{
  "best_metric": 0.794740617275238,
  "best_model_checkpoint": "saves/CADICA_qwenvl_stenosis_classily/lora/sft/checkpoint-700",
  "epoch": 1.9968586387434555,
  "eval_steps": 50,
  "global_step": 716,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013961605584642234,
      "grad_norm": 21.25276507868793,
      "learning_rate": 6.944444444444445e-06,
      "loss": 2.9908,
      "num_input_tokens_seen": 77944,
      "step": 5
    },
    {
      "epoch": 0.027923211169284468,
      "grad_norm": 21.89043285054519,
      "learning_rate": 1.388888888888889e-05,
      "loss": 3.0071,
      "num_input_tokens_seen": 155896,
      "step": 10
    },
    {
      "epoch": 0.041884816753926704,
      "grad_norm": 16.65776874449816,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 2.354,
      "num_input_tokens_seen": 233896,
      "step": 15
    },
    {
      "epoch": 0.055846422338568937,
      "grad_norm": 3.772799389266845,
      "learning_rate": 2.777777777777778e-05,
      "loss": 1.2959,
      "num_input_tokens_seen": 311840,
      "step": 20
    },
    {
      "epoch": 0.06980802792321117,
      "grad_norm": 2.5936011954385334,
      "learning_rate": 3.472222222222222e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 389816,
      "step": 25
    },
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 1.380523901017673,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 467808,
      "step": 30
    },
    {
      "epoch": 0.09773123909249563,
      "grad_norm": 0.9535971270874376,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 0.9052,
      "num_input_tokens_seen": 545776,
      "step": 35
    },
    {
      "epoch": 0.11169284467713787,
      "grad_norm": 0.7487685762175865,
      "learning_rate": 5.555555555555556e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 623744,
      "step": 40
    },
    {
      "epoch": 0.1256544502617801,
      "grad_norm": 0.9517829869317949,
      "learning_rate": 6.25e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 701720,
      "step": 45
    },
    {
      "epoch": 0.13961605584642234,
      "grad_norm": 0.5105376471286923,
      "learning_rate": 6.944444444444444e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 779728,
      "step": 50
    },
    {
      "epoch": 0.13961605584642234,
      "eval_loss": 0.9039102792739868,
      "eval_runtime": 74.9579,
      "eval_samples_per_second": 1.948,
      "eval_steps_per_second": 0.494,
      "num_input_tokens_seen": 779728,
      "step": 50
    },
    {
      "epoch": 0.15357766143106458,
      "grad_norm": 0.6125311992064874,
      "learning_rate": 7.638888888888889e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 857728,
      "step": 55
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 0.8799068808838695,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.9115,
      "num_input_tokens_seen": 935680,
      "step": 60
    },
    {
      "epoch": 0.18150087260034903,
      "grad_norm": 0.7270711909487898,
      "learning_rate": 9.027777777777779e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 1013664,
      "step": 65
    },
    {
      "epoch": 0.19546247818499127,
      "grad_norm": 0.6023654770278246,
      "learning_rate": 9.722222222222223e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 1091656,
      "step": 70
    },
    {
      "epoch": 0.2094240837696335,
      "grad_norm": 0.5698794386547648,
      "learning_rate": 9.999464569905628e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 1169664,
      "step": 75
    },
    {
      "epoch": 0.22338568935427575,
      "grad_norm": 0.32260644881875,
      "learning_rate": 9.99619291237835e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 1247672,
      "step": 80
    },
    {
      "epoch": 0.23734729493891799,
      "grad_norm": 0.41708039368778405,
      "learning_rate": 9.989949002448076e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 1325640,
      "step": 85
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 0.6145758907120942,
      "learning_rate": 9.980736554638366e-05,
      "loss": 0.9128,
      "num_input_tokens_seen": 1403688,
      "step": 90
    },
    {
      "epoch": 0.26527050610820246,
      "grad_norm": 0.30302247663937915,
      "learning_rate": 9.968561049466214e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 1481664,
