{
  "best_metric": 0.0001896605535876006,
  "best_model_checkpoint": "ckpt/llama3_8b_fuze27_no_sys/tracking_shuffled_objects_no_sys/checkpoint-700",
  "epoch": 2.8011204481792715,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.028011204481792718,
      "grad_norm": 11.68239974975586,
      "learning_rate": 2.5e-05,
      "loss": 3.5808,
      "step": 10
    },
    {
      "epoch": 0.056022408963585436,
      "grad_norm": 2.8469760417938232,
      "learning_rate": 5e-05,
      "loss": 1.4653,
      "step": 20
    },
    {
      "epoch": 0.08403361344537816,
      "grad_norm": 2.9564781188964844,
      "learning_rate": 4.9996039875197146e-05,
      "loss": 0.4603,
      "step": 30
    },
    {
      "epoch": 0.11204481792717087,
      "grad_norm": 1.9116700887680054,
      "learning_rate": 4.9984160755395636e-05,
      "loss": 0.4233,
      "step": 40
    },
    {
      "epoch": 0.1400560224089636,
      "grad_norm": 1.8749676942825317,
      "learning_rate": 4.9964366404019245e-05,
      "loss": 0.4279,
      "step": 50
    },
    {
      "epoch": 0.16806722689075632,
      "grad_norm": 2.9393470287323,
      "learning_rate": 4.9936663092116105e-05,
      "loss": 0.4078,
      "step": 60
    },
    {
      "epoch": 0.19607843137254902,
      "grad_norm": 1.5631835460662842,
      "learning_rate": 4.9901059596372036e-05,
      "loss": 0.3969,
      "step": 70
    },
    {
      "epoch": 0.22408963585434175,
      "grad_norm": 2.325085401535034,
      "learning_rate": 4.985756719632996e-05,
      "loss": 0.4476,
      "step": 80
    },
    {
      "epoch": 0.25210084033613445,
      "grad_norm": 1.209677815437317,
      "learning_rate": 4.9806199670816445e-05,
      "loss": 0.4109,
      "step": 90
    },
    {
      "epoch": 0.2801120448179272,
      "grad_norm": 1.1088253259658813,
      "learning_rate": 4.974697329357644e-05,
      "loss": 0.4079,
      "step": 100
    },
    {
      "epoch": 0.2801120448179272,
      "eval_loss": 0.408916711807251,
      "eval_runtime": 2.8461,
      "eval_samples_per_second": 52.704,
      "eval_steps_per_second": 6.676,
      "step": 100
    },
    {
      "epoch": 0.3081232492997199,
      "grad_norm": 1.0425214767456055,
      "learning_rate": 4.967990682811758e-05,
      "loss": 0.4324,
      "step": 110
    },
    {
      "epoch": 0.33613445378151263,
      "grad_norm": 1.5861892700195312,
      "learning_rate": 4.960502152176574e-05,
      "loss": 0.4116,
      "step": 120
    },
    {
      "epoch": 0.3641456582633053,
      "grad_norm": 1.4320331811904907,
      "learning_rate": 4.9522341098933635e-05,
      "loss": 0.4096,
      "step": 130
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 1.3361800909042358,
      "learning_rate": 4.943189175360472e-05,
      "loss": 0.4016,
      "step": 140
    },
    {
      "epoch": 0.42016806722689076,
      "grad_norm": 1.3678736686706543,
      "learning_rate": 4.933370214103467e-05,
      "loss": 0.4113,
      "step": 150
    },
    {
      "epoch": 0.4481792717086835,
      "grad_norm": 1.5625993013381958,
      "learning_rate": 4.922780336867309e-05,
      "loss": 0.4015,
      "step": 160
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 1.4297890663146973,
      "learning_rate": 4.9114228986308374e-05,
      "loss": 0.3892,
      "step": 170
    },
    {
      "epoch": 0.5042016806722689,
      "grad_norm": 2.283721446990967,
      "learning_rate": 4.8993014975438814e-05,
      "loss": 0.3936,
      "step": 180
    },
    {
      "epoch": 0.5322128851540616,
      "grad_norm": 3.531468629837036,
      "learning_rate": 4.8864199737873287e-05,
      "loss": 0.3261,
      "step": 190
    },
    {
      "epoch": 0.5602240896358543,
      "grad_norm": 2.091176748275757,
      "learning_rate": 4.872782408356517e-05,
      "loss": 0.3343,
      "step": 200
    },
    {
      "epoch": 0.5602240896358543,
      "eval_loss": 0.3214447498321533,
      "eval_runtime": 2.8697,
      "eval_samples_per_second": 52.27,
