{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 13.215859030837004,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04405286343612335,
      "grad_norm": 32.503074645996094,
      "learning_rate": 0.0002992070484581498,
      "loss": 18.6527,
      "step": 10
    },
    {
      "epoch": 0.0881057268722467,
      "grad_norm": 14.659343719482422,
      "learning_rate": 0.0002983259911894273,
      "loss": 6.941,
      "step": 20
    },
    {
      "epoch": 0.13215859030837004,
      "grad_norm": 13.076040267944336,
      "learning_rate": 0.00029744493392070483,
      "loss": 4.3562,
      "step": 30
    },
    {
      "epoch": 0.1762114537444934,
      "grad_norm": 21.742950439453125,
      "learning_rate": 0.00029656387665198236,
      "loss": 3.5182,
      "step": 40
    },
    {
      "epoch": 0.22026431718061673,
      "grad_norm": 19.481538772583008,
      "learning_rate": 0.0002956828193832599,
      "loss": 2.6543,
      "step": 50
    },
    {
      "epoch": 0.2643171806167401,
      "grad_norm": 8.22918701171875,
      "learning_rate": 0.0002948017621145374,
      "loss": 1.7678,
      "step": 60
    },
    {
      "epoch": 0.30837004405286345,
      "grad_norm": 5.820700168609619,
      "learning_rate": 0.00029392070484581494,
      "loss": 1.2462,
      "step": 70
    },
    {
      "epoch": 0.3524229074889868,
      "grad_norm": 2.173941135406494,
      "learning_rate": 0.00029303964757709247,
      "loss": 1.3665,
      "step": 80
    },
    {
      "epoch": 0.3964757709251101,
      "grad_norm": 1.8419830799102783,
      "learning_rate": 0.00029215859030837,
      "loss": 1.1257,
      "step": 90
    },
    {
      "epoch": 0.44052863436123346,
      "grad_norm": 1.7700544595718384,
      "learning_rate": 0.0002912775330396475,
      "loss": 1.0317,
      "step": 100
    },
    {
      "epoch": 0.4845814977973568,
      "grad_norm": 1.2313016653060913,
      "learning_rate": 0.0002903964757709251,
      "loss": 0.8453,
      "step": 110
    },
    {
      "epoch": 0.5286343612334802,
      "grad_norm": 1.3880990743637085,
      "learning_rate": 0.00028951541850220263,
      "loss": 0.8675,
      "step": 120
    },
    {
      "epoch": 0.5726872246696035,
      "grad_norm": 2.3586199283599854,
      "learning_rate": 0.00028863436123348016,
      "loss": 0.8509,
      "step": 130
    },
    {
      "epoch": 0.6167400881057269,
      "grad_norm": 1.3902242183685303,
      "learning_rate": 0.0002877533039647577,
      "loss": 0.8528,
      "step": 140
    },
    {
      "epoch": 0.6607929515418502,
      "grad_norm": 1.0430246591567993,
      "learning_rate": 0.0002868722466960352,
      "loss": 0.8726,
      "step": 150
    },
    {
      "epoch": 0.7048458149779736,
      "grad_norm": 1.3441425561904907,
      "learning_rate": 0.00028599118942731274,
      "loss": 1.1029,
      "step": 160
    },
    {
      "epoch": 0.748898678414097,
      "grad_norm": 1.18771231174469,
      "learning_rate": 0.00028511013215859026,
      "loss": 0.9229,
      "step": 170
    },
    {
      "epoch": 0.7929515418502202,
      "grad_norm": 0.8050010800361633,
      "learning_rate": 0.0002842290748898678,
      "loss": 0.8363,
      "step": 180
    },
    {
      "epoch": 0.8370044052863436,
      "grad_norm": 1.1800554990768433,
      "learning_rate": 0.0002833480176211453,
      "loss": 0.8128,
      "step": 190
    },
    {
      "epoch": 0.8810572687224669,
      "grad_norm": 1.3873122930526733,
      "learning_rate": 0.00028246696035242285,
      "loss": 0.8275,
      "step": 200
    },
    {
      "epoch": 0.9251101321585903,
      "grad_norm": 1.1561434268951416,
      "learning_rate": 0.0002815859030837004,
      "loss": 0.8961,
      "step": 210
    },
    {
      "epoch": 0.9691629955947136,
      "grad_norm": 0.6641005277633667,
      "learning_rate": 0.00028070484581497795,
      "loss": 0.6599,
      "step": 220
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6834425330162048,
      "eval_runtime": 5.6075,
      "eval_samples_per_second": 36.023,
      "eval_steps_per_second": 4.637,
      "step": 227
    },
    {
      "epoch": 1.013215859030837,
      "grad_norm": 1.0736805200576782,
      "learning_rate": 0.0002798237885462555,
      "loss": 0.7283,
      "step": 230
    },
    {
      "epoch": 1.0572687224669604,
      "grad_norm": 1.069932460784912,
      "learning_rate": 0.000278942731277533,
      "loss": 0.825,
      "step": 240
    },
    {
      "epoch": 1.1013215859030836,
      "grad_norm": 1.3098456859588623,
      "learning_rate": 0.0002780616740088106,
      "loss": 0.7152,
      "step": 250
    },
    {
      "epoch": 1.145374449339207,
      "grad_norm": 1.0547797679901123,
      "learning_rate": 0.0002771806167400881,
      "loss": 0.746,
      "step": 260
    },
    {
      "epoch": 1.1894273127753303,
      "grad_norm": 1.58526611328125,
      "learning_rate": 0.00027629955947136564,
      "loss": 0.6655,
      "step": 270
    },
    {
      "epoch": 1.2334801762114538,
      "grad_norm": 1.4090569019317627,
      "learning_rate": 0.0002754185022026431,
      "loss": 0.7397,
      "step": 280
    },
    {
      "epoch": 1.277533039647577,
      "grad_norm": 1.3417810201644897,
      "learning_rate": 0.00027453744493392064,
      "loss": 0.6534,
      "step": 290
    },
    {
      "epoch": 1.3215859030837005,
      "grad_norm": 0.7320300936698914,
      "learning_rate": 0.0002736563876651982,
      "loss": 0.719,
      "step": 300
    },
    {
      "epoch": 1.3656387665198237,
      "grad_norm": 1.7867811918258667,
      "learning_rate": 0.00027277533039647575,
      "loss": 0.8878,
      "step": 310
    },
    {
      "epoch": 1.4096916299559472,
      "grad_norm": 1.0332417488098145,
      "learning_rate": 0.0002718942731277533,
      "loss": 0.7058,
      "step": 320
    },
    {
      "epoch": 1.4537444933920705,
      "grad_norm": 1.1942096948623657,
      "learning_rate": 0.0002710132158590308,
      "loss": 0.6599,
      "step": 330
    },
    {
      "epoch": 1.497797356828194,
      "grad_norm": 1.0352708101272583,
      "learning_rate": 0.00027013215859030833,
      "loss": 0.6487,
      "step": 340
    },
    {
      "epoch": 1.5418502202643172,
      "grad_norm": 1.2984694242477417,
      "learning_rate": 0.0002692511013215859,
      "loss": 0.7225,
      "step": 350
    },
    {
      "epoch": 1.5859030837004404,
      "grad_norm": 1.1419997215270996,
      "learning_rate": 0.00026837004405286344,
      "loss": 0.7233,
      "step": 360
    },
    {
      "epoch": 1.6299559471365639,
      "grad_norm": 1.5873011350631714,
      "learning_rate": 0.00026748898678414097,
      "loss": 0.5429,
      "step": 370
    },
    {
      "epoch": 1.6740088105726874,
      "grad_norm": 1.058026671409607,
      "learning_rate": 0.0002666079295154185,
      "loss": 0.7621,
      "step": 380
