epoch": 0.4496842105263158,
      "grad_norm": 0.4019976556301117,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 5340
    },
    {
      "epoch": 0.45052631578947366,
      "grad_norm": 0.3832785189151764,
      "learning_rate": 2e-05,
      "loss": 0.6207,
      "step": 5350
    },
    {
      "epoch": 0.4513684210526316,
      "grad_norm": 0.42821282148361206,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 5360
    },
    {
      "epoch": 0.45221052631578945,
      "grad_norm": 0.39499279856681824,
      "learning_rate": 2e-05,
      "loss": 0.6379,
      "step": 5370
    },
    {
      "epoch": 0.4530526315789474,
      "grad_norm": 0.4208984673023224,
      "learning_rate": 2e-05,
      "loss": 0.6198,
      "step": 5380
    },
    {
      "epoch": 0.45389473684210524,
      "grad_norm": 0.37664926052093506,
      "learning_rate": 2e-05,
      "loss": 0.6187,
      "step": 5390
    },
    {
      "epoch": 0.45473684210526316,
      "grad_norm": 0.39407405257225037,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 5400
    },
    {
      "epoch": 0.45557894736842103,
      "grad_norm": 0.3641555905342102,
      "learning_rate": 2e-05,
      "loss": 0.615,
      "step": 5410
    },
    {
      "epoch": 0.45642105263157895,
      "grad_norm": 0.3745141923427582,
      "learning_rate": 2e-05,
      "loss": 0.6329,
      "step": 5420
    },
    {
      "epoch": 0.4572631578947368,
      "grad_norm": 0.5289539694786072,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 5430
    },
    {
      "epoch": 0.45810526315789474,
      "grad_norm": 0.3853813409805298,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 5440
    },
    {
      "epoch": 0.4589473684210526,
      "grad_norm": 0.41005533933639526,
      "learning_rate": 2e-05,
      "loss": 0.6363,
      "step": 5450
    },
    {
      "epoch": 0.4597894736842105,
      "grad_norm": 0.4378117024898529,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 5460
    },
    {
      "epoch": 0.46063157894736845,
      "grad_norm": 0.4291653037071228,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 5470
    },
    {
      "epoch": 0.4614736842105263,
      "grad_norm": 0.39368027448654175,
      "learning_rate": 2e-05,
      "loss": 0.6593,
      "step": 5480
    },
    {
      "epoch": 0.46231578947368424,
      "grad_norm": 0.40501919388771057,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 5490
    },
    {
      "epoch": 0.4631578947368421,
      "grad_norm": 0.4044288694858551,
      "learning_rate": 2e-05,
      "loss": 0.6389,
      "step": 5500
    },
    {
      "epoch": 0.464,
      "grad_norm": 0.39211922883987427,
      "learning_rate": 2e-05,
      "loss": 0.6231,
      "step": 5510
    },
    {
      "epoch": 0.4648421052631579,
      "grad_norm": 0.39945188164711,
      "learning_rate": 2e-05,
      "loss": 0.6174,
      "step": 5520
    },
    {
      "epoch": 0.4656842105263158,
      "grad_norm": 0.3761236369609833,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 5530
    },
    {
      "epoch": 0.4665263157894737,
      "grad_norm": 0.38399356603622437,
      "learning_rate": 2e-05,
      "loss": 0.6238,
      "step": 5540
    },
    {
      "epoch": 0.4673684210526316,
      "grad_norm": 0.38960474729537964,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 5550
    },
    {
      "epoch": 0.46821052631578947,
      "grad_norm": 0.4436739385128021,
      "learning_rate": 2e-05,
      "loss": 0.6209,
      "step": 5560
    },
    {
      "epoch": 0.4690526315789474,
      "grad_norm": 0.41112789511680603,
      "learning_rate": 2e-05,
      "loss": 0.6445,
      "step": 5570
    },
    {
      "epoch": 0.46989473684210525,
      "grad_norm": 0.8796234130859375,
      "learning_rate": 2e-05,
      "loss": 0.6461,
      "step": 5580
    },
    {
      "epoch": 0.4707368421052632,
      "grad_norm": 0.3961946964263916,
      "learning_rate": 2e-05,
      "loss": 0.6136,
      "step": 5590
    },
    {
      "epoch": 0.47157894736842104,
      "grad_norm": 0.4006491005420685,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 5600
    },
    {
      "epoch": 0.47242105263157896,
      "grad_norm": 0.4288517236709595,
      "learning_rate": 2e-05,
      "loss": 0.6138,
      "step": 5610
    },
    {
      "epoch": 0.47326315789473683,
      "grad_norm": 0.5256401300430298,
      "learning_rate": 2e-05,
      "loss": 0.6402,
      "step": 5620
    },
    {
      "epoch": 0.47410526315789475,
      "grad_norm": 0.40567776560783386,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 5630
    },
    {
      "epoch": 0.4749473684210526,
      "grad_norm": 0.3994002342224121,
      "learning_rate": 2e-05,
      "loss": 0.6063,
      "step": 5640
    },
    {
      "epoch": 0.47578947368421054,
      "grad_norm": 0.4029032289981842,
      "learning_rate": 2e-05,
      "loss": 0.6365,
      "step": 5650
    },
    {
      "epoch": 0.4766315789473684,
      "grad_norm": 0.39134472608566284,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 5660
    },
    {
      "epoch": 0.47747368421052633,
      "grad_norm": 0.3982046842575073,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 5670
    },
    {
      "epoch": 0.4783157894736842,
      "grad_norm": 0.3910711705684662,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 5680
    },
    {
      "epoch": 0.4791578947368421,
      "grad_norm": 0.3979829251766205,
      "learning_rate": 2e-05,
      "loss": 0.6149,
      "step": 5690
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.40359804034233093,
      "learning_rate": 2e-05,
      "loss": 0.6135,
      "step": 5700
    },
    {
      "epoch": 0.4808421052631579,
      "grad_norm": 0.6178655028343201,
      "learning_rate": 2e-05,
      "loss": 0.6481,
      "step": 5710
    },
    {
      "epoch": 0.48168421052631577,
      "grad_norm": 0.425998717546463,
      "learning_rate": 2e-05,
      "loss": 0.6172,
      "step": 5720
    },
    {
      "epoch": 0.4825263157894737,
      "grad_norm": 0.3813927173614502,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 5730
    },
    {
      "epoch": 0.48336842105263156,
      "grad_norm": 0.4358662962913513,
      "learning_rate": 2e-05,
      "loss": 0.6112,
      "step": 5740
    },
    {
      "epoch": 0.4842105263157895,
      "grad_norm": 0.41204842925071716,
      "learning_rate": 2e-05,
      "loss": 0.6209,
      "step": 5750
    },
    {
      "epoch": 0.48505263157894735,
      "grad_norm": 0.41349315643310547,
      "learning_rate": 2e-05,
      "loss": 0.6207,
      "step": 5760
    },
    {
      "epoch": 0.48589473684210527,
      "grad_norm": 0.3568525016307831,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 5770
    },
    {
      "epoch": 0.48673684210526313,
      "grad_norm": 0.45930376648902893,
      "learning_rate": 2e-05,
      "loss": 0.6452,
      "step": 5780
    },
    {
      "epoch": 0.48757894736842106,
      "grad_norm": 0.40785321593284607,
      "learning_rate": 2e-05,
      "loss": 0.6241,
      "step": 5790
    },
    {
      "epoch": 0.4884210526315789,
      "grad_norm": 0.4031713604927063,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 5800
    },
    {
      "epoch": 0.48926315789473684,
      "grad_norm": 0.3818071782588959,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 5810
    },
    {
      "epoch": 0.4901052631578947,
      "grad_norm": 0.3872465193271637,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 5820
    },
    {
      "epoch": 0.49094736842105263,
      "grad_norm": 0.41420939564704895,
      "learning_rate": 2e-05,
      "loss": 0.6504,
      "step": 5830
    },
    {
      "epoch": 0.4917894736842105,
      "grad_norm": 0.42316552996635437,
      "learning_rate": 2e-05,
      "loss": 0.6249,
      "step": 5840
    },
    {
      "epoch": 0.4926315789473684,
      "grad_norm": 0.4556106626987457,
      "learning_rate": 2e-05,
      "loss": 0.6309,
      "step": 5850
    },
    {
      "epoch": 0.49347368421052634,
      "grad_norm": 0.41279733180999756,
      "learning_rate": 2e-05,
      "loss": 0.6482,
      "step": 5860
    },
    {
      "epoch": 0.4943157894736842,
      "grad_norm": 0.4568946361541748,
      "learning_rate": 2e-05,
      "loss": 0.6021,
      "step": 5870
    },
    {
      "epoch": 0.49515789473684213,
      "grad_norm": 0.38475337624549866,
      "learning_rate": 2e-05,
      "loss": 0.6442,
      "step": 5880
    },
    {
      "epoch": 0.496,
      "grad_norm": 0.4127613306045532,
      "learning_rate": 2e-05,
      "loss": 0.6134,
      "step": 5890
    },
    {
      "epoch": 0.4968421052631579,
      "grad_norm": 0.37695586681365967,
      "learning_rate": 2e-05,
      "loss": 0.6295,
      "step": 5900
    },
    {
      "epoch": 0.4976842105263158,
      "grad_norm": 0.42272406816482544,
      "learning_rate": 2e-05,
      "loss": 0.6014,
      "step": 5910
    },
    {
      "epoch": 0.4985263157894737,
      "grad_norm": 0.41959288716316223,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 5920
    },
    {
      "epoch": 0.4993684210526316,
      "grad_norm": 0.3808814287185669,
      "learning_rate": 2e-05,
      "loss": 0.6164,
      "step": 5930
    },
    {
      "epoch": 0.5002105263157894,
      "grad_norm": 0.4124622046947479,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 5940
    },
    {
      "epoch": 0.5010526315789474,
      "grad_norm": 0.3962579667568207,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 5950
    },
    {
      "epoch": 0.5018947368421053,
      "grad_norm": 0.36633360385894775,
      "learning_rate": 2e-05,
      "loss": 0.6058,
      "step": 5960
    },
    {
      "epoch": 0.5027368421052631,
      "grad_norm": 0.3924251198768616,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 5970
    },
    {
      "epoch": 0.503578947368421,
      "grad_norm": 0.3969060182571411,
      "learning_rate": 2e-05,
      "loss": 0.6144,
      "step": 5980
    },
    {
      "epoch": 0.504421052631579,
      "grad_norm": 0.4014529287815094,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 5990
    },
    {
      "epoch": 0.5052631578947369,
      "grad_norm": 0.394632488489151,
      "learning_rate": 2e-05,
      "loss": 0.6025,
      "step": 6000
    },
    {
      "epoch": 0.5061052631578947,
      "grad_norm": 0.4615169167518616,
      "learning_rate": 2e-05,
      "loss": 0.6243,
      "step": 6010
    },
    {
      "epoch": 0.5069473684210526,
      "grad_norm": 0.4311369061470032,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 6020
    },
    {
      "epoch": 0.5077894736842106,
      "grad_norm": 0.39589923620224,
      "learning_rate": 2e-05,
      "loss": 0.6145,
      "step": 6030
    },
    {
      "epoch": 0.5086315789473684,
      "grad_norm": 0.385017991065979,
      "learning_rate": 2e-05,
      "loss": 0.6145,
      "step": 6040
    },
    {
      "epoch": 0.5094736842105263,
      "grad_norm": 0.5104061365127563,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 6050
    },
    {
      "epoch": 0.5103157894736842,
      "grad_norm": 0.3997664451599121,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 6060
    },
    {
      "epoch": 0.5111578947368421,
      "grad_norm": 0.37909621000289917,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 6070
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.4072924852371216,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 6080
    },
    {
      "epoch": 0.5128421052631579,
      "grad_norm": 0.3967787027359009,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 6090
    },
    {
      "epoch": 0.5136842105263157,
      "grad_norm": 0.46256008744239807,