      "eval_steps_per_second": 6.621,
      "step": 200
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 2.5533387660980225,
      "learning_rate": 4.858393121768335e-05,
      "loss": 0.3283,
      "step": 210
    },
    {
      "epoch": 0.6162464985994398,
      "grad_norm": 8.215011596679688,
      "learning_rate": 4.843256672692441e-05,
      "loss": 0.3155,
      "step": 220
    },
    {
      "epoch": 0.6442577030812325,
      "grad_norm": 6.319694519042969,
      "learning_rate": 4.827377856507026e-05,
      "loss": 0.2806,
      "step": 230
    },
    {
      "epoch": 0.6722689075630253,
      "grad_norm": 4.542545318603516,
      "learning_rate": 4.8107617037795965e-05,
      "loss": 0.2884,
      "step": 240
    },
    {
      "epoch": 0.7002801120448179,
      "grad_norm": 17.7308292388916,
      "learning_rate": 4.7934134786732365e-05,
      "loss": 0.286,
      "step": 250
    },
    {
      "epoch": 0.7282913165266106,
      "grad_norm": 2.3095180988311768,
      "learning_rate": 4.775338677278867e-05,
      "loss": 0.2215,
      "step": 260
    },
    {
      "epoch": 0.7563025210084033,
      "grad_norm": 5.083682060241699,
      "learning_rate": 4.756543025874034e-05,
      "loss": 0.2529,
      "step": 270
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 10.334514617919922,
      "learning_rate": 4.7370324791087616e-05,
      "loss": 0.1804,
      "step": 280
    },
    {
      "epoch": 0.8123249299719888,
      "grad_norm": 3.9857237339019775,
      "learning_rate": 4.716813218119064e-05,
      "loss": 0.1956,
      "step": 290
    },
    {
      "epoch": 0.8403361344537815,
      "grad_norm": 6.7719855308532715,
      "learning_rate": 4.6958916485686956e-05,
      "loss": 0.1914,
      "step": 300
    },
    {
      "epoch": 0.8403361344537815,
      "eval_loss": 0.1493147313594818,
      "eval_runtime": 2.8736,
      "eval_samples_per_second": 52.198,
      "eval_steps_per_second": 6.612,
      "step": 300
    },
    {
      "epoch": 0.8683473389355743,
      "grad_norm": 4.405186176300049,
      "learning_rate": 4.6742743986197757e-05,
      "loss": 0.1808,
      "step": 310
    },
    {
      "epoch": 0.896358543417367,
      "grad_norm": 6.499341011047363,
      "learning_rate": 4.65196831683292e-05,
      "loss": 0.154,
      "step": 320
    },
    {
      "epoch": 0.9243697478991597,
      "grad_norm": 3.253216505050659,
      "learning_rate": 4.628980469997547e-05,
      "loss": 0.1339,
      "step": 330
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 8.81497573852539,
      "learning_rate": 4.60531814089305e-05,
      "loss": 0.109,
      "step": 340
    },
    {
      "epoch": 0.9803921568627451,
      "grad_norm": 15.7227144241333,
      "learning_rate": 4.580988825981541e-05,
      "loss": 0.092,
      "step": 350
    },
    {
      "epoch": 1.0084033613445378,
      "grad_norm": 9.170595169067383,
      "learning_rate": 4.556000233032892e-05,
      "loss": 0.1479,
      "step": 360
    },
    {
      "epoch": 1.0364145658263306,
      "grad_norm": 14.895033836364746,
      "learning_rate": 4.530360278682842e-05,
      "loss": 0.0871,
      "step": 370
    },
    {
      "epoch": 1.0644257703081232,
      "grad_norm": 0.1763153076171875,
      "learning_rate": 4.504077085924922e-05,
      "loss": 0.0982,
      "step": 380
    },
    {
      "epoch": 1.092436974789916,
      "grad_norm": 9.751165390014648,
      "learning_rate": 4.477158981537017e-05,
      "loss": 0.0851,
      "step": 390
    },
    {
      "epoch": 1.1204481792717087,
      "grad_norm": 0.1843680888414383,
      "learning_rate": 4.449614493443354e-05,
      "loss": 0.034,
      "step": 400
    },
    {
      "epoch": 1.1204481792717087,
      "eval_loss": 0.04679349809885025,
      "eval_runtime": 2.873,
      "eval_samples_per_second": 52.21,
      "eval_steps_per_second": 6.613,
      "step": 400
    },
    {
      "epoch": 1.1484593837535013,
      "grad_norm": 6.5383501052856445,
      "learning_rate": 4.421452348012771e-05,
