{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9548387096774196,
  "eval_steps": 78,
  "global_step": 930,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.12883105396714237,
      "learning_rate": 1e-05,
      "loss": 0.4154,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 0.4359087646007538,
      "eval_runtime": 45.0668,
      "eval_samples_per_second": 35.503,
      "eval_steps_per_second": 0.555,
      "step": 1
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.13019894725030426,
      "learning_rate": 2e-05,
      "loss": 0.4102,
      "step": 2
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.12448027718685753,
      "learning_rate": 3e-05,
      "loss": 0.4339,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.13617024064519495,
      "learning_rate": 4e-05,
      "loss": 0.4463,
      "step": 4
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1192695514188061,
      "learning_rate": 5e-05,
      "loss": 0.4257,
      "step": 5
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.12484938518090542,
      "learning_rate": 6e-05,
      "loss": 0.4476,
      "step": 6
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.11589635616988343,
      "learning_rate": 7e-05,
      "loss": 0.4726,
      "step": 7
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.11326282450028814,
      "learning_rate": 8e-05,
      "loss": 0.44,
      "step": 8
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.10349552391227962,
      "learning_rate": 9e-05,
      "loss": 0.4623,
      "step": 9
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.10545214328006507,
      "learning_rate": 0.0001,
      "loss": 0.4116,
      "step": 10
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.08723759174445953,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.4216,
      "step": 11
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.08139085000714291,
      "learning_rate": 0.00012,
      "loss": 0.39,
      "step": 12
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.09205927275648706,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.434,
      "step": 13
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.10404439147125429,
      "learning_rate": 0.00014,
      "loss": 0.4511,
      "step": 14
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.08585041684076992,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.4048,
      "step": 15
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.09399258996254131,
      "learning_rate": 0.00016,
      "loss": 0.4322,
      "step": 16
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.10425830227796497,
      "learning_rate": 0.00017,
      "loss": 0.4174,
      "step": 17
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.08675512504623259,
      "learning_rate": 0.00018,
      "loss": 0.4247,
      "step": 18
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.08901871349393159,
      "learning_rate": 0.00019,
      "loss": 0.4369,
      "step": 19
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.08060121651209443,
      "learning_rate": 0.0002,
      "loss": 0.4746,
      "step": 20
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.0734185701687092,
      "learning_rate": 0.00019999940408195878,
      "loss": 0.4178,
      "step": 21
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.07838461773593017,
      "learning_rate": 0.00019999761633493753,
      "loss": 0.4321,
      "step": 22
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.07833972271815033,
      "learning_rate": 0.00019999463678024317,
      "loss": 0.4436,
      "step": 23
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.07548486089915235,
      "learning_rate": 0.0001999904654533872,
      "loss": 0.4709,
      "step": 24
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.08788637574838197,
      "learning_rate": 0.00019998510240408496,
      "loss": 0.3872,
      "step": 25
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.0812532852957508,
      "learning_rate": 0.0001999785476962552,
      "loss": 0.4058,
      "step": 26
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.08537862859222577,
      "learning_rate": 0.00019997080140801932,
      "loss": 0.4225,
      "step": 27
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.08106258078139493,
      "learning_rate": 0.00019996186363170035,
      "loss": 0.4357,
      "step": 28
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.08010684638053245,
      "learning_rate": 0.00019995173447382193,
      "loss": 0.4038,
      "step": 29
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.08709110487516528,
      "learning_rate": 0.00019994041405510705,
      "loss": 0.4724,
      "step": 30
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.07096829165155712,
      "learning_rate": 0.00019992790251047656,
      "loss": 0.4139,
      "step": 31
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.06850018644329266,
      "learning_rate": 0.00019991419998904747,
      "loss": 0.4596,
      "step": 32
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.07099101390794227,
      "learning_rate": 0.00019989930665413147,
      "loss": 0.4004,
      "step": 33
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.07941015413576936,
      "learning_rate": 0.00019988322268323268,
      "loss": 0.4299,
      "step": 34
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.07638794258403328,
      "learning_rate": 0.0001998659482680456,
      "loss": 0.4252,
      "step": 35
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.07585182258910264,
      "learning_rate": 0.00019984748361445308,
      "loss": 0.4149,
      "step": 36
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.07502482333755954,
      "learning_rate": 0.0001998278289425234,
      "loss": 0.438,
      "step": 37
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.06605988872292592,
      "learning_rate": 0.00019980698448650804,
      "loss": 0.4252,
      "step": 38
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.07295876342245719,
      "learning_rate": 0.00019978495049483884,
      "loss": 0.4132,
      "step": 39
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.06948118376282203,
      "learning_rate": 0.0001997617272301248,
      "loss": 0.4054,
      "step": 40
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.07729338055807149,
      "learning_rate": 0.00019973731496914914,
      "loss": 0.3968,
      "step": 41
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.07332342728556553,
      "learning_rate": 0.000199711714002866,
      "loss": 0.4734,
      "step": 42
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.07828161218125397,
      "learning_rate": 0.00019968492463639704,
      "loss": 0.4479,
      "step": 43
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.07194198164352739,
      "learning_rate": 0.00019965694718902745,
      "loss": 0.438,
      "step": 44
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.07115652880951175,
      "learning_rate": 0.00019962778199420265,
      "loss": 0.4227,
      "step": 45
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.0775651962190502,
      "learning_rate": 0.00019959742939952392,
      "loss": 0.4058,
      "step": 46
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.06438123421666499,
      "learning_rate": 0.00019956588976674443,
      "loss": 0.3703,
      "step": 47
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.07812636493444186,
      "learning_rate": 0.00019953316347176488,
      "loss": 0.4644,
      "step": 48
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.07727134848765603,
      "learning_rate": 0.00019949925090462909,
      "loss": 0.4314,
      "step": 49
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.07027735759113077,
      "learning_rate": 0.0001994641524695193,
      "loss": 0.4238,
      "step": 50
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.07736957822008628,
      "learning_rate": 0.00019942786858475126,
      "loss": 0.4785,
      "step": 51
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.0681570396521893,
      "learning_rate": 0.0001993903996827694,
      "loss": 0.4374,
      "step": 52
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.07143343715720442,
      "learning_rate": 0.00019935174621014173,
      "loss": 0.4194,
      "step": 53
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.06299350036763142,
      "learning_rate": 0.00019931190862755417,
      "loss": 0.4492,
      "step": 54
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.07366865611199422,
      "learning_rate": 0.0001992708874098054,
      "loss": 0.4214,
      "step": 55
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.06347135008905311,
      "learning_rate": 0.00019922868304580118,
      "loss": 0.43,
      "step": 56
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.08268927152964092,
      "learning_rate": 0.00019918529603854825,
      "loss": 0.4425,
      "step": 57
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.07034294144963513,
      "learning_rate": 0.0001991407269051487,
      "loss": 0.4105,
      "step": 58
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.06630993618399812,
      "learning_rate": 0.00019909497617679348,
      "loss": 0.3882,
      "step": 59
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.07226317955725141,
      "learning_rate": 0.00019904804439875633,
      "loss": 0.4266,
      "step": 60
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.06938919184806121,
      "learning_rate": 0.0001989999321303871,
      "loss": 0.4572,
      "step": 61
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.06492831162679569,
      "learning_rate": 0.0001989506399451051,
      "loss": 0.4318,
      "step": 62
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.06645996011351867,
      "learning_rate": 0.0001989001684303925,
      "loss": 0.417,
      "step": 63
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.08653889274769928,
      "learning_rate": 0.00019884851818778693,
      "loss": 0.4353,
      "step": 64
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.07059932526364451,
      "learning_rate": 0.00019879568983287467,
      "loss": 0.4175,
      "step": 65
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.07931101113959307,
      "learning_rate": 0.00019874168399528305,
      "loss": 0.4187,
      "step": 66
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.07125698961933573,
      "learning_rate": 0.0001986865013186732,
      "loss": 0.4159,
      "step": 67
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.08050044278206689,
      "learning_rate": 0.00019863014246073214,
      "loss": 0.4353,
      "step": 68
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.07318879088432853,
      "learning_rate": 0.0001985726080931651,
      "loss": 0.4202,
      "step": 69
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.07082720625138651,
      "learning_rate": 0.0001985138989016874,
      "loss": 0.4464,
      "step": 70
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.06698591714547134,
      "learning_rate": 0.00019845401558601634,
      "loss": 0.4251,
      "step": 71
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.07603667866936462,
      "learning_rate": 0.00019839295885986296,
      "loss": 0.4298,
      "step": 72
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.06997110349403513,
      "learning_rate": 0.00019833072945092334,
      "loss": 0.4205,
      "step": 73
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.06648812435842759,
      "learning_rate": 0.00019826732810086998,
      "loss": 0.388,
      "step": 74
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.08031943933680023,
      "learning_rate": 0.00019820275556534304,
      "loss": 0.4296,
      "step": 75
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.07292849758364575,
      "learning_rate": 0.00019813701261394136,
      "loss": 0.4693,
      "step": 76
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.06207113723375016,
      "learning_rate": 0.00019807010003021312,
      "loss": 0.4117,
      "step": 77
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.07162805849513847,
      "learning_rate": 0.00019800201861164664,
      "loss": 0.4055,
      "step": 78
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.4264196753501892,
      "eval_runtime": 45.9946,
      "eval_samples_per_second": 34.787,
      "eval_steps_per_second": 0.544,
      "step": 78
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.07162623816355349,
      "learning_rate": 0.00019793276916966083,
      "loss": 0.4203,
      "step": 79
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.06472836232949207,
      "learning_rate": 0.00019786235252959553,
      "loss": 0.4625,
      "step": 80
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.06372740407034722,
      "learning_rate": 0.00019779076953070168,
      "loss": 0.4073,
      "step": 81
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.06667839761368001,
      "learning_rate": 0.00019771802102613127,
      "loss": 0.4778,
      "step": 82
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.059574939985801316,
      "learning_rate": 0.00019764410788292722,
      "loss": 0.407,
      "step": 83
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.07648625055922231,
      "learning_rate": 0.00019756903098201308,
      "loss": 0.4295,
      "step": 84
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.06038606924130534,
      "learning_rate": 0.00019749279121818235,
      "loss": 0.41,
      "step": 85
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.07670919708474994,
      "learning_rate": 0.00019741538950008818,
      "loss": 0.4139,
      "step": 86
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.07017791304733184,
      "learning_rate": 0.00019733682675023207,
      "loss": 0.4117,
      "step": 87
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.0604723656482711,
      "learning_rate": 0.0001972571039049533,
      "loss": 0.4323,
      "step": 88
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.08127405172847653,
      "learning_rate": 0.0001971762219144174,
      "loss": 0.4396,
      "step": 89
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.07061782085169709,
      "learning_rate": 0.0001970941817426052,
      "loss": 0.3927,
      "step": 90
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.06402819168379283,
      "learning_rate": 0.00019701098436730106,
      "loss": 0.4581,
      "step": 91
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.07968710882894958,
      "learning_rate": 0.00019692663078008132,
      "loss": 0.4587,
      "step": 92
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.061691070941232884,
      "learning_rate": 0.00019684112198630244,
      "loss": 0.3996,
      "step": 93
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.07482749601200543,
      "learning_rate": 0.00019675445900508909,
      "loss": 0.4824,
      "step": 94
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.062418664967878255,
      "learning_rate": 0.00019666664286932198,
      "loss": 0.4019,
      "step": 95
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.07330319089356814,
      "learning_rate": 0.00019657767462562544,
      "loss": 0.4083,
      "step": 96
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.07046344898906809,
      "learning_rate": 0.00019648755533435518,
      "loss": 0.4487,
      "step": 97
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.06780433458947788,
      "learning_rate": 0.00019639628606958533,
      "loss": 0.428,
      "step": 98
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.07589568906196692,
      "learning_rate": 0.00019630386791909602,
      "loss": 0.4351,
      "step": 99
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.06426439686349943,
      "learning_rate": 0.00019621030198436006,
      "loss": 0.4144,
      "step": 100
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.06946467483247702,
      "learning_rate": 0.00019611558938053002,
      "loss": 0.42,
      "step": 101
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.06687266974088095,
      "learning_rate": 0.00019601973123642492,
      "loss": 0.4102,
      "step": 102
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.06476195717860939,
      "learning_rate": 0.0001959227286945167,
      "loss": 0.4307,
      "step": 103
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.06784624564247167,
      "learning_rate": 0.00019582458291091663,
      "loss": 0.437,
      "step": 104
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.07092126591759794,
      "learning_rate": 0.0001957252950553616,
      "loss": 0.4086,
      "step": 105
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.07528584467610183,
      "learning_rate": 0.00019562486631120006,
      "loss": 0.4131,
      "step": 106
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.07372322110054848,
      "learning_rate": 0.00019552329787537805,
      "loss": 0.4021,
      "step": 107
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.06862887663783342,
      "learning_rate": 0.00019542059095842485,
      "loss": 0.3967,
      "step": 108
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.06913146575239316,
      "learning_rate": 0.00019531674678443853,
      "loss": 0.3912,
      "step": 109
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.06645883203424491,
      "learning_rate": 0.00019521176659107142,
      "loss": 0.4251,
      "step": 110
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.06416682030197016,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.441,
      "step": 111
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.06876092886215074,
      "learning_rate": 0.00019499840316448673,
      "loss": 0.4391,
      "step": 112
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.06688267191994776,
      "learning_rate": 0.00019489002247421148,
      "loss": 0.4257,
      "step": 113
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.06324346733094179,
      "learning_rate": 0.00019478051085040975,
      "loss": 0.4528,
      "step": 114
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.06725879673299155,
      "learning_rate": 0.0001946698695982806,
      "loss": 0.4147,
      "step": 115
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.07241340618039839,
      "learning_rate": 0.00019455810003648637,
      "loss": 0.4527,
      "step": 116
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.06613334861150516,
      "learning_rate": 0.00019444520349713704,
      "loss": 0.4283,
      "step": 117
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.06840872345753035,