      "learning_rate": 2e-05,
      "loss": 0.6405,
      "step": 6100
    },
    {
      "epoch": 0.5145263157894737,
      "grad_norm": 0.4254210889339447,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 6110
    },
    {
      "epoch": 0.5153684210526316,
      "grad_norm": 0.4701629877090454,
      "learning_rate": 2e-05,
      "loss": 0.6073,
      "step": 6120
    },
    {
      "epoch": 0.5162105263157895,
      "grad_norm": 0.3896213173866272,
      "learning_rate": 2e-05,
      "loss": 0.6256,
      "step": 6130
    },
    {
      "epoch": 0.5170526315789473,
      "grad_norm": 0.38259029388427734,
      "learning_rate": 2e-05,
      "loss": 0.6118,
      "step": 6140
    },
    {
      "epoch": 0.5178947368421053,
      "grad_norm": 0.39990973472595215,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 6150
    },
    {
      "epoch": 0.5187368421052632,
      "grad_norm": 0.37930387258529663,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 6160
    },
    {
      "epoch": 0.519578947368421,
      "grad_norm": 0.41021496057510376,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 6170
    },
    {
      "epoch": 0.5204210526315789,
      "grad_norm": 0.43618178367614746,
      "learning_rate": 2e-05,
      "loss": 0.6162,
      "step": 6180
    },
    {
      "epoch": 0.5212631578947369,
      "grad_norm": 0.36927467584609985,
      "learning_rate": 2e-05,
      "loss": 0.6241,
      "step": 6190
    },
    {
      "epoch": 0.5221052631578947,
      "grad_norm": 0.40083691477775574,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 6200
    },
    {
      "epoch": 0.5229473684210526,
      "grad_norm": 0.4072802662849426,
      "learning_rate": 2e-05,
      "loss": 0.6172,
      "step": 6210
    },
    {
      "epoch": 0.5237894736842105,
      "grad_norm": 0.44511958956718445,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 6220
    },
    {
      "epoch": 0.5246315789473684,
      "grad_norm": 0.3812074065208435,
      "learning_rate": 2e-05,
      "loss": 0.6455,
      "step": 6230
    },
    {
      "epoch": 0.5254736842105263,
      "grad_norm": 0.3866870105266571,
      "learning_rate": 2e-05,
      "loss": 0.6243,
      "step": 6240
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 0.3964068591594696,
      "learning_rate": 2e-05,
      "loss": 0.6286,
      "step": 6250
    },
    {
      "epoch": 0.5271578947368422,
      "grad_norm": 0.3937218189239502,
      "learning_rate": 2e-05,
      "loss": 0.6249,
      "step": 6260
    },
    {
      "epoch": 0.528,
      "grad_norm": 0.43571463227272034,
      "learning_rate": 2e-05,
      "loss": 0.6482,
      "step": 6270
    },
    {
      "epoch": 0.5288421052631579,
      "grad_norm": 0.4065898358821869,
      "learning_rate": 2e-05,
      "loss": 0.638,
      "step": 6280
    },
    {
      "epoch": 0.5296842105263158,
      "grad_norm": 0.39984503388404846,
      "learning_rate": 2e-05,
      "loss": 0.6312,
      "step": 6290
    },
    {
      "epoch": 0.5305263157894737,
      "grad_norm": 0.3921568691730499,
      "learning_rate": 2e-05,
      "loss": 0.6229,
      "step": 6300
    },
    {
      "epoch": 0.5313684210526316,
      "grad_norm": 0.37051403522491455,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 6310
    },
    {
      "epoch": 0.5322105263157895,
      "grad_norm": 0.39028996229171753,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 6320
    },
    {
      "epoch": 0.5330526315789473,
      "grad_norm": 0.40236297249794006,
      "learning_rate": 2e-05,
      "loss": 0.6171,
      "step": 6330
    },
    {
      "epoch": 0.5338947368421053,
      "grad_norm": 0.37906336784362793,
      "learning_rate": 2e-05,
      "loss": 0.6008,
      "step": 6340
    },
    {
      "epoch": 0.5347368421052632,
      "grad_norm": 0.36801713705062866,
      "learning_rate": 2e-05,
      "loss": 0.6421,
      "step": 6350
    },
    {
      "epoch": 0.535578947368421,
      "grad_norm": 0.3978993892669678,
      "learning_rate": 2e-05,
      "loss": 0.6158,
      "step": 6360
    },
    {
      "epoch": 0.5364210526315789,
      "grad_norm": 0.3914985954761505,
      "learning_rate": 2e-05,
      "loss": 0.6192,
      "step": 6370
    },
    {
      "epoch": 0.5372631578947369,
      "grad_norm": 0.3718845248222351,
      "learning_rate": 2e-05,
      "loss": 0.6226,
      "step": 6380
    },
    {
      "epoch": 0.5381052631578948,
      "grad_norm": 0.39565202593803406,
      "learning_rate": 2e-05,
      "loss": 0.6182,
      "step": 6390
    },
    {
      "epoch": 0.5389473684210526,
      "grad_norm": 0.3930019736289978,
      "learning_rate": 2e-05,
      "loss": 0.6352,
      "step": 6400
    },
    {
      "epoch": 0.5397894736842105,
      "grad_norm": 0.4333121180534363,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 6410
    },
    {
      "epoch": 0.5406315789473685,
      "grad_norm": 0.37720903754234314,
      "learning_rate": 2e-05,
      "loss": 0.6191,
      "step": 6420
    },
    {
      "epoch": 0.5414736842105263,
      "grad_norm": 0.38701677322387695,
      "learning_rate": 2e-05,
      "loss": 0.6186,
      "step": 6430
    },
    {
      "epoch": 0.5423157894736842,
      "grad_norm": 0.3843369483947754,
      "learning_rate": 2e-05,
      "loss": 0.6285,
      "step": 6440
    },
    {
      "epoch": 0.5431578947368421,
      "grad_norm": 0.3840426802635193,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 6450
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.4024650752544403,
      "learning_rate": 2e-05,
      "loss": 0.6227,
      "step": 6460
    },
    {
      "epoch": 0.5448421052631579,
      "grad_norm": 0.3999703526496887,
      "learning_rate": 2e-05,
      "loss": 0.6365,
      "step": 6470
    },
    {
      "epoch": 0.5456842105263158,
      "grad_norm": 0.39638224244117737,
      "learning_rate": 2e-05,
      "loss": 0.6099,
      "step": 6480
    },
    {
      "epoch": 0.5465263157894736,
      "grad_norm": 0.42764899134635925,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 6490
    },
    {
      "epoch": 0.5473684210526316,
      "grad_norm": 0.4034484624862671,
      "learning_rate": 2e-05,
      "loss": 0.6244,
      "step": 6500
    },
    {
      "epoch": 0.5482105263157895,
      "grad_norm": 0.39687424898147583,
      "learning_rate": 2e-05,
      "loss": 0.613,
      "step": 6510
    },
    {
      "epoch": 0.5490526315789473,
      "grad_norm": 0.9770718812942505,
      "learning_rate": 2e-05,
      "loss": 0.6135,
      "step": 6520
    },
    {
      "epoch": 0.5498947368421052,
      "grad_norm": 0.5599772334098816,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 6530
    },
    {
      "epoch": 0.5507368421052632,
      "grad_norm": 0.4222373068332672,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 6540
    },
    {
      "epoch": 0.5515789473684211,
      "grad_norm": 0.38663995265960693,
      "learning_rate": 2e-05,
      "loss": 0.6294,
      "step": 6550
    },
    {
      "epoch": 0.5524210526315789,
      "grad_norm": 0.3835684061050415,
      "learning_rate": 2e-05,
      "loss": 0.6164,
      "step": 6560
    },
    {
      "epoch": 0.5532631578947368,
      "grad_norm": 0.39531210064888,
      "learning_rate": 2e-05,
      "loss": 0.6378,
      "step": 6570
    },
    {
      "epoch": 0.5541052631578948,
      "grad_norm": 0.3848089873790741,
      "learning_rate": 2e-05,
      "loss": 0.6203,
      "step": 6580
    },
    {
      "epoch": 0.5549473684210526,
      "grad_norm": 0.4379122853279114,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 6590
    },
    {
      "epoch": 0.5557894736842105,
      "grad_norm": 0.3944644629955292,
      "learning_rate": 2e-05,
      "loss": 0.6418,
      "step": 6600
    },
    {
      "epoch": 0.5566315789473684,
      "grad_norm": 0.39466187357902527,
      "learning_rate": 2e-05,
      "loss": 0.6249,
      "step": 6610
    },
    {
      "epoch": 0.5574736842105263,
      "grad_norm": 0.394088476896286,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 6620
    },
    {
      "epoch": 0.5583157894736842,
      "grad_norm": 0.4348392188549042,
      "learning_rate": 2e-05,
      "loss": 0.6256,
      "step": 6630
    },
    {
      "epoch": 0.5591578947368421,
      "grad_norm": 0.3976849913597107,
      "learning_rate": 2e-05,
      "loss": 0.6208,
      "step": 6640
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4238644540309906,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 6650
    },
    {
      "epoch": 0.5608421052631579,
      "grad_norm": 0.35792040824890137,
      "learning_rate": 2e-05,
      "loss": 0.6411,
      "step": 6660
    },
    {
      "epoch": 0.5616842105263158,
      "grad_norm": 0.41421791911125183,
      "learning_rate": 2e-05,
      "loss": 0.6344,
      "step": 6670
    },
    {
      "epoch": 0.5625263157894737,
      "grad_norm": 0.4069333076477051,
      "learning_rate": 2e-05,
      "loss": 0.6194,
      "step": 6680
    },
    {
      "epoch": 0.5633684210526316,
      "grad_norm": 0.5400185585021973,
      "learning_rate": 2e-05,
      "loss": 0.615,
      "step": 6690
    },
    {
      "epoch": 0.5642105263157895,
      "grad_norm": 0.4212380051612854,
      "learning_rate": 2e-05,
      "loss": 0.6209,
      "step": 6700
    },
    {
      "epoch": 0.5650526315789474,
      "grad_norm": 0.38524165749549866,
      "learning_rate": 2e-05,
      "loss": 0.6094,
      "step": 6710
    },
    {
      "epoch": 0.5658947368421052,
      "grad_norm": 0.400897353887558,
      "learning_rate": 2e-05,
      "loss": 0.6423,
      "step": 6720
    },
    {
      "epoch": 0.5667368421052632,
      "grad_norm": 0.3886755704879761,
      "learning_rate": 2e-05,
      "loss": 0.6267,
      "step": 6730
    },
    {
      "epoch": 0.5675789473684211,
      "grad_norm": 0.41893553733825684,
      "learning_rate": 2e-05,
      "loss": 0.6028,
      "step": 6740
    },
    {
      "epoch": 0.5684210526315789,
      "grad_norm": 0.3731280267238617,
      "learning_rate": 2e-05,
      "loss": 0.6157,
      "step": 6750
    },
    {
      "epoch": 0.5692631578947368,
      "grad_norm": 0.43054476380348206,
      "learning_rate": 2e-05,
      "loss": 0.6203,
      "step": 6760
    },
    {
      "epoch": 0.5701052631578948,
      "grad_norm": 0.5672470331192017,
      "learning_rate": 2e-05,
      "loss": 0.6291,
      "step": 6770
    },
    {
      "epoch": 0.5709473684210526,
      "grad_norm": 0.3889347016811371,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 6780
    },
    {
      "epoch": 0.5717894736842105,
      "grad_norm": 0.3816850483417511,
      "learning_rate": 2e-05,
      "loss": 0.6117,
      "step": 6790
    },
    {
      "epoch": 0.5726315789473684,
      "grad_norm": 0.42927634716033936,
      "learning_rate": 2e-05,
      "loss": 0.6187,
      "step": 6800
    },
    {
      "epoch": 0.5734736842105264,
      "grad_norm": 0.394616961479187,
      "learning_rate": 2e-05,
      "loss": 0.5918,
      "step": 6810
    },
    {
      "epoch": 0.5743157894736842,
      "grad_norm": 0.4170050621032715,
      "learning_rate": 2e-05,
      "loss": 0.6122,
      "step": 6820
    },
    {
      "epoch": 0.5751578947368421,
      "grad_norm": 0.36826762557029724,
      "learning_rate": 2e-05,
      "loss": 0.6125,
      "step": 6830
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.35624295473098755,
      "learning_rate": 2e-05,
      "loss": 0.6102,
      "step": 6840
    },
    {
      "epoch": 0.5768421052631579,
      "grad_norm": 0.41554713249206543,
      "learning_rate": 2e-05,
      "loss": 0.601,