      "step": 95
    },
    {
      "epoch": 0.2792321116928447,
      "grad_norm": 0.32920212256475023,
      "learning_rate": 9.953429730181653e-05,
      "loss": 0.9033,
      "num_input_tokens_seen": 1559632,
      "step": 100
    },
    {
      "epoch": 0.2792321116928447,
      "eval_loss": 0.9009457230567932,
      "eval_runtime": 47.3577,
      "eval_samples_per_second": 3.083,
      "eval_steps_per_second": 0.781,
      "num_input_tokens_seen": 1559632,
      "step": 100
    },
    {
      "epoch": 0.2931937172774869,
      "grad_norm": 0.3748068082841054,
      "learning_rate": 9.935351598458742e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 1637592,
      "step": 105
    },
    {
      "epoch": 0.30715532286212915,
      "grad_norm": 0.367692204424778,
      "learning_rate": 9.914337409040418e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 1715592,
      "step": 110
    },
    {
      "epoch": 0.32111692844677137,
      "grad_norm": 0.523389228578757,
      "learning_rate": 9.890399663340478e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 1793544,
      "step": 115
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 0.7666885810234405,
      "learning_rate": 9.863552602006435e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 1871520,
      "step": 120
    },
    {
      "epoch": 0.34904013961605584,
      "grad_norm": 0.45411297588089927,
      "learning_rate": 9.83381219644771e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 1949488,
      "step": 125
    },
    {
      "epoch": 0.36300174520069806,
      "grad_norm": 0.34304009173464395,
      "learning_rate": 9.801196139334195e-05,
      "loss": 0.8919,
      "num_input_tokens_seen": 2027488,
      "step": 130
    },
    {
      "epoch": 0.3769633507853403,
      "grad_norm": 0.46756876437741973,
      "learning_rate": 9.765723834070804e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 2105424,
      "step": 135
    },
    {
      "epoch": 0.39092495636998253,
      "grad_norm": 0.5067842714503226,
      "learning_rate": 9.72741638325434e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 2183432,
      "step": 140
    },
    {
      "epoch": 0.4048865619546248,
      "grad_norm": 0.45330848125911494,
      "learning_rate": 9.686296576119471e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 2261408,
      "step": 145
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 0.3379495380306588,
      "learning_rate": 9.642388874981347e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 2339368,
      "step": 150
    },
    {
      "epoch": 0.418848167539267,
      "eval_loss": 0.8987511396408081,
      "eval_runtime": 46.4204,
      "eval_samples_per_second": 3.145,
      "eval_steps_per_second": 0.797,
      "num_input_tokens_seen": 2339368,
      "step": 150
    },
    {
      "epoch": 0.4328097731239092,
      "grad_norm": 0.4705971769713144,
      "learning_rate": 9.595719400682881e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 2417328,
      "step": 155
    },
    {
      "epoch": 0.4467713787085515,
      "grad_norm": 0.2877245912486978,
      "learning_rate": 9.546315917055361e-05,
      "loss": 0.895,
      "num_input_tokens_seen": 2495328,
      "step": 160
    },
    {
      "epoch": 0.4607329842931937,
      "grad_norm": 0.3089085786477158,
      "learning_rate": 9.494207814401672e-05,
      "loss": 0.8993,
      "num_input_tokens_seen": 2573264,
      "step": 165
    },
    {
      "epoch": 0.47469458987783597,
      "grad_norm": 0.23434151328428765,
      "learning_rate": 9.439426092011875e-05,
      "loss": 0.9011,
      "num_input_tokens_seen": 2651200,
      "step": 170
    },
    {
      "epoch": 0.4886561954624782,
      "grad_norm": 0.3895079869752368,
      "learning_rate": 9.382003339721652e-05,
      "loss": 0.8943,
      "num_input_tokens_seen": 2729208,
      "step": 175
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 0.21859941380068879,
      "learning_rate": 9.321973718524472e-05,
      "loss": 0.9074,
      "num_input_tokens_seen": 2807176,
      "step": 180
    },
    {
      "epoch": 0.5165794066317626,
      "grad_norm": 0.3026569089827376,
      "learning_rate": 9.25937294024912e-05,
      "loss": 0.8979,