      "learning_rate": 0.0001943311813257743,
      "loss": 0.4131,
      "step": 118
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.0704617274405355,
      "learning_rate": 0.00019421603488135557,
      "loss": 0.4379,
      "step": 119
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.06787816360755813,
      "learning_rate": 0.00019409976553623766,
      "loss": 0.4177,
      "step": 120
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.06972714992884815,
      "learning_rate": 0.0001939823746761606,
      "loss": 0.3795,
      "step": 121
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.06779189257040956,
      "learning_rate": 0.00019386386370023103,
      "loss": 0.4323,
      "step": 122
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.07014006424788279,
      "learning_rate": 0.00019374423402090553,
      "loss": 0.3989,
      "step": 123
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.07661659071907612,
      "learning_rate": 0.00019362348706397373,
      "loss": 0.4422,
      "step": 124
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.07584620511227261,
      "learning_rate": 0.0001935016242685415,
      "loss": 0.4512,
      "step": 125
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.07525185527526344,
      "learning_rate": 0.00019337864708701357,
      "loss": 0.4298,
      "step": 126
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.0710375411666665,
      "learning_rate": 0.00019325455698507638,
      "loss": 0.3878,
      "step": 127
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.0673112696784915,
      "learning_rate": 0.00019312935544168048,
      "loss": 0.3705,
      "step": 128
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.0668828668657636,
      "learning_rate": 0.00019300304394902313,
      "loss": 0.4017,
      "step": 129
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.07246208549885681,
      "learning_rate": 0.00019287562401253022,
      "loss": 0.3806,
      "step": 130
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.06437628312475605,
      "learning_rate": 0.0001927470971508386,
      "loss": 0.4033,
      "step": 131
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.062491146220490415,
      "learning_rate": 0.00019261746489577765,
      "loss": 0.4136,
      "step": 132
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.06460575253319188,
      "learning_rate": 0.0001924867287923515,
      "loss": 0.4285,
      "step": 133
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.06783309737426804,
      "learning_rate": 0.0001923548903987201,
      "loss": 0.4211,
      "step": 134
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.07213052803073557,
      "learning_rate": 0.00019222195128618106,
      "loss": 0.4416,
      "step": 135
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.0677809803143048,
      "learning_rate": 0.00019208791303915063,
      "loss": 0.4384,
      "step": 136
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.07088268081293747,
      "learning_rate": 0.0001919527772551451,
      "loss": 0.4099,
      "step": 137
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.06916606268106777,
      "learning_rate": 0.0001918165455447614,
      "loss": 0.4299,
      "step": 138
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.07309072410021103,
      "learning_rate": 0.00019167921953165825,
      "loss": 0.4122,
      "step": 139
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.07900186130531832,
      "learning_rate": 0.00019154080085253666,
      "loss": 0.4599,
      "step": 140
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.06570679783619483,
      "learning_rate": 0.00019140129115712034,
      "loss": 0.3662,
      "step": 141
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.08015148942504076,
      "learning_rate": 0.0001912606921081362,
      "loss": 0.4002,
      "step": 142
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.06917459897842565,
      "learning_rate": 0.00019111900538129443,
      "loss": 0.4339,
      "step": 143
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.07295609866871429,
      "learning_rate": 0.0001909762326652686,
      "loss": 0.4252,
      "step": 144
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.0808022860773497,
      "learning_rate": 0.0001908323756616754,
      "loss": 0.4424,
      "step": 145
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.06342264967015554,
      "learning_rate": 0.00019068743608505455,
      "loss": 0.4039,
      "step": 146
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.06858427993056751,
      "learning_rate": 0.0001905414156628482,
      "loss": 0.4157,
      "step": 147
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.06968670755842428,
      "learning_rate": 0.00019039431613538047,
      "loss": 0.413,
      "step": 148
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.07867845357193032,
      "learning_rate": 0.00019024613925583652,
      "loss": 0.4086,
      "step": 149
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.0737734252643262,
      "learning_rate": 0.0001900968867902419,
      "loss": 0.42,
      "step": 150
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.06795299857235934,
      "learning_rate": 0.0001899465605174414,
      "loss": 0.4186,
      "step": 151
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.07145917531667596,
      "learning_rate": 0.00018979516222907775,
      "loss": 0.4451,
      "step": 152
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.07680401394397088,
      "learning_rate": 0.00018964269372957038,
      "loss": 0.4206,
      "step": 153
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.06594889233352544,
      "learning_rate": 0.00018948915683609388,
      "loss": 0.4234,
      "step": 154
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.07340962027871342,
      "learning_rate": 0.00018933455337855632,
      "loss": 0.4543,
      "step": 155
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.07514208916494701,
      "learning_rate": 0.00018917888519957754,
      "loss": 0.4281,
      "step": 156
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.42719101905822754,
      "eval_runtime": 31.4768,
      "eval_samples_per_second": 50.831,
      "eval_steps_per_second": 0.794,
      "step": 156
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.06549409287056523,
      "learning_rate": 0.000189022154154467,
      "loss": 0.4324,
      "step": 157
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.0682462164394273,
      "learning_rate": 0.00018886436211120193,
      "loss": 0.433,
      "step": 158
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.08588011407859532,
      "learning_rate": 0.00018870551095040477,
      "loss": 0.4304,
      "step": 159
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.06987370558027692,
      "learning_rate": 0.000188545602565321,
      "loss": 0.4482,
      "step": 160
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.06240247030946795,
      "learning_rate": 0.00018838463886179644,
      "loss": 0.3907,
      "step": 161
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.06891290467646147,
      "learning_rate": 0.00018822262175825462,
      "loss": 0.4267,
      "step": 162
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.06542585845674868,
      "learning_rate": 0.0001880595531856738,
      "loss": 0.4284,
      "step": 163
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.0629730206122836,
      "learning_rate": 0.00018789543508756408,
      "loss": 0.4003,
      "step": 164
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.06871544541236552,
      "learning_rate": 0.0001877302694199442,
      "loss": 0.4216,
      "step": 165
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.07149666774302382,
      "learning_rate": 0.00018756405815131813,
      "loss": 0.3836,
      "step": 166
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.06997442761322878,
      "learning_rate": 0.0001873968032626518,
      "loss": 0.4188,
      "step": 167
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.084225812122246,
      "learning_rate": 0.00018722850674734927,
      "loss": 0.4298,
      "step": 168
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.06761216783540702,
      "learning_rate": 0.00018705917061122916,
      "loss": 0.361,
      "step": 169
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.07234030808073647,
      "learning_rate": 0.00018688879687250067,
      "loss": 0.4097,
      "step": 170
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.06626071479471723,
      "learning_rate": 0.00018671738756173944,
      "loss": 0.4082,
      "step": 171
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.06855387266060617,
      "learning_rate": 0.0001865449447218635,
      "loss": 0.4192,
      "step": 172
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.06828142558387336,
      "learning_rate": 0.00018637147040810885,
      "loss": 0.4425,
      "step": 173
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.06573023022659233,
      "learning_rate": 0.00018619696668800492,
      "loss": 0.436,
      "step": 174
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.07607545925103458,
      "learning_rate": 0.0001860214356413501,
      "loss": 0.4538,
      "step": 175
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.06694123802468883,
      "learning_rate": 0.00018584487936018661,
      "loss": 0.4172,
      "step": 176
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.06815569028292222,
      "learning_rate": 0.00018566729994877603,
      "loss": 0.4514,
      "step": 177
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.10184702083851915,
      "learning_rate": 0.0001854886995235738,
      "loss": 0.4442,
      "step": 178
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.07024340413477916,
      "learning_rate": 0.00018530908021320425,
      "loss": 0.4424,
      "step": 179
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.07366377295970097,
      "learning_rate": 0.00018512844415843514,
      "loss": 0.4149,
      "step": 180
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.07181140519657206,
      "learning_rate": 0.0001849467935121521,
      "loss": 0.4355,
      "step": 181
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.06898359648739377,
      "learning_rate": 0.00018476413043933313,
      "loss": 0.4275,
      "step": 182
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.06680808433374633,
      "learning_rate": 0.00018458045711702264,
      "loss": 0.4172,
      "step": 183
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.06630828587681081,
      "learning_rate": 0.00018439577573430555,
      "loss": 0.3927,
      "step": 184
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.06251103706062068,
      "learning_rate": 0.00018421008849228118,
      "loss": 0.4255,
      "step": 185
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.06270748210458083,
      "learning_rate": 0.00018402339760403713,
      "loss": 0.3854,
      "step": 186
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.06864674994414362,
      "learning_rate": 0.00018383570529462273,
      "loss": 0.4186,
      "step": 187
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.06636160864684977,
      "learning_rate": 0.00018364701380102266,
      "loss": 0.3948,
      "step": 188
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.06783696558409527,
      "learning_rate": 0.00018345732537213027,
      "loss": 0.4157,
      "step": 189
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.06273783195633519,
      "learning_rate": 0.00018326664226872065,
      "loss": 0.42,
      "step": 190
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.0635460399988344,
      "learning_rate": 0.00018307496676342385,
      "loss": 0.3863,
      "step": 191
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.06892459709466887,
      "learning_rate": 0.00018288230114069765,
      "loss": 0.3972,
      "step": 192
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.06554573002953047,
      "learning_rate": 0.00018268864769680054,
      "loss": 0.4235,
      "step": 193
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.07841135830884972,
      "learning_rate": 0.0001824940087397641,
      "loss": 0.4345,
      "step": 194
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.07099852734871953,
      "learning_rate": 0.00018229838658936564,
      "loss": 0.4582,
      "step": 195
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.07334719442008945,
      "learning_rate": 0.00018210178357710058,
      "loss": 0.4247,
      "step": 196
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.06401877277457285,
      "learning_rate": 0.0001819042020461545,
      "loss": 0.4233,
      "step": 197
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.05794553222990041,
      "learning_rate": 0.0001817056443513754,
      "loss": 0.3993,
      "step": 198
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.06734254228013473,
      "learning_rate": 0.00018150611285924556,
      "loss": 0.4501,
      "step": 199
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.061167398242663894,
      "learning_rate": 0.00018130560994785325,
      "loss": 0.4791,
      "step": 200
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.06368439834981215,
      "learning_rate": 0.00018110413800686456,
      "loss": 0.4031,
      "step": 201
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.06396454527108245,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.4333,
      "step": 202
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.07036915456962346,
      "learning_rate": 0.00018069829665247976,
      "loss": 0.4089,
      "step": 203
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.07566355163015755,
      "learning_rate": 0.00018049393207604733,
      "loss": 0.4125,
      "step": 204
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.07110213568246515,
      "learning_rate": 0.00018028860814388827,
      "loss": 0.4423,
      "step": 205
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.06609288074156261,
      "learning_rate": 0.00018008232730312723,
      "loss": 0.3893,
      "step": 206
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.06540430345204035,
      "learning_rate": 0.00017987509201229378,
      "loss": 0.3645,
      "step": 207
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.06607405584993503,
      "learning_rate": 0.00017966690474129285,
      "loss": 0.4117,
      "step": 208
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.0662155287622059,
      "learning_rate": 0.00017945776797137543,
      "loss": 0.4269,
      "step": 209
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0617368116387909,
      "learning_rate": 0.00017924768419510904,
      "loss": 0.3799,
      "step": 210
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0627987397717418,
      "learning_rate": 0.00017903665591634794,
      "loss": 0.371,
      "step": 211
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.06749499008783807,
      "learning_rate": 0.00017882468565020326,
      "loss": 0.3992,
      "step": 212
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.06631945098396477,
      "learning_rate": 0.00017861177592301317,
      "loss": 0.4095,
      "step": 213
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.06685762710051685,
      "learning_rate": 0.00017839792927231254,
      "loss": 0.4349,
      "step": 214
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.07502266024951904,
      "learning_rate": 0.000178183148246803,
      "loss": 0.4197,
      "step": 215
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.08306628299652864,
      "learning_rate": 0.00017796743540632223,
      "loss": 0.4286,
      "step": 216
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.06166158787464328,
      "learning_rate": 0.0001777507933218138,
      "loss": 0.3879,
      "step": 217
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.06410797862871172,
      "learning_rate": 0.00017753322457529614,
      "loss": 0.4082,
      "step": 218
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.07279590060361223,
      "learning_rate": 0.00017731473175983212,
      "loss": 0.4491,
      "step": 219
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.06330655145519701,
      "learning_rate": 0.00017709531747949796,
      "loss": 0.3696,
      "step": 220
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.06265173315476923,
      "learning_rate": 0.00017687498434935223,
      "loss": 0.4107,
      "step": 221
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.06448654912056977,
      "learning_rate": 0.00017665373499540463,
      "loss": 0.4038,
      "step": 222
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.08291630469395751,
      "learning_rate": 0.00017643157205458483,
      "loss": 0.422,
      "step": 223
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.06632609649023061,
      "learning_rate": 0.00017620849817471092,
      "loss": 0.4271,
      "step": 224
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.07513194076660681,
      "learning_rate": 0.0001759845160144579,
      "loss": 0.4529,
      "step": 225
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.0687911159938889,
      "learning_rate": 0.00017575962824332596,
      "loss": 0.3814,
      "step": 226
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.06806851223680013,
      "learning_rate": 0.00017553383754160865,
      "loss": 0.4038,
      "step": 227
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.060354615878050986,
      "learning_rate": 0.00017530714660036112,
      "loss": 0.4139,
      "step": 228
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.06771319705049085,
      "learning_rate": 0.00017507955812136775,
      "loss": 0.4178,
      "step": 229
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.06554465726049169,
      "learning_rate": 0.00017485107481711012,
      "loss": 0.4008,
      "step": 230
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.06628707050461834,
      "learning_rate": 0.00017462169941073475,
      "loss": 0.3704,
      "step": 231
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.0704660527100064,
      "learning_rate": 0.0001743914346360205,
      "loss": 0.4096,
      "step": 232
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.06714268412305492,
      "learning_rate": 0.00017416028323734598,
      "loss": 0.4047,
      "step": 233
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.06708061829484278,
      "learning_rate": 0.00017392824796965702,
      "loss": 0.3995,
      "step": 234