      "step": 6850
    },
    {
      "epoch": 0.5776842105263158,
      "grad_norm": 0.4252447187900543,
      "learning_rate": 2e-05,
      "loss": 0.6103,
      "step": 6860
    },
    {
      "epoch": 0.5785263157894737,
      "grad_norm": 0.38633179664611816,
      "learning_rate": 2e-05,
      "loss": 0.602,
      "step": 6870
    },
    {
      "epoch": 0.5793684210526315,
      "grad_norm": 0.4050305187702179,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 6880
    },
    {
      "epoch": 0.5802105263157895,
      "grad_norm": 0.39746952056884766,
      "learning_rate": 2e-05,
      "loss": 0.6294,
      "step": 6890
    },
    {
      "epoch": 0.5810526315789474,
      "grad_norm": 0.4022492468357086,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 6900
    },
    {
      "epoch": 0.5818947368421052,
      "grad_norm": 0.4315609633922577,
      "learning_rate": 2e-05,
      "loss": 0.6012,
      "step": 6910
    },
    {
      "epoch": 0.5827368421052631,
      "grad_norm": 0.4079793095588684,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 6920
    },
    {
      "epoch": 0.5835789473684211,
      "grad_norm": 0.4974655508995056,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 6930
    },
    {
      "epoch": 0.584421052631579,
      "grad_norm": 0.43974098563194275,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 6940
    },
    {
      "epoch": 0.5852631578947368,
      "grad_norm": 0.3788515329360962,
      "learning_rate": 2e-05,
      "loss": 0.6197,
      "step": 6950
    },
    {
      "epoch": 0.5861052631578947,
      "grad_norm": 0.4458039402961731,
      "learning_rate": 2e-05,
      "loss": 0.6091,
      "step": 6960
    },
    {
      "epoch": 0.5869473684210527,
      "grad_norm": 0.3835600018501282,
      "learning_rate": 2e-05,
      "loss": 0.6298,
      "step": 6970
    },
    {
      "epoch": 0.5877894736842105,
      "grad_norm": 0.4597538411617279,
      "learning_rate": 2e-05,
      "loss": 0.6463,
      "step": 6980
    },
    {
      "epoch": 0.5886315789473684,
      "grad_norm": 0.4088018536567688,
      "learning_rate": 2e-05,
      "loss": 0.6578,
      "step": 6990
    },
    {
      "epoch": 0.5894736842105263,
      "grad_norm": 0.3554892838001251,
      "learning_rate": 2e-05,
      "loss": 0.6141,
      "step": 7000
    },
    {
      "epoch": 0.5903157894736842,
      "grad_norm": 0.39358317852020264,
      "learning_rate": 2e-05,
      "loss": 0.6141,
      "step": 7010
    },
    {
      "epoch": 0.5911578947368421,
      "grad_norm": 0.3419573903083801,
      "learning_rate": 2e-05,
      "loss": 0.6218,
      "step": 7020
    },
    {
      "epoch": 0.592,
      "grad_norm": 0.3773942291736603,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 7030
    },
    {
      "epoch": 0.592842105263158,
      "grad_norm": 0.3934532701969147,
      "learning_rate": 2e-05,
      "loss": 0.6142,
      "step": 7040
    },
    {
      "epoch": 0.5936842105263158,
      "grad_norm": 0.3976365029811859,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 7050
    },
    {
      "epoch": 0.5945263157894737,
      "grad_norm": 0.375040203332901,
      "learning_rate": 2e-05,
      "loss": 0.6241,
      "step": 7060
    },
    {
      "epoch": 0.5953684210526315,
      "grad_norm": 0.37573471665382385,
      "learning_rate": 2e-05,
      "loss": 0.616,
      "step": 7070
    },
    {
      "epoch": 0.5962105263157895,
      "grad_norm": 0.4249876141548157,
      "learning_rate": 2e-05,
      "loss": 0.6194,
      "step": 7080
    },
    {
      "epoch": 0.5970526315789474,
      "grad_norm": 0.3869653046131134,
      "learning_rate": 2e-05,
      "loss": 0.6442,
      "step": 7090
    },
    {
      "epoch": 0.5978947368421053,
      "grad_norm": 0.4078359007835388,
      "learning_rate": 2e-05,
      "loss": 0.6265,
      "step": 7100
    },
    {
      "epoch": 0.5987368421052631,
      "grad_norm": 0.43414589762687683,
      "learning_rate": 2e-05,
      "loss": 0.6108,
      "step": 7110
    },
    {
      "epoch": 0.5995789473684211,
      "grad_norm": 0.378732830286026,
      "learning_rate": 2e-05,
      "loss": 0.6158,
      "step": 7120
    },
    {
      "epoch": 0.600421052631579,
      "grad_norm": 0.39870065450668335,
      "learning_rate": 2e-05,
      "loss": 0.6314,
      "step": 7130
    },
    {
      "epoch": 0.6012631578947368,
      "grad_norm": 0.3718423545360565,
      "learning_rate": 2e-05,
      "loss": 0.6251,
      "step": 7140
    },
    {
      "epoch": 0.6021052631578947,
      "grad_norm": 0.4216981530189514,
      "learning_rate": 2e-05,
      "loss": 0.6057,
      "step": 7150
    },
    {
      "epoch": 0.6029473684210527,
      "grad_norm": 0.35097551345825195,
      "learning_rate": 2e-05,
      "loss": 0.5891,
      "step": 7160
    },
    {
      "epoch": 0.6037894736842105,
      "grad_norm": 0.41022756695747375,
      "learning_rate": 2e-05,
      "loss": 0.6081,
      "step": 7170
    },
    {
      "epoch": 0.6046315789473684,
      "grad_norm": 0.4306015968322754,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 7180
    },
    {
      "epoch": 0.6054736842105263,
      "grad_norm": 0.3911057710647583,
      "learning_rate": 2e-05,
      "loss": 0.6171,
      "step": 7190
    },
    {
      "epoch": 0.6063157894736843,
      "grad_norm": 0.3893387019634247,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 7200
    },
    {
      "epoch": 0.6071578947368421,
      "grad_norm": 0.4134942293167114,
      "learning_rate": 2e-05,
      "loss": 0.6181,
      "step": 7210
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.39391592144966125,
      "learning_rate": 2e-05,
      "loss": 0.6343,
      "step": 7220
    },
    {
      "epoch": 0.6088421052631579,
      "grad_norm": 0.4187861382961273,
      "learning_rate": 2e-05,
      "loss": 0.6248,
      "step": 7230
    },
    {
      "epoch": 0.6096842105263158,
      "grad_norm": 0.3935670554637909,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 7240
    },
    {
      "epoch": 0.6105263157894737,
      "grad_norm": 0.3879421353340149,
      "learning_rate": 2e-05,
      "loss": 0.615,
      "step": 7250
    },
    {
      "epoch": 0.6113684210526316,
      "grad_norm": 0.38003504276275635,
      "learning_rate": 2e-05,
      "loss": 0.6137,
      "step": 7260
    },
    {
      "epoch": 0.6122105263157894,
      "grad_norm": 0.3767760396003723,
      "learning_rate": 2e-05,
      "loss": 0.6135,
      "step": 7270
    },
    {
      "epoch": 0.6130526315789474,
      "grad_norm": 0.39842909574508667,
      "learning_rate": 2e-05,
      "loss": 0.6198,
      "step": 7280
    },
    {
      "epoch": 0.6138947368421053,
      "grad_norm": 0.4637824594974518,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 7290
    },
    {
      "epoch": 0.6147368421052631,
      "grad_norm": 0.37352022528648376,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 7300
    },
    {
      "epoch": 0.615578947368421,
      "grad_norm": 0.3793522119522095,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 7310
    },
    {
      "epoch": 0.616421052631579,
      "grad_norm": 0.38974273204803467,
      "learning_rate": 2e-05,
      "loss": 0.5876,
      "step": 7320
    },
    {
      "epoch": 0.6172631578947368,
      "grad_norm": 0.4033825993537903,
      "learning_rate": 2e-05,
      "loss": 0.5867,
      "step": 7330
    },
    {
      "epoch": 0.6181052631578947,
      "grad_norm": 0.37120676040649414,
      "learning_rate": 2e-05,
      "loss": 0.6018,
      "step": 7340
    },
    {
      "epoch": 0.6189473684210526,
      "grad_norm": 0.3990624248981476,
      "learning_rate": 2e-05,
      "loss": 0.6177,
      "step": 7350
    },
    {
      "epoch": 0.6197894736842106,
      "grad_norm": 0.3737449645996094,
      "learning_rate": 2e-05,
      "loss": 0.6105,
      "step": 7360
    },
    {
      "epoch": 0.6206315789473684,
      "grad_norm": 0.43965956568717957,
      "learning_rate": 2e-05,
      "loss": 0.6234,
      "step": 7370
    },
    {
      "epoch": 0.6214736842105263,
      "grad_norm": 0.3872416913509369,
      "learning_rate": 2e-05,
      "loss": 0.6186,
      "step": 7380
    },
    {
      "epoch": 0.6223157894736842,
      "grad_norm": 0.3955213129520416,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 7390
    },
    {
      "epoch": 0.6231578947368421,
      "grad_norm": 0.3895691931247711,
      "learning_rate": 2e-05,
      "loss": 0.6457,
      "step": 7400
    },
    {
      "epoch": 0.624,
      "grad_norm": 0.3872500956058502,
      "learning_rate": 2e-05,
      "loss": 0.6218,
      "step": 7410
    },
    {
      "epoch": 0.6248421052631579,
      "grad_norm": 0.38227546215057373,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 7420
    },
    {
      "epoch": 0.6256842105263158,
      "grad_norm": 0.39810335636138916,
      "learning_rate": 2e-05,
      "loss": 0.639,
      "step": 7430
    },
    {
      "epoch": 0.6265263157894737,
      "grad_norm": 0.4022630751132965,
      "learning_rate": 2e-05,
      "loss": 0.6205,
      "step": 7440
    },
    {
      "epoch": 0.6273684210526316,
      "grad_norm": 0.389854371547699,
      "learning_rate": 2e-05,
      "loss": 0.6057,
      "step": 7450
    },
    {
      "epoch": 0.6282105263157894,
      "grad_norm": 0.3680373430252075,
      "learning_rate": 2e-05,
      "loss": 0.6136,
      "step": 7460
    },
    {
      "epoch": 0.6290526315789474,
      "grad_norm": 0.43438953161239624,
      "learning_rate": 2e-05,
      "loss": 0.6177,
      "step": 7470
    },
    {
      "epoch": 0.6298947368421053,
      "grad_norm": 0.34751957654953003,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 7480
    },
    {
      "epoch": 0.6307368421052632,
      "grad_norm": 0.3754175901412964,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 7490
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": 0.37520673871040344,
      "learning_rate": 2e-05,
      "loss": 0.635,
      "step": 7500
    },
    {
      "epoch": 0.632421052631579,
      "grad_norm": 0.3866310119628906,
      "learning_rate": 2e-05,
      "loss": 0.6122,
      "step": 7510
    },
    {
      "epoch": 0.6332631578947369,
      "grad_norm": 0.42295631766319275,
      "learning_rate": 2e-05,
      "loss": 0.5981,
      "step": 7520
    },
    {
      "epoch": 0.6341052631578947,
      "grad_norm": 0.40831121802330017,
      "learning_rate": 2e-05,
      "loss": 0.6277,
      "step": 7530
    },
    {
      "epoch": 0.6349473684210526,
      "grad_norm": 0.40017178654670715,
      "learning_rate": 2e-05,
      "loss": 0.5961,
      "step": 7540
    },
    {
      "epoch": 0.6357894736842106,
      "grad_norm": 0.4313274919986725,
      "learning_rate": 2e-05,
      "loss": 0.6255,
      "step": 7550
    },
    {
      "epoch": 0.6366315789473684,
      "grad_norm": 0.424806147813797,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 7560
    },
    {
      "epoch": 0.6374736842105263,
      "grad_norm": 0.4185604155063629,
      "learning_rate": 2e-05,
      "loss": 0.6385,
      "step": 7570
    },
    {
      "epoch": 0.6383157894736842,
      "grad_norm": 0.3939436078071594,
      "learning_rate": 2e-05,
      "loss": 0.6057,
      "step": 7580
    },
    {
      "epoch": 0.6391578947368421,
      "grad_norm": 0.3819863498210907,
      "learning_rate": 2e-05,
      "loss": 0.596,
      "step": 7590
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3709205985069275,
      "learning_rate": 2e-05,
      "loss": 0.6213,
      "step": 7600
    },
    {
      "epoch": 0.6408421052631579,
      "grad_norm": 0.37696346640586853,