      "num_input_tokens_seen": 2885136,
      "step": 185
    },
    {
      "epoch": 0.5305410122164049,
      "grad_norm": 0.24552828812005026,
      "learning_rate": 9.194238246314599e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 2963120,
      "step": 190
    },
    {
      "epoch": 0.5445026178010471,
      "grad_norm": 0.31370225105827704,
      "learning_rate": 9.126608385575076e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 3041096,
      "step": 195
    },
    {
      "epoch": 0.5584642233856894,
      "grad_norm": 0.19991243424614802,
      "learning_rate": 9.056523591268064e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 3119064,
      "step": 200
    },
    {
      "epoch": 0.5584642233856894,
      "eval_loss": 0.9003660678863525,
      "eval_runtime": 46.4656,
      "eval_samples_per_second": 3.142,
      "eval_steps_per_second": 0.796,
      "num_input_tokens_seen": 3119064,
      "step": 200
    },
    {
      "epoch": 0.5724258289703316,
      "grad_norm": 0.29215232660029566,
      "learning_rate": 8.984025557079523e-05,
      "loss": 0.9016,
      "num_input_tokens_seen": 3197048,
      "step": 205
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 0.4850357234182881,
      "learning_rate": 8.90915741234015e-05,
      "loss": 0.907,
      "num_input_tokens_seen": 3275024,
      "step": 210
    },
    {
      "epoch": 0.6003490401396161,
      "grad_norm": 0.3567000129034729,
      "learning_rate": 8.831963696367581e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 3353024,
      "step": 215
    },
    {
      "epoch": 0.6143106457242583,
      "grad_norm": 0.26853087654006846,
      "learning_rate": 8.752490331969807e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 3430936,
      "step": 220
    },
    {
      "epoch": 0.6282722513089005,
      "grad_norm": 0.20742364801678845,
      "learning_rate": 8.670784598125533e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 3508920,
      "step": 225
    },
    {
      "epoch": 0.6422338568935427,
      "grad_norm": 0.2838565453202246,
      "learning_rate": 8.586895101857747e-05,
      "loss": 0.8936,
      "num_input_tokens_seen": 3586920,
      "step": 230
    },
    {
      "epoch": 0.6561954624781849,
      "grad_norm": 0.4892803498442385,
      "learning_rate": 8.500871749317243e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 3664896,
      "step": 235
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 0.37976656488421395,
      "learning_rate": 8.412765716093272e-05,
      "loss": 0.9034,
      "num_input_tokens_seen": 3742832,
      "step": 240
    },
    {
      "epoch": 0.6841186736474695,
      "grad_norm": 0.291645150101734,
      "learning_rate": 8.322629416769006e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 3820792,
      "step": 245
    },
    {
      "epoch": 0.6980802792321117,
      "grad_norm": 0.30837187813275513,
      "learning_rate": 8.230516473739935e-05,
      "loss": 0.8933,
      "num_input_tokens_seen": 3898784,
      "step": 250
    },
    {
      "epoch": 0.6980802792321117,
      "eval_loss": 0.9052047729492188,
      "eval_runtime": 46.4894,
      "eval_samples_per_second": 3.141,
      "eval_steps_per_second": 0.796,
      "num_input_tokens_seen": 3898784,
      "step": 250
    },
    {
      "epoch": 0.7120418848167539,
      "grad_norm": 0.30727171157251326,
      "learning_rate": 8.1364816853137e-05,
      "loss": 0.9079,
      "num_input_tokens_seen": 3976824,
      "step": 255
    },
    {
      "epoch": 0.7260034904013961,
      "grad_norm": 0.24728667308223853,
      "learning_rate": 8.040580993110404e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 4054752,
      "step": 260
    },
    {
      "epoch": 0.7399650959860384,
      "grad_norm": 0.154782478333375,
      "learning_rate": 7.942871448782748e-05,
      "loss": 0.895,
      "num_input_tokens_seen": 4132664,
      "step": 265
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 0.21937062747939326,
      "learning_rate": 7.843411180075794e-05,
      "loss": 0.8984,
      "num_input_tokens_seen": 4210656,
      "step": 270
    },
    {
      "epoch": 0.7678883071553229,
      "grad_norm": 0.28854882430140594,
      "learning_rate": 7.742259356246593e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 4288664,