    },
    {
      "epoch": 1.7180616740088106,
      "grad_norm": 1.424886703491211,
      "learning_rate": 0.000265726872246696,
      "loss": 0.7103,
      "step": 390
    },
    {
      "epoch": 1.7621145374449338,
      "grad_norm": 0.9987337589263916,
      "learning_rate": 0.00026484581497797355,
      "loss": 0.6882,
      "step": 400
    },
    {
      "epoch": 1.8061674008810573,
      "grad_norm": 1.0241808891296387,
      "learning_rate": 0.0002639647577092511,
      "loss": 0.6754,
      "step": 410
    },
    {
      "epoch": 1.8502202643171806,
      "grad_norm": 0.7069824934005737,
      "learning_rate": 0.0002630837004405286,
      "loss": 0.6426,
      "step": 420
    },
    {
      "epoch": 1.894273127753304,
      "grad_norm": 1.18909752368927,
      "learning_rate": 0.00026220264317180613,
      "loss": 0.7879,
      "step": 430
    },
    {
      "epoch": 1.9383259911894273,
      "grad_norm": 0.8950007557868958,
      "learning_rate": 0.00026132158590308366,
      "loss": 0.7197,
      "step": 440
    },
    {
      "epoch": 1.9823788546255505,
      "grad_norm": 1.3497512340545654,
      "learning_rate": 0.00026044052863436124,
      "loss": 0.6892,
      "step": 450
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6154947876930237,
      "eval_runtime": 5.3228,
      "eval_samples_per_second": 37.95,
      "eval_steps_per_second": 4.885,
      "step": 454
    },
    {
      "epoch": 2.026431718061674,
      "grad_norm": 1.0325676202774048,
      "learning_rate": 0.00025955947136563877,
      "loss": 0.7104,
      "step": 460
    },
    {
      "epoch": 2.0704845814977975,
      "grad_norm": 1.4701021909713745,
      "learning_rate": 0.0002586784140969163,
      "loss": 0.6249,
      "step": 470
    },
    {
      "epoch": 2.1145374449339207,
      "grad_norm": 1.2472504377365112,
      "learning_rate": 0.0002577973568281938,
      "loss": 0.7115,
      "step": 480
    },
    {
      "epoch": 2.158590308370044,
      "grad_norm": 1.01516592502594,
      "learning_rate": 0.00025691629955947135,
      "loss": 0.6039,
      "step": 490
    },
    {
      "epoch": 2.202643171806167,
      "grad_norm": 1.3985668420791626,
      "learning_rate": 0.0002560352422907489,
      "loss": 0.5976,
      "step": 500
    },
    {
      "epoch": 2.246696035242291,
      "grad_norm": 0.6047684550285339,
      "learning_rate": 0.0002551541850220264,
      "loss": 0.5158,
      "step": 510
    },
    {
      "epoch": 2.290748898678414,
      "grad_norm": 0.8428493142127991,
      "learning_rate": 0.00025427312775330393,
      "loss": 0.6338,
      "step": 520
    },
    {
      "epoch": 2.3348017621145374,
      "grad_norm": 1.0199517011642456,
      "learning_rate": 0.00025339207048458146,
      "loss": 0.469,
      "step": 530
    },
    {
      "epoch": 2.3788546255506606,
      "grad_norm": 0.8641414642333984,
      "learning_rate": 0.000252511013215859,
      "loss": 0.583,
      "step": 540
    },
    {
      "epoch": 2.4229074889867843,
      "grad_norm": 0.8442863821983337,
      "learning_rate": 0.00025162995594713657,
      "loss": 0.6108,
      "step": 550
    },
    {
      "epoch": 2.4669603524229076,
      "grad_norm": 0.8864941000938416,
      "learning_rate": 0.0002507488986784141,
      "loss": 0.5572,
      "step": 560
    },
    {
      "epoch": 2.511013215859031,
      "grad_norm": 0.9025411605834961,
      "learning_rate": 0.0002498678414096916,
      "loss": 0.6174,
      "step": 570
    },
    {
      "epoch": 2.555066079295154,
      "grad_norm": 0.8481220602989197,
      "learning_rate": 0.00024898678414096915,
      "loss": 0.6118,
      "step": 580
    },
    {
      "epoch": 2.5991189427312777,
      "grad_norm": 0.9391738772392273,
      "learning_rate": 0.0002481057268722467,
      "loss": 0.569,
      "step": 590
    },
    {
      "epoch": 2.643171806167401,
      "grad_norm": 1.0381453037261963,
      "learning_rate": 0.0002472246696035242,
      "loss": 0.4904,
      "step": 600
    },
    {
      "epoch": 2.6872246696035242,
      "grad_norm": 1.023573398590088,
      "learning_rate": 0.00024634361233480173,
      "loss": 0.5969,
      "step": 610
    },
    {
      "epoch": 2.7312775330396475,
      "grad_norm": 1.2105042934417725,
      "learning_rate": 0.00024546255506607926,
      "loss": 0.6589,
      "step": 620
    },
    {
      "epoch": 2.7753303964757707,
      "grad_norm": 1.1160320043563843,
      "learning_rate": 0.0002445814977973568,
      "loss": 0.5242,
      "step": 630
    },
    {
      "epoch": 2.8193832599118944,
      "grad_norm": 1.1934391260147095,
      "learning_rate": 0.00024370044052863436,
      "loss": 0.6548,
      "step": 640
    },
    {
      "epoch": 2.8634361233480177,
      "grad_norm": 1.1788102388381958,
      "learning_rate": 0.0002428193832599119,
      "loss": 0.5828,
      "step": 650
    },
    {
      "epoch": 2.907488986784141,
      "grad_norm": 1.2889748811721802,
      "learning_rate": 0.00024193832599118942,
      "loss": 0.6468,
      "step": 660
    },
    {
      "epoch": 2.951541850220264,
      "grad_norm": 0.8837119340896606,
      "learning_rate": 0.00024105726872246695,
      "loss": 0.6694,
      "step": 670
    },
    {
      "epoch": 2.995594713656388,
      "grad_norm": 1.0216575860977173,
      "learning_rate": 0.00024017621145374447,
      "loss": 0.633,
      "step": 680
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.592506468296051,
      "eval_runtime": 5.3852,
      "eval_samples_per_second": 37.51,
      "eval_steps_per_second": 4.828,
      "step": 681
    },
    {
      "epoch": 3.039647577092511,
      "grad_norm": 1.58785080909729,
      "learning_rate": 0.00023929515418502203,
      "loss": 0.4512,
      "step": 690
    },
    {
      "epoch": 3.0837004405286343,
      "grad_norm": 1.0036600828170776,
      "learning_rate": 0.00023841409691629955,
      "loss": 0.9613,
      "step": 700
    },
    {
      "epoch": 3.1277533039647576,
      "grad_norm": 0.9956134557723999,
      "learning_rate": 0.00023753303964757708,
      "loss": 0.479,
      "step": 710
    },
    {
      "epoch": 3.171806167400881,
      "grad_norm": 1.1154946088790894,
      "learning_rate": 0.0002366519823788546,
      "loss": 0.5444,
      "step": 720
    },
    {
      "epoch": 3.2158590308370045,
      "grad_norm": 1.3544610738754272,
      "learning_rate": 0.00023577092511013214,
      "loss": 0.5163,
      "step": 730
    },
    {
      "epoch": 3.2599118942731278,
      "grad_norm": 0.7720727920532227,
      "learning_rate": 0.0002348898678414097,
      "loss": 0.5317,
      "step": 740
    },
    {
      "epoch": 3.303964757709251,
      "grad_norm": 0.9804306030273438,
      "learning_rate": 0.00023400881057268722,
      "loss": 0.5179,
      "step": 750
    },
    {
      "epoch": 3.3480176211453743,