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.42397916316986084,
      "eval_runtime": 31.4123,
      "eval_samples_per_second": 50.936,
      "eval_steps_per_second": 0.796,
      "step": 234
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.0692311421790837,
      "learning_rate": 0.00017369533159843369,
      "loss": 0.4493,
      "step": 235
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.07010639301506344,
      "learning_rate": 0.00017346153689965727,
      "loss": 0.4134,
      "step": 236
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.0657681493331333,
      "learning_rate": 0.00017322686665977737,
      "loss": 0.3958,
      "step": 237
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.06417143561447558,
      "learning_rate": 0.00017299132367567857,
      "loss": 0.4104,
      "step": 238
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.08433041973271442,
      "learning_rate": 0.00017275491075464716,
      "loss": 0.4417,
      "step": 239
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.06309959644633113,
      "learning_rate": 0.00017251763071433765,
      "loss": 0.4027,
      "step": 240
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.06840734501694612,
      "learning_rate": 0.00017227948638273916,
      "loss": 0.3961,
      "step": 241
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.06502571584749511,
      "learning_rate": 0.00017204048059814175,
      "loss": 0.3961,
      "step": 242
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.06645154810491374,
      "learning_rate": 0.00017180061620910263,
      "loss": 0.4034,
      "step": 243
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.07169067898808924,
      "learning_rate": 0.00017155989607441213,
      "loss": 0.4578,
      "step": 244
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.06951259218049313,
      "learning_rate": 0.00017131832306305965,
      "loss": 0.4222,
      "step": 245
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.06936292391783441,
      "learning_rate": 0.0001710759000541995,
      "loss": 0.4412,
      "step": 246
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.0641932957849969,
      "learning_rate": 0.00017083262993711662,
      "loss": 0.4122,
      "step": 247
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.06519962544514646,
      "learning_rate": 0.00017058851561119198,
      "loss": 0.3949,
      "step": 248
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.06139777990217894,
      "learning_rate": 0.00017034355998586827,
      "loss": 0.4533,
      "step": 249
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.06382389409436602,
      "learning_rate": 0.00017009776598061495,
      "loss": 0.4225,
      "step": 250
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.0745112215377997,
      "learning_rate": 0.00016985113652489374,
      "loss": 0.4172,
      "step": 251
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.06202295907342136,
      "learning_rate": 0.00016960367455812336,
      "loss": 0.4156,
      "step": 252
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.0601321734977661,
      "learning_rate": 0.00016935538302964494,
      "loss": 0.4094,
      "step": 253
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.07237506968634055,
      "learning_rate": 0.00016910626489868649,
      "loss": 0.4254,
      "step": 254
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.06626388773117148,
      "learning_rate": 0.0001688563231343277,
      "loss": 0.4277,
      "step": 255
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.06262183979430723,
      "learning_rate": 0.0001686055607154648,
      "loss": 0.4143,
      "step": 256
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.07788401630185697,
      "learning_rate": 0.00016835398063077474,
      "loss": 0.4328,
      "step": 257
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.06846682414537354,
      "learning_rate": 0.00016810158587867973,
      "loss": 0.418,
      "step": 258
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.0689288634381005,
      "learning_rate": 0.00016784837946731148,
      "loss": 0.4209,
      "step": 259
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.06820042044945746,
      "learning_rate": 0.00016759436441447545,
      "loss": 0.3902,
      "step": 260
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.07119552472188734,
      "learning_rate": 0.0001673395437476146,
      "loss": 0.4306,
      "step": 261
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.06555865871255981,
      "learning_rate": 0.00016708392050377363,
      "loss": 0.3975,
      "step": 262
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.0712809166108178,
      "learning_rate": 0.0001668274977295626,
      "loss": 0.3934,
      "step": 263
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.07147432449436239,
      "learning_rate": 0.00016657027848112062,
      "loss": 0.3742,
      "step": 264
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.06455739935278262,
      "learning_rate": 0.00016631226582407952,
      "loss": 0.4168,
      "step": 265
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.06793873318686483,
      "learning_rate": 0.00016605346283352727,
      "loss": 0.4116,
      "step": 266
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.058261601463606476,
      "learning_rate": 0.00016579387259397127,
      "loss": 0.3845,
      "step": 267
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.062017087628894275,
      "learning_rate": 0.00016553349819930165,
      "loss": 0.3933,
      "step": 268
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.059279362535743535,
      "learning_rate": 0.00016527234275275445,
      "loss": 0.3607,
      "step": 269
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.0615298158124674,
      "learning_rate": 0.00016501040936687443,
      "loss": 0.3978,
      "step": 270
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.05968002667313829,
      "learning_rate": 0.00016474770116347824,
      "loss": 0.4241,
      "step": 271
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.06836235857310884,
      "learning_rate": 0.00016448422127361706,
      "loss": 0.4213,
      "step": 272
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.06328807134715098,
      "learning_rate": 0.00016421997283753927,
      "loss": 0.3835,
      "step": 273
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.07185050583270135,
      "learning_rate": 0.00016395495900465304,
      "loss": 0.4397,
      "step": 274
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.06731895516948408,
      "learning_rate": 0.00016368918293348892,
      "loss": 0.4266,
      "step": 275
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.06154277474915469,
      "learning_rate": 0.000163422647791662,
      "loss": 0.405,
      "step": 276
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.060450974009086104,
      "learning_rate": 0.00016315535675583424,
      "loss": 0.4149,
      "step": 277
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.0607120928467259,
      "learning_rate": 0.00016288731301167668,
      "loss": 0.42,
      "step": 278
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.06605542298126882,
      "learning_rate": 0.00016261851975383137,
      "loss": 0.4092,
      "step": 279
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.06717022241182037,
      "learning_rate": 0.00016234898018587337,
      "loss": 0.4373,
      "step": 280
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.06027581892234991,
      "learning_rate": 0.00016207869752027246,
      "loss": 0.3889,
      "step": 281
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.06756360223208042,
      "learning_rate": 0.00016180767497835503,
      "loss": 0.411,
      "step": 282
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.062275969305366226,
      "learning_rate": 0.00016153591579026546,
      "loss": 0.4146,
      "step": 283
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.06405007529726114,
      "learning_rate": 0.00016126342319492784,
      "loss": 0.3982,
      "step": 284
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.06859548867939594,
      "learning_rate": 0.00016099020044000727,
      "loss": 0.3841,
      "step": 285
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.06040247687681776,
      "learning_rate": 0.00016071625078187114,
      "loss": 0.4261,
      "step": 286
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.0705993351573918,
      "learning_rate": 0.00016044157748555026,
      "loss": 0.4264,
      "step": 287
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.0728951497630979,
      "learning_rate": 0.00016016618382470012,
      "loss": 0.4225,
      "step": 288
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.058326948226576664,
      "learning_rate": 0.00015989007308156173,
      "loss": 0.3817,
      "step": 289
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.06467973723572644,
      "learning_rate": 0.00015961324854692254,
      "loss": 0.4132,
      "step": 290
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.0650952066772671,
      "learning_rate": 0.0001593357135200773,
      "loss": 0.4185,
      "step": 291
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.06768263235822244,
      "learning_rate": 0.0001590574713087885,
      "loss": 0.4362,
      "step": 292
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.05967945902895899,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.3907,
      "step": 293
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.05879444877054761,
      "learning_rate": 0.00015849887860603374,
      "loss": 0.398,
      "step": 294
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.06052224942840143,
      "learning_rate": 0.00015821853477207708,
      "loss": 0.3788,
      "step": 295
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.058292067258336945,
      "learning_rate": 0.00015793749706861636,
      "loss": 0.3658,
      "step": 296
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.06616138091865216,
      "learning_rate": 0.00015765576884516031,
      "loss": 0.4136,
      "step": 297
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.06466816301584516,
      "learning_rate": 0.00015737335345944757,
      "loss": 0.3972,
      "step": 298
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.06889146736527235,
      "learning_rate": 0.00015709025427740661,
      "loss": 0.4038,
      "step": 299
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.07028354273665399,
      "learning_rate": 0.00015680647467311557,
      "loss": 0.4375,
      "step": 300
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.06913117274097347,
      "learning_rate": 0.00015652201802876227,
      "loss": 0.4067,
      "step": 301
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.06933114545676364,
      "learning_rate": 0.00015623688773460357,
      "loss": 0.3561,
      "step": 302
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.0732735312469693,
      "learning_rate": 0.0001559510871889252,
      "loss": 0.4192,
      "step": 303
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.06095351289998777,
      "learning_rate": 0.00015566461979800122,
      "loss": 0.4002,
      "step": 304
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.06279122778820988,
      "learning_rate": 0.0001553774889760533,
      "loss": 0.4052,
      "step": 305
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.06346871620982147,
      "learning_rate": 0.00015508969814521025,
      "loss": 0.4115,
      "step": 306
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.06265655296923098,
      "learning_rate": 0.00015480125073546704,
      "loss": 0.4103,
      "step": 307
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.06650336261140556,
      "learning_rate": 0.00015451215018464387,
      "loss": 0.4208,
      "step": 308
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.062225515787237784,
      "learning_rate": 0.00015422239993834552,
      "loss": 0.3965,
      "step": 309
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.0705406540443855,
      "learning_rate": 0.00015393200344991995,
      "loss": 0.4105,
      "step": 310
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.06075229905592138,
      "learning_rate": 0.00015364096418041726,
      "loss": 0.3781,
      "step": 311
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.06703452615801342,
      "learning_rate": 0.0001533492855985485,
      "loss": 0.3828,
      "step": 312
    },
    {
      "epoch": 1.01,
      "eval_loss": 0.42175912857055664,
      "eval_runtime": 31.52,
      "eval_samples_per_second": 50.761,
      "eval_steps_per_second": 0.793,
      "step": 312
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.07504940119607073,
      "learning_rate": 0.00015305697118064428,
      "loss": 0.4538,
      "step": 313
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.06730008921576576,
      "learning_rate": 0.0001527640244106133,
      "loss": 0.4308,
      "step": 314
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.06514147722515332,
      "learning_rate": 0.0001524704487799008,
      "loss": 0.4237,
      "step": 315
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.07257799027926257,
      "learning_rate": 0.00015217624778744718,
      "loss": 0.4121,
      "step": 316
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.061674598401760324,
      "learning_rate": 0.00015188142493964595,
      "loss": 0.4026,
      "step": 317
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.06932398196022131,
      "learning_rate": 0.00015158598375030217,
      "loss": 0.3375,
      "step": 318
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.06215611676822286,
      "learning_rate": 0.00015128992774059063,
      "loss": 0.3474,
      "step": 319
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.07308539456115362,
      "learning_rate": 0.0001509932604390136,
      "loss": 0.3743,
      "step": 320
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.08101673846459073,
      "learning_rate": 0.00015069598538135906,
      "loss": 0.3923,
      "step": 321
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.0781854723300375,
      "learning_rate": 0.0001503981061106584,
      "loss": 0.3548,
      "step": 322
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.07970059476577994,
      "learning_rate": 0.00015009962617714424,
      "loss": 0.3765,
      "step": 323
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.07371660759085928,
      "learning_rate": 0.00014980054913820814,
      "loss": 0.3726,
      "step": 324
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.0785483838771728,
      "learning_rate": 0.00014950087855835815,
      "loss": 0.3604,
      "step": 325
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.07692841651155574,
      "learning_rate": 0.00014920061800917638,
      "loss": 0.343,
      "step": 326
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.08648668987773721,
      "learning_rate": 0.0001488997710692764,
      "loss": 0.3902,
      "step": 327
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.0865161229608944,
      "learning_rate": 0.0001485983413242606,
      "loss": 0.3469,
      "step": 328
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.08794754622273447,
      "learning_rate": 0.00014829633236667747,
      "loss": 0.3763,
      "step": 329
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.08251409399865628,
      "learning_rate": 0.00014799374779597867,
      "loss": 0.3817,
      "step": 330
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.08397228131714292,
      "learning_rate": 0.0001476905912184763,
      "loss": 0.4025,
      "step": 331
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.08505107767597189,
      "learning_rate": 0.00014738686624729986,
      "loss": 0.4011,
      "step": 332
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.07659854660730629,
      "learning_rate": 0.0001470825765023532,
      "loss": 0.3865,
      "step": 333
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.07772072859171653,
      "learning_rate": 0.0001467777256102712,
      "loss": 0.3605,
      "step": 334
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.0836194254090897,
      "learning_rate": 0.00014647231720437686,
      "loss": 0.3526,
      "step": 335
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.08020420151550436,
      "learning_rate": 0.00014616635492463776,
      "loss": 0.3848,
      "step": 336
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.0823616317274472,
      "learning_rate": 0.00014585984241762267,
      "loss": 0.3737,
      "step": 337
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.08804564315641605,
      "learning_rate": 0.00014555278333645833,
      "loss": 0.3738,
      "step": 338
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.07907938057510025,
      "learning_rate": 0.00014524518134078563,
      "loss": 0.3472,
      "step": 339
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.07856882508739876,
      "learning_rate": 0.00014493704009671613,
      "loss": 0.3782,
      "step": 340
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.07868957075703749,
      "learning_rate": 0.0001446283632767884,
      "loss": 0.3575,
      "step": 341
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.08071752593754973,
      "learning_rate": 0.00014431915455992414,
      "loss": 0.3503,
      "step": 342
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.07598399871189387,
      "learning_rate": 0.0001440094176313844,
      "loss": 0.3281,
      "step": 343
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.07681293813372528,
      "learning_rate": 0.00014369915618272567,
      "loss": 0.4022,
      "step": 344
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.08145783028585485,
      "learning_rate": 0.00014338837391175582,
      "loss": 0.391,
      "step": 345
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.07557613065220387,
      "learning_rate": 0.00014307707452249012,
      "loss": 0.3687,
      "step": 346
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.08185926073157018,
      "learning_rate": 0.000142765261725107,
      "loss": 0.3988,
      "step": 347
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.07738407882567049,
      "learning_rate": 0.0001424529392359039,
      "loss": 0.3866,
      "step": 348
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.08703938835229595,
      "learning_rate": 0.00014214011077725292,
      "loss": 0.3447,
      "step": 349
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.07135637580394055,