      "step": 275
    },
    {
      "epoch": 0.7818499127399651,
      "grad_norm": 0.23726372374470991,
      "learning_rate": 7.639476152864162e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 4366608,
      "step": 280
    },
    {
      "epoch": 0.7958115183246073,
      "grad_norm": 0.34106252388262337,
      "learning_rate": 7.535122716010849e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 4444568,
      "step": 285
    },
    {
      "epoch": 0.8097731239092496,
      "grad_norm": 0.37561723629929783,
      "learning_rate": 7.42926112590631e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 4522512,
      "step": 290
    },
    {
      "epoch": 0.8237347294938918,
      "grad_norm": 0.26053439132207656,
      "learning_rate": 7.321954359975776e-05,
      "loss": 0.9002,
      "num_input_tokens_seen": 4600504,
      "step": 295
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 0.248855595995717,
      "learning_rate": 7.21326625538456e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 4678472,
      "step": 300
    },
    {
      "epoch": 0.837696335078534,
      "eval_loss": 0.9003945589065552,
      "eval_runtime": 46.3308,
      "eval_samples_per_second": 3.151,
      "eval_steps_per_second": 0.799,
      "num_input_tokens_seen": 4678472,
      "step": 300
    },
    {
      "epoch": 0.8516579406631762,
      "grad_norm": 0.18200381843269203,
      "learning_rate": 7.103261471061116e-05,
      "loss": 0.9088,
      "num_input_tokens_seen": 4756440,
      "step": 305
    },
    {
      "epoch": 0.8656195462478184,
      "grad_norm": 0.19255528640902111,
      "learning_rate": 6.992005449231208e-05,
      "loss": 0.899,
      "num_input_tokens_seen": 4834424,
      "step": 310
    },
    {
      "epoch": 0.8795811518324608,
      "grad_norm": 0.3960850388870267,
      "learning_rate": 6.879564376486114e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 4912376,
      "step": 315
    },
    {
      "epoch": 0.893542757417103,
      "grad_norm": 0.3472454916197344,
      "learning_rate": 6.76600514440799e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 4990328,
      "step": 320
    },
    {
      "epoch": 0.9075043630017452,
      "grad_norm": 0.42131468150264795,
      "learning_rate": 6.651395309775837e-05,
      "loss": 0.8916,
      "num_input_tokens_seen": 5068304,
      "step": 325
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 0.7373865772840422,
      "learning_rate": 6.535803054375738e-05,
      "loss": 0.8937,
      "num_input_tokens_seen": 5146272,
      "step": 330
    },
    {
      "epoch": 0.9354275741710296,
      "grad_norm": 0.6075085371236206,
      "learning_rate": 6.419297144439283e-05,
      "loss": 0.8965,
      "num_input_tokens_seen": 5224232,
      "step": 335
    },
    {
      "epoch": 0.9493891797556719,
      "grad_norm": 0.5336896324950464,
      "learning_rate": 6.301946889734302e-05,
      "loss": 0.8957,
      "num_input_tokens_seen": 5302200,
      "step": 340
    },
    {
      "epoch": 0.9633507853403142,
      "grad_norm": 0.7535717304812762,
      "learning_rate": 6.183822102332234e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 5380168,
      "step": 345
    },
    {
      "epoch": 0.9773123909249564,
      "grad_norm": 1.9146693047973427,
      "learning_rate": 6.064993055076698e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 5458104,
      "step": 350
    },
    {
      "epoch": 0.9773123909249564,
      "eval_loss": 0.9016226530075073,
      "eval_runtime": 46.29,
      "eval_samples_per_second": 3.154,
      "eval_steps_per_second": 0.799,
      "num_input_tokens_seen": 5458104,
      "step": 350
    },
    {
      "epoch": 0.9912739965095986,
      "grad_norm": 0.4275320872134287,
      "learning_rate": 5.945530439777923e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 5536072,
      "step": 355
    },
    {
      "epoch": 1.0027923211169285,
      "grad_norm": 1.5609644797837416,
      "learning_rate": 5.8255053251579616e-05,
      "loss": 0.7347,
      "num_input_tokens_seen": 5600392,
      "step": 360
    },
    {
      "epoch": 1.0167539267015706,
      "grad_norm": 0.4880192687307677,
      "learning_rate": 5.704989114571648e-05,
      "loss": 0.8899,