      "grad_norm": 1.0230934619903564,
      "learning_rate": 0.00023312775330396474,
      "loss": 0.5261,
      "step": 760
    },
    {
      "epoch": 3.392070484581498,
      "grad_norm": 0.8620821237564087,
      "learning_rate": 0.00023224669603524227,
      "loss": 0.4998,
      "step": 770
    },
    {
      "epoch": 3.436123348017621,
      "grad_norm": 0.8884461522102356,
      "learning_rate": 0.0002313656387665198,
      "loss": 0.5527,
      "step": 780
    },
    {
      "epoch": 3.4801762114537445,
      "grad_norm": 0.7721192836761475,
      "learning_rate": 0.00023048458149779735,
      "loss": 0.5279,
      "step": 790
    },
    {
      "epoch": 3.5242290748898677,
      "grad_norm": 1.0769802331924438,
      "learning_rate": 0.00022960352422907488,
      "loss": 0.5851,
      "step": 800
    },
    {
      "epoch": 3.568281938325991,
      "grad_norm": 1.3999199867248535,
      "learning_rate": 0.0002287224669603524,
      "loss": 0.44,
      "step": 810
    },
    {
      "epoch": 3.6123348017621146,
      "grad_norm": 0.9963156580924988,
      "learning_rate": 0.00022784140969162993,
      "loss": 0.6028,
      "step": 820
    },
    {
      "epoch": 3.656387665198238,
      "grad_norm": 0.9077759981155396,
      "learning_rate": 0.00022696035242290746,
      "loss": 0.6824,
      "step": 830
    },
    {
      "epoch": 3.700440528634361,
      "grad_norm": 0.9758647680282593,
      "learning_rate": 0.00022607929515418502,
      "loss": 0.5424,
      "step": 840
    },
    {
      "epoch": 3.744493392070485,
      "grad_norm": 0.9838646054267883,
      "learning_rate": 0.00022519823788546254,
      "loss": 0.5588,
      "step": 850
    },
    {
      "epoch": 3.788546255506608,
      "grad_norm": 1.1924773454666138,
      "learning_rate": 0.00022431718061674007,
      "loss": 0.6215,
      "step": 860
    },
    {
      "epoch": 3.8325991189427313,
      "grad_norm": 1.27988600730896,
      "learning_rate": 0.0002234361233480176,
      "loss": 0.5336,
      "step": 870
    },
    {
      "epoch": 3.8766519823788546,
      "grad_norm": 1.0098719596862793,
      "learning_rate": 0.00022255506607929512,
      "loss": 0.6623,
      "step": 880
    },
    {
      "epoch": 3.920704845814978,
      "grad_norm": 1.301437497138977,
      "learning_rate": 0.00022167400881057268,
      "loss": 0.4837,
      "step": 890
    },
    {
      "epoch": 3.964757709251101,
      "grad_norm": 1.3062794208526611,
      "learning_rate": 0.0002207929515418502,
      "loss": 0.4399,
      "step": 900
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5791140198707581,
      "eval_runtime": 5.2768,
      "eval_samples_per_second": 38.281,
      "eval_steps_per_second": 4.927,
      "step": 908
    },
    {
      "epoch": 4.008810572687224,
      "grad_norm": 1.2243441343307495,
      "learning_rate": 0.00021991189427312773,
      "loss": 0.5225,
      "step": 910
    },
    {
      "epoch": 4.052863436123348,
      "grad_norm": 1.0874862670898438,
      "learning_rate": 0.00021903083700440526,
      "loss": 0.5878,
      "step": 920
    },
    {
      "epoch": 4.096916299559472,
      "grad_norm": 1.1561787128448486,
      "learning_rate": 0.0002181497797356828,
      "loss": 0.4172,
      "step": 930
    },
    {
      "epoch": 4.140969162995595,
      "grad_norm": 0.9504215121269226,
      "learning_rate": 0.00021726872246696034,
      "loss": 0.454,
      "step": 940
    },
    {
      "epoch": 4.185022026431718,
      "grad_norm": 1.0901755094528198,
      "learning_rate": 0.00021638766519823787,
      "loss": 0.5222,
      "step": 950
    },
    {
      "epoch": 4.229074889867841,
      "grad_norm": 0.7518570423126221,
      "learning_rate": 0.0002155066079295154,
      "loss": 0.4048,
      "step": 960
    },
    {
      "epoch": 4.273127753303965,
      "grad_norm": 0.9933887720108032,
      "learning_rate": 0.00021462555066079292,
      "loss": 0.4545,
      "step": 970
    },
    {
      "epoch": 4.317180616740088,
      "grad_norm": 0.8956694006919861,
      "learning_rate": 0.00021374449339207048,
      "loss": 0.5703,
      "step": 980
    },
    {
      "epoch": 4.361233480176211,
      "grad_norm": 1.0768828392028809,
      "learning_rate": 0.000212863436123348,
      "loss": 0.411,
      "step": 990
    },
    {
      "epoch": 4.405286343612334,
      "grad_norm": 1.3219349384307861,
      "learning_rate": 0.00021198237885462553,
      "loss": 0.5096,
      "step": 1000
    },
    {
      "epoch": 4.4493392070484585,
      "grad_norm": 0.6028145551681519,
      "learning_rate": 0.00021110132158590306,
      "loss": 0.5427,
      "step": 1010
    },
    {
      "epoch": 4.493392070484582,
      "grad_norm": 0.6015641689300537,
      "learning_rate": 0.00021022026431718059,
      "loss": 0.4855,
      "step": 1020
    },
    {
      "epoch": 4.537444933920705,
      "grad_norm": 0.7184689044952393,
      "learning_rate": 0.00020933920704845814,
      "loss": 0.4893,
      "step": 1030
    },
    {
      "epoch": 4.581497797356828,
      "grad_norm": 1.445830225944519,
      "learning_rate": 0.00020845814977973567,
      "loss": 0.4412,
      "step": 1040
    },
    {
      "epoch": 4.6255506607929515,
      "grad_norm": 0.9506711959838867,
      "learning_rate": 0.0002075770925110132,
      "loss": 0.5596,
      "step": 1050
    },
    {
      "epoch": 4.669603524229075,
      "grad_norm": 0.9642265439033508,
      "learning_rate": 0.00020669603524229072,
      "loss": 0.3944,
      "step": 1060
    },
    {
      "epoch": 4.713656387665198,
      "grad_norm": 0.9548330307006836,
      "learning_rate": 0.00020581497797356825,
      "loss": 0.4925,
      "step": 1070
    },
    {
      "epoch": 4.757709251101321,
      "grad_norm": 1.5850030183792114,
      "learning_rate": 0.0002049339207048458,
      "loss": 0.5114,
      "step": 1080
    },
    {
      "epoch": 4.8017621145374445,
      "grad_norm": 0.7429970502853394,
      "learning_rate": 0.00020405286343612333,
      "loss": 0.5556,
      "step": 1090
    },
    {
      "epoch": 4.845814977973569,
      "grad_norm": 0.9865929484367371,
      "learning_rate": 0.00020317180616740086,
      "loss": 0.4612,
      "step": 1100
    },
    {
      "epoch": 4.889867841409692,
      "grad_norm": 0.8113177418708801,
      "learning_rate": 0.00020229074889867838,
      "loss": 0.5196,
      "step": 1110
    },
    {
      "epoch": 4.933920704845815,
      "grad_norm": 1.1767125129699707,
      "learning_rate": 0.0002014096916299559,
      "loss": 0.5321,
      "step": 1120
    },
    {
      "epoch": 4.977973568281938,
      "grad_norm": 0.8367587327957153,
      "learning_rate": 0.00020052863436123347,
      "loss": 0.506,
      "step": 1130
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.5730367302894592,
      "eval_runtime": 6.1337,
      "eval_samples_per_second": 32.933,