      "learning_rate": 2e-05,
      "loss": 0.5985,
      "step": 7610
    },
    {
      "epoch": 0.6416842105263157,
      "grad_norm": 0.40637457370758057,
      "learning_rate": 2e-05,
      "loss": 0.6293,
      "step": 7620
    },
    {
      "epoch": 0.6425263157894737,
      "grad_norm": 0.4028889536857605,
      "learning_rate": 2e-05,
      "loss": 0.6225,
      "step": 7630
    },
    {
      "epoch": 0.6433684210526316,
      "grad_norm": 0.3886878490447998,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 7640
    },
    {
      "epoch": 0.6442105263157895,
      "grad_norm": 0.42368072271347046,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 7650
    },
    {
      "epoch": 0.6450526315789473,
      "grad_norm": 0.37349072098731995,
      "learning_rate": 2e-05,
      "loss": 0.6198,
      "step": 7660
    },
    {
      "epoch": 0.6458947368421053,
      "grad_norm": 0.40476658940315247,
      "learning_rate": 2e-05,
      "loss": 0.6066,
      "step": 7670
    },
    {
      "epoch": 0.6467368421052632,
      "grad_norm": 0.39988085627555847,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 7680
    },
    {
      "epoch": 0.647578947368421,
      "grad_norm": 0.6748995184898376,
      "learning_rate": 2e-05,
      "loss": 0.6118,
      "step": 7690
    },
    {
      "epoch": 0.6484210526315789,
      "grad_norm": 0.41185885667800903,
      "learning_rate": 2e-05,
      "loss": 0.6229,
      "step": 7700
    },
    {
      "epoch": 0.6492631578947369,
      "grad_norm": 0.40317633748054504,
      "learning_rate": 2e-05,
      "loss": 0.6225,
      "step": 7710
    },
    {
      "epoch": 0.6501052631578947,
      "grad_norm": 0.3782958388328552,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 7720
    },
    {
      "epoch": 0.6509473684210526,
      "grad_norm": 0.37397336959838867,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 7730
    },
    {
      "epoch": 0.6517894736842105,
      "grad_norm": 0.4152805805206299,
      "learning_rate": 2e-05,
      "loss": 0.6139,
      "step": 7740
    },
    {
      "epoch": 0.6526315789473685,
      "grad_norm": 0.4015020728111267,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 7750
    },
    {
      "epoch": 0.6534736842105263,
      "grad_norm": 0.4377886652946472,
      "learning_rate": 2e-05,
      "loss": 0.6375,
      "step": 7760
    },
    {
      "epoch": 0.6543157894736842,
      "grad_norm": 0.38124382495880127,
      "learning_rate": 2e-05,
      "loss": 0.6099,
      "step": 7770
    },
    {
      "epoch": 0.655157894736842,
      "grad_norm": 0.40309011936187744,
      "learning_rate": 2e-05,
      "loss": 0.606,
      "step": 7780
    },
    {
      "epoch": 0.656,
      "grad_norm": 0.36981794238090515,
      "learning_rate": 2e-05,
      "loss": 0.6226,
      "step": 7790
    },
    {
      "epoch": 0.6568421052631579,
      "grad_norm": 0.6922754049301147,
      "learning_rate": 2e-05,
      "loss": 0.6092,
      "step": 7800
    },
    {
      "epoch": 0.6576842105263158,
      "grad_norm": 0.39232248067855835,
      "learning_rate": 2e-05,
      "loss": 0.6287,
      "step": 7810
    },
    {
      "epoch": 0.6585263157894737,
      "grad_norm": 0.39232611656188965,
      "learning_rate": 2e-05,
      "loss": 0.586,
      "step": 7820
    },
    {
      "epoch": 0.6593684210526316,
      "grad_norm": 0.42964327335357666,
      "learning_rate": 2e-05,
      "loss": 0.617,
      "step": 7830
    },
    {
      "epoch": 0.6602105263157895,
      "grad_norm": 0.5250550508499146,
      "learning_rate": 2e-05,
      "loss": 0.6129,
      "step": 7840
    },
    {
      "epoch": 0.6610526315789473,
      "grad_norm": 0.44053763151168823,
      "learning_rate": 2e-05,
      "loss": 0.6222,
      "step": 7850
    },
    {
      "epoch": 0.6618947368421053,
      "grad_norm": 0.3583446443080902,
      "learning_rate": 2e-05,
      "loss": 0.6124,
      "step": 7860
    },
    {
      "epoch": 0.6627368421052632,
      "grad_norm": 0.426673024892807,
      "learning_rate": 2e-05,
      "loss": 0.6315,
      "step": 7870
    },
    {
      "epoch": 0.663578947368421,
      "grad_norm": 0.37713444232940674,
      "learning_rate": 2e-05,
      "loss": 0.6139,
      "step": 7880
    },
    {
      "epoch": 0.6644210526315789,
      "grad_norm": 0.4152722656726837,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 7890
    },
    {
      "epoch": 0.6652631578947369,
      "grad_norm": 0.37450534105300903,
      "learning_rate": 2e-05,
      "loss": 0.5873,
      "step": 7900
    },
    {
      "epoch": 0.6661052631578948,
      "grad_norm": 0.4055245816707611,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 7910
    },
    {
      "epoch": 0.6669473684210526,
      "grad_norm": 0.3808107078075409,
      "learning_rate": 2e-05,
      "loss": 0.6055,
      "step": 7920
    },
    {
      "epoch": 0.6677894736842105,
      "grad_norm": 0.4053349494934082,
      "learning_rate": 2e-05,
      "loss": 0.6172,
      "step": 7930
    },
    {
      "epoch": 0.6686315789473685,
      "grad_norm": 0.3777830898761749,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 7940
    },
    {
      "epoch": 0.6694736842105263,
      "grad_norm": 0.38636314868927,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 7950
    },
    {
      "epoch": 0.6703157894736842,
      "grad_norm": 0.40595880150794983,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 7960
    },
    {
      "epoch": 0.6711578947368421,
      "grad_norm": 0.4244743287563324,
      "learning_rate": 2e-05,
      "loss": 0.6216,
      "step": 7970
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.4173278212547302,
      "learning_rate": 2e-05,
      "loss": 0.6117,
      "step": 7980
    },
    {
      "epoch": 0.6728421052631579,
      "grad_norm": 0.3786630928516388,
      "learning_rate": 2e-05,
      "loss": 0.6265,
      "step": 7990
    },
    {
      "epoch": 0.6736842105263158,
      "grad_norm": 0.3942064046859741,
      "learning_rate": 2e-05,
      "loss": 0.5994,
      "step": 8000
    },
    {
      "epoch": 0.6745263157894736,
      "grad_norm": 0.4017740786075592,
      "learning_rate": 2e-05,
      "loss": 0.618,
      "step": 8010
    },
    {
      "epoch": 0.6753684210526316,
      "grad_norm": 0.42018380761146545,
      "learning_rate": 2e-05,
      "loss": 0.6086,
      "step": 8020
    },
    {
      "epoch": 0.6762105263157895,
      "grad_norm": 0.3867743909358978,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 8030
    },
    {
      "epoch": 0.6770526315789474,
      "grad_norm": 0.3890368342399597,
      "learning_rate": 2e-05,
      "loss": 0.6224,
      "step": 8040
    },
    {
      "epoch": 0.6778947368421052,
      "grad_norm": 0.38337814807891846,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 8050
    },
    {
      "epoch": 0.6787368421052632,
      "grad_norm": 0.4176730811595917,
      "learning_rate": 2e-05,
      "loss": 0.6118,
      "step": 8060
    },
    {
      "epoch": 0.6795789473684211,
      "grad_norm": 0.3962196111679077,
      "learning_rate": 2e-05,
      "loss": 0.6097,
      "step": 8070
    },
    {
      "epoch": 0.6804210526315789,
      "grad_norm": 0.4186159372329712,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 8080
    },
    {
      "epoch": 0.6812631578947368,
      "grad_norm": 0.4013622999191284,
      "learning_rate": 2e-05,
      "loss": 0.6093,
      "step": 8090
    },
    {
      "epoch": 0.6821052631578948,
      "grad_norm": 0.3850988447666168,
      "learning_rate": 2e-05,
      "loss": 0.6071,
      "step": 8100
    },
    {
      "epoch": 0.6829473684210526,
      "grad_norm": 0.4025191068649292,
      "learning_rate": 2e-05,
      "loss": 0.6136,
      "step": 8110
    },
    {
      "epoch": 0.6837894736842105,
      "grad_norm": 0.3868328928947449,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 8120
    },
    {
      "epoch": 0.6846315789473684,
      "grad_norm": 0.38254281878471375,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 8130
    },
    {
      "epoch": 0.6854736842105263,
      "grad_norm": 0.41677263379096985,
      "learning_rate": 2e-05,
      "loss": 0.6228,
      "step": 8140
    },
    {
      "epoch": 0.6863157894736842,
      "grad_norm": 0.38550490140914917,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 8150
    },
    {
      "epoch": 0.6871578947368421,
      "grad_norm": 0.41530483961105347,
      "learning_rate": 2e-05,
      "loss": 0.6047,
      "step": 8160
    },
    {
      "epoch": 0.688,
      "grad_norm": 0.41003069281578064,
      "learning_rate": 2e-05,
      "loss": 0.5965,
      "step": 8170
    },
    {
      "epoch": 0.6888421052631579,
      "grad_norm": 0.40380245447158813,
      "learning_rate": 2e-05,
      "loss": 0.6081,
      "step": 8180
    },
    {
      "epoch": 0.6896842105263158,
      "grad_norm": 0.4276886582374573,
      "learning_rate": 2e-05,
      "loss": 0.641,
      "step": 8190
    },
    {
      "epoch": 0.6905263157894737,
      "grad_norm": 0.42920419573783875,
      "learning_rate": 2e-05,
      "loss": 0.6163,
      "step": 8200
    },
    {
      "epoch": 0.6913684210526316,
      "grad_norm": 0.38665059208869934,
      "learning_rate": 2e-05,
      "loss": 0.592,
      "step": 8210
    },
    {
      "epoch": 0.6922105263157895,
      "grad_norm": 0.4015626311302185,
      "learning_rate": 2e-05,
      "loss": 0.6191,
      "step": 8220
    },
    {
      "epoch": 0.6930526315789474,
      "grad_norm": 0.3644522428512573,
      "learning_rate": 2e-05,
      "loss": 0.6072,
      "step": 8230
    },
    {
      "epoch": 0.6938947368421052,
      "grad_norm": 0.4537876844406128,
      "learning_rate": 2e-05,
      "loss": 0.5882,
      "step": 8240
    },
    {
      "epoch": 0.6947368421052632,
      "grad_norm": 0.38193899393081665,
      "learning_rate": 2e-05,
      "loss": 0.6052,
      "step": 8250
    },
    {
      "epoch": 0.6955789473684211,
      "grad_norm": 0.3975464403629303,
      "learning_rate": 2e-05,
      "loss": 0.594,
      "step": 8260
    },
    {
      "epoch": 0.6964210526315789,
      "grad_norm": 0.6420904994010925,
      "learning_rate": 2e-05,
      "loss": 0.6085,
      "step": 8270
    },
    {
      "epoch": 0.6972631578947368,
      "grad_norm": 0.4209294021129608,
      "learning_rate": 2e-05,
      "loss": 0.6286,
      "step": 8280
    },
    {
      "epoch": 0.6981052631578948,
      "grad_norm": 0.41728150844573975,
      "learning_rate": 2e-05,
      "loss": 0.6354,
      "step": 8290
    },
    {
      "epoch": 0.6989473684210527,
      "grad_norm": 0.3867236077785492,
      "learning_rate": 2e-05,
      "loss": 0.6044,
      "step": 8300
    },
    {
      "epoch": 0.6997894736842105,
      "grad_norm": 0.4111960828304291,
      "learning_rate": 2e-05,
      "loss": 0.6235,
      "step": 8310
    },
    {
      "epoch": 0.7006315789473684,
      "grad_norm": 0.4291996657848358,
      "learning_rate": 2e-05,
      "loss": 0.6126,
      "step": 8320
    },
    {
      "epoch": 0.7014736842105264,
      "grad_norm": 0.38982871174812317,
      "learning_rate": 2e-05,
      "loss": 0.6238,
      "step": 8330
    },
    {
      "epoch": 0.7023157894736842,
      "grad_norm": 0.3766862452030182,
      "learning_rate": 2e-05,
      "loss": 0.6241,
      "step": 8340
    },
    {
      "epoch": 0.7031578947368421,
      "grad_norm": 0.41615402698516846,
      "learning_rate": 2e-05,
      "loss": 0.6239,
      "step": 8350
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.39169448614120483,
      "learning_rate": 2e-05,
      "loss": 0.6404,
