{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4713263157894737,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008421052631578948,
      "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": 1.9999810479633703e-05,
      "loss": 0.6033,
      "step": 5090
    },
    {
      "epoch": 0.42947368421052634,
      "grad_norm": 0.3895128667354584,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.6074,
      "step": 5100
    },
    {
      "epoch": 0.4303157894736842,
      "grad_norm": 0.41490331292152405,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.6278,
      "step": 5110
    },
    {
      "epoch": 0.43115789473684213,
      "grad_norm": 0.38978466391563416,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.6274,
      "step": 5120
    },
    {
      "epoch": 0.432,
      "grad_norm": 0.41340598464012146,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.6086,
      "step": 5130
    },
    {
      "epoch": 0.4328421052631579,
      "grad_norm": 0.4304012656211853,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.627,
      "step": 5140
    },
    {
      "epoch": 0.4336842105263158,
      "grad_norm": 0.38799095153808594,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.6134,
      "step": 5150
    },
    {
      "epoch": 0.4345263157894737,
      "grad_norm": 0.38585421442985535,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.6339,
      "step": 5160
    },
    {
      "epoch": 0.43536842105263157,
      "grad_norm": 0.39215758442878723,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.6357,
      "step": 5170
    },
    {
      "epoch": 0.4362105263157895,
      "grad_norm": 0.38533928990364075,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.6383,
      "step": 5180
    },
    {
      "epoch": 0.43705263157894736,
      "grad_norm": 0.3715132474899292,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.6312,
      "step": 5190
    },
    {
      "epoch": 0.4378947368421053,
      "grad_norm": 0.3948341906070709,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.6453,
      "step": 5200
    },
    {
      "epoch": 0.43873684210526315,
      "grad_norm": 0.43796950578689575,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.637,
      "step": 5210
    },
    {
      "epoch": 0.43957894736842107,
      "grad_norm": 0.402692973613739,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.6435,
      "step": 5220
    },
    {
      "epoch": 0.44042105263157894,
      "grad_norm": 0.38847073912620544,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6214,
      "step": 5230
    },
    {
      "epoch": 0.44126315789473686,
      "grad_norm": 0.3658401668071747,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.6444,
      "step": 5240
    },
    {
      "epoch": 0.4421052631578947,
      "grad_norm": 0.3508830666542053,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.62,
      "step": 5250
    },
    {
      "epoch": 0.44294736842105265,
      "grad_norm": 0.43886348605155945,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.6568,
      "step": 5260
    },
    {
      "epoch": 0.4437894736842105,
      "grad_norm": 0.40810438990592957,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.6547,
      "step": 5270
    },
    {
      "epoch": 0.44463157894736843,
      "grad_norm": 0.5229907035827637,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.6232,
      "step": 5280
    },
    {
      "epoch": 0.4454736842105263,
      "grad_norm": 0.45484665036201477,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.6371,
      "step": 5290
    },
    {
      "epoch": 0.4463157894736842,
      "grad_norm": 0.4980567395687103,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.6179,
      "step": 5300
    },
    {
      "epoch": 0.4471578947368421,
      "grad_norm": 0.400259792804718,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.6228,
      "step": 5310
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.4110405147075653,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.6405,
      "step": 5320
    },
    {
      "epoch": 0.4488421052631579,
      "grad_norm": 0.37979406118392944,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.6179,
      "step": 5330
    },
    {
      "epoch": 0.4496842105263158,
      "grad_norm": 0.4020436406135559,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.6153,
      "step": 5340
    },
    {
      "epoch": 0.45052631578947366,
      "grad_norm": 0.40083691477775574,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.6147,
      "step": 5350
    },
    {
      "epoch": 0.4513684210526316,
      "grad_norm": 0.44168534874916077,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6369,
      "step": 5360
    },
    {
      "epoch": 0.45221052631578945,
      "grad_norm": 0.3963114023208618,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.6311,
      "step": 5370
    },
    {
      "epoch": 0.4530526315789474,
      "grad_norm": 0.39288169145584106,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6127,
      "step": 5380
    },
    {
      "epoch": 0.45389473684210524,
      "grad_norm": 0.4298765957355499,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.6107,
      "step": 5390
    },
    {
      "epoch": 0.45473684210526316,
      "grad_norm": 0.40992313623428345,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.6509,
      "step": 5400
    },
    {
      "epoch": 0.45557894736842103,
      "grad_norm": 0.3770509660243988,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.6064,
      "step": 5410
    },
    {
      "epoch": 0.45642105263157895,
      "grad_norm": 0.43356555700302124,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.6232,
      "step": 5420
    },
    {
      "epoch": 0.4572631578947368,
      "grad_norm": 0.4027460217475891,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.6385,
      "step": 5430
    },
    {
      "epoch": 0.45810526315789474,
      "grad_norm": 0.39758753776550293,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.6194,
      "step": 5440
    },
    {
      "epoch": 0.4589473684210526,
      "grad_norm": 0.39061155915260315,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.6262,
      "step": 5450
    },
    {
      "epoch": 0.4597894736842105,
      "grad_norm": 0.4132283926010132,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.6057,
      "step": 5460
    },
    {
      "epoch": 0.46063157894736845,
      "grad_norm": 0.40713775157928467,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.621,
      "step": 5470
    },
    {
      "epoch": 0.4614736842105263,
      "grad_norm": 0.4089098870754242,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.6483,
      "step": 5480
    },
    {
      "epoch": 0.46231578947368424,
      "grad_norm": 0.47794732451438904,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.6355,
      "step": 5490
    },
    {
      "epoch": 0.4631578947368421,
      "grad_norm": 0.3756923973560333,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.628,
      "step": 5500
    },
    {
      "epoch": 0.464,
      "grad_norm": 0.3797348737716675,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.6117,
      "step": 5510
    },
    {
      "epoch": 0.4648421052631579,
      "grad_norm": 0.44111719727516174,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.6081,
      "step": 5520
    },
    {
      "epoch": 0.4656842105263158,
      "grad_norm": 0.3731600344181061,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.6168,
      "step": 5530
    },
    {
      "epoch": 0.4665263157894737,
      "grad_norm": 0.3821868300437927,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.6118,
      "step": 5540
    },
    {
      "epoch": 0.4673684210526316,
      "grad_norm": 0.43992534279823303,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.5917,
      "step": 5550
    },
    {
      "epoch": 0.46821052631578947,
      "grad_norm": 0.39101478457450867,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.6089,
      "step": 5560
    },
    {
      "epoch": 0.4690526315789474,
      "grad_norm": 0.4084720313549042,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.632,
      "step": 5570
    },
    {
      "epoch": 0.46989473684210525,
      "grad_norm": 0.4356926679611206,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6347,
      "step": 5580
    },
    {
      "epoch": 0.4707368421052632,
      "grad_norm": 0.4925185441970825,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.6021,
      "step": 5590
    },
    {
      "epoch": 0.4713263157894737,
      "step": 5597,
      "total_flos": 2.4985159616803373e+19,
      "train_loss": 0.056818318810019086,
      "train_runtime": 5768.7068,
      "train_samples_per_second": 11.643,
      "train_steps_per_second": 0.97
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "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": 2.4985159616803373e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}