      "eval_steps_per_second": 4.239,
      "step": 1135
    },
    {
      "epoch": 5.022026431718062,
      "grad_norm": 1.1242823600769043,
      "learning_rate": 0.000199647577092511,
      "loss": 0.4678,
      "step": 1140
    },
    {
      "epoch": 5.066079295154185,
      "grad_norm": 1.0385881662368774,
      "learning_rate": 0.00019876651982378852,
      "loss": 0.3968,
      "step": 1150
    },
    {
      "epoch": 5.110132158590308,
      "grad_norm": 0.9282165765762329,
      "learning_rate": 0.00019788546255506605,
      "loss": 0.5089,
      "step": 1160
    },
    {
      "epoch": 5.154185022026431,
      "grad_norm": 1.401548147201538,
      "learning_rate": 0.00019700440528634357,
      "loss": 0.4457,
      "step": 1170
    },
    {
      "epoch": 5.1982378854625555,
      "grad_norm": 0.6676862835884094,
      "learning_rate": 0.00019612334801762113,
      "loss": 0.3175,
      "step": 1180
    },
    {
      "epoch": 5.242290748898679,
      "grad_norm": 1.1318411827087402,
      "learning_rate": 0.00019524229074889866,
      "loss": 0.4468,
      "step": 1190
    },
    {
      "epoch": 5.286343612334802,
      "grad_norm": 0.706200361251831,
      "learning_rate": 0.00019436123348017618,
      "loss": 0.3954,
      "step": 1200
    },
    {
      "epoch": 5.330396475770925,
      "grad_norm": 0.6558952927589417,
      "learning_rate": 0.0001934801762114537,
      "loss": 0.4318,
      "step": 1210
    },
    {
      "epoch": 5.3744493392070485,
      "grad_norm": 0.59174644947052,
      "learning_rate": 0.00019259911894273124,
      "loss": 0.3962,
      "step": 1220
    },
    {
      "epoch": 5.418502202643172,
      "grad_norm": 0.9306423664093018,
      "learning_rate": 0.0001917180616740088,
      "loss": 0.4161,
      "step": 1230
    },
    {
      "epoch": 5.462555066079295,
      "grad_norm": 1.2412904500961304,
      "learning_rate": 0.00019083700440528632,
      "loss": 0.4259,
      "step": 1240
    },
    {
      "epoch": 5.506607929515418,
      "grad_norm": 0.8949795961380005,
      "learning_rate": 0.00018995594713656385,
      "loss": 0.5512,
      "step": 1250
    },
    {
      "epoch": 5.5506607929515415,
      "grad_norm": 0.9977787733078003,
      "learning_rate": 0.00018907488986784137,
      "loss": 0.4497,
      "step": 1260
    },
    {
      "epoch": 5.594713656387665,
      "grad_norm": 1.0676085948944092,
      "learning_rate": 0.0001881938325991189,
      "loss": 0.4344,
      "step": 1270
    },
    {
      "epoch": 5.638766519823789,
      "grad_norm": 0.6446275115013123,
      "learning_rate": 0.00018731277533039648,
      "loss": 0.4184,
      "step": 1280
    },
    {
      "epoch": 5.682819383259912,
      "grad_norm": 1.3255438804626465,
      "learning_rate": 0.000186431718061674,
      "loss": 0.5441,
      "step": 1290
    },
    {
      "epoch": 5.726872246696035,
      "grad_norm": 0.823581337928772,
      "learning_rate": 0.0001855506607929515,
      "loss": 0.5028,
      "step": 1300
    },
    {
      "epoch": 5.770925110132159,
      "grad_norm": 1.0471981763839722,
      "learning_rate": 0.00018466960352422904,
      "loss": 0.4407,
      "step": 1310
    },
    {
      "epoch": 5.814977973568282,
      "grad_norm": 1.0394315719604492,
      "learning_rate": 0.00018378854625550662,
      "loss": 0.4788,
      "step": 1320
    },
    {
      "epoch": 5.859030837004405,
      "grad_norm": 1.4738258123397827,
      "learning_rate": 0.00018290748898678414,
      "loss": 0.5495,
      "step": 1330
    },
    {
      "epoch": 5.903083700440528,
      "grad_norm": 1.2812182903289795,
      "learning_rate": 0.00018202643171806167,
      "loss": 0.4335,
      "step": 1340
    },
    {
      "epoch": 5.9471365638766525,
      "grad_norm": 1.4929533004760742,
      "learning_rate": 0.0001811453744493392,
      "loss": 0.5097,
      "step": 1350
    },
    {
      "epoch": 5.991189427312776,
      "grad_norm": 1.2788587808609009,
      "learning_rate": 0.00018026431718061673,
      "loss": 0.4702,
      "step": 1360
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.5740869045257568,
      "eval_runtime": 4.9653,
      "eval_samples_per_second": 40.682,
      "eval_steps_per_second": 5.236,
      "step": 1362
    },
    {
      "epoch": 6.035242290748899,
      "grad_norm": 0.9543855786323547,
      "learning_rate": 0.00017938325991189428,
      "loss": 0.4232,
      "step": 1370
    },
    {
      "epoch": 6.079295154185022,
      "grad_norm": 1.0528812408447266,
      "learning_rate": 0.0001785022026431718,
      "loss": 0.4025,
      "step": 1380
    },
    {
      "epoch": 6.1233480176211454,
      "grad_norm": 0.9573265910148621,
      "learning_rate": 0.00017762114537444933,
      "loss": 0.4127,
      "step": 1390
    },
    {
      "epoch": 6.167400881057269,
      "grad_norm": 1.7806532382965088,
      "learning_rate": 0.00017674008810572686,
      "loss": 0.4646,
      "step": 1400
    },
    {
      "epoch": 6.211453744493392,
      "grad_norm": 1.0559179782867432,
      "learning_rate": 0.0001758590308370044,
      "loss": 0.3222,
      "step": 1410
    },
    {
      "epoch": 6.255506607929515,
      "grad_norm": 0.9502829313278198,
      "learning_rate": 0.00017497797356828194,
      "loss": 0.4697,
      "step": 1420
    },
    {
      "epoch": 6.299559471365638,
      "grad_norm": 0.6869007349014282,
      "learning_rate": 0.00017409691629955947,
      "loss": 0.4155,
      "step": 1430
    },
    {
      "epoch": 6.343612334801762,
      "grad_norm": 0.6793345808982849,
      "learning_rate": 0.000173215859030837,
      "loss": 0.4236,
      "step": 1440
    },
    {
      "epoch": 6.387665198237886,
      "grad_norm": 1.067975640296936,
      "learning_rate": 0.00017233480176211452,
      "loss": 0.3558,
      "step": 1450
    },
    {
      "epoch": 6.431718061674009,
      "grad_norm": 1.0968421697616577,
      "learning_rate": 0.00017145374449339205,
      "loss": 0.4453,
      "step": 1460
    },
    {
      "epoch": 6.475770925110132,
      "grad_norm": 1.1832313537597656,
      "learning_rate": 0.0001705726872246696,
      "loss": 0.5115,
      "step": 1470
    },
    {
      "epoch": 6.5198237885462555,
      "grad_norm": 0.9857836365699768,
      "learning_rate": 0.00016969162995594713,
      "loss": 0.4274,
      "step": 1480
    },
    {
      "epoch": 6.563876651982379,
      "grad_norm": 0.9006336331367493,
      "learning_rate": 0.00016881057268722466,
      "loss": 0.3865,
      "step": 1490
    },
    {
      "epoch": 6.607929515418502,
      "grad_norm": 1.1091986894607544,
      "learning_rate": 0.0001679295154185022,
      "loss": 0.3988,
      "step": 1500
    },
    {
      "epoch": 6.651982378854625,
      "grad_norm": 1.423886775970459,
      "learning_rate": 0.00016704845814977971,
      "loss": 0.5057,
      "step": 1510