      "step": 8360
    },
    {
      "epoch": 0.7048421052631579,
      "grad_norm": 0.39166396856307983,
      "learning_rate": 2e-05,
      "loss": 0.6138,
      "step": 8370
    },
    {
      "epoch": 0.7056842105263158,
      "grad_norm": 0.3843355178833008,
      "learning_rate": 2e-05,
      "loss": 0.6313,
      "step": 8380
    },
    {
      "epoch": 0.7065263157894737,
      "grad_norm": 0.38467076420783997,
      "learning_rate": 2e-05,
      "loss": 0.608,
      "step": 8390
    },
    {
      "epoch": 0.7073684210526315,
      "grad_norm": 0.3867534399032593,
      "learning_rate": 2e-05,
      "loss": 0.6233,
      "step": 8400
    },
    {
      "epoch": 0.7082105263157895,
      "grad_norm": 0.39145153760910034,
      "learning_rate": 2e-05,
      "loss": 0.6338,
      "step": 8410
    },
    {
      "epoch": 0.7090526315789474,
      "grad_norm": 0.3922288119792938,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 8420
    },
    {
      "epoch": 0.7098947368421052,
      "grad_norm": 0.4388987720012665,
      "learning_rate": 2e-05,
      "loss": 0.6115,
      "step": 8430
    },
    {
      "epoch": 0.7107368421052631,
      "grad_norm": 0.4052526354789734,
      "learning_rate": 2e-05,
      "loss": 0.5961,
      "step": 8440
    },
    {
      "epoch": 0.7115789473684211,
      "grad_norm": 0.3902601897716522,
      "learning_rate": 2e-05,
      "loss": 0.6238,
      "step": 8450
    },
    {
      "epoch": 0.712421052631579,
      "grad_norm": 0.4350232779979706,
      "learning_rate": 2e-05,
      "loss": 0.6283,
      "step": 8460
    },
    {
      "epoch": 0.7132631578947368,
      "grad_norm": 0.40290361642837524,
      "learning_rate": 2e-05,
      "loss": 0.6091,
      "step": 8470
    },
    {
      "epoch": 0.7141052631578947,
      "grad_norm": 0.38193315267562866,
      "learning_rate": 2e-05,
      "loss": 0.607,
      "step": 8480
    },
    {
      "epoch": 0.7149473684210527,
      "grad_norm": 0.3941733241081238,
      "learning_rate": 2e-05,
      "loss": 0.6484,
      "step": 8490
    },
    {
      "epoch": 0.7157894736842105,
      "grad_norm": 0.39484620094299316,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 8500
    },
    {
      "epoch": 0.7166315789473684,
      "grad_norm": 0.3932025730609894,
      "learning_rate": 2e-05,
      "loss": 0.6131,
      "step": 8510
    },
    {
      "epoch": 0.7174736842105263,
      "grad_norm": 0.3656826913356781,
      "learning_rate": 2e-05,
      "loss": 0.6233,
      "step": 8520
    },
    {
      "epoch": 0.7183157894736842,
      "grad_norm": 0.3975438177585602,
      "learning_rate": 2e-05,
      "loss": 0.6269,
      "step": 8530
    },
    {
      "epoch": 0.7191578947368421,
      "grad_norm": 0.39981529116630554,
      "learning_rate": 2e-05,
      "loss": 0.5867,
      "step": 8540
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4163762927055359,
      "learning_rate": 2e-05,
      "loss": 0.6227,
      "step": 8550
    },
    {
      "epoch": 0.7208421052631578,
      "grad_norm": 0.3717016875743866,
      "learning_rate": 2e-05,
      "loss": 0.6079,
      "step": 8560
    },
    {
      "epoch": 0.7216842105263158,
      "grad_norm": 0.3649222254753113,
      "learning_rate": 2e-05,
      "loss": 0.6008,
      "step": 8570
    },
    {
      "epoch": 0.7225263157894737,
      "grad_norm": 0.3871695399284363,
      "learning_rate": 2e-05,
      "loss": 0.5958,
      "step": 8580
    },
    {
      "epoch": 0.7233684210526315,
      "grad_norm": 0.43108946084976196,
      "learning_rate": 2e-05,
      "loss": 0.6237,
      "step": 8590
    },
    {
      "epoch": 0.7242105263157895,
      "grad_norm": 0.39820754528045654,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 8600
    },
    {
      "epoch": 0.7250526315789474,
      "grad_norm": 0.3852376341819763,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 8610
    },
    {
      "epoch": 0.7258947368421053,
      "grad_norm": 0.4331970512866974,
      "learning_rate": 2e-05,
      "loss": 0.5972,
      "step": 8620
    },
    {
      "epoch": 0.7267368421052631,
      "grad_norm": 0.3896394371986389,
      "learning_rate": 2e-05,
      "loss": 0.5993,
      "step": 8630
    },
    {
      "epoch": 0.7275789473684211,
      "grad_norm": 0.3932952880859375,
      "learning_rate": 2e-05,
      "loss": 0.6467,
      "step": 8640
    },
    {
      "epoch": 0.728421052631579,
      "grad_norm": 0.3805220127105713,
      "learning_rate": 2e-05,
      "loss": 0.6216,
      "step": 8650
    },
    {
      "epoch": 0.7292631578947368,
      "grad_norm": 0.46920347213745117,
      "learning_rate": 2e-05,
      "loss": 0.6159,
      "step": 8660
    },
    {
      "epoch": 0.7301052631578947,
      "grad_norm": 0.4276695251464844,
      "learning_rate": 2e-05,
      "loss": 0.6052,
      "step": 8670
    },
    {
      "epoch": 0.7309473684210527,
      "grad_norm": 0.4183971583843231,
      "learning_rate": 2e-05,
      "loss": 0.6151,
      "step": 8680
    },
    {
      "epoch": 0.7317894736842105,
      "grad_norm": 0.3821810781955719,
      "learning_rate": 2e-05,
      "loss": 0.6076,
      "step": 8690
    },
    {
      "epoch": 0.7326315789473684,
      "grad_norm": 0.5641628503799438,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 8700
    },
    {
      "epoch": 0.7334736842105263,
      "grad_norm": 0.4875067472457886,
      "learning_rate": 2e-05,
      "loss": 0.624,
      "step": 8710
    },
    {
      "epoch": 0.7343157894736843,
      "grad_norm": 0.3897894024848938,
      "learning_rate": 2e-05,
      "loss": 0.6049,
      "step": 8720
    },
    {
      "epoch": 0.7351578947368421,
      "grad_norm": 0.5182130336761475,
      "learning_rate": 2e-05,
      "loss": 0.6064,
      "step": 8730
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.3910011649131775,
      "learning_rate": 2e-05,
      "loss": 0.6106,
      "step": 8740
    },
    {
      "epoch": 0.7368421052631579,
      "grad_norm": 0.38370877504348755,
      "learning_rate": 2e-05,
      "loss": 0.6412,
      "step": 8750
    },
    {
      "epoch": 0.7376842105263158,
      "grad_norm": 0.37845680117607117,
      "learning_rate": 2e-05,
      "loss": 0.6143,
      "step": 8760
    },
    {
      "epoch": 0.7385263157894737,
      "grad_norm": 0.4932238757610321,
      "learning_rate": 2e-05,
      "loss": 0.6198,
      "step": 8770
    },
    {
      "epoch": 0.7393684210526316,
      "grad_norm": 0.3976691663265228,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 8780
    },
    {
      "epoch": 0.7402105263157894,
      "grad_norm": 0.3707191050052643,
      "learning_rate": 2e-05,
      "loss": 0.5797,
      "step": 8790
    },
    {
      "epoch": 0.7410526315789474,
      "grad_norm": 0.3922211527824402,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 8800
    },
    {
      "epoch": 0.7418947368421053,
      "grad_norm": 0.3702085018157959,
      "learning_rate": 2e-05,
      "loss": 0.6333,
      "step": 8810
    },
    {
      "epoch": 0.7427368421052631,
      "grad_norm": 0.3686163127422333,
      "learning_rate": 2e-05,
      "loss": 0.5913,
      "step": 8820
    },
    {
      "epoch": 0.743578947368421,
      "grad_norm": 0.3886764645576477,
      "learning_rate": 2e-05,
      "loss": 0.5892,
      "step": 8830
    },
    {
      "epoch": 0.744421052631579,
      "grad_norm": 0.3690405786037445,
      "learning_rate": 2e-05,
      "loss": 0.6004,
      "step": 8840
    },
    {
      "epoch": 0.7452631578947368,
      "grad_norm": 0.4076629877090454,
      "learning_rate": 2e-05,
      "loss": 0.6005,
      "step": 8850
    },
    {
      "epoch": 0.7461052631578947,
      "grad_norm": 0.3714067339897156,
      "learning_rate": 2e-05,
      "loss": 0.5977,
      "step": 8860
    },
    {
      "epoch": 0.7469473684210526,
      "grad_norm": 0.35408058762550354,
      "learning_rate": 2e-05,
      "loss": 0.6114,
      "step": 8870
    },
    {
      "epoch": 0.7477894736842106,
      "grad_norm": 0.561436116695404,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 8880
    },
    {
      "epoch": 0.7486315789473684,
      "grad_norm": 0.34749865531921387,
      "learning_rate": 2e-05,
      "loss": 0.5934,
      "step": 8890
    },
    {
      "epoch": 0.7494736842105263,
      "grad_norm": 0.37814122438430786,
      "learning_rate": 2e-05,
      "loss": 0.6009,
      "step": 8900
    },
    {
      "epoch": 0.7503157894736842,
      "grad_norm": 0.4139373302459717,
      "learning_rate": 2e-05,
      "loss": 0.5959,
      "step": 8910
    },
    {
      "epoch": 0.7511578947368421,
      "grad_norm": 0.374818354845047,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 8920
    },
    {
      "epoch": 0.752,
      "grad_norm": 0.4098844528198242,
      "learning_rate": 2e-05,
      "loss": 0.6056,
      "step": 8930
    },
    {
      "epoch": 0.7528421052631579,
      "grad_norm": 0.3790528178215027,
      "learning_rate": 2e-05,
      "loss": 0.6155,
      "step": 8940
    },
    {
      "epoch": 0.7536842105263157,
      "grad_norm": 0.44824013113975525,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 8950
    },
    {
      "epoch": 0.7545263157894737,
      "grad_norm": 0.40466606616973877,
      "learning_rate": 2e-05,
      "loss": 0.607,
      "step": 8960
    },
    {
      "epoch": 0.7553684210526316,
      "grad_norm": 0.3990813195705414,
      "learning_rate": 2e-05,
      "loss": 0.6068,
      "step": 8970
    },
    {
      "epoch": 0.7562105263157894,
      "grad_norm": 0.3804592788219452,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 8980
    },
    {
      "epoch": 0.7570526315789474,
      "grad_norm": 0.3686272203922272,
      "learning_rate": 2e-05,
      "loss": 0.6101,
      "step": 8990
    },
    {
      "epoch": 0.7578947368421053,
      "grad_norm": 0.4029053747653961,
      "learning_rate": 2e-05,
      "loss": 0.6307,
      "step": 9000
    },
    {
      "epoch": 0.7587368421052632,
      "grad_norm": 0.3778361976146698,
      "learning_rate": 2e-05,
      "loss": 0.6284,
      "step": 9010
    },
    {
      "epoch": 0.759578947368421,
      "grad_norm": 0.42222627997398376,
      "learning_rate": 2e-05,
      "loss": 0.6333,
      "step": 9020
    },
    {
      "epoch": 0.760421052631579,
      "grad_norm": 0.40035736560821533,
      "learning_rate": 2e-05,
      "loss": 0.654,
      "step": 9030
    },
    {
      "epoch": 0.7612631578947369,
      "grad_norm": 0.4328446686267853,
      "learning_rate": 2e-05,
      "loss": 0.6275,
      "step": 9040
    },
    {
      "epoch": 0.7621052631578947,
      "grad_norm": 0.380464106798172,
      "learning_rate": 2e-05,
      "loss": 0.6154,
      "step": 9050
    },
    {
      "epoch": 0.7629473684210526,
      "grad_norm": 0.4303261637687683,
      "learning_rate": 2e-05,
      "loss": 0.6189,
      "step": 9060
    },
    {
      "epoch": 0.7637894736842106,
      "grad_norm": 0.38995397090911865,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 9070
    },
    {
      "epoch": 0.7646315789473684,
      "grad_norm": 0.4443624019622803,
      "learning_rate": 2e-05,
      "loss": 0.62,
      "step": 9080
    },
    {
      "epoch": 0.7654736842105263,
      "grad_norm": 0.44582244753837585,
      "learning_rate": 2e-05,
      "loss": 0.618,
      "step": 9090
    },
    {
      "epoch": 0.7663157894736842,
      "grad_norm": 0.38731810450553894,
      "learning_rate": 2e-05,
      "loss": 0.6137,
      "step": 9100
    },
    {
      "epoch": 0.7671578947368421,
      "grad_norm": 0.44161754846572876,
      "learning_rate": 2e-05,