      "num_input_tokens_seen": 5678424,
      "step": 365
    },
    {
      "epoch": 1.030715532286213,
      "grad_norm": 0.37744049750587216,
      "learning_rate": 5.5840535035285025e-05,
      "loss": 0.8929,
      "num_input_tokens_seen": 5756400,
      "step": 370
    },
    {
      "epoch": 1.0446771378708553,
      "grad_norm": 0.9395658719072697,
      "learning_rate": 5.4627704370408236e-05,
      "loss": 0.8904,
      "num_input_tokens_seen": 5834352,
      "step": 375
    },
    {
      "epoch": 1.0586387434554974,
      "grad_norm": 0.5490353927737941,
      "learning_rate": 5.341212066823355e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 5912320,
      "step": 380
    },
    {
      "epoch": 1.0726003490401397,
      "grad_norm": 1.1114978460946199,
      "learning_rate": 5.219450708369977e-05,
      "loss": 0.8843,
      "num_input_tokens_seen": 5990312,
      "step": 385
    },
    {
      "epoch": 1.0865619546247818,
      "grad_norm": 1.1913979980069938,
      "learning_rate": 5.0975587979329734e-05,
      "loss": 0.8879,
      "num_input_tokens_seen": 6068280,
      "step": 390
    },
    {
      "epoch": 1.100523560209424,
      "grad_norm": 2.222951083897475,
      "learning_rate": 4.9756088494304504e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 6146288,
      "step": 395
    },
    {
      "epoch": 1.1144851657940662,
      "grad_norm": 1.5657191660795382,
      "learning_rate": 4.853673411307564e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 6224248,
      "step": 400
    },
    {
      "epoch": 1.1144851657940662,
      "eval_loss": 0.8960007429122925,
      "eval_runtime": 46.5432,
      "eval_samples_per_second": 3.137,
      "eval_steps_per_second": 0.795,
      "num_input_tokens_seen": 6224248,
      "step": 400
    },
    {
      "epoch": 1.1284467713787085,
      "grad_norm": 2.66254478509122,
      "learning_rate": 4.731825023377192e-05,
      "loss": 0.8631,
      "num_input_tokens_seen": 6302208,
      "step": 405
    },
    {
      "epoch": 1.1424083769633508,
      "grad_norm": 1.7871963773512198,
      "learning_rate": 4.610136173665751e-05,
      "loss": 0.8722,
      "num_input_tokens_seen": 6380096,
      "step": 410
    },
    {
      "epoch": 1.156369982547993,
      "grad_norm": 2.511479150461674,
      "learning_rate": 4.4886792552898286e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 6458096,
      "step": 415
    },
    {
      "epoch": 1.1703315881326353,
      "grad_norm": 2.9185608745372837,
      "learning_rate": 4.367526523389253e-05,
      "loss": 0.8446,
      "num_input_tokens_seen": 6536064,
      "step": 420
    },
    {
      "epoch": 1.1842931937172776,
      "grad_norm": 3.7053458609530403,
      "learning_rate": 4.24675005214227e-05,
      "loss": 0.8576,
      "num_input_tokens_seen": 6614048,
      "step": 425
    },
    {
      "epoch": 1.1982547993019197,
      "grad_norm": 2.3073667949552124,
      "learning_rate": 4.1264216918883656e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 6691984,
      "step": 430
    },
    {
      "epoch": 1.212216404886562,
      "grad_norm": 2.597389084730036,
      "learning_rate": 4.006613026384249e-05,
      "loss": 0.8708,
      "num_input_tokens_seen": 6769984,
      "step": 435
    },
    {
      "epoch": 1.2261780104712041,
      "grad_norm": 1.889225145403208,
      "learning_rate": 3.887395330218429e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 6847976,
      "step": 440
    },
    {
      "epoch": 1.2401396160558464,
      "grad_norm": 2.61428130233799,
      "learning_rate": 3.768839526409718e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 6925944,
      "step": 445
    },
    {
      "epoch": 1.2541012216404885,
      "grad_norm": 5.825082846471074,
      "learning_rate": 3.651016144214878e-05,
      "loss": 0.8127,
      "num_input_tokens_seen": 7003904,
      "step": 450
    },
    {
      "epoch": 1.2541012216404885,
      "eval_loss": 0.8821887373924255,
      "eval_runtime": 46.527,
      "eval_samples_per_second": 3.138,
      "eval_steps_per_second": 0.795,
      "num_input_tokens_seen": 7003904,
      "step": 450
    },
    {
      "epoch": 1.2680628272251309,
      "grad_norm": 3.745385127572465,
      "learning_rate": 3.533995277170532e-05,
      "loss": 0.837,