    },
    {
      "epoch": 6.6960352422907485,
      "grad_norm": 0.9245197176933289,
      "learning_rate": 0.00016616740088105727,
      "loss": 0.3966,
      "step": 1520
    },
    {
      "epoch": 6.740088105726873,
      "grad_norm": 0.944870114326477,
      "learning_rate": 0.0001652863436123348,
      "loss": 0.4521,
      "step": 1530
    },
    {
      "epoch": 6.784140969162996,
      "grad_norm": 0.8870773315429688,
      "learning_rate": 0.00016440528634361232,
      "loss": 0.4425,
      "step": 1540
    },
    {
      "epoch": 6.828193832599119,
      "grad_norm": 0.7404115200042725,
      "learning_rate": 0.00016352422907488985,
      "loss": 0.3207,
      "step": 1550
    },
    {
      "epoch": 6.872246696035242,
      "grad_norm": 0.9958137273788452,
      "learning_rate": 0.00016264317180616738,
      "loss": 0.4244,
      "step": 1560
    },
    {
      "epoch": 6.916299559471366,
      "grad_norm": 1.0651079416275024,
      "learning_rate": 0.00016176211453744493,
      "loss": 0.4075,
      "step": 1570
    },
    {
      "epoch": 6.960352422907489,
      "grad_norm": 0.9528789520263672,
      "learning_rate": 0.00016088105726872246,
      "loss": 0.493,
      "step": 1580
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.5702072381973267,
      "eval_runtime": 5.1787,
      "eval_samples_per_second": 39.006,
      "eval_steps_per_second": 5.021,
      "step": 1589
    },
    {
      "epoch": 7.004405286343612,
      "grad_norm": 1.0486853122711182,
      "learning_rate": 0.00015999999999999999,
      "loss": 0.4131,
      "step": 1590
    },
    {
      "epoch": 7.048458149779735,
      "grad_norm": 1.2176262140274048,
      "learning_rate": 0.0001591189427312775,
      "loss": 0.4417,
      "step": 1600
    },
    {
      "epoch": 7.092511013215859,
      "grad_norm": 1.187107801437378,
      "learning_rate": 0.00015823788546255504,
      "loss": 0.4372,
      "step": 1610
    },
    {
      "epoch": 7.136563876651983,
      "grad_norm": 0.9459372758865356,
      "learning_rate": 0.0001573568281938326,
      "loss": 0.356,
      "step": 1620
    },
    {
      "epoch": 7.180616740088106,
      "grad_norm": 0.8114103078842163,
      "learning_rate": 0.00015647577092511012,
      "loss": 0.308,
      "step": 1630
    },
    {
      "epoch": 7.224669603524229,
      "grad_norm": 1.035370945930481,
      "learning_rate": 0.00015559471365638765,
      "loss": 0.3738,
      "step": 1640
    },
    {
      "epoch": 7.2687224669603525,
      "grad_norm": 1.0260848999023438,
      "learning_rate": 0.00015471365638766518,
      "loss": 0.342,
      "step": 1650
    },
    {
      "epoch": 7.312775330396476,
      "grad_norm": 0.8079932928085327,
      "learning_rate": 0.00015383259911894273,
      "loss": 0.4381,
      "step": 1660
    },
    {
      "epoch": 7.356828193832599,
      "grad_norm": 1.318695068359375,
      "learning_rate": 0.00015295154185022026,
      "loss": 0.3685,
      "step": 1670
    },
    {
      "epoch": 7.400881057268722,
      "grad_norm": 1.3181859254837036,
      "learning_rate": 0.00015207048458149778,
      "loss": 0.3465,
      "step": 1680
    },
    {
      "epoch": 7.4449339207048455,
      "grad_norm": 1.0277948379516602,
      "learning_rate": 0.0001511894273127753,
      "loss": 0.3659,
      "step": 1690
    },
    {
      "epoch": 7.48898678414097,
      "grad_norm": 1.1619762182235718,
      "learning_rate": 0.00015030837004405284,
      "loss": 0.4304,
      "step": 1700
    },
    {
      "epoch": 7.533039647577093,
      "grad_norm": 1.2854048013687134,
      "learning_rate": 0.0001494273127753304,
      "loss": 0.4372,
      "step": 1710
    },
    {
      "epoch": 7.577092511013216,
      "grad_norm": 1.032459020614624,
      "learning_rate": 0.00014854625550660792,
      "loss": 0.3687,
      "step": 1720
    },
    {
      "epoch": 7.621145374449339,
      "grad_norm": 0.9430228471755981,
      "learning_rate": 0.00014766519823788545,
      "loss": 0.3967,
      "step": 1730
    },
    {
      "epoch": 7.665198237885463,
      "grad_norm": 1.2012503147125244,
      "learning_rate": 0.00014678414096916297,
      "loss": 0.4028,
      "step": 1740
    },
    {
      "epoch": 7.709251101321586,
      "grad_norm": 0.9703013896942139,
      "learning_rate": 0.00014590308370044053,
      "loss": 0.4037,
      "step": 1750
    },
    {
      "epoch": 7.753303964757709,
      "grad_norm": 1.2811229228973389,
      "learning_rate": 0.00014502202643171806,
      "loss": 0.3725,
      "step": 1760
    },
    {
      "epoch": 7.797356828193832,
      "grad_norm": 0.9879553914070129,
      "learning_rate": 0.00014414096916299558,
      "loss": 0.4385,
      "step": 1770
    },
    {
      "epoch": 7.841409691629956,
      "grad_norm": 1.4015151262283325,
      "learning_rate": 0.0001432599118942731,
      "loss": 0.4046,
      "step": 1780
    },
    {
      "epoch": 7.885462555066079,
      "grad_norm": 0.9369928240776062,
      "learning_rate": 0.00014237885462555064,
      "loss": 0.4232,
      "step": 1790
    },
    {
      "epoch": 7.929515418502203,
      "grad_norm": 0.7787442803382874,
      "learning_rate": 0.0001414977973568282,
      "loss": 0.3679,
      "step": 1800
    },
    {
      "epoch": 7.973568281938326,
      "grad_norm": 0.7212619781494141,
      "learning_rate": 0.00014061674008810572,
      "loss": 0.4299,
      "step": 1810
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.57987380027771,
      "eval_runtime": 4.8534,
      "eval_samples_per_second": 41.62,
      "eval_steps_per_second": 5.357,
      "step": 1816
    },
    {
      "epoch": 8.017621145374449,
      "grad_norm": 1.1815301179885864,
      "learning_rate": 0.00013973568281938325,
      "loss": 0.4058,
      "step": 1820
    },
    {
      "epoch": 8.061674008810572,
      "grad_norm": 0.7913572192192078,
      "learning_rate": 0.00013885462555066077,
      "loss": 0.2876,
      "step": 1830
    },
    {
      "epoch": 8.105726872246697,
      "grad_norm": 0.9591747522354126,
      "learning_rate": 0.0001379735682819383,
      "loss": 0.2801,
      "step": 1840
    },
    {
      "epoch": 8.14977973568282,
      "grad_norm": 1.2883862257003784,
      "learning_rate": 0.00013709251101321585,
      "loss": 0.3435,
      "step": 1850
    },
    {
      "epoch": 8.193832599118943,
      "grad_norm": 1.2138097286224365,
      "learning_rate": 0.00013621145374449338,
      "loss": 0.4603,
      "step": 1860
    },
    {
      "epoch": 8.237885462555067,
      "grad_norm": 0.9017927050590515,
      "learning_rate": 0.0001353303964757709,
      "loss": 0.328,
      "step": 1870
    },
    {
      "epoch": 8.28193832599119,
      "grad_norm": 1.0213032960891724,
      "learning_rate": 0.00013444933920704844,
      "loss": 0.4241,
      "step": 1880
    },
    {