      "learning_rate": 0.0001418267800775565,
      "loss": 0.3279,
      "step": 350
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.08051955565921015,
      "learning_rate": 0.00014151295087120306,
      "loss": 0.3599,
      "step": 351
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.07193112928698735,
      "learning_rate": 0.00014119862689852223,
      "loss": 0.3481,
      "step": 352
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.07479779956270906,
      "learning_rate": 0.00014088381190574051,
      "loss": 0.3745,
      "step": 353
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.0751825668630308,
      "learning_rate": 0.0001405685096449367,
      "loss": 0.3685,
      "step": 354
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.07642737965285674,
      "learning_rate": 0.00014025272387399674,
      "loss": 0.376,
      "step": 355
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.0798422899504642,
      "learning_rate": 0.00013993645835656953,
      "loss": 0.3523,
      "step": 356
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.08064560321068621,
      "learning_rate": 0.00013961971686202163,
      "loss": 0.4053,
      "step": 357
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.07569549836084653,
      "learning_rate": 0.00013930250316539238,
      "loss": 0.3612,
      "step": 358
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.07994677074783292,
      "learning_rate": 0.0001389848210473491,
      "loss": 0.3482,
      "step": 359
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.08019343947198912,
      "learning_rate": 0.0001386666742941419,
      "loss": 0.3372,
      "step": 360
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.087138980594316,
      "learning_rate": 0.0001383480666975586,
      "loss": 0.3655,
      "step": 361
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.07628509775359035,
      "learning_rate": 0.00013802900205487948,
      "loss": 0.3474,
      "step": 362
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.08306725000867372,
      "learning_rate": 0.00013770948416883205,
      "loss": 0.3585,
      "step": 363
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.07457563276380513,
      "learning_rate": 0.00013738951684754585,
      "loss": 0.3193,
      "step": 364
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.08622596022590924,
      "learning_rate": 0.00013706910390450677,
      "loss": 0.3805,
      "step": 365
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.07962595816519315,
      "learning_rate": 0.00013674824915851192,
      "loss": 0.3612,
      "step": 366
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.07753518500830336,
      "learning_rate": 0.000136426956433624,
      "loss": 0.4072,
      "step": 367
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.07575904715111116,
      "learning_rate": 0.0001361052295591255,
      "loss": 0.3726,
      "step": 368
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.07232655392409623,
      "learning_rate": 0.00013578307236947348,
      "loss": 0.3487,
      "step": 369
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.08315953805458862,
      "learning_rate": 0.00013546048870425356,
      "loss": 0.4122,
      "step": 370
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.07725321318543174,
      "learning_rate": 0.0001351374824081343,
      "loss": 0.3994,
      "step": 371
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.06801941671303215,
      "learning_rate": 0.00013481405733082116,
      "loss": 0.3572,
      "step": 372
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.07601041676987139,
      "learning_rate": 0.00013449021732701106,
      "loss": 0.3623,
      "step": 373
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.08156069545740995,
      "learning_rate": 0.00013416596625634593,
      "loss": 0.3668,
      "step": 374
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.0811687981854912,
      "learning_rate": 0.00013384130798336705,
      "loss": 0.3733,
      "step": 375
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.07837688253315354,
      "learning_rate": 0.00013351624637746886,
      "loss": 0.3444,
      "step": 376
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.07841335962415118,
      "learning_rate": 0.00013319078531285285,
      "loss": 0.3553,
      "step": 377
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.0885143980175204,
      "learning_rate": 0.00013286492866848142,
      "loss": 0.3542,
      "step": 378
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.08695366578649416,
      "learning_rate": 0.00013253868032803173,
      "loss": 0.3571,
      "step": 379
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.09224273069746307,
      "learning_rate": 0.00013221204417984908,
      "loss": 0.3701,
      "step": 380
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.08465395748798342,
      "learning_rate": 0.000131885024116901,
      "loss": 0.3649,
      "step": 381
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.08136521455604329,
      "learning_rate": 0.00013155762403673063,
      "loss": 0.3635,
      "step": 382
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.08232751491278589,
      "learning_rate": 0.0001312298478414102,
      "loss": 0.383,
      "step": 383
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.08042003166038371,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.3854,
      "step": 384
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.07517595475460019,
      "learning_rate": 0.0001305731827359753,
      "loss": 0.3753,
      "step": 385
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.08259426772986962,
      "learning_rate": 0.00013024430165223244,
      "loss": 0.3743,
      "step": 386
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.08249155111922718,
      "learning_rate": 0.00012991506010598964,
      "loss": 0.4007,
      "step": 387
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.07736305672570372,
      "learning_rate": 0.0001295854620212664,
      "loss": 0.363,
      "step": 388
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.07326027328989387,
      "learning_rate": 0.0001292555113263316,
      "loss": 0.3605,
      "step": 389
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.08267552681029215,
      "learning_rate": 0.00012892521195365678,
      "loss": 0.3811,
      "step": 390
    },
    {
      "epoch": 1.24,
      "eval_loss": 0.4272485077381134,
      "eval_runtime": 31.4867,
      "eval_samples_per_second": 50.815,
      "eval_steps_per_second": 0.794,
      "step": 390
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.07836996060236551,
      "learning_rate": 0.00012859456783986893,
      "loss": 0.3375,
      "step": 391
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.08132634828350899,
      "learning_rate": 0.00012826358292570398,
      "loss": 0.3803,
      "step": 392
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.07855371030733681,
      "learning_rate": 0.00012793226115595952,
      "loss": 0.3333,
      "step": 393
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.07761778090681633,
      "learning_rate": 0.00012760060647944795,
      "loss": 0.3249,
      "step": 394
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.08653543003834345,
      "learning_rate": 0.00012726862284894938,
      "loss": 0.3754,
      "step": 395
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.08368500314474855,
      "learning_rate": 0.00012693631422116454,
      "loss": 0.3497,
      "step": 396
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.08863281952751606,
      "learning_rate": 0.00012660368455666752,
      "loss": 0.3951,
      "step": 397
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.08372219881292209,
      "learning_rate": 0.0001262707378198587,
      "loss": 0.3793,
      "step": 398
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.08088256185598346,
      "learning_rate": 0.00012593747797891742,
      "loss": 0.3541,
      "step": 399
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.0809646416504841,
      "learning_rate": 0.0001256039090057547,
      "loss": 0.3351,
      "step": 400
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.0834243840140192,
      "learning_rate": 0.000125270034875966,
      "loss": 0.3653,
      "step": 401
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.08569597310919161,
      "learning_rate": 0.00012493585956878354,
      "loss": 0.3869,
      "step": 402
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.07886614027929317,
      "learning_rate": 0.0001246013870670293,
      "loss": 0.3551,
      "step": 403
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.08109270204021289,
      "learning_rate": 0.0001242666213570672,
      "loss": 0.3693,
      "step": 404
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.07898699446219974,
      "learning_rate": 0.0001239315664287558,
      "loss": 0.3469,
      "step": 405
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.07880897037364112,
      "learning_rate": 0.00012359622627540058,
      "loss": 0.3739,
      "step": 406
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.08289417173418122,
      "learning_rate": 0.00012326060489370653,
      "loss": 0.3727,
      "step": 407
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.0842462438087756,
      "learning_rate": 0.00012292470628373037,
      "loss": 0.3647,
      "step": 408
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.08537904855482682,
      "learning_rate": 0.00012258853444883295,
      "loss": 0.3593,
      "step": 409
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.07031356294146404,
      "learning_rate": 0.00012225209339563145,
      "loss": 0.3109,
      "step": 410
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.08271226158300135,
      "learning_rate": 0.00012191538713395178,
      "loss": 0.341,
      "step": 411
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.081638753906054,
      "learning_rate": 0.00012157841967678063,
      "loss": 0.3721,
      "step": 412
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.08282882522768446,
      "learning_rate": 0.00012124119504021775,
      "loss": 0.3868,
      "step": 413
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.07632165634443522,
      "learning_rate": 0.00012090371724342804,
      "loss": 0.3596,
      "step": 414
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.07723277347269442,
      "learning_rate": 0.00012056599030859366,
      "loss": 0.3623,
      "step": 415
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.07600198653763515,
      "learning_rate": 0.00012022801826086609,
      "loss": 0.3838,
      "step": 416
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.0737660398002958,
      "learning_rate": 0.00011988980512831809,
      "loss": 0.3406,
      "step": 417
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.0827463463847524,
      "learning_rate": 0.00011955135494189588,
      "loss": 0.3797,
      "step": 418
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.07835261149802049,
      "learning_rate": 0.00011921267173537086,
      "loss": 0.3556,
      "step": 419
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.07869323076948366,
      "learning_rate": 0.00011887375954529168,
      "loss": 0.3732,
      "step": 420
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.07528530037269936,
      "learning_rate": 0.00011853462241093613,
      "loss": 0.3586,
      "step": 421
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.08540035485538978,
      "learning_rate": 0.00011819526437426298,
      "loss": 0.3744,
      "step": 422
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.08162324916698332,
      "learning_rate": 0.00011785568947986367,
      "loss": 0.3453,
      "step": 423
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.08520031508547123,
      "learning_rate": 0.0001175159017749144,
      "loss": 0.3661,
      "step": 424
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.08622581171267268,
      "learning_rate": 0.00011717590530912763,
      "loss": 0.34,
      "step": 425
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.08293740708713171,
      "learning_rate": 0.00011683570413470383,
      "loss": 0.3525,
      "step": 426
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.08630674052013089,
      "learning_rate": 0.0001164953023062835,
      "loss": 0.3663,
      "step": 427
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.08157627202282051,
      "learning_rate": 0.00011615470388089835,
      "loss": 0.3496,
      "step": 428
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.08155874853440571,
      "learning_rate": 0.00011581391291792336,
      "loss": 0.367,
      "step": 429
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.08112096087877045,
      "learning_rate": 0.00011547293347902812,
      "loss": 0.3969,
      "step": 430
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.0788722068697115,
      "learning_rate": 0.0001151317696281287,
      "loss": 0.3669,
      "step": 431
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.08202544827607615,
      "learning_rate": 0.00011479042543133895,
      "loss": 0.3604,
      "step": 432
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.07688818566027143,
      "learning_rate": 0.00011444890495692213,
      "loss": 0.3553,
      "step": 433
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.08043229761191359,
      "learning_rate": 0.00011410721227524255,
      "loss": 0.3829,
      "step": 434
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.07714097071379519,
      "learning_rate": 0.00011376535145871684,
      "loss": 0.3578,
      "step": 435
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.0745231009772769,
      "learning_rate": 0.00011342332658176555,
      "loss": 0.3325,
      "step": 436
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.07732479544740335,
      "learning_rate": 0.00011308114172076462,
      "loss": 0.3674,
      "step": 437
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.08520859762973104,
      "learning_rate": 0.00011273880095399667,
      "loss": 0.4001,
      "step": 438
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.0765417910322464,
      "learning_rate": 0.00011239630836160246,
      "loss": 0.3821,
      "step": 439
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.07773772452915924,
      "learning_rate": 0.0001120536680255323,
      "loss": 0.3667,
      "step": 440
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.07826316953532447,
      "learning_rate": 0.00011171088402949739,
      "loss": 0.3785,
      "step": 441
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.0807755587572588,
      "learning_rate": 0.00011136796045892102,
      "loss": 0.3379,
      "step": 442
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.0769264038040213,
      "learning_rate": 0.00011102490140089009,
      "loss": 0.3849,
      "step": 443
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.07698654136683346,
      "learning_rate": 0.00011068171094410618,
      "loss": 0.3566,
      "step": 444
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.0767333052243508,
      "learning_rate": 0.00011033839317883701,
      "loss": 0.3344,
      "step": 445
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.08336825371769054,
      "learning_rate": 0.00010999495219686762,
      "loss": 0.3912,
      "step": 446
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.08777594502071236,
      "learning_rate": 0.00010965139209145152,
      "loss": 0.3919,
      "step": 447
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.08075368961243723,
      "learning_rate": 0.00010930771695726201,
      "loss": 0.3355,
      "step": 448
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.0817581524046764,
      "learning_rate": 0.00010896393089034336,
      "loss": 0.3837,
      "step": 449
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.07809571969176528,
      "learning_rate": 0.00010862003798806196,
      "loss": 0.3461,
      "step": 450
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.08275587894806521,
      "learning_rate": 0.00010827604234905748,
      "loss": 0.3226,
      "step": 451
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.08006879996457565,
      "learning_rate": 0.00010793194807319408,
      "loss": 0.3817,
      "step": 452
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.0740637352494351,
      "learning_rate": 0.00010758775926151154,
      "loss": 0.3574,
      "step": 453
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.07910943875130351,
      "learning_rate": 0.00010724348001617625,
      "loss": 0.3581,
      "step": 454
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.0753786712518676,
      "learning_rate": 0.00010689911444043248,
      "loss": 0.3655,
      "step": 455
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.08280644113490662,
      "learning_rate": 0.00010655466663855349,
      "loss": 0.3774,
      "step": 456
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.08547898182915604,
      "learning_rate": 0.0001062101407157924,
      "loss": 0.4035,
      "step": 457
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.08698240550610552,
      "learning_rate": 0.00010586554077833347,
      "loss": 0.3688,
      "step": 458
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.08242367645301783,
      "learning_rate": 0.00010552087093324315,
      "loss": 0.3754,
      "step": 459
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.07376544324977327,
      "learning_rate": 0.00010517613528842097,
      "loss": 0.3587,
      "step": 460
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.08100703861143939,
      "learning_rate": 0.00010483133795255071,
      "loss": 0.3812,
      "step": 461
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.08170556800441375,
      "learning_rate": 0.00010448648303505151,
      "loss": 0.3667,
      "step": 462
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.08085852463592974,
      "learning_rate": 0.00010414157464602866,
      "loss": 0.3514,
      "step": 463
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.0789315049905107,
      "learning_rate": 0.00010379661689622477,
      "loss": 0.3991,
      "step": 464
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.07803841799948108,
      "learning_rate": 0.00010345161389697082,
      "loss": 0.3181,
      "step": 465
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.08070026028876534,
      "learning_rate": 0.00010310656976013705,
      "loss": 0.3837,
      "step": 466