      "num_input_tokens_seen": 7081856,
      "step": 455
    },
    {
      "epoch": 1.2820244328097732,
      "grad_norm": 3.1595331481084896,
      "learning_rate": 3.4178465413942625e-05,
      "loss": 0.8631,
      "num_input_tokens_seen": 7159776,
      "step": 460
    },
    {
      "epoch": 1.2959860383944153,
      "grad_norm": 3.241335528071935,
      "learning_rate": 3.3026390341697576e-05,
      "loss": 0.8511,
      "num_input_tokens_seen": 7237720,
      "step": 465
    },
    {
      "epoch": 1.3099476439790576,
      "grad_norm": 2.065109380918334,
      "learning_rate": 3.188441292840587e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 7315704,
      "step": 470
    },
    {
      "epoch": 1.3239092495637,
      "grad_norm": 3.25037907997737,
      "learning_rate": 3.075321254037112e-05,
      "loss": 0.872,
      "num_input_tokens_seen": 7393672,
      "step": 475
    },
    {
      "epoch": 1.337870855148342,
      "grad_norm": 4.772644937407586,
      "learning_rate": 2.963346213260737e-05,
      "loss": 0.8397,
      "num_input_tokens_seen": 7471632,
      "step": 480
    },
    {
      "epoch": 1.3518324607329844,
      "grad_norm": 4.371699254241009,
      "learning_rate": 2.8525827848495913e-05,
      "loss": 0.8254,
      "num_input_tokens_seen": 7549624,
      "step": 485
    },
    {
      "epoch": 1.3657940663176265,
      "grad_norm": 3.1620853722419797,
      "learning_rate": 2.743096862349427e-05,
      "loss": 0.8236,
      "num_input_tokens_seen": 7627568,
      "step": 490
    },
    {
      "epoch": 1.3797556719022688,
      "grad_norm": 5.330846549870451,
      "learning_rate": 2.6349535793133196e-05,
      "loss": 0.8561,
      "num_input_tokens_seen": 7705512,
      "step": 495
    },
    {
      "epoch": 1.3937172774869109,
      "grad_norm": 6.723230790293138,
      "learning_rate": 2.5282172705535013e-05,
      "loss": 0.8198,
      "num_input_tokens_seen": 7783528,
      "step": 500
    },
    {
      "epoch": 1.3937172774869109,
      "eval_loss": 0.846021294593811,
      "eval_runtime": 46.2996,
      "eval_samples_per_second": 3.153,
      "eval_steps_per_second": 0.799,
      "num_input_tokens_seen": 7783528,
      "step": 500
    },
    {
      "epoch": 1.4076788830715532,
      "grad_norm": 3.08479773422221,
      "learning_rate": 2.4229514338683458e-05,
      "loss": 0.8498,
      "num_input_tokens_seen": 7861512,
      "step": 505
    },
    {
      "epoch": 1.4216404886561955,
      "grad_norm": 6.6639723675246385,
      "learning_rate": 2.3192186922673186e-05,
      "loss": 0.8195,
      "num_input_tokens_seen": 7939480,
      "step": 510
    },
    {
      "epoch": 1.4356020942408376,
      "grad_norm": 5.041383098958532,
      "learning_rate": 2.2170807567163294e-05,
      "loss": 0.8428,
      "num_input_tokens_seen": 8017496,
      "step": 515
    },
    {
      "epoch": 1.44956369982548,
      "grad_norm": 6.547003709523484,
      "learning_rate": 2.1165983894256647e-05,
      "loss": 0.8534,
      "num_input_tokens_seen": 8095504,
      "step": 520
    },
    {
      "epoch": 1.4635253054101223,
      "grad_norm": 6.396371201229991,
      "learning_rate": 2.0178313677023425e-05,
      "loss": 0.8113,
      "num_input_tokens_seen": 8173440,
      "step": 525
    },
    {
      "epoch": 1.4774869109947644,
      "grad_norm": 3.194803428296668,
      "learning_rate": 1.9208384483883817e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 8251400,
      "step": 530
    },
    {
      "epoch": 1.4914485165794067,
      "grad_norm": 3.8201513107026552,
      "learning_rate": 1.8256773329061567e-05,
      "loss": 0.8158,
      "num_input_tokens_seen": 8329384,
      "step": 535
    },
    {
      "epoch": 1.505410122164049,
      "grad_norm": 7.081372136918514,
      "learning_rate": 1.732404632931625e-05,
      "loss": 0.8183,
      "num_input_tokens_seen": 8407384,
      "step": 540
    },
    {
      "epoch": 1.5193717277486911,
      "grad_norm": 5.594321718168384,
      "learning_rate": 1.6410758367158385e-05,
      "loss": 0.8364,
      "num_input_tokens_seen": 8485328,
      "step": 545
    },
    {
      "epoch": 1.5333333333333332,
      "grad_norm": 6.74399246622262,
      "learning_rate": 1.5517452760747975e-05,
      "loss": 0.832,
