{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 5.0,
  "global_step": 79,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "advantage": -3.92578125,
      "epoch": 0.012779552715654952,
      "grad_norm": 1766.3862790437397,
      "importance_weight": 1.0,
      "learning_rate": 3.333333333333333e-07,
      "loss": 48.151039123535156,
      "reverse_kl": 3.92578125,
      "reward": 3.92578125,
      "step": 1
    },
    {
      "advantage": -2.927734375,
      "epoch": 0.025559105431309903,
      "grad_norm": 2509.801170673312,
      "importance_weight": 1.0,
      "learning_rate": 6.666666666666666e-07,
      "loss": 41.717323303222656,
      "reverse_kl": 2.927734375,
      "reward": 2.927734375,
      "step": 2
    },
    {
      "advantage": -3.84375,
      "epoch": 0.038338658146964855,
      "grad_norm": 2440.9886995025977,
      "importance_weight": 1.0,
      "learning_rate": 1e-06,
      "loss": 45.08679962158203,
      "reverse_kl": 3.84375,
      "reward": 3.84375,
      "step": 3
    },
    {
      "advantage": -3.13671875,
      "epoch": 0.051118210862619806,
      "grad_norm": 3114.6850101681343,
      "importance_weight": 1.0,
      "learning_rate": 9.995728791936505e-07,
      "loss": 45.681129455566406,
      "reverse_kl": 3.13671875,
      "reward": 3.13671875,
      "step": 4
    },
    {
      "advantage": -3.94921875,
      "epoch": 0.06389776357827476,
      "grad_norm": 4276.076605236566,
      "importance_weight": 1.0,
      "learning_rate": 9.982922465033348e-07,
      "loss": 46.051109313964844,
      "reverse_kl": 3.94921875,
      "reward": 3.94921875,
      "step": 5
    },
    {
      "advantage": -3.9453125,
      "epoch": 0.07667731629392971,
      "grad_norm": 1775.2570763305876,
      "importance_weight": 1.0,
      "learning_rate": 9.961602898685223e-07,
      "loss": 46.91913986206055,
      "reverse_kl": 3.9453125,
      "reward": 3.9453125,
      "step": 6
    },
    {
      "advantage": -3.24609375,
      "epoch": 0.08945686900958466,
      "grad_norm": 1263.5670027747772,
      "importance_weight": 1.0,
      "learning_rate": 9.931806517013612e-07,
      "loss": 45.04924011230469,
      "reverse_kl": 3.24609375,
      "reward": 3.24609375,
      "step": 7
    },
    {
      "advantage": -4.11328125,
      "epoch": 0.10223642172523961,
      "grad_norm": 3392.4545892833858,
      "importance_weight": 1.0,
      "learning_rate": 9.893584226636772e-07,
      "loss": 47.20243835449219,
      "reverse_kl": 4.11328125,
      "reward": 4.11328125,
      "step": 8
    },
    {
      "advantage": -3.07421875,
      "epoch": 0.11501597444089456,
      "grad_norm": 2288.276537721676,
      "importance_weight": 1.0,
      "learning_rate": 9.847001329696652e-07,
      "loss": 45.77381134033203,
      "reverse_kl": 3.07421875,
      "reward": 3.07421875,
      "step": 9
    },
    {
      "advantage": -3.74609375,
      "epoch": 0.12779552715654952,
      "grad_norm": 1868.1392071694281,
      "importance_weight": 1.0,
      "learning_rate": 9.792137412291263e-07,
      "loss": 43.440433502197266,
      "reverse_kl": 3.74609375,
      "reward": 3.74609375,
      "step": 10
    },
    {
      "advantage": -3.85546875,
      "epoch": 0.14057507987220447,
      "grad_norm": 2507.1600102448983,
      "importance_weight": 1.0,
      "learning_rate": 9.729086208503173e-07,
      "loss": 45.46769332885742,
      "reverse_kl": 3.85546875,
      "reward": 3.85546875,
      "step": 11
    },
    {
      "advantage": -2.63623046875,
      "epoch": 0.15335463258785942,
      "grad_norm": 2451.1971108618945,
      "importance_weight": 1.0,
      "learning_rate": 9.657955440256395e-07,
      "loss": 45.64551544189453,
      "reverse_kl": 2.63623046875,
      "reward": 2.63623046875,
      "step": 12
    },
    {
      "advantage": -3.16796875,
      "epoch": 0.16613418530351437,
      "grad_norm": 2632.70424289851,
      "importance_weight": 1.0,
      "learning_rate": 9.578866633275286e-07,
      "loss": 37.2404899597168,
      "reverse_kl": 3.16796875,
      "reward": 3.16796875,
      "step": 13
    },
    {
      "advantage": -2.37890625,
      "epoch": 0.17891373801916932,
      "grad_norm": 2583.407007373188,
      "importance_weight": 1.0,
      "learning_rate": 9.491954909459894e-07,
      "loss": 31.012496948242188,