      "loss": 0.0584,
      "step": 410
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.003353311913087964,
      "learning_rate": 4.392681467294117e-05,
      "loss": 0.0344,
      "step": 420
    },
    {
      "epoch": 1.2044817927170868,
      "grad_norm": 0.0053819697350263596,
      "learning_rate": 4.3633109661896595e-05,
      "loss": 0.0111,
      "step": 430
    },
    {
      "epoch": 1.2324929971988796,
      "grad_norm": 32.45022201538086,
      "learning_rate": 4.33335014956739e-05,
      "loss": 0.1152,
      "step": 440
    },
    {
      "epoch": 1.2605042016806722,
      "grad_norm": 7.157898902893066,
      "learning_rate": 4.30280850931315e-05,
      "loss": 0.0524,
      "step": 450
    },
    {
      "epoch": 1.2885154061624648,
      "grad_norm": 0.7617452144622803,
      "learning_rate": 4.271695721323506e-05,
      "loss": 0.0121,
      "step": 460
    },
    {
      "epoch": 1.3165266106442577,
      "grad_norm": 23.683460235595703,
      "learning_rate": 4.240021642440333e-05,
      "loss": 0.0947,
      "step": 470
    },
    {
      "epoch": 1.3445378151260505,
      "grad_norm": 0.007763392757624388,
      "learning_rate": 4.207796307328059e-05,
      "loss": 0.0011,
      "step": 480
    },
    {
      "epoch": 1.3725490196078431,
      "grad_norm": 0.14272698760032654,
      "learning_rate": 4.175029925294595e-05,
      "loss": 0.0433,
      "step": 490
    },
    {
      "epoch": 1.4005602240896358,
      "grad_norm": 0.22110269963741302,
      "learning_rate": 4.141732877056915e-05,
      "loss": 0.0276,
      "step": 500
    },
    {
      "epoch": 1.4005602240896358,
      "eval_loss": 0.008394874632358551,
      "eval_runtime": 2.8698,
      "eval_samples_per_second": 52.269,
      "eval_steps_per_second": 6.621,
      "step": 500
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.014582260511815548,
      "learning_rate": 4.107915711452347e-05,
      "loss": 0.0156,
      "step": 510
    },
    {
      "epoch": 1.4565826330532212,
      "grad_norm": 0.0064322990365326405,
      "learning_rate": 4.073589142096592e-05,
      "loss": 0.0092,
      "step": 520
    },
    {
      "epoch": 1.484593837535014,
      "grad_norm": 0.006550009828060865,
      "learning_rate": 4.038764043989548e-05,
      "loss": 0.0056,
      "step": 530
    },
    {
      "epoch": 1.5126050420168067,
      "grad_norm": 0.00899260863661766,
      "learning_rate": 4.003451450069994e-05,
      "loss": 0.0001,
      "step": 540
    },
    {
      "epoch": 1.5406162464985993,
      "grad_norm": 0.0009019902790896595,
      "learning_rate": 3.9676625477202554e-05,
      "loss": 0.0081,
      "step": 550
    },
    {
      "epoch": 1.5686274509803921,
      "grad_norm": 0.005392414517700672,
      "learning_rate": 3.9314086752219195e-05,
      "loss": 0.0259,
      "step": 560
    },
    {
      "epoch": 1.596638655462185,
      "grad_norm": 0.0008008493459783494,
      "learning_rate": 3.8947013181637624e-05,
      "loss": 0.0008,
      "step": 570
    },
    {
      "epoch": 1.6246498599439776,
      "grad_norm": 0.014637362211942673,
      "learning_rate": 3.857552105802994e-05,
      "loss": 0.0006,
      "step": 580
    },
    {
      "epoch": 1.6526610644257702,
      "grad_norm": 0.0004757639835588634,
      "learning_rate": 3.819972807380995e-05,
      "loss": 0.0053,
      "step": 590
    },
    {
      "epoch": 1.680672268907563,
      "grad_norm": 0.0002963497245218605,
      "learning_rate": 3.781975328394708e-05,
      "loss": 0.0001,
      "step": 600
    },
    {
      "epoch": 1.680672268907563,
      "eval_loss": 0.01559663750231266,
      "eval_runtime": 2.8686,
      "eval_samples_per_second": 52.29,
      "eval_steps_per_second": 6.623,
      "step": 600
    },
    {
      "epoch": 1.708683473389356,
      "grad_norm": 3.8546576499938965,
      "learning_rate": 3.74357170682485e-05,
      "loss": 0.0094,
      "step": 610
    },
    {
      "epoch": 1.7366946778711485,