      "epoch": 8.325991189427313,
      "grad_norm": 0.782507598400116,
      "learning_rate": 0.00013356828193832596,
      "loss": 0.287,
      "step": 1890
    },
    {
      "epoch": 8.370044052863436,
      "grad_norm": 0.8239027261734009,
      "learning_rate": 0.00013268722466960352,
      "loss": 0.3471,
      "step": 1900
    },
    {
      "epoch": 8.41409691629956,
      "grad_norm": 0.9952473044395447,
      "learning_rate": 0.00013180616740088104,
      "loss": 0.325,
      "step": 1910
    },
    {
      "epoch": 8.458149779735683,
      "grad_norm": 0.7988440990447998,
      "learning_rate": 0.00013092511013215857,
      "loss": 0.3397,
      "step": 1920
    },
    {
      "epoch": 8.502202643171806,
      "grad_norm": 1.2881464958190918,
      "learning_rate": 0.0001300440528634361,
      "loss": 0.4655,
      "step": 1930
    },
    {
      "epoch": 8.54625550660793,
      "grad_norm": 0.9545268416404724,
      "learning_rate": 0.00012916299559471365,
      "loss": 0.4031,
      "step": 1940
    },
    {
      "epoch": 8.590308370044053,
      "grad_norm": 1.550424337387085,
      "learning_rate": 0.00012828193832599118,
      "loss": 0.3697,
      "step": 1950
    },
    {
      "epoch": 8.634361233480176,
      "grad_norm": 1.2041224241256714,
      "learning_rate": 0.0001274008810572687,
      "loss": 0.43,
      "step": 1960
    },
    {
      "epoch": 8.678414096916299,
      "grad_norm": 0.8280724287033081,
      "learning_rate": 0.00012651982378854626,
      "loss": 0.4045,
      "step": 1970
    },
    {
      "epoch": 8.722466960352422,
      "grad_norm": 0.8164283037185669,
      "learning_rate": 0.00012563876651982376,
      "loss": 0.4001,
      "step": 1980
    },
    {
      "epoch": 8.766519823788546,
      "grad_norm": 0.9470929503440857,
      "learning_rate": 0.00012475770925110132,
      "loss": 0.3767,
      "step": 1990
    },
    {
      "epoch": 8.810572687224669,
      "grad_norm": 0.7390472888946533,
      "learning_rate": 0.00012387665198237884,
      "loss": 0.4206,
      "step": 2000
    },
    {
      "epoch": 8.854625550660792,
      "grad_norm": 0.8382723927497864,
      "learning_rate": 0.00012299559471365637,
      "loss": 0.3061,
      "step": 2010
    },
    {
      "epoch": 8.898678414096917,
      "grad_norm": 1.060539722442627,
      "learning_rate": 0.00012211453744493392,
      "loss": 0.4921,
      "step": 2020
    },
    {
      "epoch": 8.94273127753304,
      "grad_norm": 0.6955994367599487,
      "learning_rate": 0.00012123348017621144,
      "loss": 0.4077,
      "step": 2030
    },
    {
      "epoch": 8.986784140969164,
      "grad_norm": 0.8158656358718872,
      "learning_rate": 0.00012035242290748898,
      "loss": 0.3759,
      "step": 2040
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.582844614982605,
      "eval_runtime": 4.8405,
      "eval_samples_per_second": 41.732,
      "eval_steps_per_second": 5.371,
      "step": 2043
    },
    {
      "epoch": 9.030837004405287,
      "grad_norm": 0.9192315936088562,
      "learning_rate": 0.0001194713656387665,
      "loss": 0.3809,
      "step": 2050
    },
    {
      "epoch": 9.07488986784141,
      "grad_norm": 1.0536017417907715,
      "learning_rate": 0.00011859030837004403,
      "loss": 0.3321,
      "step": 2060
    },
    {
      "epoch": 9.118942731277533,
      "grad_norm": 1.1080108880996704,
      "learning_rate": 0.00011770925110132157,
      "loss": 0.407,
      "step": 2070
    },
    {
      "epoch": 9.162995594713657,
      "grad_norm": 0.9956775903701782,
      "learning_rate": 0.0001168281938325991,
      "loss": 0.3423,
      "step": 2080
    },
    {
      "epoch": 9.20704845814978,
      "grad_norm": 0.746013343334198,
      "learning_rate": 0.00011594713656387664,
      "loss": 0.3794,
      "step": 2090
    },
    {
      "epoch": 9.251101321585903,
      "grad_norm": 1.126372218132019,
      "learning_rate": 0.00011506607929515417,
      "loss": 0.4121,
      "step": 2100
    },
    {
      "epoch": 9.295154185022026,
      "grad_norm": 1.4978642463684082,
      "learning_rate": 0.00011418502202643172,
      "loss": 0.3358,
      "step": 2110
    },
    {
      "epoch": 9.33920704845815,
      "grad_norm": 0.7826859951019287,
      "learning_rate": 0.00011330396475770924,
      "loss": 0.2931,
      "step": 2120
    },
    {
      "epoch": 9.383259911894273,
      "grad_norm": 1.1644082069396973,
      "learning_rate": 0.00011242290748898676,
      "loss": 0.377,
      "step": 2130
    },
    {
      "epoch": 9.427312775330396,
      "grad_norm": 0.8106231093406677,
      "learning_rate": 0.00011154185022026432,
      "loss": 0.3562,
      "step": 2140
    },
    {
      "epoch": 9.47136563876652,
      "grad_norm": 1.162919282913208,
      "learning_rate": 0.00011066079295154183,
      "loss": 0.3441,
      "step": 2150
    },
    {
      "epoch": 9.515418502202643,
      "grad_norm": 0.7184136509895325,
      "learning_rate": 0.00010977973568281939,
      "loss": 0.3254,
      "step": 2160
    },
    {
      "epoch": 9.559471365638766,
      "grad_norm": 0.9587578177452087,
      "learning_rate": 0.00010889867841409691,
      "loss": 0.3533,
      "step": 2170
    },
    {
      "epoch": 9.603524229074889,
      "grad_norm": 0.8703950643539429,
      "learning_rate": 0.00010801762114537444,
      "loss": 0.3366,
      "step": 2180
    },
    {
      "epoch": 9.647577092511014,
      "grad_norm": 0.7304671406745911,
      "learning_rate": 0.00010713656387665198,
      "loss": 0.3608,
      "step": 2190
    },
    {
      "epoch": 9.691629955947137,
      "grad_norm": 1.1611542701721191,
      "learning_rate": 0.00010625550660792951,
      "loss": 0.3353,
      "step": 2200
    },
    {
      "epoch": 9.73568281938326,
      "grad_norm": 0.7281723022460938,
      "learning_rate": 0.00010537444933920705,
      "loss": 0.3082,
      "step": 2210
    },
    {
      "epoch": 9.779735682819384,
      "grad_norm": 1.1435456275939941,
      "learning_rate": 0.00010449339207048458,
      "loss": 0.4317,
      "step": 2220
    },
    {
      "epoch": 9.823788546255507,
      "grad_norm": 0.9928381443023682,
      "learning_rate": 0.0001036123348017621,
      "loss": 0.3564,
      "step": 2230
    },
    {
      "epoch": 9.86784140969163,
      "grad_norm": 0.8395977020263672,
      "learning_rate": 0.00010273127753303964,
      "loss": 0.3531,
      "step": 2240
    },
    {
      "epoch": 9.911894273127754,
      "grad_norm": 1.0142395496368408,
      "learning_rate": 0.00010185022026431717,
      "loss": 0.3896,
      "step": 2250
    },
    {
      "epoch": 9.955947136563877,
      "grad_norm": 0.6916971802711487,
      "learning_rate": 0.00010096916299559471,
      "loss": 0.3667,
      "step": 2260
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.7665943503379822,