      "reverse_kl": 2.37890625,
      "reward": 2.37890625,
      "step": 14
    },
    {
      "advantage": -2.8515625,
      "epoch": 0.19169329073482427,
      "grad_norm": 2263.2260347179504,
      "importance_weight": 1.0,
      "learning_rate": 9.397368756032444e-07,
      "loss": 25.766422271728516,
      "reverse_kl": 2.8515625,
      "reward": 2.8515625,
      "step": 15
    },
    {
      "advantage": -1.126220703125,
      "epoch": 0.20447284345047922,
      "grad_norm": 2678.5821915540587,
      "importance_weight": 1.0,
      "learning_rate": 9.295269771849425e-07,
      "loss": 25.585540771484375,
      "reverse_kl": 1.126220703125,
      "reward": 1.126220703125,
      "step": 16
    },
    {
      "advantage": -0.64129638671875,
      "epoch": 0.21725239616613418,
      "grad_norm": 2429.312747247921,
      "importance_weight": 1.0,
      "learning_rate": 9.185832391312642e-07,
      "loss": 16.185579299926758,
      "reverse_kl": 0.64129638671875,
      "reward": 0.64129638671875,
      "step": 17
    },
    {
      "advantage": -0.745819091796875,
      "epoch": 0.23003194888178913,
      "grad_norm": 2155.968556544084,
      "importance_weight": 1.0,
      "learning_rate": 9.069243586350975e-07,
      "loss": 13.251764297485352,
      "reverse_kl": 0.745819091796875,
      "reward": 0.745819091796875,
      "step": 18
    },
    {
      "advantage": -0.5244140625,
      "epoch": 0.24281150159744408,
      "grad_norm": 2233.5797863173366,
      "importance_weight": 1.0,
      "learning_rate": 8.945702546981968e-07,
      "loss": 14.515754699707031,
      "reverse_kl": 0.5244140625,
      "reward": 0.5244140625,
      "step": 19
    },
    {
      "advantage": -0.5096435546875,
      "epoch": 0.25559105431309903,
      "grad_norm": 1326.620853496907,
      "importance_weight": 1.0,
      "learning_rate": 8.815420340999033e-07,
      "loss": 9.711706161499023,
      "reverse_kl": 0.5096435546875,
      "reward": 0.5096435546875,
      "step": 20
    },
    {
      "advantage": -0.7093505859375,
      "epoch": 0.268370607028754,
      "grad_norm": 918.7432079807351,
      "importance_weight": 1.0,
      "learning_rate": 8.678619553365658e-07,
      "loss": 4.557295799255371,
      "reverse_kl": 0.7093505859375,
      "reward": 0.7093505859375,
      "step": 21
    },
    {
      "advantage": -0.5733413696289062,
      "epoch": 0.28115015974440893,
      "grad_norm": 959.052727538278,
      "importance_weight": 1.0,
      "learning_rate": 8.535533905932737e-07,
      "loss": 5.451420783996582,
      "reverse_kl": 0.5733413696289062,
      "reward": 0.5733413696289062,
      "step": 22
    },
    {
      "advantage": 0.018230915069580078,
      "epoch": 0.2939297124600639,
      "grad_norm": 3131.793657478231,
      "importance_weight": 1.0,
      "learning_rate": 8.386407858128706e-07,
      "loss": 7.077840328216553,
      "reverse_kl": -0.018230915069580078,
      "reward": -0.018230915069580078,
      "step": 23
    },
    {
      "advantage": 0.022310495376586914,
      "epoch": 0.30670926517571884,
      "grad_norm": 480.814504695948,
      "importance_weight": 1.0,
      "learning_rate": 8.231496189304704e-07,
      "loss": 2.99735164642334,
      "reverse_kl": -0.022310495376586914,
      "reward": -0.022310495376586914,
      "step": 24
    },
    {
      "advantage": -0.143890380859375,
      "epoch": 0.3194888178913738,
      "grad_norm": 1645.0396015205918,
      "importance_weight": 1.0,
      "learning_rate": 8.071063563448339e-07,
      "loss": 0.4990193545818329,
      "reverse_kl": 0.143890380859375,
      "reward": 0.143890380859375,
      "step": 25
    },
    {
      "advantage": -0.053829342126846313,
      "epoch": 0.33226837060702874,
      "grad_norm": 1272.7167133433024,
      "importance_weight": 1.0,
      "learning_rate": 7.905384077009692e-07,
      "loss": 1.8080556392669678,
      "reverse_kl": 0.053829342126846313,
      "reward": 0.053829342126846313,
      "step": 26
    },
    {
      "advantage": 0.032070159912109375,
      "epoch": 0.3450479233226837,
      "grad_norm": 899.6348865035224,
      "importance_weight": 1.0,
      "learning_rate": 7.734740790612136e-07,
      "loss": 3.544450283050537,
      "reverse_kl": -0.032070159912109375,
      "reward": -0.032070159912109375,
      "step": 27