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.08631416699901101,
      "learning_rate": 0.0001027614885980839,
      "loss": 0.3562,
      "step": 467
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.07989902063759162,
      "learning_rate": 0.00010241637452361323,
      "loss": 0.3738,
      "step": 468
    },
    {
      "epoch": 1.49,
      "eval_loss": 0.4268099069595337,
      "eval_runtime": 31.5034,
      "eval_samples_per_second": 50.788,
      "eval_steps_per_second": 0.794,
      "step": 468
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.08278261941170967,
      "learning_rate": 0.00010207123164991911,
      "loss": 0.3798,
      "step": 469
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.07848169734840553,
      "learning_rate": 0.00010172606409053886,
      "loss": 0.3807,
      "step": 470
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.07862977038589984,
      "learning_rate": 0.00010138087595930395,
      "loss": 0.3701,
      "step": 471
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.08217076517383108,
      "learning_rate": 0.0001010356713702911,
      "loss": 0.3903,
      "step": 472
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.0802021316689387,
      "learning_rate": 0.00010069045443777318,
      "loss": 0.3634,
      "step": 473
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.07338173288420062,
      "learning_rate": 0.00010034522927617014,
      "loss": 0.3811,
      "step": 474
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.08200044564472758,
      "learning_rate": 0.0001,
      "loss": 0.3468,
      "step": 475
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.07839544989553099,
      "learning_rate": 9.96547707238299e-05,
      "loss": 0.3511,
      "step": 476
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.07952288871273722,
      "learning_rate": 9.930954556222683e-05,
      "loss": 0.3552,
      "step": 477
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.08699081546573348,
      "learning_rate": 9.896432862970892e-05,
      "loss": 0.3526,
      "step": 478
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.08423355750874421,
      "learning_rate": 9.861912404069608e-05,
      "loss": 0.3765,
      "step": 479
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.07759825635438429,
      "learning_rate": 9.827393590946116e-05,
      "loss": 0.3575,
      "step": 480
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.08333101182425925,
      "learning_rate": 9.792876835008091e-05,
      "loss": 0.3393,
      "step": 481
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.07612722261829721,
      "learning_rate": 9.75836254763868e-05,
      "loss": 0.3471,
      "step": 482
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.07873796857875963,
      "learning_rate": 9.723851140191613e-05,
      "loss": 0.3481,
      "step": 483
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.07711176362729558,
      "learning_rate": 9.689343023986302e-05,
      "loss": 0.3371,
      "step": 484
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.08146330945626319,
      "learning_rate": 9.654838610302923e-05,
      "loss": 0.3685,
      "step": 485
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.07904971154072554,
      "learning_rate": 9.620338310377525e-05,
      "loss": 0.3851,
      "step": 486
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.07790237835005973,
      "learning_rate": 9.58584253539714e-05,
      "loss": 0.3499,
      "step": 487
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.07832489963425983,
      "learning_rate": 9.551351696494854e-05,
      "loss": 0.3727,
      "step": 488
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.08248518618426699,
      "learning_rate": 9.516866204744931e-05,
      "loss": 0.3784,
      "step": 489
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.08050275079657898,
      "learning_rate": 9.482386471157904e-05,
      "loss": 0.3362,
      "step": 490
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.07770689006350201,
      "learning_rate": 9.447912906675686e-05,
      "loss": 0.3609,
      "step": 491
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.07175481533546442,
      "learning_rate": 9.413445922166653e-05,
      "loss": 0.3544,
      "step": 492
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.08064147746188607,
      "learning_rate": 9.378985928420762e-05,
      "loss": 0.3466,
      "step": 493
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.07743882202608632,
      "learning_rate": 9.344533336144652e-05,
      "loss": 0.3763,
      "step": 494
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.07329593275985641,
      "learning_rate": 9.31008855595675e-05,
      "loss": 0.3503,
      "step": 495
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.0813108612869619,
      "learning_rate": 9.275651998382377e-05,
      "loss": 0.3582,
      "step": 496
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.07766612713514506,
      "learning_rate": 9.241224073848848e-05,
      "loss": 0.3736,
      "step": 497
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.08022030269140305,
      "learning_rate": 9.206805192680593e-05,
      "loss": 0.3651,
      "step": 498
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.0810514185142845,
      "learning_rate": 9.172395765094254e-05,
      "loss": 0.3734,
      "step": 499
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.08336908596716675,
      "learning_rate": 9.137996201193805e-05,
      "loss": 0.3764,
      "step": 500
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.07708910580019225,
      "learning_rate": 9.103606910965666e-05,
      "loss": 0.3432,
      "step": 501
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.07690985523300399,
      "learning_rate": 9.069228304273802e-05,
      "loss": 0.3366,
      "step": 502
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.07814120482229833,
      "learning_rate": 9.034860790854849e-05,
      "loss": 0.38,
      "step": 503
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.07754525008604259,
      "learning_rate": 9.00050478031324e-05,
      "loss": 0.3746,
      "step": 504
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.08174093565738355,
      "learning_rate": 8.9661606821163e-05,
      "loss": 0.3521,
      "step": 505
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.0874470830671821,
      "learning_rate": 8.931828905589385e-05,
      "loss": 0.3576,
      "step": 506
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.0791505967737016,
      "learning_rate": 8.897509859910995e-05,
      "loss": 0.3623,
      "step": 507
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.08010612556499642,
      "learning_rate": 8.863203954107902e-05,
      "loss": 0.3596,
      "step": 508
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.07639082594002607,
      "learning_rate": 8.828911597050263e-05,
      "loss": 0.3754,
      "step": 509
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.0812907343779127,
      "learning_rate": 8.79463319744677e-05,
      "loss": 0.3569,
      "step": 510
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.07981116292396602,
      "learning_rate": 8.760369163839758e-05,
      "loss": 0.369,
      "step": 511
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.08007339719446442,
      "learning_rate": 8.726119904600336e-05,
      "loss": 0.3619,
      "step": 512
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.08188074360240942,
      "learning_rate": 8.691885827923541e-05,
      "loss": 0.3588,
      "step": 513
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.07330400401389027,
      "learning_rate": 8.657667341823448e-05,
      "loss": 0.361,
      "step": 514
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.07580473341808247,
      "learning_rate": 8.62346485412832e-05,
      "loss": 0.3332,
      "step": 515
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.0751836352756251,
      "learning_rate": 8.589278772475749e-05,
      "loss": 0.3275,
      "step": 516
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.076324517800282,
      "learning_rate": 8.55510950430779e-05,
      "loss": 0.3692,
      "step": 517
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.08062731434583217,
      "learning_rate": 8.520957456866107e-05,
      "loss": 0.3388,
      "step": 518
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.07385163740127851,
      "learning_rate": 8.486823037187129e-05,
      "loss": 0.3478,
      "step": 519
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.07490293522468204,
      "learning_rate": 8.452706652097186e-05,
      "loss": 0.3636,
      "step": 520
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.0809647940628489,
      "learning_rate": 8.418608708207667e-05,
      "loss": 0.3675,
      "step": 521
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.07923274582937138,
      "learning_rate": 8.384529611910163e-05,
      "loss": 0.3552,
      "step": 522
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.08216616911262736,
      "learning_rate": 8.35046976937165e-05,
      "loss": 0.3275,
      "step": 523
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.07936137998207138,
      "learning_rate": 8.316429586529615e-05,
      "loss": 0.3582,
      "step": 524
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.08058250439811132,
      "learning_rate": 8.282409469087239e-05,
      "loss": 0.3676,
      "step": 525
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.0792633274598118,
      "learning_rate": 8.248409822508561e-05,
      "loss": 0.3606,
      "step": 526
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.08160855744923941,
      "learning_rate": 8.214431052013634e-05,
      "loss": 0.378,
      "step": 527
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.08244144254881117,
      "learning_rate": 8.180473562573705e-05,
      "loss": 0.3602,
      "step": 528
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.08332491219956155,
      "learning_rate": 8.146537758906388e-05,
      "loss": 0.3853,
      "step": 529
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.0787381785277075,
      "learning_rate": 8.112624045470835e-05,
      "loss": 0.3578,
      "step": 530
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.07465375920723784,
      "learning_rate": 8.078732826462915e-05,
      "loss": 0.3553,
      "step": 531
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.07914126541842607,
      "learning_rate": 8.044864505810414e-05,
      "loss": 0.38,
      "step": 532
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.0762325619765211,
      "learning_rate": 8.011019487168192e-05,
      "loss": 0.3522,
      "step": 533
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.08205023330200409,
      "learning_rate": 7.977198173913394e-05,
      "loss": 0.3578,
      "step": 534
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.07440912776013751,
      "learning_rate": 7.943400969140635e-05,
      "loss": 0.3349,
      "step": 535
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.08653541120426957,
      "learning_rate": 7.909628275657198e-05,
      "loss": 0.3878,
      "step": 536
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.08271269508372392,
      "learning_rate": 7.875880495978227e-05,
      "loss": 0.3687,
      "step": 537
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.07990749427850796,
      "learning_rate": 7.84215803232194e-05,
      "loss": 0.3428,
      "step": 538
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.08010343988871826,
      "learning_rate": 7.808461286604827e-05,
      "loss": 0.3363,
      "step": 539
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.08385707389143623,
      "learning_rate": 7.774790660436858e-05,
      "loss": 0.3795,
      "step": 540
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.08101037261788462,
      "learning_rate": 7.741146555116708e-05,
      "loss": 0.3781,
      "step": 541
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.08278683251141064,
      "learning_rate": 7.707529371626965e-05,
      "loss": 0.3406,
      "step": 542
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.08593501282474013,
      "learning_rate": 7.673939510629349e-05,
      "loss": 0.3664,
      "step": 543
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.07732960398229517,
      "learning_rate": 7.640377372459945e-05,
      "loss": 0.359,
      "step": 544
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.07698197755802726,
      "learning_rate": 7.606843357124426e-05,
      "loss": 0.3525,
      "step": 545
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.07748553112157543,
      "learning_rate": 7.573337864293283e-05,
      "loss": 0.3538,
      "step": 546
    },
    {
      "epoch": 1.74,
      "eval_loss": 0.42420506477355957,
      "eval_runtime": 31.5034,
      "eval_samples_per_second": 50.788,
      "eval_steps_per_second": 0.794,
      "step": 546
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.0773849964940893,
      "learning_rate": 7.539861293297072e-05,
      "loss": 0.3442,
      "step": 547
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.08408291713234628,
      "learning_rate": 7.506414043121647e-05,
      "loss": 0.3502,
      "step": 548
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.08137948671656789,
      "learning_rate": 7.472996512403403e-05,
      "loss": 0.3834,
      "step": 549
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.07884987774361263,
      "learning_rate": 7.43960909942453e-05,
      "loss": 0.3535,
      "step": 550
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.08601619357770882,
      "learning_rate": 7.406252202108258e-05,
      "loss": 0.3354,
      "step": 551
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.08277228544080684,
      "learning_rate": 7.372926218014131e-05,
      "loss": 0.3658,
      "step": 552
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.08055402198167343,
      "learning_rate": 7.339631544333249e-05,
      "loss": 0.3933,
      "step": 553
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.08231940351547128,
      "learning_rate": 7.306368577883547e-05,
      "loss": 0.389,
      "step": 554
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.07982419038477599,
      "learning_rate": 7.273137715105063e-05,
      "loss": 0.3373,
      "step": 555
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.07946502223724,
      "learning_rate": 7.239939352055208e-05,
      "loss": 0.355,
      "step": 556
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.07851925552097333,
      "learning_rate": 7.206773884404051e-05,
      "loss": 0.3424,
      "step": 557
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.0803764068033967,
      "learning_rate": 7.173641707429606e-05,
      "loss": 0.3563,
      "step": 558
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.0761192213153037,
      "learning_rate": 7.14054321601311e-05,
      "loss": 0.3636,
      "step": 559
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.08367999776936366,
      "learning_rate": 7.107478804634325e-05,
      "loss": 0.3653,
      "step": 560
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.08261688121892191,
      "learning_rate": 7.07444886736684e-05,
      "loss": 0.3486,
      "step": 561
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.08105782422350354,
      "learning_rate": 7.041453797873363e-05,
      "loss": 0.3748,
      "step": 562
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.07917828136130849,
      "learning_rate": 7.008493989401039e-05,
      "loss": 0.3621,
      "step": 563
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.07203487013224297,
      "learning_rate": 6.975569834776758e-05,
      "loss": 0.3233,
      "step": 564
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.08093269486229393,
      "learning_rate": 6.942681726402473e-05,
      "loss": 0.3582,
      "step": 565
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.08678311759977655,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.3794,
      "step": 566
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.08084081388372036,
      "learning_rate": 6.877015215858981e-05,
      "loss": 0.3422,
      "step": 567
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.08553392571374521,
      "learning_rate": 6.844237596326941e-05,
      "loss": 0.3761,
      "step": 568
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.07912394124644004,
      "learning_rate": 6.811497588309901e-05,
      "loss": 0.355,
      "step": 569
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.07620057152346986,
      "learning_rate": 6.778795582015097e-05,
      "loss": 0.3413,
      "step": 570
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.07513161847440154,
      "learning_rate": 6.746131967196834e-05,
      "loss": 0.3568,
      "step": 571
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.07561185266854122,
      "learning_rate": 6.713507133151857e-05,
      "loss": 0.3081,
      "step": 572
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.07460533026271897,
      "learning_rate": 6.680921468714719e-05,
      "loss": 0.359,
      "step": 573
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.07903492734652774,
      "learning_rate": 6.648375362253118e-05,
      "loss": 0.3524,
      "step": 574
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.07549694179386848,
      "learning_rate": 6.615869201663296e-05,
      "loss": 0.341,
      "step": 575
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.07701423597853588,
      "learning_rate": 6.583403374365405e-05,
      "loss": 0.3441,
      "step": 576
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.07155223176721602,
      "learning_rate": 6.550978267298893e-05,
      "loss": 0.3269,
      "step": 577
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.08474429506026998,
      "learning_rate": 6.518594266917882e-05,
      "loss": 0.3767,
      "step": 578
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.07508251474073095,
      "learning_rate": 6.486251759186572e-05,
      "loss": 0.3851,
      "step": 579
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.08827225430407397,
      "learning_rate": 6.453951129574644e-05,
      "loss": 0.3738,
      "step": 580
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.08202839176557213,
      "learning_rate": 6.421692763052653e-05,
      "loss": 0.3499,
      "step": 581
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.07833005102594903,
      "learning_rate": 6.389477044087452e-05,
      "loss": 0.3662,
      "step": 582
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.08063203854850547,