      "num_input_tokens_seen": 8563264,
      "step": 550
    },
    {
      "epoch": 1.5333333333333332,
      "eval_loss": 0.8187811374664307,
      "eval_runtime": 46.4274,
      "eval_samples_per_second": 3.145,
      "eval_steps_per_second": 0.797,
      "num_input_tokens_seen": 8563264,
      "step": 550
    },
    {
      "epoch": 1.5472949389179755,
      "grad_norm": 3.6364413823808968,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 8641240,
      "step": 555
    },
    {
      "epoch": 1.5612565445026179,
      "grad_norm": 4.8432505694035495,
      "learning_rate": 1.3792902133797692e-05,
      "loss": 0.8372,
      "num_input_tokens_seen": 8719256,
      "step": 560
    },
    {
      "epoch": 1.57521815008726,
      "grad_norm": 3.5469277881436545,
      "learning_rate": 1.2962683054376373e-05,
      "loss": 0.8107,
      "num_input_tokens_seen": 8797240,
      "step": 565
    },
    {
      "epoch": 1.5891797556719023,
      "grad_norm": 3.916756694748693,
      "learning_rate": 1.2154497602603703e-05,
      "loss": 0.8472,
      "num_input_tokens_seen": 8875208,
      "step": 570
    },
    {
      "epoch": 1.6031413612565446,
      "grad_norm": 5.159527110013197,
      "learning_rate": 1.13688265707936e-05,
      "loss": 0.8221,
      "num_input_tokens_seen": 8953176,
      "step": 575
    },
    {
      "epoch": 1.6171029668411867,
      "grad_norm": 3.9357730966522637,
      "learning_rate": 1.060613735735384e-05,
      "loss": 0.8076,
      "num_input_tokens_seen": 9031192,
      "step": 580
    },
    {
      "epoch": 1.6310645724258288,
      "grad_norm": 4.296138435306084,
      "learning_rate": 9.86688368872919e-06,
      "loss": 0.7987,
      "num_input_tokens_seen": 9109184,
      "step": 585
    },
    {
      "epoch": 1.6450261780104714,
      "grad_norm": 5.917164796099687,
      "learning_rate": 9.151505349477902e-06,
      "loss": 0.7814,
      "num_input_tokens_seen": 9187136,
      "step": 590
    },
    {
      "epoch": 1.6589877835951135,
      "grad_norm": 6.642916615289702,
      "learning_rate": 8.460427920642423e-06,
      "loss": 0.7907,
      "num_input_tokens_seen": 9265112,
      "step": 595
    },
    {
      "epoch": 1.6729493891797556,
      "grad_norm": 6.455529863705343,
      "learning_rate": 7.794062526569734e-06,
      "loss": 0.786,
      "num_input_tokens_seen": 9343120,
      "step": 600
    },
    {
      "epoch": 1.6729493891797556,
      "eval_loss": 0.8021153211593628,
      "eval_runtime": 46.5141,
      "eval_samples_per_second": 3.139,
      "eval_steps_per_second": 0.795,
      "num_input_tokens_seen": 9343120,
      "step": 600
    },
    {
      "epoch": 1.6869109947643979,
      "grad_norm": 18.00061211274046,
      "learning_rate": 7.152805590332079e-06,
      "loss": 0.7702,
      "num_input_tokens_seen": 9421080,
      "step": 605
    },
    {
      "epoch": 1.7008726003490402,
      "grad_norm": 6.515893934614007,
      "learning_rate": 6.53703859789348e-06,
      "loss": 0.7447,
      "num_input_tokens_seen": 9499048,
      "step": 610
    },
    {
      "epoch": 1.7148342059336823,
      "grad_norm": 8.393536107737633,
      "learning_rate": 5.947127871162456e-06,
      "loss": 0.7943,
      "num_input_tokens_seen": 9577048,
      "step": 615
    },
    {
      "epoch": 1.7287958115183246,
      "grad_norm": 10.522936719634327,
      "learning_rate": 5.383424350065824e-06,
      "loss": 0.7784,
      "num_input_tokens_seen": 9655032,
      "step": 620
    },
    {
      "epoch": 1.742757417102967,
      "grad_norm": 4.303878340242376,
      "learning_rate": 4.846263383773364e-06,
      "loss": 0.8188,
      "num_input_tokens_seen": 9733000,
      "step": 625
    },
    {
      "epoch": 1.756719022687609,
      "grad_norm": 7.6073494826846035,
      "learning_rate": 4.335964531197401e-06,
      "loss": 0.7514,
      "num_input_tokens_seen": 9810984,
      "step": 630
    },
    {
      "epoch": 1.7706806282722511,
      "grad_norm": 4.962378312676078,
      "learning_rate": 3.8528313708861174e-06,
      "loss": 0.8165,
      "num_input_tokens_seen": 9888984,
      "step": 635
    },
    {
      "epoch": 1.7846422338568937,
      "grad_norm": 5.475236961963435,
      "learning_rate": 3.397151320423647e-06,
      "loss": 0.7778,