      "learning_rate": 0.00010008810572687224,
      "loss": 0.3075,
      "step": 2270
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.579430878162384,
      "eval_runtime": 4.7891,
      "eval_samples_per_second": 42.179,
      "eval_steps_per_second": 5.429,
      "step": 2270
    },
    {
      "epoch": 10.044052863436123,
      "grad_norm": 1.4675018787384033,
      "learning_rate": 9.920704845814978e-05,
      "loss": 0.3251,
      "step": 2280
    },
    {
      "epoch": 10.088105726872246,
      "grad_norm": 0.6954736709594727,
      "learning_rate": 9.83259911894273e-05,
      "loss": 0.3656,
      "step": 2290
    },
    {
      "epoch": 10.13215859030837,
      "grad_norm": 1.4188182353973389,
      "learning_rate": 9.744493392070483e-05,
      "loss": 0.335,
      "step": 2300
    },
    {
      "epoch": 10.176211453744493,
      "grad_norm": 0.9333553910255432,
      "learning_rate": 9.656387665198237e-05,
      "loss": 0.2888,
      "step": 2310
    },
    {
      "epoch": 10.220264317180616,
      "grad_norm": 0.886482834815979,
      "learning_rate": 9.56828193832599e-05,
      "loss": 0.3122,
      "step": 2320
    },
    {
      "epoch": 10.26431718061674,
      "grad_norm": 0.6795399188995361,
      "learning_rate": 9.480176211453744e-05,
      "loss": 0.3765,
      "step": 2330
    },
    {
      "epoch": 10.308370044052863,
      "grad_norm": 1.3046603202819824,
      "learning_rate": 9.392070484581497e-05,
      "loss": 0.3316,
      "step": 2340
    },
    {
      "epoch": 10.352422907488986,
      "grad_norm": 1.0006519556045532,
      "learning_rate": 9.30396475770925e-05,
      "loss": 0.3659,
      "step": 2350
    },
    {
      "epoch": 10.396475770925111,
      "grad_norm": 1.1640467643737793,
      "learning_rate": 9.215859030837004e-05,
      "loss": 0.346,
      "step": 2360
    },
    {
      "epoch": 10.440528634361234,
      "grad_norm": 0.9744365811347961,
      "learning_rate": 9.127753303964756e-05,
      "loss": 0.3317,
      "step": 2370
    },
    {
      "epoch": 10.484581497797357,
      "grad_norm": 1.039802074432373,
      "learning_rate": 9.03964757709251e-05,
      "loss": 0.3162,
      "step": 2380
    },
    {
      "epoch": 10.52863436123348,
      "grad_norm": 0.9926576614379883,
      "learning_rate": 8.951541850220263e-05,
      "loss": 0.3559,
      "step": 2390
    },
    {
      "epoch": 10.572687224669604,
      "grad_norm": 1.0141366720199585,
      "learning_rate": 8.863436123348016e-05,
      "loss": 0.3196,
      "step": 2400
    },
    {
      "epoch": 10.616740088105727,
      "grad_norm": 0.5856879353523254,
      "learning_rate": 8.77533039647577e-05,
      "loss": 0.2919,
      "step": 2410
    },
    {
      "epoch": 10.66079295154185,
      "grad_norm": 0.9484356045722961,
      "learning_rate": 8.687224669603523e-05,
      "loss": 0.339,
      "step": 2420
    },
    {
      "epoch": 10.704845814977974,
      "grad_norm": 0.9014990925788879,
      "learning_rate": 8.599118942731277e-05,
      "loss": 0.3089,
      "step": 2430
    },
    {
      "epoch": 10.748898678414097,
      "grad_norm": 0.9830072522163391,
      "learning_rate": 8.51101321585903e-05,
      "loss": 0.3461,
      "step": 2440
    },
    {
      "epoch": 10.79295154185022,
      "grad_norm": 1.051647424697876,
      "learning_rate": 8.422907488986782e-05,
      "loss": 0.292,
      "step": 2450
    },
    {
      "epoch": 10.837004405286343,
      "grad_norm": 1.0580625534057617,
      "learning_rate": 8.334801762114536e-05,
      "loss": 0.4052,
      "step": 2460
    },
    {
      "epoch": 10.881057268722467,
      "grad_norm": 1.01996648311615,
      "learning_rate": 8.246696035242289e-05,
      "loss": 0.3927,
      "step": 2470
    },
    {
      "epoch": 10.92511013215859,
      "grad_norm": 0.6538860201835632,
      "learning_rate": 8.158590308370044e-05,
      "loss": 0.3451,
      "step": 2480
    },
    {
      "epoch": 10.969162995594713,
      "grad_norm": 0.9368380308151245,
      "learning_rate": 8.070484581497796e-05,
      "loss": 0.3932,
      "step": 2490
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.5825287103652954,
      "eval_runtime": 4.811,
      "eval_samples_per_second": 41.987,
      "eval_steps_per_second": 5.404,
      "step": 2497
    },
    {
      "epoch": 11.013215859030836,
      "grad_norm": 0.9590967893600464,
      "learning_rate": 7.982378854625551e-05,
      "loss": 0.32,
      "step": 2500
    },
    {
      "epoch": 11.05726872246696,
      "grad_norm": 0.9905742406845093,
      "learning_rate": 7.894273127753304e-05,
      "loss": 0.3029,
      "step": 2510
    },
    {
      "epoch": 11.101321585903083,
      "grad_norm": 1.2009577751159668,
      "learning_rate": 7.806167400881057e-05,
      "loss": 0.3626,
      "step": 2520
    },
    {
      "epoch": 11.145374449339206,
      "grad_norm": 1.0607908964157104,
      "learning_rate": 7.718061674008811e-05,
      "loss": 0.314,
      "step": 2530
    },
    {
      "epoch": 11.189427312775331,
      "grad_norm": 1.1098504066467285,
      "learning_rate": 7.629955947136563e-05,
      "loss": 0.3062,
      "step": 2540
    },
    {
      "epoch": 11.233480176211454,
      "grad_norm": 0.6961995959281921,
      "learning_rate": 7.541850220264317e-05,
      "loss": 0.3499,
      "step": 2550
    },
    {
      "epoch": 11.277533039647578,
      "grad_norm": 1.0727498531341553,
      "learning_rate": 7.45374449339207e-05,
      "loss": 0.2559,
      "step": 2560
    },
    {
      "epoch": 11.321585903083701,
      "grad_norm": 1.064344048500061,
      "learning_rate": 7.365638766519823e-05,
      "loss": 0.3011,
      "step": 2570
    },
    {
      "epoch": 11.365638766519824,
      "grad_norm": 1.1059036254882812,
      "learning_rate": 7.277533039647577e-05,
      "loss": 0.3415,
      "step": 2580
    },
    {
      "epoch": 11.409691629955947,
      "grad_norm": 0.8815020322799683,
      "learning_rate": 7.18942731277533e-05,
      "loss": 0.3164,
      "step": 2590
    },
    {
      "epoch": 11.45374449339207,
      "grad_norm": 0.9667496085166931,
      "learning_rate": 7.101321585903082e-05,
      "loss": 0.3642,
      "step": 2600
    },
    {
      "epoch": 11.497797356828194,
      "grad_norm": 0.942876935005188,
      "learning_rate": 7.013215859030836e-05,
      "loss": 0.3624,
      "step": 2610
    },
    {
      "epoch": 11.541850220264317,
      "grad_norm": 1.022675633430481,
      "learning_rate": 6.925110132158589e-05,
      "loss": 0.3351,
      "step": 2620
    },
    {
      "epoch": 11.58590308370044,
      "grad_norm": 0.9919267892837524,
      "learning_rate": 6.837004405286343e-05,
      "loss": 0.3335,
      "step": 2630
    },
    {
      "epoch": 11.629955947136564,
      "grad_norm": 0.9724282026290894,
      "learning_rate": 6.748898678414096e-05,
      "loss": 0.3154,