      "learning_rate": 6.357304356637606e-05,
      "loss": 0.3849,
      "step": 583
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.08474453958289388,
      "learning_rate": 6.325175084148809e-05,
      "loss": 0.3579,
      "step": 584
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.08269426648339273,
      "learning_rate": 6.293089609549325e-05,
      "loss": 0.382,
      "step": 585
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.07526222880479648,
      "learning_rate": 6.261048315245419e-05,
      "loss": 0.3112,
      "step": 586
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.07725562423865082,
      "learning_rate": 6.229051583116796e-05,
      "loss": 0.3248,
      "step": 587
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.07265450260153018,
      "learning_rate": 6.197099794512056e-05,
      "loss": 0.3304,
      "step": 588
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.08008761756737857,
      "learning_rate": 6.165193330244143e-05,
      "loss": 0.3519,
      "step": 589
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.0837137135807681,
      "learning_rate": 6.133332570585812e-05,
      "loss": 0.3338,
      "step": 590
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.08294791697322053,
      "learning_rate": 6.101517895265094e-05,
      "loss": 0.3514,
      "step": 591
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.08535305333963247,
      "learning_rate": 6.069749683460765e-05,
      "loss": 0.3722,
      "step": 592
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.0842929704042906,
      "learning_rate": 6.0380283137978396e-05,
      "loss": 0.383,
      "step": 593
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.0815197289066111,
      "learning_rate": 6.006354164343046e-05,
      "loss": 0.3879,
      "step": 594
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.07818811966248758,
      "learning_rate": 5.9747276126003257e-05,
      "loss": 0.3975,
      "step": 595
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.07781593462629392,
      "learning_rate": 5.943149035506337e-05,
      "loss": 0.3611,
      "step": 596
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.07759999060222111,
      "learning_rate": 5.9116188094259516e-05,
      "loss": 0.3494,
      "step": 597
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.0786183900535505,
      "learning_rate": 5.880137310147782e-05,
      "loss": 0.3407,
      "step": 598
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.08562034093492009,
      "learning_rate": 5.848704912879699e-05,
      "loss": 0.3474,
      "step": 599
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.08260942393320048,
      "learning_rate": 5.817321992244351e-05,
      "loss": 0.3329,
      "step": 600
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.07904114888657499,
      "learning_rate": 5.785988922274711e-05,
      "loss": 0.3524,
      "step": 601
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.07582527865163859,
      "learning_rate": 5.754706076409613e-05,
      "loss": 0.3433,
      "step": 602
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.0869013852078456,
      "learning_rate": 5.723473827489301e-05,
      "loss": 0.3676,
      "step": 603
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.07866051540263401,
      "learning_rate": 5.692292547750988e-05,
      "loss": 0.3548,
      "step": 604
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.08298313952251557,
      "learning_rate": 5.6611626088244194e-05,
      "loss": 0.3624,
      "step": 605
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.08120082412434228,
      "learning_rate": 5.630084381727434e-05,
      "loss": 0.35,
      "step": 606
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.08364501577678882,
      "learning_rate": 5.599058236861558e-05,
      "loss": 0.3143,
      "step": 607
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.07755177727522106,
      "learning_rate": 5.568084544007588e-05,
      "loss": 0.326,
      "step": 608
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.0839037119714208,
      "learning_rate": 5.5371636723211606e-05,
      "loss": 0.3656,
      "step": 609
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.08240199041279521,
      "learning_rate": 5.506295990328385e-05,
      "loss": 0.378,
      "step": 610
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.07648943764975623,
      "learning_rate": 5.4754818659214405e-05,
      "loss": 0.3411,
      "step": 611
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.0794911884788703,
      "learning_rate": 5.444721666354169e-05,
      "loss": 0.3334,
      "step": 612
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.07612986812407667,
      "learning_rate": 5.4140157582377336e-05,
      "loss": 0.3375,
      "step": 613
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.07371427661365673,
      "learning_rate": 5.383364507536229e-05,
      "loss": 0.3637,
      "step": 614
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.07428710700805227,
      "learning_rate": 5.3527682795623146e-05,
      "loss": 0.3338,
      "step": 615
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.07983783106037953,
      "learning_rate": 5.32222743897288e-05,
      "loss": 0.3398,
      "step": 616
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.07908192143487973,
      "learning_rate": 5.291742349764683e-05,
      "loss": 0.3582,
      "step": 617
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.08383309245612144,
      "learning_rate": 5.261313375270014e-05,
      "loss": 0.3816,
      "step": 618
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.07254861503772095,
      "learning_rate": 5.23094087815237e-05,
      "loss": 0.3414,
      "step": 619
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.06986123884274745,
      "learning_rate": 5.200625220402139e-05,
      "loss": 0.329,
      "step": 620
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.0805963229496782,
      "learning_rate": 5.170366763332257e-05,
      "loss": 0.3869,
      "step": 621
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.07858086444818861,
      "learning_rate": 5.14016586757394e-05,
      "loss": 0.3478,
      "step": 622
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.07616355434846495,
      "learning_rate": 5.1100228930723616e-05,
      "loss": 0.3337,
      "step": 623
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.08384597541604742,
      "learning_rate": 5.0799381990823636e-05,
      "loss": 0.3657,
      "step": 624
    },
    {
      "epoch": 1.99,
      "eval_loss": 0.420468270778656,
      "eval_runtime": 31.4902,
      "eval_samples_per_second": 50.81,
      "eval_steps_per_second": 0.794,
      "step": 624
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.07327253984323831,
      "learning_rate": 5.0499121441641864e-05,
      "loss": 0.329,
      "step": 625
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.08453781723805916,
      "learning_rate": 5.0199450861791916e-05,
      "loss": 0.388,
      "step": 626
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.07882486326711363,
      "learning_rate": 4.990037382285581e-05,
      "loss": 0.3541,
      "step": 627
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.08003482578396849,
      "learning_rate": 4.960189388934163e-05,
      "loss": 0.3386,
      "step": 628
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.0838769458500269,
      "learning_rate": 4.9304014618640995e-05,
      "loss": 0.3768,
      "step": 629
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.09332671571833408,
      "learning_rate": 4.900673956098644e-05,
      "loss": 0.3781,
      "step": 630
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.07910170672839167,
      "learning_rate": 4.87100722594094e-05,
      "loss": 0.3587,
      "step": 631
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.08098524635003682,
      "learning_rate": 4.841401624969782e-05,
      "loss": 0.3711,
      "step": 632
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.08047679772919877,
      "learning_rate": 4.811857506035407e-05,
      "loss": 0.3826,
      "step": 633
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.08459068332714359,
      "learning_rate": 4.782375221255285e-05,
      "loss": 0.335,
      "step": 634
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.09069536699997757,
      "learning_rate": 4.75295512200992e-05,
      "loss": 0.2951,
      "step": 635
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.08635001444449246,
      "learning_rate": 4.723597558938672e-05,
      "loss": 0.2609,
      "step": 636
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.09231785290881173,
      "learning_rate": 4.694302881935574e-05,
      "loss": 0.311,
      "step": 637
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.09216262109904506,
      "learning_rate": 4.66507144014515e-05,
      "loss": 0.2868,
      "step": 638
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.07998659264172378,
      "learning_rate": 4.6359035819582754e-05,
      "loss": 0.287,
      "step": 639
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.0872310232712246,
      "learning_rate": 4.606799655008009e-05,
      "loss": 0.2935,
      "step": 640
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.09462786970171194,
      "learning_rate": 4.5777600061654505e-05,
      "loss": 0.2902,
      "step": 641
    },
    {
      "epoch": 2.03,
      "grad_norm": 0.08889106283048188,
      "learning_rate": 4.548784981535614e-05,
      "loss": 0.2823,
      "step": 642
    },
    {
      "epoch": 2.03,
      "grad_norm": 0.09539933548381978,
      "learning_rate": 4.519874926453302e-05,
      "loss": 0.2801,
      "step": 643
    },
    {
      "epoch": 2.03,
      "grad_norm": 0.10073254835349207,
      "learning_rate": 4.491030185478976e-05,
      "loss": 0.2778,
      "step": 644
    },
    {
      "epoch": 2.04,
      "grad_norm": 0.12360798343872831,
      "learning_rate": 4.462251102394669e-05,
      "loss": 0.2837,
      "step": 645
    },
    {
      "epoch": 2.04,
      "grad_norm": 0.11958371690764477,
      "learning_rate": 4.433538020199882e-05,
      "loss": 0.2912,
      "step": 646
    },
    {
      "epoch": 2.04,
      "grad_norm": 0.1685182100963429,
      "learning_rate": 4.404891281107482e-05,
      "loss": 0.2793,
      "step": 647
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.14755862833725772,
      "learning_rate": 4.376311226539644e-05,
      "loss": 0.2993,
      "step": 648
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.13672529613627066,
      "learning_rate": 4.3477981971237767e-05,
      "loss": 0.3005,
      "step": 649
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.11493571015328757,
      "learning_rate": 4.3193525326884435e-05,
      "loss": 0.297,
      "step": 650
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.10967347597474586,
      "learning_rate": 4.290974572259342e-05,
      "loss": 0.2788,
      "step": 651
    },
    {
      "epoch": 2.06,
      "grad_norm": 0.10752386540019615,
      "learning_rate": 4.262664654055246e-05,
      "loss": 0.273,
      "step": 652
    },
    {
      "epoch": 2.06,
      "grad_norm": 0.12307571648351004,
      "learning_rate": 4.234423115483971e-05,
      "loss": 0.2983,
      "step": 653
    },
    {
      "epoch": 2.06,
      "grad_norm": 0.10021318421284302,
      "learning_rate": 4.2062502931383664e-05,
      "loss": 0.2561,
      "step": 654
    },
    {
      "epoch": 2.07,
      "grad_norm": 0.0913433477997775,
      "learning_rate": 4.1781465227922957e-05,
      "loss": 0.2528,
      "step": 655
    },
    {
      "epoch": 2.07,
      "grad_norm": 0.09457879486552605,
      "learning_rate": 4.1501121393966305e-05,
      "loss": 0.2856,
      "step": 656
    },
    {
      "epoch": 2.07,
      "grad_norm": 0.1027903701566837,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.2971,
      "step": 657
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.1008928625344023,
      "learning_rate": 4.0942528691211526e-05,
      "loss": 0.2736,
      "step": 658
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.11631218322544735,
      "learning_rate": 4.0664286479922754e-05,
      "loss": 0.2937,
      "step": 659
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.10157122482824303,
      "learning_rate": 4.038675145307747e-05,
      "loss": 0.276,
      "step": 660
    },
    {
      "epoch": 2.09,
      "grad_norm": 0.10094004400490046,
      "learning_rate": 4.010992691843829e-05,
      "loss": 0.2692,
      "step": 661
    },
    {
      "epoch": 2.09,
      "grad_norm": 0.09865119970425237,
      "learning_rate": 3.98338161752999e-05,
      "loss": 0.2678,
      "step": 662
    },
    {
      "epoch": 2.09,
      "grad_norm": 0.11518683099293696,
      "learning_rate": 3.9558422514449774e-05,
      "loss": 0.2812,
      "step": 663
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.10980691099409233,
      "learning_rate": 3.9283749218128885e-05,
      "loss": 0.2883,
      "step": 664
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.11039367226797306,
      "learning_rate": 3.900979955999271e-05,
      "loss": 0.2692,
      "step": 665
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.10807597680719257,
      "learning_rate": 3.873657680507217e-05,
      "loss": 0.264,
      "step": 666
    },
    {
      "epoch": 2.11,
      "grad_norm": 0.12807158979333827,
      "learning_rate": 3.846408420973455e-05,
      "loss": 0.2883,
      "step": 667
    },
    {
      "epoch": 2.11,
      "grad_norm": 0.11641944484045623,
      "learning_rate": 3.819232502164499e-05,
      "loss": 0.2883,
      "step": 668
    },
    {
      "epoch": 2.11,
      "grad_norm": 0.10787128594205424,
      "learning_rate": 3.792130247972756e-05,
      "loss": 0.2716,
      "step": 669
    },
    {
      "epoch": 2.12,
      "grad_norm": 0.11612766953136715,
      "learning_rate": 3.7651019814126654e-05,
      "loss": 0.28,
      "step": 670
    },
    {
      "epoch": 2.12,
      "grad_norm": 0.11583875313121475,
      "learning_rate": 3.738148024616863e-05,
      "loss": 0.2821,
      "step": 671
    },
    {
      "epoch": 2.12,
      "grad_norm": 0.10871704207101558,
      "learning_rate": 3.711268698832335e-05,
      "loss": 0.2717,
      "step": 672
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.11207061090758715,
      "learning_rate": 3.684464324416578e-05,
      "loss": 0.2719,
      "step": 673
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.11635664212457444,
      "learning_rate": 3.6577352208338015e-05,
      "loss": 0.2578,
      "step": 674
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.11062630068019738,
      "learning_rate": 3.6310817066511105e-05,
      "loss": 0.2728,
      "step": 675
    },
    {
      "epoch": 2.14,
      "grad_norm": 0.10316685184952971,
      "learning_rate": 3.604504099534696e-05,
      "loss": 0.2851,
      "step": 676
    },
    {
      "epoch": 2.14,
      "grad_norm": 0.10947256760334166,
      "learning_rate": 3.5780027162460736e-05,
      "loss": 0.2922,
      "step": 677
    },
    {
      "epoch": 2.14,
      "grad_norm": 0.10674401902170978,
      "learning_rate": 3.5515778726382966e-05,
      "loss": 0.2708,
      "step": 678
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.1047734281548304,
      "learning_rate": 3.5252298836521766e-05,
      "loss": 0.2794,
      "step": 679
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.11074787895607416,
      "learning_rate": 3.498959063312558e-05,
      "loss": 0.2679,
      "step": 680
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.10885916134526002,
      "learning_rate": 3.47276572472456e-05,
      "loss": 0.2583,
      "step": 681
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.109437549961412,
      "learning_rate": 3.446650180069837e-05,
      "loss": 0.2691,
      "step": 682
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.11099241066345907,
      "learning_rate": 3.4206127406028745e-05,
      "loss": 0.2567,
      "step": 683
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.10966550615353308,
      "learning_rate": 3.394653716647277e-05,
      "loss": 0.2805,
      "step": 684
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.11169657053562163,
      "learning_rate": 3.36877341759205e-05,
      "loss": 0.2616,
      "step": 685
    },
    {
      "epoch": 2.17,
      "grad_norm": 0.11873657470655434,
      "learning_rate": 3.342972151887941e-05,
      "loss": 0.3129,
      "step": 686
    },
    {
      "epoch": 2.17,
      "grad_norm": 0.10971252694540876,
      "learning_rate": 3.3172502270437454e-05,
      "loss": 0.2592,
      "step": 687
    },
    {
      "epoch": 2.17,
      "grad_norm": 0.10649209787962353,
      "learning_rate": 3.29160794962264e-05,
      "loss": 0.2613,
      "step": 688
    },
    {
      "epoch": 2.18,
      "grad_norm": 0.11308270352603313,
      "learning_rate": 3.266045625238538e-05,
      "loss": 0.2694,
      "step": 689
    },
    {
      "epoch": 2.18,
      "grad_norm": 0.11998647850816677,
      "learning_rate": 3.2405635585524565e-05,
      "loss": 0.2731,
      "step": 690
    },
    {
      "epoch": 2.18,
      "grad_norm": 0.12104691091782901,
      "learning_rate": 3.21516205326885e-05,
      "loss": 0.2888,
      "step": 691
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.1179672880587401,
      "learning_rate": 3.1898414121320276e-05,
      "loss": 0.2669,
      "step": 692
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.1147115290101146,
      "learning_rate": 3.164601936922528e-05,
      "loss": 0.2682,
      "step": 693
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.1168680406413447,
      "learning_rate": 3.1394439284535205e-05,
      "loss": 0.2746,
      "step": 694
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.11342242914504051,
      "learning_rate": 3.114367686567228e-05,
      "loss": 0.2761,
      "step": 695
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.10991320429442564,
      "learning_rate": 3.089373510131354e-05,
      "loss": 0.2718,
      "step": 696
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.11228406550935485,
      "learning_rate": 3.0644616970355054e-05,
      "loss": 0.2744,
      "step": 697
    },
    {
      "epoch": 2.21,
      "grad_norm": 0.10890177977539851,
      "learning_rate": 3.039632544187663e-05,
      "loss": 0.2887,
      "step": 698
    },
    {
      "epoch": 2.21,
      "grad_norm": 0.11076095479218435,
      "learning_rate": 3.0148863475106314e-05,
      "loss": 0.2637,
      "step": 699
    },
    {
      "epoch": 2.21,