      "num_input_tokens_seen": 9966984,
      "step": 640
    },
    {
      "epoch": 1.7986038394415358,
      "grad_norm": 7.658362208235343,
      "learning_rate": 2.9691954654443355e-06,
      "loss": 0.7745,
      "num_input_tokens_seen": 10044928,
      "step": 645
    },
    {
      "epoch": 1.812565445026178,
      "grad_norm": 6.541910443575109,
      "learning_rate": 2.5692183983629713e-06,
      "loss": 0.8312,
      "num_input_tokens_seen": 10122936,
      "step": 650
    },
    {
      "epoch": 1.812565445026178,
      "eval_loss": 0.7986289858818054,
      "eval_runtime": 46.4273,
      "eval_samples_per_second": 3.145,
      "eval_steps_per_second": 0.797,
      "num_input_tokens_seen": 10122936,
      "step": 650
    },
    {
      "epoch": 1.8265270506108202,
      "grad_norm": 9.39953604203485,
      "learning_rate": 2.197458066916891e-06,
      "loss": 0.794,
      "num_input_tokens_seen": 10200848,
      "step": 655
    },
    {
      "epoch": 1.8404886561954625,
      "grad_norm": 5.45740552150688,
      "learning_rate": 1.8541356326100433e-06,
      "loss": 0.8129,
      "num_input_tokens_seen": 10278848,
      "step": 660
    },
    {
      "epoch": 1.8544502617801046,
      "grad_norm": 5.2960185422888575,
      "learning_rate": 1.5394553391432143e-06,
      "loss": 0.8057,
      "num_input_tokens_seen": 10356800,
      "step": 665
    },
    {
      "epoch": 1.868411867364747,
      "grad_norm": 4.792059943516234,
      "learning_rate": 1.2536043909088191e-06,
      "loss": 0.7678,
      "num_input_tokens_seen": 10434768,
      "step": 670
    },
    {
      "epoch": 1.8823734729493893,
      "grad_norm": 4.924682676533088,
      "learning_rate": 9.967528416222838e-07,
      "loss": 0.7851,
      "num_input_tokens_seen": 10512752,
      "step": 675
    },
    {
      "epoch": 1.8963350785340314,
      "grad_norm": 7.6405932797853096,
      "learning_rate": 7.690534931565518e-07,
      "loss": 0.7381,
      "num_input_tokens_seen": 10590760,
      "step": 680
    },
    {
      "epoch": 1.9102966841186735,
      "grad_norm": 4.326706788996488,
      "learning_rate": 5.706418046396989e-07,
      "loss": 0.7495,
      "num_input_tokens_seen": 10668752,
      "step": 685
    },
    {
      "epoch": 1.924258289703316,
      "grad_norm": 5.0004961580783815,
      "learning_rate": 4.0163581186984935e-07,
      "loss": 0.7406,
      "num_input_tokens_seen": 10746696,
      "step": 690
    },
    {
      "epoch": 1.9382198952879581,
      "grad_norm": 7.870737498179825,
      "learning_rate": 2.62136057095258e-07,
      "loss": 0.7823,
      "num_input_tokens_seen": 10824664,
      "step": 695
    },
    {
      "epoch": 1.9521815008726002,
      "grad_norm": 6.748561230572036,
      "learning_rate": 1.5222552920138856e-07,
      "loss": 0.7797,
      "num_input_tokens_seen": 10902632,
      "step": 700
    },
    {
      "epoch": 1.9521815008726002,
      "eval_loss": 0.794740617275238,
      "eval_runtime": 46.5422,
      "eval_samples_per_second": 3.137,
      "eval_steps_per_second": 0.795,
      "num_input_tokens_seen": 10902632,
      "step": 700
    },
    {
      "epoch": 1.9661431064572426,
      "grad_norm": 23.84213538429929,
      "learning_rate": 7.196961434052796e-08,
      "loss": 0.8029,
      "num_input_tokens_seen": 10980576,
      "step": 705
    },
    {
      "epoch": 1.9801047120418849,
      "grad_norm": 7.4190814811781145,
      "learning_rate": 2.1416057033352144e-08,
      "loss": 0.7942,
      "num_input_tokens_seen": 11058552,
      "step": 710
    },
    {
      "epoch": 1.994066317626527,
      "grad_norm": 6.100028794501354,
      "learning_rate": 5.949317655462583e-10,
      "loss": 0.7964,
      "num_input_tokens_seen": 11136520,
      "step": 715
    },
    {
      "epoch": 1.9968586387434555,
      "num_input_tokens_seen": 11152104,
      "step": 716,
      "total_flos": 754095660204032.0,
      "train_loss": 0.9049516569136241,
      "train_runtime": 17154.0309,
      "train_samples_per_second": 1.336,
      "train_steps_per_second": 0.042
    }
  ],
  "logging_steps": 5,
  "max_steps": 716,
  "num_input_tokens_seen": 11152104,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 754095660204032.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}