      "loss": 0.5973,
      "step": 9110
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.36884984374046326,
      "learning_rate": 2e-05,
      "loss": 0.6071,
      "step": 9120
    },
    {
      "epoch": 0.7688421052631579,
      "grad_norm": 0.4917847216129303,
      "learning_rate": 2e-05,
      "loss": 0.6032,
      "step": 9130
    },
    {
      "epoch": 0.7696842105263157,
      "grad_norm": 0.4165985882282257,
      "learning_rate": 2e-05,
      "loss": 0.6076,
      "step": 9140
    },
    {
      "epoch": 0.7705263157894737,
      "grad_norm": 0.3916205167770386,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 9150
    },
    {
      "epoch": 0.7713684210526316,
      "grad_norm": 0.3963650166988373,
      "learning_rate": 2e-05,
      "loss": 0.6228,
      "step": 9160
    },
    {
      "epoch": 0.7722105263157895,
      "grad_norm": 0.36129698157310486,
      "learning_rate": 2e-05,
      "loss": 0.6164,
      "step": 9170
    },
    {
      "epoch": 0.7730526315789473,
      "grad_norm": 0.4237774908542633,
      "learning_rate": 2e-05,
      "loss": 0.6021,
      "step": 9180
    },
    {
      "epoch": 0.7738947368421053,
      "grad_norm": 0.4200058579444885,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 9190
    },
    {
      "epoch": 0.7747368421052632,
      "grad_norm": 0.3934716582298279,
      "learning_rate": 2e-05,
      "loss": 0.6015,
      "step": 9200
    },
    {
      "epoch": 0.775578947368421,
      "grad_norm": 0.37617427110671997,
      "learning_rate": 2e-05,
      "loss": 0.6162,
      "step": 9210
    },
    {
      "epoch": 0.7764210526315789,
      "grad_norm": 0.39147791266441345,
      "learning_rate": 2e-05,
      "loss": 0.6011,
      "step": 9220
    },
    {
      "epoch": 0.7772631578947369,
      "grad_norm": 0.3889387845993042,
      "learning_rate": 2e-05,
      "loss": 0.5993,
      "step": 9230
    },
    {
      "epoch": 0.7781052631578947,
      "grad_norm": 0.3912642300128937,
      "learning_rate": 2e-05,
      "loss": 0.6386,
      "step": 9240
    },
    {
      "epoch": 0.7789473684210526,
      "grad_norm": 0.39061713218688965,
      "learning_rate": 2e-05,
      "loss": 0.6154,
      "step": 9250
    },
    {
      "epoch": 0.7797894736842105,
      "grad_norm": 0.3704598844051361,
      "learning_rate": 2e-05,
      "loss": 0.6219,
      "step": 9260
    },
    {
      "epoch": 0.7806315789473685,
      "grad_norm": 0.3853263854980469,
      "learning_rate": 2e-05,
      "loss": 0.5962,
      "step": 9270
    },
    {
      "epoch": 0.7814736842105263,
      "grad_norm": 0.3756018877029419,
      "learning_rate": 2e-05,
      "loss": 0.5894,
      "step": 9280
    },
    {
      "epoch": 0.7823157894736842,
      "grad_norm": 0.39318525791168213,
      "learning_rate": 2e-05,
      "loss": 0.6142,
      "step": 9290
    },
    {
      "epoch": 0.783157894736842,
      "grad_norm": 0.40024733543395996,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 9300
    },
    {
      "epoch": 0.784,
      "grad_norm": 0.42342421412467957,
      "learning_rate": 2e-05,
      "loss": 0.6419,
      "step": 9310
    },
    {
      "epoch": 0.7848421052631579,
      "grad_norm": 0.43104588985443115,
      "learning_rate": 2e-05,
      "loss": 0.603,
      "step": 9320
    },
    {
      "epoch": 0.7856842105263158,
      "grad_norm": 0.38286805152893066,
      "learning_rate": 2e-05,
      "loss": 0.6188,
      "step": 9330
    },
    {
      "epoch": 0.7865263157894736,
      "grad_norm": 0.3894636332988739,
      "learning_rate": 2e-05,
      "loss": 0.5912,
      "step": 9340
    },
    {
      "epoch": 0.7873684210526316,
      "grad_norm": 0.3928429186344147,
      "learning_rate": 2e-05,
      "loss": 0.6024,
      "step": 9350
    },
    {
      "epoch": 0.7882105263157895,
      "grad_norm": 0.382275253534317,
      "learning_rate": 2e-05,
      "loss": 0.5953,
      "step": 9360
    },
    {
      "epoch": 0.7890526315789473,
      "grad_norm": 0.5603330135345459,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 9370
    },
    {
      "epoch": 0.7898947368421053,
      "grad_norm": 0.3889352083206177,
      "learning_rate": 2e-05,
      "loss": 0.5981,
      "step": 9380
    },
    {
      "epoch": 0.7907368421052632,
      "grad_norm": 0.3761619031429291,
      "learning_rate": 2e-05,
      "loss": 0.6156,
      "step": 9390
    },
    {
      "epoch": 0.791578947368421,
      "grad_norm": 0.3792596459388733,
      "learning_rate": 2e-05,
      "loss": 0.5937,
      "step": 9400
    },
    {
      "epoch": 0.7924210526315789,
      "grad_norm": 0.3793589174747467,
      "learning_rate": 2e-05,
      "loss": 0.6088,
      "step": 9410
    },
    {
      "epoch": 0.7932631578947369,
      "grad_norm": 0.38541755080223083,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 9420
    },
    {
      "epoch": 0.7941052631578948,
      "grad_norm": 0.3757480978965759,
      "learning_rate": 2e-05,
      "loss": 0.6055,
      "step": 9430
    },
    {
      "epoch": 0.7949473684210526,
      "grad_norm": 0.3930641710758209,
      "learning_rate": 2e-05,
      "loss": 0.6069,
      "step": 9440
    },
    {
      "epoch": 0.7957894736842105,
      "grad_norm": 0.3951413333415985,
      "learning_rate": 2e-05,
      "loss": 0.6305,
      "step": 9450
    },
    {
      "epoch": 0.7966315789473685,
      "grad_norm": 0.44933682680130005,
      "learning_rate": 2e-05,
      "loss": 0.5927,
      "step": 9460
    },
    {
      "epoch": 0.7974736842105263,
      "grad_norm": 0.40417245030403137,
      "learning_rate": 2e-05,
      "loss": 0.6089,
      "step": 9470
    },
    {
      "epoch": 0.7983157894736842,
      "grad_norm": 0.4282498061656952,
      "learning_rate": 2e-05,
      "loss": 0.6218,
      "step": 9480
    },
    {
      "epoch": 0.7991578947368421,
      "grad_norm": 0.39783138036727905,
      "learning_rate": 2e-05,
      "loss": 0.6013,
      "step": 9490
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3762165307998657,
      "learning_rate": 2e-05,
      "loss": 0.5848,
      "step": 9500
    },
    {
      "epoch": 0.8008421052631579,
      "grad_norm": 0.3964996635913849,
      "learning_rate": 2e-05,
      "loss": 0.6168,
      "step": 9510
    },
    {
      "epoch": 0.8016842105263158,
      "grad_norm": 0.4069039225578308,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 9520
    },
    {
      "epoch": 0.8025263157894736,
      "grad_norm": 0.4078322649002075,
      "learning_rate": 2e-05,
      "loss": 0.5862,
      "step": 9530
    },
    {
      "epoch": 0.8033684210526316,
      "grad_norm": 0.39243605732917786,
      "learning_rate": 2e-05,
      "loss": 0.6034,
      "step": 9540
    },
    {
      "epoch": 0.8042105263157895,
      "grad_norm": 0.36534327268600464,
      "learning_rate": 2e-05,
      "loss": 0.6052,
      "step": 9550
    },
    {
      "epoch": 0.8050526315789474,
      "grad_norm": 0.3864537179470062,
      "learning_rate": 2e-05,
      "loss": 0.5962,
      "step": 9560
    },
    {
      "epoch": 0.8058947368421052,
      "grad_norm": 0.4002450704574585,
      "learning_rate": 2e-05,
      "loss": 0.6042,
      "step": 9570
    },
    {
      "epoch": 0.8067368421052632,
      "grad_norm": 0.39283686876296997,
      "learning_rate": 2e-05,
      "loss": 0.605,
      "step": 9580
    },
    {
      "epoch": 0.8075789473684211,
      "grad_norm": 0.35580024123191833,
      "learning_rate": 2e-05,
      "loss": 0.6221,
      "step": 9590
    },
    {
      "epoch": 0.8084210526315789,
      "grad_norm": 0.43879732489585876,
      "learning_rate": 2e-05,
      "loss": 0.5996,
      "step": 9600
    },
    {
      "epoch": 0.8092631578947368,
      "grad_norm": 0.4227176606655121,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 9610
    },
    {
      "epoch": 0.8101052631578948,
      "grad_norm": 0.38655734062194824,
      "learning_rate": 2e-05,
      "loss": 0.6084,
      "step": 9620
    },
    {
      "epoch": 0.8109473684210526,
      "grad_norm": 0.45072346925735474,
      "learning_rate": 2e-05,
      "loss": 0.6161,
      "step": 9630
    },
    {
      "epoch": 0.8117894736842105,
      "grad_norm": 0.396930068731308,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 9640
    },
    {
      "epoch": 0.8126315789473684,
      "grad_norm": 0.39923062920570374,
      "learning_rate": 2e-05,
      "loss": 0.6103,
      "step": 9650
    },
    {
      "epoch": 0.8134736842105263,
      "grad_norm": 0.414527028799057,
      "learning_rate": 2e-05,
      "loss": 0.6209,
      "step": 9660
    },
    {
      "epoch": 0.8143157894736842,
      "grad_norm": 0.40754929184913635,
      "learning_rate": 2e-05,
      "loss": 0.6055,
      "step": 9670
    },
    {
      "epoch": 0.8151578947368421,
      "grad_norm": 0.3787885308265686,
      "learning_rate": 2e-05,
      "loss": 0.617,
      "step": 9680
    },
    {
      "epoch": 0.816,
      "grad_norm": 0.3715304136276245,
      "learning_rate": 2e-05,
      "loss": 0.5946,
      "step": 9690
    },
    {
      "epoch": 0.8168421052631579,
      "grad_norm": 0.43339547514915466,
      "learning_rate": 2e-05,
      "loss": 0.6173,
      "step": 9700
    },
    {
      "epoch": 0.8176842105263158,
      "grad_norm": 0.3883928656578064,
      "learning_rate": 2e-05,
      "loss": 0.6217,
      "step": 9710
    },
    {
      "epoch": 0.8185263157894737,
      "grad_norm": 0.4657495617866516,
      "learning_rate": 2e-05,
      "loss": 0.6273,
      "step": 9720
    },
    {
      "epoch": 0.8193684210526315,
      "grad_norm": 0.39182767271995544,
      "learning_rate": 2e-05,
      "loss": 0.5814,
      "step": 9730
    },
    {
      "epoch": 0.8202105263157895,
      "grad_norm": 0.39359068870544434,
      "learning_rate": 2e-05,
      "loss": 0.6212,
      "step": 9740
    },
    {
      "epoch": 0.8210526315789474,
      "grad_norm": 0.428130179643631,
      "learning_rate": 2e-05,
      "loss": 0.6067,
      "step": 9750
    },
    {
      "epoch": 0.8218947368421052,
      "grad_norm": 0.4010125398635864,
      "learning_rate": 2e-05,
      "loss": 0.627,
      "step": 9760
    },
    {
      "epoch": 0.8227368421052632,
      "grad_norm": 0.38525041937828064,
      "learning_rate": 2e-05,
      "loss": 0.6155,
      "step": 9770
    },
    {
      "epoch": 0.8235789473684211,
      "grad_norm": 0.4096451997756958,
      "learning_rate": 2e-05,
      "loss": 0.6245,
      "step": 9780
    },
    {
      "epoch": 0.8244210526315789,
      "grad_norm": 0.42166978120803833,
      "learning_rate": 2e-05,
      "loss": 0.5918,
      "step": 9790
    },
    {
      "epoch": 0.8252631578947368,
      "grad_norm": 0.3755349814891815,
      "learning_rate": 2e-05,
      "loss": 0.5776,
      "step": 9800
    },
    {
      "epoch": 0.8261052631578948,
      "grad_norm": 0.39658355712890625,
      "learning_rate": 2e-05,
      "loss": 0.5735,
      "step": 9810
    },
    {
      "epoch": 0.8269473684210527,
      "grad_norm": 0.4165673851966858,
      "learning_rate": 2e-05,
      "loss": 0.6067,
      "step": 9820
    },
    {
      "epoch": 0.8277894736842105,
      "grad_norm": 0.3732060492038727,
      "learning_rate": 2e-05,
      "loss": 0.6098,
      "step": 9830
    },
    {
      "epoch": 0.8286315789473684,
      "grad_norm": 0.5100370645523071,
      "learning_rate": 2e-05,
      "loss": 0.619,
      "step": 9840
    },
    {
      "epoch": 0.8294736842105264,
      "grad_norm": 0.43233969807624817,
      "learning_rate": 2e-05,
      "loss": 0.6243,
      "step": 9850
    },
    {
      "epoch": 0.8303157894736842,