      "step": 2640
    },
    {
      "epoch": 11.674008810572687,
      "grad_norm": 1.3246617317199707,
      "learning_rate": 6.660792951541849e-05,
      "loss": 0.4366,
      "step": 2650
    },
    {
      "epoch": 11.71806167400881,
      "grad_norm": 1.0111949443817139,
      "learning_rate": 6.572687224669603e-05,
      "loss": 0.3324,
      "step": 2660
    },
    {
      "epoch": 11.762114537444933,
      "grad_norm": 0.8399791717529297,
      "learning_rate": 6.484581497797357e-05,
      "loss": 0.2669,
      "step": 2670
    },
    {
      "epoch": 11.806167400881057,
      "grad_norm": 0.917736828327179,
      "learning_rate": 6.39647577092511e-05,
      "loss": 0.324,
      "step": 2680
    },
    {
      "epoch": 11.85022026431718,
      "grad_norm": 0.9939138293266296,
      "learning_rate": 6.308370044052864e-05,
      "loss": 0.2888,
      "step": 2690
    },
    {
      "epoch": 11.894273127753303,
      "grad_norm": 0.9510142803192139,
      "learning_rate": 6.220264317180616e-05,
      "loss": 0.3428,
      "step": 2700
    },
    {
      "epoch": 11.938325991189426,
      "grad_norm": 1.3216148614883423,
      "learning_rate": 6.132158590308369e-05,
      "loss": 0.3254,
      "step": 2710
    },
    {
      "epoch": 11.982378854625551,
      "grad_norm": 1.2755056619644165,
      "learning_rate": 6.0440528634361224e-05,
      "loss": 0.3188,
      "step": 2720
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.5862967371940613,
      "eval_runtime": 5.4591,
      "eval_samples_per_second": 37.002,
      "eval_steps_per_second": 4.763,
      "step": 2724
    },
    {
      "epoch": 12.026431718061675,
      "grad_norm": 0.8617092967033386,
      "learning_rate": 5.955947136563876e-05,
      "loss": 0.2953,
      "step": 2730
    },
    {
      "epoch": 12.070484581497798,
      "grad_norm": 0.7434670329093933,
      "learning_rate": 5.86784140969163e-05,
      "loss": 0.312,
      "step": 2740
    },
    {
      "epoch": 12.114537444933921,
      "grad_norm": 0.9274753332138062,
      "learning_rate": 5.779735682819383e-05,
      "loss": 0.2664,
      "step": 2750
    },
    {
      "epoch": 12.158590308370044,
      "grad_norm": 1.058923363685608,
      "learning_rate": 5.691629955947135e-05,
      "loss": 0.3236,
      "step": 2760
    },
    {
      "epoch": 12.202643171806168,
      "grad_norm": 0.7601414918899536,
      "learning_rate": 5.6035242290748894e-05,
      "loss": 0.3076,
      "step": 2770
    },
    {
      "epoch": 12.246696035242291,
      "grad_norm": 0.7787047624588013,
      "learning_rate": 5.515418502202643e-05,
      "loss": 0.2618,
      "step": 2780
    },
    {
      "epoch": 12.290748898678414,
      "grad_norm": 0.9064326882362366,
      "learning_rate": 5.427312775330396e-05,
      "loss": 0.3305,
      "step": 2790
    },
    {
      "epoch": 12.334801762114537,
      "grad_norm": 1.0712478160858154,
      "learning_rate": 5.3392070484581496e-05,
      "loss": 0.341,
      "step": 2800
    },
    {
      "epoch": 12.37885462555066,
      "grad_norm": 0.6585920453071594,
      "learning_rate": 5.251101321585903e-05,
      "loss": 0.3476,
      "step": 2810
    },
    {
      "epoch": 12.422907488986784,
      "grad_norm": 1.1152169704437256,
      "learning_rate": 5.162995594713656e-05,
      "loss": 0.3418,
      "step": 2820
    },
    {
      "epoch": 12.466960352422907,
      "grad_norm": 0.926008403301239,
      "learning_rate": 5.074889867841409e-05,
      "loss": 0.2543,
      "step": 2830
    },
    {
      "epoch": 12.51101321585903,
      "grad_norm": 1.1506083011627197,
      "learning_rate": 4.9867841409691625e-05,
      "loss": 0.2895,
      "step": 2840
    },
    {
      "epoch": 12.555066079295154,
      "grad_norm": 0.8726121783256531,
      "learning_rate": 4.898678414096916e-05,
      "loss": 0.2917,
      "step": 2850
    },
    {
      "epoch": 12.599118942731277,
      "grad_norm": 1.1620839834213257,
      "learning_rate": 4.810572687224669e-05,
      "loss": 0.3585,
      "step": 2860
    },
    {
      "epoch": 12.6431718061674,
      "grad_norm": 1.1911215782165527,
      "learning_rate": 4.7224669603524226e-05,
      "loss": 0.3177,
      "step": 2870
    },
    {
      "epoch": 12.687224669603523,
      "grad_norm": 0.9236161708831787,
      "learning_rate": 4.6343612334801754e-05,
      "loss": 0.3203,
      "step": 2880
    },
    {
      "epoch": 12.731277533039648,
      "grad_norm": 1.0384935140609741,
      "learning_rate": 4.546255506607929e-05,
      "loss": 0.3264,
      "step": 2890
    },
    {
      "epoch": 12.775330396475772,
      "grad_norm": 1.3048256635665894,
      "learning_rate": 4.458149779735682e-05,
      "loss": 0.3544,
      "step": 2900
    },
    {
      "epoch": 12.819383259911895,
      "grad_norm": 1.127678394317627,
      "learning_rate": 4.370044052863436e-05,
      "loss": 0.3768,
      "step": 2910
    },
    {
      "epoch": 12.863436123348018,
      "grad_norm": 0.9425409436225891,
      "learning_rate": 4.2819383259911896e-05,
      "loss": 0.2778,
      "step": 2920
    },
    {
      "epoch": 12.907488986784141,
      "grad_norm": 1.2469598054885864,
      "learning_rate": 4.1938325991189416e-05,
      "loss": 0.3532,
      "step": 2930
    },
    {
      "epoch": 12.951541850220265,
      "grad_norm": 0.7975876927375793,
      "learning_rate": 4.105726872246696e-05,
      "loss": 0.3189,
      "step": 2940
    },
    {
      "epoch": 12.995594713656388,
      "grad_norm": 0.8869457840919495,
      "learning_rate": 4.017621145374449e-05,
      "loss": 0.3282,
      "step": 2950
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.5989018678665161,
      "eval_runtime": 4.9822,
      "eval_samples_per_second": 40.544,
      "eval_steps_per_second": 5.219,
      "step": 2951
    },
    {
      "epoch": 13.039647577092511,
      "grad_norm": 1.1934689283370972,
      "learning_rate": 3.9295154185022025e-05,
      "loss": 0.2919,
      "step": 2960
    },
    {
      "epoch": 13.083700440528634,
      "grad_norm": 1.1812618970870972,
      "learning_rate": 3.841409691629956e-05,
      "loss": 0.3219,
      "step": 2970
    },
    {
      "epoch": 13.127753303964758,
      "grad_norm": 1.2065187692642212,
      "learning_rate": 3.753303964757709e-05,
      "loss": 0.29,
      "step": 2980
    },
    {
      "epoch": 13.17180616740088,
      "grad_norm": 0.8890476822853088,
      "learning_rate": 3.665198237885462e-05,
      "loss": 0.2462,
      "step": 2990
    },
    {
      "epoch": 13.215859030837004,
      "grad_norm": 1.2433491945266724,
      "learning_rate": 3.5770925110132154e-05,
      "loss": 0.2614,
      "step": 3000
    }
  ],
  "logging_steps": 10,
  "max_steps": 3405,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4451705806651392.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}