      "grad_norm": 0.38633355498313904,
      "learning_rate": 3.704774109322166e-05,
      "loss": 0.034,
      "step": 620
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.0008682355983182788,
      "learning_rate": 3.665594827352908e-05,
      "loss": 0.0004,
      "step": 630
    },
    {
      "epoch": 1.7927170868347337,
      "grad_norm": 0.01699518784880638,
      "learning_rate": 3.626046273304779e-05,
      "loss": 0.0146,
      "step": 640
    },
    {
      "epoch": 1.8207282913165266,
      "grad_norm": 0.0713791772723198,
      "learning_rate": 3.586140976554564e-05,
      "loss": 0.0001,
      "step": 650
    },
    {
      "epoch": 1.8487394957983194,
      "grad_norm": 0.0003395811072550714,
      "learning_rate": 3.545891579498695e-05,
      "loss": 0.0001,
      "step": 660
    },
    {
      "epoch": 1.876750700280112,
      "grad_norm": 0.0010548827704042196,
      "learning_rate": 3.5053108335480206e-05,
      "loss": 0.0022,
      "step": 670
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 9.065557242138311e-05,
      "learning_rate": 3.4644115950880254e-05,
      "loss": 0.0389,
      "step": 680
    },
    {
      "epoch": 1.9327731092436975,
      "grad_norm": 0.010628440417349339,
      "learning_rate": 3.423206821405799e-05,
      "loss": 0.0002,
      "step": 690
    },
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 0.0003508552326820791,
      "learning_rate": 3.381709566585044e-05,
      "loss": 0.0,
      "step": 700
    },
    {
      "epoch": 1.9607843137254903,
      "eval_loss": 0.0001896605535876006,
      "eval_runtime": 2.8723,
      "eval_samples_per_second": 52.223,
      "eval_steps_per_second": 6.615,
      "step": 700
    },
    {
      "epoch": 1.988795518207283,
      "grad_norm": 0.00045208673691377044,
      "learning_rate": 3.339932977370404e-05,
      "loss": 0.0548,
      "step": 710
    },
    {
      "epoch": 2.0168067226890756,
      "grad_norm": 0.0002967917826026678,
      "learning_rate": 3.297890289002451e-05,
      "loss": 0.0669,
      "step": 720
    },
    {
      "epoch": 2.044817927170868,
      "grad_norm": 0.0014003561809659004,
      "learning_rate": 3.255594821024622e-05,
      "loss": 0.0137,
      "step": 730
    },
    {
      "epoch": 2.0728291316526612,
      "grad_norm": 0.029797283932566643,
      "learning_rate": 3.2130599730634606e-05,
      "loss": 0.005,
      "step": 740
    },
    {
      "epoch": 2.100840336134454,
      "grad_norm": 0.0008940272382460535,
      "learning_rate": 3.170299220583479e-05,
      "loss": 0.0006,
      "step": 750
    },
    {
      "epoch": 2.1288515406162465,
      "grad_norm": 0.0002603277680464089,
      "learning_rate": 3.127326110617996e-05,
      "loss": 0.0002,
      "step": 760
    },
    {
      "epoch": 2.156862745098039,
      "grad_norm": 0.000504830852150917,
      "learning_rate": 3.084154257477301e-05,
      "loss": 0.0,
      "step": 770
    },
    {
      "epoch": 2.184873949579832,
      "grad_norm": 0.0018955356208607554,
      "learning_rate": 3.0407973384355088e-05,
      "loss": 0.0165,
      "step": 780
    },
    {
      "epoch": 2.212885154061625,
      "grad_norm": 0.0010648163734003901,
      "learning_rate": 2.997269089397455e-05,
      "loss": 0.0,
      "step": 790
    },
    {
      "epoch": 2.2408963585434174,
      "grad_norm": 0.0006610855343751609,
      "learning_rate": 2.9535833005470315e-05,
      "loss": 0.0022,
      "step": 800
    },
    {
      "epoch": 2.2408963585434174,
      "eval_loss": 0.012090143747627735,
      "eval_runtime": 2.8741,
      "eval_samples_per_second": 52.189,
      "eval_steps_per_second": 6.611,
      "step": 800
    },
    {
      "epoch": 2.26890756302521,
      "grad_norm": 0.0001721924199955538,
      "learning_rate": 2.9097538119783152e-05,
      "loss": 0.0,
      "step": 810
    },
    {
      "epoch": 2.2969187675070026,
      "grad_norm": 0.0006904829642735422,
      "learning_rate": 2.8657945093108886e-05,
      "loss": 0.0161,