      "grad_norm": 0.3745150566101074,
      "learning_rate": 2e-05,
      "loss": 0.6321,
      "step": 9860
    },
    {
      "epoch": 0.8311578947368421,
      "grad_norm": 0.44246339797973633,
      "learning_rate": 2e-05,
      "loss": 0.6,
      "step": 9870
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.37267735600471497,
      "learning_rate": 2e-05,
      "loss": 0.6212,
      "step": 9880
    },
    {
      "epoch": 0.8328421052631579,
      "grad_norm": 0.40941670536994934,
      "learning_rate": 2e-05,
      "loss": 0.6225,
      "step": 9890
    },
    {
      "epoch": 0.8336842105263158,
      "grad_norm": 0.40743348002433777,
      "learning_rate": 2e-05,
      "loss": 0.6048,
      "step": 9900
    },
    {
      "epoch": 0.8345263157894737,
      "grad_norm": 0.3952489197254181,
      "learning_rate": 2e-05,
      "loss": 0.619,
      "step": 9910
    },
    {
      "epoch": 0.8353684210526315,
      "grad_norm": 0.4360242784023285,
      "learning_rate": 2e-05,
      "loss": 0.6052,
      "step": 9920
    },
    {
      "epoch": 0.8362105263157895,
      "grad_norm": 0.3822256326675415,
      "learning_rate": 2e-05,
      "loss": 0.6116,
      "step": 9930
    },
    {
      "epoch": 0.8370526315789474,
      "grad_norm": 0.4218772053718567,
      "learning_rate": 2e-05,
      "loss": 0.6081,
      "step": 9940
    },
    {
      "epoch": 0.8378947368421052,
      "grad_norm": 0.41678670048713684,
      "learning_rate": 2e-05,
      "loss": 0.6043,
      "step": 9950
    },
    {
      "epoch": 0.8387368421052631,
      "grad_norm": 0.4994248151779175,
      "learning_rate": 2e-05,
      "loss": 0.6166,
      "step": 9960
    },
    {
      "epoch": 0.8395789473684211,
      "grad_norm": 0.431614488363266,
      "learning_rate": 2e-05,
      "loss": 0.5976,
      "step": 9970
    },
    {
      "epoch": 0.840421052631579,
      "grad_norm": 0.43207889795303345,
      "learning_rate": 2e-05,
      "loss": 0.6163,
      "step": 9980
    },
    {
      "epoch": 0.8412631578947368,
      "grad_norm": 0.4098546802997589,
      "learning_rate": 2e-05,
      "loss": 0.5939,
      "step": 9990
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 0.7963051199913025,
      "learning_rate": 2e-05,
      "loss": 0.6009,
      "step": 10000
    },
    {
      "epoch": 0.8429473684210527,
      "grad_norm": 0.37439993023872375,
      "learning_rate": 2e-05,
      "loss": 0.6148,
      "step": 10010
    },
    {
      "epoch": 0.8437894736842105,
      "grad_norm": 0.411466121673584,
      "learning_rate": 2e-05,
      "loss": 0.6162,
      "step": 10020
    },
    {
      "epoch": 0.8446315789473684,
      "grad_norm": 0.37725070118904114,
      "learning_rate": 2e-05,
      "loss": 0.5869,
      "step": 10030
    },
    {
      "epoch": 0.8454736842105263,
      "grad_norm": 0.4031277894973755,
      "learning_rate": 2e-05,
      "loss": 0.5979,
      "step": 10040
    },
    {
      "epoch": 0.8463157894736842,
      "grad_norm": 0.3987767994403839,
      "learning_rate": 2e-05,
      "loss": 0.6023,
      "step": 10050
    },
    {
      "epoch": 0.8471578947368421,
      "grad_norm": 0.3944873809814453,
      "learning_rate": 2e-05,
      "loss": 0.5988,
      "step": 10060
    },
    {
      "epoch": 0.848,
      "grad_norm": 0.41239312291145325,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 10070
    },
    {
      "epoch": 0.8488421052631578,
      "grad_norm": 0.3916851580142975,
      "learning_rate": 2e-05,
      "loss": 0.6022,
      "step": 10080
    },
    {
      "epoch": 0.8496842105263158,
      "grad_norm": 0.40699607133865356,
      "learning_rate": 2e-05,
      "loss": 0.6043,
      "step": 10090
    },
    {
      "epoch": 0.8505263157894737,
      "grad_norm": 0.47179755568504333,
      "learning_rate": 2e-05,
      "loss": 0.6148,
      "step": 10100
    },
    {
      "epoch": 0.8513684210526316,
      "grad_norm": 0.43656548857688904,
      "learning_rate": 2e-05,
      "loss": 0.6151,
      "step": 10110
    },
    {
      "epoch": 0.8522105263157894,
      "grad_norm": 0.3864140510559082,
      "learning_rate": 2e-05,
      "loss": 0.5984,
      "step": 10120
    },
    {
      "epoch": 0.8530526315789474,
      "grad_norm": 0.3854968547821045,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 10130
    },
    {
      "epoch": 0.8538947368421053,
      "grad_norm": 0.3961552679538727,
      "learning_rate": 2e-05,
      "loss": 0.637,
      "step": 10140
    },
    {
      "epoch": 0.8547368421052631,
      "grad_norm": 0.41292205452919006,
      "learning_rate": 2e-05,
      "loss": 0.6092,
      "step": 10150
    },
    {
      "epoch": 0.8555789473684211,
      "grad_norm": 0.41433319449424744,
      "learning_rate": 2e-05,
      "loss": 0.598,
      "step": 10160
    },
    {
      "epoch": 0.856421052631579,
      "grad_norm": 0.3900005519390106,
      "learning_rate": 2e-05,
      "loss": 0.6068,
      "step": 10170
    },
    {
      "epoch": 0.8572631578947368,
      "grad_norm": 0.41450434923171997,
      "learning_rate": 1.999957274187931e-05,
      "loss": 0.6013,
      "step": 10180
    },
    {
      "epoch": 0.8581052631578947,
      "grad_norm": 0.3888067603111267,
      "learning_rate": 1.9991978062682906e-05,
      "loss": 0.6015,
      "step": 10190
    },
    {
      "epoch": 0.8589473684210527,
      "grad_norm": 0.4253007173538208,
      "learning_rate": 1.997489709987378e-05,
      "loss": 0.5948,
      "step": 10200
    },
    {
      "epoch": 0.8597894736842105,
      "grad_norm": 0.3854413330554962,
      "learning_rate": 1.9948346151502135e-05,
      "loss": 0.617,
      "step": 10210
    },
    {
      "epoch": 0.8606315789473684,
      "grad_norm": 0.43955618143081665,
      "learning_rate": 1.991235055154325e-05,
      "loss": 0.6143,
      "step": 10220
    },
    {
      "epoch": 0.8614736842105263,
      "grad_norm": 0.3526889979839325,
      "learning_rate": 1.9866944645724693e-05,
      "loss": 0.6079,
      "step": 10230
    },
    {
      "epoch": 0.8623157894736843,
      "grad_norm": 0.4268966019153595,
      "learning_rate": 1.9812171758754842e-05,
      "loss": 0.6057,
      "step": 10240
    },
    {
      "epoch": 0.8631578947368421,
      "grad_norm": 0.3522892892360687,
      "learning_rate": 1.9748084152983987e-05,
      "loss": 0.5957,
      "step": 10250
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.36722132563591003,
      "learning_rate": 1.967474297853744e-05,
      "loss": 0.5925,
      "step": 10260
    },
    {
      "epoch": 0.8648421052631579,
      "grad_norm": 0.3856015205383301,
      "learning_rate": 1.9592218214968256e-05,
      "loss": 0.603,
      "step": 10270
    },
    {
      "epoch": 0.8656842105263158,
      "grad_norm": 0.42420274019241333,
      "learning_rate": 1.9500588604485213e-05,
      "loss": 0.6279,
      "step": 10280
    },
    {
      "epoch": 0.8665263157894737,
      "grad_norm": 0.4391865134239197,
      "learning_rate": 1.939994157681979e-05,
      "loss": 0.6218,
      "step": 10290
    },
    {
      "epoch": 0.8673684210526316,
      "grad_norm": 0.3904142379760742,
      "learning_rate": 1.9290373165803803e-05,
      "loss": 0.6102,
      "step": 10300
    },
    {
      "epoch": 0.8682105263157894,
      "grad_norm": 0.3880654573440552,
      "learning_rate": 1.9171987917737328e-05,
      "loss": 0.6093,
      "step": 10310
    },
    {
      "epoch": 0.8690526315789474,
      "grad_norm": 0.3856695592403412,
      "learning_rate": 1.9044898791634322e-05,
      "loss": 0.595,
      "step": 10320
    },
    {
      "epoch": 0.8698947368421053,
      "grad_norm": 0.39545565843582153,
      "learning_rate": 1.8909227051441126e-05,
      "loss": 0.5972,
      "step": 10330
    },
    {
      "epoch": 0.8707368421052631,
      "grad_norm": 0.3605109453201294,
      "learning_rate": 1.8765102150330692e-05,
      "loss": 0.5914,
      "step": 10340
    },
    {
      "epoch": 0.871578947368421,
      "grad_norm": 0.38808298110961914,
      "learning_rate": 1.861266160718297e-05,
      "loss": 0.5974,
      "step": 10350
    },
    {
      "epoch": 0.872421052631579,
      "grad_norm": 0.38688620924949646,
      "learning_rate": 1.845205087536923e-05,
      "loss": 0.6138,
      "step": 10360
    },
    {
      "epoch": 0.8732631578947369,
      "grad_norm": 0.37399938702583313,
      "learning_rate": 1.8283423203965633e-05,
      "loss": 0.617,
      "step": 10370
    },
    {
      "epoch": 0.8741052631578947,
      "grad_norm": 0.3891572952270508,
      "learning_rate": 1.810693949152835e-05,
      "loss": 0.613,
      "step": 10380
    },
    {
      "epoch": 0.8749473684210526,
      "grad_norm": 0.40963906049728394,
      "learning_rate": 1.7922768132569902e-05,
      "loss": 0.6221,
      "step": 10390
    },
    {
      "epoch": 0.8757894736842106,
      "grad_norm": 0.39830684661865234,
      "learning_rate": 1.773108485688302e-05,
      "loss": 0.6174,
      "step": 10400
    },
    {
      "epoch": 0.8766315789473684,
      "grad_norm": 0.37325605750083923,
      "learning_rate": 1.7532072561865537e-05,
      "loss": 0.6172,
      "step": 10410
    },
    {
      "epoch": 0.8774736842105263,
      "grad_norm": 0.36259591579437256,
      "learning_rate": 1.732592113800616e-05,
      "loss": 0.5997,
      "step": 10420
    },
    {
      "epoch": 0.8783157894736842,
      "grad_norm": 0.38770145177841187,
      "learning_rate": 1.7112827287697668e-05,
      "loss": 0.6218,
      "step": 10430
    },
    {
      "epoch": 0.8791578947368421,
      "grad_norm": 0.4243277907371521,
      "learning_rate": 1.6892994337550452e-05,
      "loss": 0.6231,
      "step": 10440
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.36625993251800537,
      "learning_rate": 1.6666632044385487e-05,
      "loss": 0.6038,
      "step": 10450
    },
    {
      "epoch": 0.8808421052631579,
      "grad_norm": 0.39976075291633606,
      "learning_rate": 1.6433956395091757e-05,
      "loss": 0.5912,
      "step": 10460
    },
    {
      "epoch": 0.8816842105263157,
      "grad_norm": 0.4071301817893982,
      "learning_rate": 1.6195189400539238e-05,
      "loss": 0.594,
      "step": 10470
    },
    {
      "epoch": 0.8825263157894737,
      "grad_norm": 0.3901573717594147,
      "learning_rate": 1.595055888374398e-05,
      "loss": 0.627,
      "step": 10480
    },
    {
      "epoch": 0.8833684210526316,
      "grad_norm": 0.41846510767936707,
      "learning_rate": 1.5700298262487434e-05,
      "loss": 0.6001,
      "step": 10490
    },
    {
      "epoch": 0.8842105263157894,
      "grad_norm": 0.36507144570350647,
      "learning_rate": 1.5444646326597467e-05,
      "loss": 0.6107,
      "step": 10500
    },
    {
      "epoch": 0.8850526315789473,
      "grad_norm": 0.39777687191963196,
      "learning_rate": 1.5183847010103586e-05,
      "loss": 0.6074,
      "step": 10510
    },
    {
      "epoch": 0.8858947368421053,
      "grad_norm": 0.4392140209674835,
      "learning_rate": 1.4918149158483686e-05,
      "loss": 0.6143,
      "step": 10520
    },
    {
      "epoch": 0.8867368421052632,
      "grad_norm": 0.3884710371494293,
      "learning_rate": 1.464780629122452e-05,
      "loss": 0.591,
      "step": 10530
    },
    {
      "epoch": 0.887578947368421,
      "grad_norm": 0.40131768584251404,
      "learning_rate": 1.4373076359922374e-05,
      "loss": 0.604,
      "step": 10540
    },
    {
      "epoch": 0.888421052631579,
      "grad_norm": 0.369512677192688,
      "learning_rate": 1.4094221502154774e-05,
      "loss": 0.5975,
      "step": 10550
    },
    {
      "epoch": 0.8892631578947369,