      "grad_norm": 0.1143508108804589,
      "learning_rate": 2.9902234019385057e-05,
      "loss": 0.2525,
      "step": 700
    },
    {
      "epoch": 2.22,
      "grad_norm": 0.11821080694052107,
      "learning_rate": 2.9656440014131737e-05,
      "loss": 0.2803,
      "step": 701
    },
    {
      "epoch": 2.22,
      "grad_norm": 0.12283194025891259,
      "learning_rate": 2.941148438880803e-05,
      "loss": 0.287,
      "step": 702
    },
    {
      "epoch": 2.22,
      "eval_loss": 0.4607037901878357,
      "eval_runtime": 31.4671,
      "eval_samples_per_second": 50.847,
      "eval_steps_per_second": 0.794,
      "step": 702
    },
    {
      "epoch": 2.22,
      "grad_norm": 0.10618017142486377,
      "learning_rate": 2.9167370062883403e-05,
      "loss": 0.2431,
      "step": 703
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.11274524190900949,
      "learning_rate": 2.892409994580053e-05,
      "loss": 0.2755,
      "step": 704
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.11208676465742909,
      "learning_rate": 2.8681676936940393e-05,
      "loss": 0.2725,
      "step": 705
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.1080271574221204,
      "learning_rate": 2.84401039255879e-05,
      "loss": 0.2757,
      "step": 706
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.11003973883142379,
      "learning_rate": 2.8199383790897404e-05,
      "loss": 0.2589,
      "step": 707
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.12169213652263852,
      "learning_rate": 2.795951940185827e-05,
      "loss": 0.2704,
      "step": 708
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.10899008988877626,
      "learning_rate": 2.7720513617260856e-05,
      "loss": 0.2568,
      "step": 709
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.11162434871062812,
      "learning_rate": 2.7482369285662378e-05,
      "loss": 0.2631,
      "step": 710
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.1156078283987831,
      "learning_rate": 2.7245089245352862e-05,
      "loss": 0.2656,
      "step": 711
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.10639613003997833,
      "learning_rate": 2.7008676324321447e-05,
      "loss": 0.2396,
      "step": 712
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.10704066955577284,
      "learning_rate": 2.677313334022268e-05,
      "loss": 0.2641,
      "step": 713
    },
    {
      "epoch": 2.26,
      "grad_norm": 0.10510223778516452,
      "learning_rate": 2.653846310034277e-05,
      "loss": 0.262,
      "step": 714
    },
    {
      "epoch": 2.26,
      "grad_norm": 0.11621876698253043,
      "learning_rate": 2.6304668401566335e-05,
      "loss": 0.2551,
      "step": 715
    },
    {
      "epoch": 2.26,
      "grad_norm": 0.11374741040415604,
      "learning_rate": 2.607175203034299e-05,
      "loss": 0.2632,
      "step": 716
    },
    {
      "epoch": 2.27,
      "grad_norm": 0.11878165118132909,
      "learning_rate": 2.5839716762653998e-05,
      "loss": 0.2818,
      "step": 717
    },
    {
      "epoch": 2.27,
      "grad_norm": 0.11411847478631887,
      "learning_rate": 2.56085653639795e-05,
      "loss": 0.2679,
      "step": 718
    },
    {
      "epoch": 2.27,
      "grad_norm": 0.11927953423562035,
      "learning_rate": 2.5378300589265257e-05,
      "loss": 0.2718,
      "step": 719
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.11292387644233305,
      "learning_rate": 2.514892518288988e-05,
      "loss": 0.2644,
      "step": 720
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.12997215467957038,
      "learning_rate": 2.4920441878632272e-05,
      "loss": 0.2618,
      "step": 721
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.11477522784542633,
      "learning_rate": 2.4692853399638917e-05,
      "loss": 0.2615,
      "step": 722
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.11801399206569434,
      "learning_rate": 2.446616245839136e-05,
      "loss": 0.2911,
      "step": 723
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.11678416460466445,
      "learning_rate": 2.4240371756674064e-05,
      "loss": 0.2732,
      "step": 724
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.10663418680535579,
      "learning_rate": 2.401548398554213e-05,
      "loss": 0.2729,
      "step": 725
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.12210427384023882,
      "learning_rate": 2.3791501825289087e-05,
      "loss": 0.2677,
      "step": 726
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.12038247346356847,
      "learning_rate": 2.356842794541516e-05,
      "loss": 0.2755,
      "step": 727
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.1274827035223249,
      "learning_rate": 2.3346265004595392e-05,
      "loss": 0.2808,
      "step": 728
    },
    {
      "epoch": 2.31,
      "grad_norm": 0.12753822770363968,
      "learning_rate": 2.3125015650647797e-05,
      "loss": 0.2914,
      "step": 729
    },
    {
      "epoch": 2.31,
      "grad_norm": 0.12364919397261674,
      "learning_rate": 2.290468252050204e-05,
      "loss": 0.2826,
      "step": 730
    },
    {
      "epoch": 2.31,
      "grad_norm": 0.12135969844852122,
      "learning_rate": 2.2685268240167902e-05,
      "loss": 0.2628,
      "step": 731
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.11304483325514314,
      "learning_rate": 2.246677542470388e-05,
      "loss": 0.2818,
      "step": 732
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.12122915373246147,
      "learning_rate": 2.224920667818622e-05,
      "loss": 0.2785,
      "step": 733
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.10927091056289465,
      "learning_rate": 2.2032564593677774e-05,
      "loss": 0.259,
      "step": 734
    },
    {
      "epoch": 2.33,
      "grad_norm": 0.10836119914289992,
      "learning_rate": 2.181685175319702e-05,
      "loss": 0.2308,
      "step": 735
    },
    {
      "epoch": 2.33,
      "grad_norm": 0.11719069476541212,
      "learning_rate": 2.160207072768746e-05,
      "loss": 0.2705,
      "step": 736
    },
    {
      "epoch": 2.33,
      "grad_norm": 0.11182760407461201,
      "learning_rate": 2.1388224076986874e-05,
      "loss": 0.2477,
      "step": 737
    },
    {
      "epoch": 2.34,
      "grad_norm": 0.10616491729332692,
      "learning_rate": 2.117531434979675e-05,
      "loss": 0.2601,
      "step": 738
    },
    {
      "epoch": 2.34,
      "grad_norm": 0.12111208761482414,
      "learning_rate": 2.096334408365207e-05,
      "loss": 0.2683,
      "step": 739
    },
    {
      "epoch": 2.34,
      "grad_norm": 0.12124840796792066,
      "learning_rate": 2.0752315804890977e-05,
      "loss": 0.2642,
      "step": 740
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.11287575528112874,
      "learning_rate": 2.0542232028624586e-05,
      "loss": 0.246,
      "step": 741
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.11651709242939365,
      "learning_rate": 2.033309525870717e-05,
      "loss": 0.2711,
      "step": 742
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.12067094176278668,
      "learning_rate": 2.012490798770624e-05,
      "loss": 0.2674,
      "step": 743
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.11427558043480016,
      "learning_rate": 1.991767269687278e-05,
      "loss": 0.2704,
      "step": 744
    },
    {
      "epoch": 2.36,
      "grad_norm": 0.11321595381869821,
      "learning_rate": 1.971139185611176e-05,
      "loss": 0.2635,
      "step": 745
    },
    {
      "epoch": 2.36,
      "grad_norm": 0.12474627747289992,
      "learning_rate": 1.9506067923952676e-05,
      "loss": 0.2961,
      "step": 746
    },
    {
      "epoch": 2.36,
      "grad_norm": 0.11418438451204156,
      "learning_rate": 1.930170334752025e-05,
      "loss": 0.2796,
      "step": 747
    },
    {
      "epoch": 2.37,
      "grad_norm": 0.11646749722139911,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.2507,
      "step": 748
    },
    {
      "epoch": 2.37,
      "grad_norm": 0.12562474204506652,
      "learning_rate": 1.8895861993135444e-05,
      "loss": 0.275,
      "step": 749
    },
    {
      "epoch": 2.37,
      "grad_norm": 0.12444807183638722,
      "learning_rate": 1.8694390052146737e-05,
      "loss": 0.2913,
      "step": 750
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.12022056627946713,
      "learning_rate": 1.8493887140754464e-05,
      "loss": 0.2951,
      "step": 751
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.11563361238932722,
      "learning_rate": 1.8294355648624606e-05,
      "loss": 0.2789,
      "step": 752
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.11307908796738765,
      "learning_rate": 1.8095797953845505e-05,
      "loss": 0.256,
      "step": 753
    },
    {
      "epoch": 2.39,
      "grad_norm": 0.11304898539499737,
      "learning_rate": 1.789821642289945e-05,
      "loss": 0.2564,
      "step": 754
    },
    {
      "epoch": 2.39,
      "grad_norm": 0.11688020470193958,
      "learning_rate": 1.7701613410634365e-05,
      "loss": 0.2924,
      "step": 755
    },
    {
      "epoch": 2.39,
      "grad_norm": 0.11172231938588859,
      "learning_rate": 1.750599126023591e-05,
      "loss": 0.2611,
      "step": 756
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.10987074932571032,
      "learning_rate": 1.731135230319948e-05,
      "loss": 0.2516,
      "step": 757
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.12464848654951363,
      "learning_rate": 1.7117698859302357e-05,
      "loss": 0.292,
      "step": 758
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.11800113195648207,
      "learning_rate": 1.6925033236576172e-05,
      "loss": 0.2741,
      "step": 759
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.10942401809369491,
      "learning_rate": 1.6733357731279377e-05,
      "loss": 0.2515,
      "step": 760
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.1190601686940271,
      "learning_rate": 1.6542674627869737e-05,
      "loss": 0.278,
      "step": 761
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.11445268494192679,
      "learning_rate": 1.6352986198977325e-05,
      "loss": 0.2704,
      "step": 762
    },
    {
      "epoch": 2.42,
      "grad_norm": 0.11720715199041416,
      "learning_rate": 1.6164294705377292e-05,
      "loss": 0.2848,
      "step": 763
    },
    {
      "epoch": 2.42,
      "grad_norm": 0.12070811301483302,
      "learning_rate": 1.5976602395962893e-05,
      "loss": 0.2783,
      "step": 764
    },
    {
      "epoch": 2.42,
      "grad_norm": 0.13468157178794404,
      "learning_rate": 1.5789911507718826e-05,
      "loss": 0.3039,
      "step": 765
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.12061677331493204,
      "learning_rate": 1.560422426569449e-05,
      "loss": 0.2662,
      "step": 766
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.12673405305579089,
      "learning_rate": 1.5419542882977368e-05,
      "loss": 0.2822,
      "step": 767
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.11814809328732971,
      "learning_rate": 1.523586956066686e-05,
      "loss": 0.26,
      "step": 768
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.11898546981683543,
      "learning_rate": 1.5053206487847914e-05,
      "loss": 0.2405,
      "step": 769
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.11742711207169722,
      "learning_rate": 1.4871555841564887e-05,
      "loss": 0.305,
      "step": 770
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.11871453998066525,
      "learning_rate": 1.4690919786795764e-05,
      "loss": 0.2798,
      "step": 771
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.12298585120264952,
      "learning_rate": 1.4511300476426227e-05,
      "loss": 0.2808,
      "step": 772
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.12615015209022892,
      "learning_rate": 1.4332700051223991e-05,
      "loss": 0.2974,
      "step": 773
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.11430665952738557,
      "learning_rate": 1.415512063981339e-05,
      "loss": 0.2671,
      "step": 774
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.1171650666381251,
      "learning_rate": 1.3978564358649927e-05,
      "loss": 0.2731,
      "step": 775
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.12222819227556621,
      "learning_rate": 1.3803033311995072e-05,
      "loss": 0.2834,
      "step": 776
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.1163107259252253,
      "learning_rate": 1.3628529591891181e-05,
      "loss": 0.2669,
      "step": 777
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.12540992932847908,
      "learning_rate": 1.3455055278136519e-05,
      "loss": 0.2621,
      "step": 778
    },
    {
      "epoch": 2.47,
      "grad_norm": 0.11680539141579303,
      "learning_rate": 1.3282612438260578e-05,
      "loss": 0.259,
      "step": 779
    },
    {
      "epoch": 2.47,
      "grad_norm": 0.11336206151669477,
      "learning_rate": 1.311120312749935e-05,
      "loss": 0.2472,
      "step": 780
    },
    {
      "epoch": 2.47,
      "eval_loss": 0.4616280198097229,
      "eval_runtime": 31.5042,
      "eval_samples_per_second": 50.787,
      "eval_steps_per_second": 0.794,
      "step": 780
    },
    {
      "epoch": 2.47,
      "grad_norm": 0.11231755721944928,
      "learning_rate": 1.2940829388770836e-05,
      "loss": 0.254,
      "step": 781
    },
    {
      "epoch": 2.48,
      "grad_norm": 0.11613657118626357,
      "learning_rate": 1.2771493252650723e-05,
      "loss": 0.2569,
      "step": 782
    },
    {
      "epoch": 2.48,
      "grad_norm": 0.1205692005065408,
      "learning_rate": 1.260319673734821e-05,
      "loss": 0.2808,
      "step": 783
    },
    {
      "epoch": 2.48,
      "grad_norm": 0.13429235201104567,
      "learning_rate": 1.2435941848681864e-05,
      "loss": 0.2992,
      "step": 784
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.12209777141726344,
      "learning_rate": 1.2269730580055805e-05,
      "loss": 0.292,
      "step": 785
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.11984230781692078,
      "learning_rate": 1.2104564912435923e-05,
      "loss": 0.2545,
      "step": 786
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.11064840160176172,
      "learning_rate": 1.19404468143262e-05,
      "loss": 0.2662,
      "step": 787
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.11384032261333481,
      "learning_rate": 1.1777378241745385e-05,
      "loss": 0.243,
      "step": 788
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.11761268728737763,
      "learning_rate": 1.1615361138203573e-05,
      "loss": 0.2566,
      "step": 789
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.11333949137476289,
      "learning_rate": 1.1454397434679021e-05,
      "loss": 0.2621,
      "step": 790
    },
    {
      "epoch": 2.51,
      "grad_norm": 0.11888615604478664,
      "learning_rate": 1.1294489049595248e-05,
      "loss": 0.2571,
      "step": 791
    },
    {
      "epoch": 2.51,
      "grad_norm": 0.14555553319408016,
      "learning_rate": 1.11356378887981e-05,
      "loss": 0.2501,
      "step": 792
    },
    {
      "epoch": 2.51,
      "grad_norm": 0.11806896023949925,
      "learning_rate": 1.0977845845533008e-05,
      "loss": 0.256,
      "step": 793
    },
    {
      "epoch": 2.52,
      "grad_norm": 0.12305227525153528,
      "learning_rate": 1.0821114800422483e-05,
      "loss": 0.2757,
      "step": 794
    },
    {
      "epoch": 2.52,
      "grad_norm": 0.11698046515681931,
      "learning_rate": 1.0665446621443708e-05,
      "loss": 0.2463,
      "step": 795
    },
    {
      "epoch": 2.52,
      "grad_norm": 0.12106256698760157,
      "learning_rate": 1.0510843163906148e-05,
      "loss": 0.2738,
      "step": 796
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.11713059951298724,
      "learning_rate": 1.0357306270429624e-05,
      "loss": 0.266,
      "step": 797
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.11310905490381504,
      "learning_rate": 1.020483777092226e-05,
      "loss": 0.2528,
      "step": 798
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.11442184085785533,
      "learning_rate": 1.0053439482558601e-05,
      "loss": 0.2482,
      "step": 799
    },
    {
      "epoch": 2.54,
      "grad_norm": 0.11907659044892492,
      "learning_rate": 9.903113209758096e-06,
      "loss": 0.2621,
      "step": 800
    },
    {
      "epoch": 2.54,
      "grad_norm": 0.11344451484586272,
      "learning_rate": 9.753860744163524e-06,
      "loss": 0.2738,
      "step": 801
    },
    {
      "epoch": 2.54,
      "grad_norm": 0.1268044481466165,
      "learning_rate": 9.605683864619574e-06,
      "loss": 0.2729,
      "step": 802
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.12313277788211967,
      "learning_rate": 9.458584337151811e-06,
      "loss": 0.2784,
      "step": 803
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.1113812545688282,
      "learning_rate": 9.31256391494546e-06,
      "loss": 0.2621,
      "step": 804
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.11960885016029069,
      "learning_rate": 9.1676243383246e-06,
      "loss": 0.2605,
      "step": 805
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.12129984514402677,
      "learning_rate": 9.023767334731425e-06,
      "loss": 0.2649,
      "step": 806
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.11493833954251931,
      "learning_rate": 8.880994618705574e-06,
      "loss": 0.2627,
      "step": 807
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.11588267035569787,
      "learning_rate": 8.739307891863813e-06,
      "loss": 0.2686,
      "step": 808
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.11796713805834204,
      "learning_rate": 8.598708842879689e-06,
      "loss": 0.2596,
      "step": 809
    },
    {
      "epoch": 2.57,
      "grad_norm": 0.12059030167195925,
      "learning_rate": 8.45919914746337e-06,
      "loss": 0.2426,
      "step": 810
    },
    {
      "epoch": 2.57,
      "grad_norm": 0.12863692411726227,
      "learning_rate": 8.32078046834176e-06,
      "loss": 0.28,
      "step": 811
    },
    {
      "epoch": 2.57,
      "grad_norm": 0.11855028417664637,
      "learning_rate": 8.183454455238636e-06,
      "loss": 0.2759,
      "step": 812
    },
    {
      "epoch": 2.58,
      "grad_norm": 0.13518389463224478,
      "learning_rate": 8.047222744854943e-06,
      "loss": 0.2797,
      "step": 813
    },
    {
      "epoch": 2.58,
      "grad_norm": 0.11896717625386971,
      "learning_rate": 7.912086960849374e-06,
      "loss": 0.2679,
      "step": 814
    },
    {
      "epoch": 2.58,
      "grad_norm": 0.11945593649110982,