      "step": 820
    },
    {
      "epoch": 2.3249299719887957,
      "grad_norm": 31.687515258789062,
      "learning_rate": 2.821719319290736e-05,
      "loss": 0.0684,
      "step": 830
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.002737673930823803,
      "learning_rate": 2.777542205378113e-05,
      "loss": 0.0001,
      "step": 840
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.005815234035253525,
      "learning_rate": 2.733277163323782e-05,
      "loss": 0.0038,
      "step": 850
    },
    {
      "epoch": 2.4089635854341735,
      "grad_norm": 0.0005296830786392093,
      "learning_rate": 2.6889382167350154e-05,
      "loss": 0.0001,
      "step": 860
    },
    {
      "epoch": 2.4369747899159666,
      "grad_norm": 0.001167680835351348,
      "learning_rate": 2.6445394126327865e-05,
      "loss": 0.0,
      "step": 870
    },
    {
      "epoch": 2.4649859943977592,
      "grad_norm": 0.0011337065370753407,
      "learning_rate": 2.6000948170015205e-05,
      "loss": 0.016,
      "step": 880
    },
    {
      "epoch": 2.492997198879552,
      "grad_norm": 0.002113680588081479,
      "learning_rate": 2.555618510332859e-05,
      "loss": 0.0,
      "step": 890
    },
    {
      "epoch": 2.5210084033613445,
      "grad_norm": 0.00770913390442729,
      "learning_rate": 2.511124583164816e-05,
      "loss": 0.0,
      "step": 900
    },
    {
      "epoch": 2.5210084033613445,
      "eval_loss": 0.005274456925690174,
      "eval_runtime": 2.873,
      "eval_samples_per_second": 52.21,
      "eval_steps_per_second": 6.613,
      "step": 900
    },
    {
      "epoch": 2.549019607843137,
      "grad_norm": 0.0025718784891068935,
      "learning_rate": 2.4666271316177557e-05,
      "loss": 0.0001,
      "step": 910
    },
    {
      "epoch": 2.5770308123249297,
      "grad_norm": 0.00034481892362236977,
      "learning_rate": 2.422140252928601e-05,
      "loss": 0.0002,
      "step": 920
    },
    {
      "epoch": 2.6050420168067228,
      "grad_norm": 0.00020677850989159197,
      "learning_rate": 2.3776780409846888e-05,
      "loss": 0.0001,
      "step": 930
    },
    {
      "epoch": 2.6330532212885154,
      "grad_norm": 0.0001574026100570336,
      "learning_rate": 2.3332545818586827e-05,
      "loss": 0.0,
      "step": 940
    },
    {
      "epoch": 2.661064425770308,
      "grad_norm": 0.017475171014666557,
      "learning_rate": 2.2888839493459684e-05,
      "loss": 0.0,
      "step": 950
    },
    {
      "epoch": 2.689075630252101,
      "grad_norm": 0.006330742035061121,
      "learning_rate": 2.2445802005059328e-05,
      "loss": 0.0,
      "step": 960
    },
    {
      "epoch": 2.7170868347338937,
      "grad_norm": 0.0004552874306682497,
      "learning_rate": 2.2003573712085457e-05,
      "loss": 0.0,
      "step": 970
    },
    {
      "epoch": 2.7450980392156863,
      "grad_norm": 0.0009258923237212002,
      "learning_rate": 2.1562294716876596e-05,
      "loss": 0.0,
      "step": 980
    },
    {
      "epoch": 2.773109243697479,
      "grad_norm": 0.0004848680691793561,
      "learning_rate": 2.112210482102428e-05,
      "loss": 0.0,
      "step": 990
    },
    {
      "epoch": 2.8011204481792715,
      "grad_norm": 0.0008731464040465653,
      "learning_rate": 2.0683143481082452e-05,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 2.8011204481792715,
      "eval_loss": 0.005379769951105118,
      "eval_runtime": 2.8678,
      "eval_samples_per_second": 52.304,
      "eval_steps_per_second": 6.625,
      "step": 1000
    },
    {
      "epoch": 2.8011204481792715,
      "step": 1000,
      "total_flos": 8.23229487073198e+16,
      "train_loss": 0.1688887168551737,
      "train_runtime": 530.2932,
      "train_samples_per_second": 26.872,
      "train_steps_per_second": 3.366
    }
  ],
  "logging_steps": 10,
  "max_steps": 1785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 8.23229487073198e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}