      "grad_norm": 0.3728256821632385,
      "learning_rate": 1.3811507791358113e-05,
      "loss": 0.6074,
      "step": 10560
    },
    {
      "epoch": 0.8901052631578947,
      "grad_norm": 0.36384326219558716,
      "learning_rate": 1.3525204982949753e-05,
      "loss": 0.5824,
      "step": 10570
    },
    {
      "epoch": 0.8909473684210526,
      "grad_norm": 0.3682207763195038,
      "learning_rate": 1.3235586256936996e-05,
      "loss": 0.5995,
      "step": 10580
    },
    {
      "epoch": 0.8917894736842106,
      "grad_norm": 0.4122195243835449,
      "learning_rate": 1.294292795725836e-05,
      "loss": 0.595,
      "step": 10590
    },
    {
      "epoch": 0.8926315789473684,
      "grad_norm": 0.3915764391422272,
      "learning_rate": 1.2647509328105978e-05,
      "loss": 0.6088,
      "step": 10600
    },
    {
      "epoch": 0.8934736842105263,
      "grad_norm": 0.40447014570236206,
      "learning_rate": 1.2349612247480686e-05,
      "loss": 0.6043,
      "step": 10610
    },
    {
      "epoch": 0.8943157894736842,
      "grad_norm": 0.4350722134113312,
      "learning_rate": 1.2049520958234e-05,
      "loss": 0.6069,
      "step": 10620
    },
    {
      "epoch": 0.8951578947368422,
      "grad_norm": 0.41973501443862915,
      "learning_rate": 1.1747521796853638e-05,
      "loss": 0.6139,
      "step": 10630
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.39660730957984924,
      "learning_rate": 1.1443902920251395e-05,
      "loss": 0.5983,
      "step": 10640
    },
    {
      "epoch": 0.8968421052631579,
      "grad_norm": 0.39003786444664,
      "learning_rate": 1.1138954030814009e-05,
      "loss": 0.6047,
      "step": 10650
    },
    {
      "epoch": 0.8976842105263158,
      "grad_norm": 0.44097790122032166,
      "learning_rate": 1.0832966099979388e-05,
      "loss": 0.612,
      "step": 10660
    },
    {
      "epoch": 0.8985263157894737,
      "grad_norm": 0.3655848205089569,
      "learning_rate": 1.0526231090601969e-05,
      "loss": 0.599,
      "step": 10670
    },
    {
      "epoch": 0.8993684210526316,
      "grad_norm": 0.3842141628265381,
      "learning_rate": 1.0219041678372106e-05,
      "loss": 0.5881,
      "step": 10680
    },
    {
      "epoch": 0.9002105263157895,
      "grad_norm": 0.4212636351585388,
      "learning_rate": 9.911690972555268e-06,
      "loss": 0.6071,
      "step": 10690
    },
    {
      "epoch": 0.9010526315789473,
      "grad_norm": 0.4402164816856384,
      "learning_rate": 9.604472236317541e-06,
      "loss": 0.5847,
      "step": 10700
    },
    {
      "epoch": 0.9018947368421053,
      "grad_norm": 0.4362366497516632,
      "learning_rate": 9.297678606904335e-06,
      "loss": 0.6128,
      "step": 10710
    },
    {
      "epoch": 0.9027368421052632,
      "grad_norm": 0.5107752084732056,
      "learning_rate": 8.991602815939155e-06,
      "loss": 0.5952,
      "step": 10720
    },
    {
      "epoch": 0.903578947368421,
      "grad_norm": 0.39686527848243713,
      "learning_rate": 8.686536910109464e-06,
      "loss": 0.5946,
      "step": 10730
    },
    {
      "epoch": 0.9044210526315789,
      "grad_norm": 0.40367424488067627,
      "learning_rate": 8.382771972506084e-06,
      "loss": 0.6035,
      "step": 10740
    },
    {
      "epoch": 0.9052631578947369,
      "grad_norm": 0.38632747530937195,
      "learning_rate": 8.08059784488198e-06,
      "loss": 0.5945,
      "step": 10750
    },
    {
      "epoch": 0.9061052631578947,
      "grad_norm": 0.36890578269958496,
      "learning_rate": 7.780302851095544e-06,
      "loss": 0.6161,
      "step": 10760
    },
    {
      "epoch": 0.9069473684210526,
      "grad_norm": 0.37951570749282837,
      "learning_rate": 7.482173522002166e-06,
      "loss": 0.5935,
      "step": 10770
    },
    {
      "epoch": 0.9077894736842105,
      "grad_norm": 0.4353990852832794,
      "learning_rate": 7.186494322056624e-06,
      "loss": 0.5962,
      "step": 10780
    },
    {
      "epoch": 0.9086315789473685,
      "grad_norm": 0.39137041568756104,
      "learning_rate": 6.893547377887183e-06,
      "loss": 0.5901,
      "step": 10790
    },
    {
      "epoch": 0.9094736842105263,
      "grad_norm": 0.37068524956703186,
      "learning_rate": 6.6036122091003365e-06,
      "loss": 0.5853,
      "step": 10800
    },
    {
      "epoch": 0.9103157894736842,
      "grad_norm": 0.3575446307659149,
      "learning_rate": 6.316965461573097e-06,
      "loss": 0.5932,
      "step": 10810
    },
    {
      "epoch": 0.9111578947368421,
      "grad_norm": 0.37402066588401794,
      "learning_rate": 6.0338806434872965e-06,
      "loss": 0.6172,
      "step": 10820
    },
    {
      "epoch": 0.912,
      "grad_norm": 0.3950884938240051,
      "learning_rate": 5.754627864357749e-06,
      "loss": 0.5728,
      "step": 10830
    },
    {
      "epoch": 0.9128421052631579,
      "grad_norm": 0.3723188042640686,
      "learning_rate": 5.479473577303319e-06,
      "loss": 0.5872,
      "step": 10840
    },
    {
      "epoch": 0.9136842105263158,
      "grad_norm": 0.3793545365333557,
      "learning_rate": 5.208680324806778e-06,
      "loss": 0.5896,
      "step": 10850
    },
    {
      "epoch": 0.9145263157894736,
      "grad_norm": 0.38453802466392517,
      "learning_rate": 4.94250648820606e-06,
      "loss": 0.5955,
      "step": 10860
    },
    {
      "epoch": 0.9153684210526316,
      "grad_norm": 0.35649433732032776,
      "learning_rate": 4.681206041155951e-06,
      "loss": 0.5826,
      "step": 10870
    },
    {
      "epoch": 0.9162105263157895,
      "grad_norm": 0.3954254686832428,
      "learning_rate": 4.42502830729538e-06,
      "loss": 0.5987,
      "step": 10880
    },
    {
      "epoch": 0.9170526315789473,
      "grad_norm": 0.3525383770465851,
      "learning_rate": 4.174217722351669e-06,
      "loss": 0.5781,
      "step": 10890
    },
    {
      "epoch": 0.9178947368421052,
      "grad_norm": 0.3583209812641144,
      "learning_rate": 3.929013600908575e-06,
      "loss": 0.5972,
      "step": 10900
    },
    {
      "epoch": 0.9187368421052632,
      "grad_norm": 0.3958036005496979,
      "learning_rate": 3.689649908060807e-06,
      "loss": 0.5948,
      "step": 10910
    },
    {
      "epoch": 0.919578947368421,
      "grad_norm": 0.400116503238678,
      "learning_rate": 3.4563550361728028e-06,
      "loss": 0.5816,
      "step": 10920
    },
    {
      "epoch": 0.9204210526315789,
      "grad_norm": 0.4585050344467163,
      "learning_rate": 3.2293515869548216e-06,
      "loss": 0.6005,
      "step": 10930
    },
    {
      "epoch": 0.9212631578947369,
      "grad_norm": 0.3764231503009796,
      "learning_rate": 3.0088561590642746e-06,
      "loss": 0.6036,
      "step": 10940
    },
    {
      "epoch": 0.9221052631578948,
      "grad_norm": 0.3646579682826996,
      "learning_rate": 2.795079141434941e-06,
      "loss": 0.5878,
      "step": 10950
    },
    {
      "epoch": 0.9229473684210526,
      "grad_norm": 0.392438679933548,
      "learning_rate": 2.588224512531299e-06,
      "loss": 0.5962,
      "step": 10960
    },
    {
      "epoch": 0.9237894736842105,
      "grad_norm": 0.40966305136680603,
      "learning_rate": 2.388489645719495e-06,
      "loss": 0.6006,
      "step": 10970
    },
    {
      "epoch": 0.9246315789473685,
      "grad_norm": 0.3910335302352905,
      "learning_rate": 2.196065120940663e-06,
      "loss": 0.5914,
      "step": 10980
    },
    {
      "epoch": 0.9254736842105263,
      "grad_norm": 0.4602462351322174,
      "learning_rate": 2.0111345428662808e-06,
      "loss": 0.5967,
      "step": 10990
    },
    {
      "epoch": 0.9263157894736842,
      "grad_norm": 0.33170175552368164,
      "learning_rate": 1.8338743657091102e-06,
      "loss": 0.573,
      "step": 11000
    },
    {
      "epoch": 0.9271578947368421,
      "grad_norm": 0.3983405530452728,
      "learning_rate": 1.6644537248568113e-06,
      "loss": 0.6048,
      "step": 11010
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.3635168969631195,
      "learning_rate": 1.503034275488951e-06,
      "loss": 0.5833,
      "step": 11020
    },
    {
      "epoch": 0.9288421052631579,
      "grad_norm": 0.4094637930393219,
      "learning_rate": 1.3497700383313495e-06,
      "loss": 0.6019,
      "step": 11030
    },
    {
      "epoch": 0.9296842105263158,
      "grad_norm": 0.39057308435440063,
      "learning_rate": 1.2048072526949562e-06,
      "loss": 0.5758,
      "step": 11040
    },
    {
      "epoch": 0.9305263157894736,
      "grad_norm": 0.37934526801109314,
      "learning_rate": 1.0682842369394756e-06,
      "loss": 0.5975,
      "step": 11050
    },
    {
      "epoch": 0.9313684210526316,
      "grad_norm": 0.38531434535980225,
      "learning_rate": 9.403312564948792e-07,
      "loss": 0.5873,
      "step": 11060
    },
    {
      "epoch": 0.9322105263157895,
      "grad_norm": 0.37032586336135864,
      "learning_rate": 8.210703995667471e-07,
      "loss": 0.5934,
      "step": 11070
    },
    {
      "epoch": 0.9330526315789474,
      "grad_norm": 0.48567065596580505,
      "learning_rate": 7.106154606440245e-07,
      "loss": 0.5839,
      "step": 11080
    },
    {
      "epoch": 0.9338947368421052,
      "grad_norm": 0.5229890942573547,
      "learning_rate": 6.090718319203393e-07,
      "loss": 0.6133,
      "step": 11090
    },
    {
      "epoch": 0.9347368421052632,
      "grad_norm": 0.38894179463386536,
      "learning_rate": 5.165364027325089e-07,
      "loss": 0.6151,
      "step": 11100
    },
    {
      "epoch": 0.9355789473684211,
      "grad_norm": 0.40042078495025635,
      "learning_rate": 4.330974671121585e-07,
      "loss": 0.5642,
      "step": 11110
    },
    {
      "epoch": 0.9364210526315789,
      "grad_norm": 0.3740803301334381,
      "learning_rate": 3.588346395386846e-07,
      "loss": 0.582,
      "step": 11120
    },
    {
      "epoch": 0.9372631578947368,
      "grad_norm": 0.37116748094558716,
      "learning_rate": 2.938187789739395e-07,
      "loss": 0.5913,
      "step": 11130
    },
    {
      "epoch": 0.9381052631578948,
      "grad_norm": 0.38756856322288513,
      "learning_rate": 2.3811192125110724e-07,
      "loss": 0.5919,
      "step": 11140
    },
    {
      "epoch": 0.9389473684210526,
      "grad_norm": 0.4080219268798828,
      "learning_rate": 1.917672198823177e-07,
      "loss": 0.5914,
      "step": 11150
    },
    {
      "epoch": 0.9397894736842105,
      "grad_norm": 0.3706664443016052,
      "learning_rate": 1.5482889534143795e-07,
      "loss": 0.5995,
      "step": 11160
    },
    {
      "epoch": 0.9406315789473684,
      "grad_norm": 0.4166904091835022,
      "learning_rate": 1.273321928704694e-07,
      "loss": 0.5972,
      "step": 11170
    },
    {
      "epoch": 0.9414736842105264,
      "grad_norm": 0.35638365149497986,
      "learning_rate": 1.0930334884977866e-07,
      "loss": 0.5972,
      "step": 11180
    },
    {
      "epoch": 0.9423157894736842,
      "grad_norm": 0.38753822445869446,
      "learning_rate": 1.0075956576427487e-07,
      "loss": 0.5539,
      "step": 11190
    },
    {
      "epoch": 0.9425684210526316,
      "step": 11193,
      "total_flos": 4.9965855215861105e+19,
      "train_loss": 0.05441043181595316,
      "train_runtime": 11536.679,
      "train_samples_per_second": 11.643,
      "train_steps_per_second": 0.97
    }
  ],
  "logging_steps": 10,
  "max_steps": 11193,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 636,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9965855215861105e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}