      "learning_rate": 7.778048713818975e-06,
      "loss": 0.2575,
      "step": 815
    },
    {
      "epoch": 2.59,
      "grad_norm": 0.1314685860445357,
      "learning_rate": 7.645109601279921e-06,
      "loss": 0.2927,
      "step": 816
    },
    {
      "epoch": 2.59,
      "grad_norm": 0.11335570667280918,
      "learning_rate": 7.513271207648532e-06,
      "loss": 0.2802,
      "step": 817
    },
    {
      "epoch": 2.59,
      "grad_norm": 0.12344467860485761,
      "learning_rate": 7.382535104222366e-06,
      "loss": 0.2719,
      "step": 818
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.11739097652512145,
      "learning_rate": 7.252902849161436e-06,
      "loss": 0.228,
      "step": 819
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.1120903789895003,
      "learning_rate": 7.124375987469767e-06,
      "loss": 0.2356,
      "step": 820
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.11506670132462016,
      "learning_rate": 6.996956050976877e-06,
      "loss": 0.2603,
      "step": 821
    },
    {
      "epoch": 2.61,
      "grad_norm": 0.11840704691214253,
      "learning_rate": 6.870644558319528e-06,
      "loss": 0.2603,
      "step": 822
    },
    {
      "epoch": 2.61,
      "grad_norm": 0.1280231716660491,
      "learning_rate": 6.745443014923658e-06,
      "loss": 0.2366,
      "step": 823
    },
    {
      "epoch": 2.61,
      "grad_norm": 0.12478948539357017,
      "learning_rate": 6.621352912986467e-06,
      "loss": 0.2798,
      "step": 824
    },
    {
      "epoch": 2.62,
      "grad_norm": 0.1143850775224043,
      "learning_rate": 6.498375731458528e-06,
      "loss": 0.2589,
      "step": 825
    },
    {
      "epoch": 2.62,
      "grad_norm": 0.10991498790006785,
      "learning_rate": 6.37651293602628e-06,
      "loss": 0.2581,
      "step": 826
    },
    {
      "epoch": 2.62,
      "grad_norm": 0.11851915490030443,
      "learning_rate": 6.2557659790945185e-06,
      "loss": 0.2681,
      "step": 827
    },
    {
      "epoch": 2.63,
      "grad_norm": 0.11726129513065525,
      "learning_rate": 6.136136299768991e-06,
      "loss": 0.2631,
      "step": 828
    },
    {
      "epoch": 2.63,
      "grad_norm": 0.11653911908981955,
      "learning_rate": 6.017625323839415e-06,
      "loss": 0.2623,
      "step": 829
    },
    {
      "epoch": 2.63,
      "grad_norm": 0.12276112968416866,
      "learning_rate": 5.900234463762366e-06,
      "loss": 0.2512,
      "step": 830
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.12156192199599036,
      "learning_rate": 5.783965118644441e-06,
      "loss": 0.2708,
      "step": 831
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.1179784862340501,
      "learning_rate": 5.668818674225685e-06,
      "loss": 0.254,
      "step": 832
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.11429129215980628,
      "learning_rate": 5.554796502862958e-06,
      "loss": 0.2588,
      "step": 833
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.11855933778871278,
      "learning_rate": 5.44189996351363e-06,
      "loss": 0.2668,
      "step": 834
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.12848156271807984,
      "learning_rate": 5.3301304017194135e-06,
      "loss": 0.2624,
      "step": 835
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.14802409628408783,
      "learning_rate": 5.219489149590251e-06,
      "loss": 0.2635,
      "step": 836
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.11370611265083304,
      "learning_rate": 5.109977525788512e-06,
      "loss": 0.2492,
      "step": 837
    },
    {
      "epoch": 2.66,
      "grad_norm": 0.12515230139267264,
      "learning_rate": 5.0015968355132555e-06,
      "loss": 0.2667,
      "step": 838
    },
    {
      "epoch": 2.66,
      "grad_norm": 0.11654124880557705,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.2583,
      "step": 839
    },
    {
      "epoch": 2.66,
      "grad_norm": 0.1198065276015506,
      "learning_rate": 4.788233408928589e-06,
      "loss": 0.2744,
      "step": 840
    },
    {
      "epoch": 2.67,
      "grad_norm": 0.11967721469724298,
      "learning_rate": 4.68325321556149e-06,
      "loss": 0.2657,
      "step": 841
    },
    {
      "epoch": 2.67,
      "grad_norm": 0.12018383317195055,
      "learning_rate": 4.5794090415751665e-06,
      "loss": 0.2561,
      "step": 842
    },
    {
      "epoch": 2.67,
      "grad_norm": 0.11383394947087137,
      "learning_rate": 4.476702124621956e-06,
      "loss": 0.2652,
      "step": 843
    },
    {
      "epoch": 2.68,
      "grad_norm": 0.12120983112824826,
      "learning_rate": 4.375133688799959e-06,
      "loss": 0.2745,
      "step": 844
    },
    {
      "epoch": 2.68,
      "grad_norm": 0.12116730670938095,
      "learning_rate": 4.27470494463843e-06,
      "loss": 0.2839,
      "step": 845
    },
    {
      "epoch": 2.68,
      "grad_norm": 0.12118621725920352,
      "learning_rate": 4.175417089083378e-06,
      "loss": 0.2787,
      "step": 846
    },
    {
      "epoch": 2.69,
      "grad_norm": 0.12210517385508135,
      "learning_rate": 4.0772713054833216e-06,
      "loss": 0.2577,
      "step": 847
    },
    {
      "epoch": 2.69,
      "grad_norm": 0.1155349279124714,
      "learning_rate": 3.980268763575079e-06,
      "loss": 0.2504,
      "step": 848
    },
    {
      "epoch": 2.69,
      "grad_norm": 0.11927896816158835,
      "learning_rate": 3.8844106194699696e-06,
      "loss": 0.2766,
      "step": 849
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.11757637906600046,
      "learning_rate": 3.789698015639953e-06,
      "loss": 0.2641,
      "step": 850
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.13294257778294574,
      "learning_rate": 3.6961320809039912e-06,
      "loss": 0.2723,
      "step": 851
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.1121175159927733,
      "learning_rate": 3.6037139304146762e-06,
      "loss": 0.2508,
      "step": 852
    },
    {
      "epoch": 2.71,
      "grad_norm": 0.1226487379142707,
      "learning_rate": 3.512444665644865e-06,
      "loss": 0.2657,
      "step": 853
    },
    {
      "epoch": 2.71,
      "grad_norm": 0.11983665022528354,
      "learning_rate": 3.4223253743745797e-06,
      "loss": 0.2785,
      "step": 854
    },
    {
      "epoch": 2.71,
      "grad_norm": 0.12373769689340608,
      "learning_rate": 3.3333571306780497e-06,
      "loss": 0.2856,
      "step": 855
    },
    {
      "epoch": 2.72,
      "grad_norm": 0.12467618813268888,
      "learning_rate": 3.245540994910934e-06,
      "loss": 0.2814,
      "step": 856
    },
    {
      "epoch": 2.72,
      "grad_norm": 0.11409160148057285,
      "learning_rate": 3.158878013697586e-06,
      "loss": 0.2454,
      "step": 857
    },
    {
      "epoch": 2.72,
      "grad_norm": 0.11419063614806882,
      "learning_rate": 3.073369219918698e-06,
      "loss": 0.2541,
      "step": 858
    },
    {
      "epoch": 2.72,
      "eval_loss": 0.4631785452365875,
      "eval_runtime": 31.4806,
      "eval_samples_per_second": 50.825,
      "eval_steps_per_second": 0.794,
      "step": 858
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.12407471505381608,
      "learning_rate": 2.9890156326989437e-06,
      "loss": 0.269,
      "step": 859
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.12386640609812333,
      "learning_rate": 2.905818257394799e-06,
      "loss": 0.2698,
      "step": 860
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.12962050229779049,
      "learning_rate": 2.8237780855825956e-06,
      "loss": 0.2673,
      "step": 861
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.13353409801742866,
      "learning_rate": 2.742896095046732e-06,
      "loss": 0.3056,
      "step": 862
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.11801600046803927,
      "learning_rate": 2.663173249767936e-06,
      "loss": 0.264,
      "step": 863
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.11632643097642553,
      "learning_rate": 2.5846104999118325e-06,
      "loss": 0.2386,
      "step": 864
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.11945225654820267,
      "learning_rate": 2.5072087818176382e-06,
      "loss": 0.2636,
      "step": 865
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.11564876112492893,
      "learning_rate": 2.4309690179869503e-06,
      "loss": 0.2482,
      "step": 866
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.11111842030418848,
      "learning_rate": 2.3558921170727888e-06,
      "loss": 0.2509,
      "step": 867
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.1258453602627421,
      "learning_rate": 2.281978973868748e-06,
      "loss": 0.2612,
      "step": 868
    },
    {
      "epoch": 2.76,
      "grad_norm": 0.1319781842893948,
      "learning_rate": 2.20923046929834e-06,
      "loss": 0.2568,
      "step": 869
    },
    {
      "epoch": 2.76,
      "grad_norm": 0.11896606461595723,
      "learning_rate": 2.137647470404469e-06,
      "loss": 0.281,
      "step": 870
    },
    {
      "epoch": 2.76,
      "grad_norm": 0.10770903698788507,
      "learning_rate": 2.067230830339184e-06,
      "loss": 0.2528,
      "step": 871
    },
    {
      "epoch": 2.77,
      "grad_norm": 0.11886023139093475,
      "learning_rate": 1.997981388353376e-06,
      "loss": 0.2681,
      "step": 872
    },
    {
      "epoch": 2.77,
      "grad_norm": 0.13175528849885204,
      "learning_rate": 1.9298999697868967e-06,
      "loss": 0.2896,
      "step": 873
    },
    {
      "epoch": 2.77,
      "grad_norm": 0.12410952741514393,
      "learning_rate": 1.8629873860586566e-06,
      "loss": 0.286,
      "step": 874
    },
    {
      "epoch": 2.78,
      "grad_norm": 0.1216075403194452,
      "learning_rate": 1.797244434656975e-06,
      "loss": 0.2703,
      "step": 875
    },
    {
      "epoch": 2.78,
      "grad_norm": 0.12930878011055508,
      "learning_rate": 1.7326718991300562e-06,
      "loss": 0.2882,
      "step": 876
    },
    {
      "epoch": 2.78,
      "grad_norm": 0.13072221599455136,
      "learning_rate": 1.6692705490766957e-06,
      "loss": 0.2778,
      "step": 877
    },
    {
      "epoch": 2.79,
      "grad_norm": 0.11117499678860303,
      "learning_rate": 1.6070411401370334e-06,
      "loss": 0.2573,
      "step": 878
    },
    {
      "epoch": 2.79,
      "grad_norm": 0.11359863379630329,
      "learning_rate": 1.5459844139836477e-06,
      "loss": 0.2606,
      "step": 879
    },
    {
      "epoch": 2.79,
      "grad_norm": 0.1211143637616324,
      "learning_rate": 1.48610109831262e-06,
      "loss": 0.2497,
      "step": 880
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.11611223436915132,
      "learning_rate": 1.4273919068349184e-06,
      "loss": 0.2797,
      "step": 881
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.12085756217913753,
      "learning_rate": 1.3698575392678492e-06,
      "loss": 0.2574,
      "step": 882
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.12766042429380325,
      "learning_rate": 1.313498681326797e-06,
      "loss": 0.2909,
      "step": 883
    },
    {
      "epoch": 2.81,
      "grad_norm": 0.13388090392325733,
      "learning_rate": 1.258316004716953e-06,
      "loss": 0.2963,
      "step": 884
    },
    {
      "epoch": 2.81,
      "grad_norm": 0.11517933969885684,
      "learning_rate": 1.2043101671253554e-06,
      "loss": 0.2658,
      "step": 885
    },
    {
      "epoch": 2.81,
      "grad_norm": 0.11888714309122816,
      "learning_rate": 1.1514818122130845e-06,
      "loss": 0.2495,
      "step": 886
    },
    {
      "epoch": 2.82,
      "grad_norm": 0.14775431419260776,
      "learning_rate": 1.0998315696075122e-06,
      "loss": 0.3034,
      "step": 887
    },
    {
      "epoch": 2.82,
      "grad_norm": 0.11965685148229928,
      "learning_rate": 1.0493600548948878e-06,
      "loss": 0.274,
      "step": 888
    },
    {
      "epoch": 2.82,
      "grad_norm": 0.12011355107703363,
      "learning_rate": 1.0000678696129306e-06,
      "loss": 0.2525,
      "step": 889
    },
    {
      "epoch": 2.83,
      "grad_norm": 0.1303662002730925,
      "learning_rate": 9.519556012436815e-07,
      "loss": 0.2674,
      "step": 890
    },
    {
      "epoch": 2.83,
      "grad_norm": 0.11123034298883819,
      "learning_rate": 9.0502382320653e-07,
      "loss": 0.2731,
      "step": 891
    },
    {
      "epoch": 2.83,
      "grad_norm": 0.12047670417284058,
      "learning_rate": 8.592730948513206e-07,
      "loss": 0.2689,
      "step": 892
    },
    {
      "epoch": 2.84,
      "grad_norm": 0.12338785202182712,
      "learning_rate": 8.147039614517571e-07,
      "loss": 0.2827,
      "step": 893
    },
    {
      "epoch": 2.84,
      "grad_norm": 0.11357914052124327,
      "learning_rate": 7.71316954198853e-07,
      "loss": 0.2519,
      "step": 894
    },
    {
      "epoch": 2.84,
      "grad_norm": 0.12123954748173423,
      "learning_rate": 7.291125901946027e-07,
      "loss": 0.276,
      "step": 895
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.1193372635837929,
      "learning_rate": 6.880913724458538e-07,
      "loss": 0.2612,
      "step": 896
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.1143863028858849,
      "learning_rate": 6.482537898582885e-07,
      "loss": 0.2477,
      "step": 897
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.12787241782240982,
      "learning_rate": 6.096003172305742e-07,
      "loss": 0.2885,
      "step": 898
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.11481521229169728,
      "learning_rate": 5.721314152487556e-07,
      "loss": 0.2482,
      "step": 899
    },
    {
      "epoch": 2.86,
      "grad_norm": 0.13026291577401214,
      "learning_rate": 5.358475304807375e-07,
      "loss": 0.2709,
      "step": 900
    },
    {
      "epoch": 2.86,
      "grad_norm": 0.12344201915653148,
      "learning_rate": 5.007490953709226e-07,
      "loss": 0.2879,
      "step": 901
    },
    {
      "epoch": 2.86,
      "grad_norm": 0.1240036415461378,
      "learning_rate": 4.668365282351372e-07,
      "loss": 0.2732,
      "step": 902
    },
    {
      "epoch": 2.87,
      "grad_norm": 0.11613523598152595,
      "learning_rate": 4.341102332556024e-07,
      "loss": 0.2612,
      "step": 903
    },
    {
      "epoch": 2.87,
      "grad_norm": 0.12395979532480268,
      "learning_rate": 4.025706004760932e-07,
      "loss": 0.2784,
      "step": 904
    },
    {
      "epoch": 2.87,
      "grad_norm": 0.1262399891151353,
      "learning_rate": 3.7221800579735346e-07,
      "loss": 0.2794,
      "step": 905
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.11892782029439829,
      "learning_rate": 3.4305281097254393e-07,
      "loss": 0.2631,
      "step": 906
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.12463823510439939,
      "learning_rate": 3.1507536360299016e-07,
      "loss": 0.2766,
      "step": 907
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.11935642896867422,
      "learning_rate": 2.882859971339857e-07,
      "loss": 0.274,
      "step": 908
    },
    {
      "epoch": 2.89,
      "grad_norm": 0.11498686508134462,
      "learning_rate": 2.6268503085089547e-07,
      "loss": 0.2502,
      "step": 909
    },
    {
      "epoch": 2.89,
      "grad_norm": 0.12231215743234737,
      "learning_rate": 2.382727698752474e-07,
      "loss": 0.2438,
      "step": 910
    },
    {
      "epoch": 2.89,
      "grad_norm": 0.11327830672551155,
      "learning_rate": 2.1504950516118006e-07,
      "loss": 0.279,
      "step": 911
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.12622774204128723,
      "learning_rate": 1.9301551349195646e-07,
      "loss": 0.2774,
      "step": 912
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.11612765349173348,
      "learning_rate": 1.721710574766333e-07,
      "loss": 0.2714,
      "step": 913
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.1232154841443732,
      "learning_rate": 1.5251638554694136e-07,
      "loss": 0.2758,
      "step": 914
    },
    {
      "epoch": 2.91,
      "grad_norm": 0.11296948331776904,
      "learning_rate": 1.340517319543877e-07,
      "loss": 0.2585,
      "step": 915
    },
    {
      "epoch": 2.91,
      "grad_norm": 0.11627335434941227,
      "learning_rate": 1.1677731676733584e-07,
      "loss": 0.2558,
      "step": 916
    },
    {
      "epoch": 2.91,
      "grad_norm": 0.11047691213893575,
      "learning_rate": 1.0069334586854107e-07,
      "loss": 0.2405,
      "step": 917
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.13251790163054394,
      "learning_rate": 8.580001095253032e-08,
      "loss": 0.2465,
      "step": 918
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.10765278879866313,
      "learning_rate": 7.20974895234705e-08,
      "loss": 0.2464,
      "step": 919
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.11765995156594018,
      "learning_rate": 5.958594489295921e-08,
      "loss": 0.2539,
      "step": 920
    },
    {
      "epoch": 2.93,
      "grad_norm": 0.11822150544482929,
      "learning_rate": 4.8265526178070676e-08,
      "loss": 0.2662,
      "step": 921
    },
    {
      "epoch": 2.93,
      "grad_norm": 0.11462613998587723,
      "learning_rate": 3.813636829966827e-08,
      "loss": 0.248,
      "step": 922
    },
    {
      "epoch": 2.93,
      "grad_norm": 0.12454233728062138,
      "learning_rate": 2.9198591980705848e-08,
      "loss": 0.2658,
      "step": 923
    },
    {
      "epoch": 2.94,
      "grad_norm": 0.12035515326393312,
      "learning_rate": 2.145230374481777e-08,
      "loss": 0.2623,
      "step": 924
    },
    {
      "epoch": 2.94,
      "grad_norm": 0.11829501375571466,
      "learning_rate": 1.4897595915053242e-08,
      "loss": 0.2722,
      "step": 925
    },
    {
      "epoch": 2.94,
      "grad_norm": 0.11388392284548388,
      "learning_rate": 9.534546612810502e-09,
      "loss": 0.2664,
      "step": 926
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.12315953864328971,
      "learning_rate": 5.363219756837623e-09,
      "loss": 0.2474,
      "step": 927
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.11747013420761931,
      "learning_rate": 2.3836650624997624e-09,
      "loss": 0.2619,
      "step": 928
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.12055917384165643,
      "learning_rate": 5.959180412129506e-10,
      "loss": 0.2471,
      "step": 929
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.12657240487417656,
      "learning_rate": 0.0,
      "loss": 0.2499,
      "step": 930
    }
  ],
  "logging_steps": 1,
  "max_steps": 930,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 310,
  "total_flos": 5.097207143250002e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}