    },
    {
      "advantage": -0.072998046875,
      "epoch": 0.35782747603833864,
      "grad_norm": 2149.9142909936318,
      "importance_weight": 1.0,
      "learning_rate": 7.559425245448005e-07,
      "loss": 4.3909807205200195,
      "reverse_kl": 0.072998046875,
      "reward": 0.072998046875,
      "step": 28
    },
    {
      "advantage": -1.0737258195877075,
      "epoch": 0.3706070287539936,
      "grad_norm": 6026.541569255804,
      "importance_weight": 1.0,
      "learning_rate": 7.379736965185368e-07,
      "loss": 2.915527582168579,
      "reverse_kl": 1.0737258195877075,
      "reward": 1.0737258195877075,
      "step": 29
    },
    {
      "advantage": -0.3123030960559845,
      "epoch": 0.38338658146964855,
      "grad_norm": 1283.800763346129,
      "importance_weight": 1.0,
      "learning_rate": 7.195982944236852e-07,
      "loss": 3.1713125705718994,
      "reverse_kl": 0.3123030960559845,
      "reward": 0.3123030960559845,
      "step": 30
    },
    {
      "advantage": -0.1704559326171875,
      "epoch": 0.3961661341853035,
      "grad_norm": 1137.5039769653895,
      "importance_weight": 1.0,
      "learning_rate": 7.008477123264847e-07,
      "loss": 2.7144887447357178,
      "reverse_kl": 0.1704559326171875,
      "reward": 0.1704559326171875,
      "step": 31
    },
    {
      "advantage": -0.14324569702148438,
      "epoch": 0.40894568690095845,
      "grad_norm": 9005.346122172677,
      "importance_weight": 1.0,
      "learning_rate": 6.817539852819148e-07,
      "loss": 4.307586669921875,
      "reverse_kl": 0.14324569702148438,
      "reward": 0.14324569702148438,
      "step": 32
    },
    {
      "advantage": 0.014892578125,
      "epoch": 0.4217252396166134,
      "grad_norm": 1033.1215742066095,
      "importance_weight": 1.0,
      "learning_rate": 6.623497346023417e-07,
      "loss": 1.4840457439422607,
      "reverse_kl": -0.014892578125,
      "reward": -0.014892578125,
      "step": 33
    },
    {
      "advantage": 0.1416250839829445,
      "epoch": 0.43450479233226835,
      "grad_norm": 1430.7158391206654,
      "importance_weight": 1.0,
      "learning_rate": 6.426681121245527e-07,
      "loss": 1.9607133865356445,
      "reverse_kl": -0.1416250839829445,
      "reward": -0.1416250839829445,
      "step": 34
    },
    {
      "advantage": -0.24991704523563385,
      "epoch": 0.4472843450479233,
      "grad_norm": 1266.6505487937372,
      "importance_weight": 1.0,
      "learning_rate": 6.227427435703995e-07,
      "loss": 4.122557640075684,
      "reverse_kl": 0.24991704523563385,
      "reward": 0.24991704523563385,
      "step": 35
    },
    {
      "advantage": -0.7578125,
      "epoch": 0.46006389776357826,
      "grad_norm": 2803.3547918878144,
      "importance_weight": 1.0,
      "learning_rate": 6.026076710978171e-07,
      "loss": 4.029263019561768,
      "reverse_kl": 0.7578125,
      "reward": 0.7578125,
      "step": 36
    },
    {
      "advantage": -0.524169921875,
      "epoch": 0.4728434504792332,
      "grad_norm": 940.5745157654247,
      "importance_weight": 1.0,
      "learning_rate": 5.82297295140367e-07,
      "loss": 3.470961093902588,
      "reverse_kl": 0.524169921875,
      "reward": 0.524169921875,
      "step": 37
    },
    {
      "advantage": -0.16888809204101562,
      "epoch": 0.48562300319488816,
      "grad_norm": 2673.884020259494,
      "importance_weight": 1.0,
      "learning_rate": 5.618463156346739e-07,
      "loss": 4.502054214477539,
      "reverse_kl": 0.16888809204101562,
      "reward": 0.16888809204101562,
      "step": 38
    },
    {
      "advantage": -0.58636474609375,
      "epoch": 0.4984025559105431,
      "grad_norm": 2193.250131807019,
      "importance_weight": 1.0,
      "learning_rate": 5.412896727361662e-07,
      "loss": 2.874572992324829,
      "reverse_kl": 0.58636474609375,
      "reward": 0.58636474609375,
      "step": 39
    },
    {
      "advantage": 0.0633249282836914,
      "epoch": 0.5111821086261981,
      "grad_norm": 943.4161743228711,
      "importance_weight": 1.0,
      "learning_rate": 5.206624871244065e-07,
      "loss": 1.3326880931854248,
      "reverse_kl": -0.0633249282836914,
      "reward": -0.0633249282836914,
      "step": 40
    },
    {
      "advantage": -0.25299072265625,
      "epoch": 0.5239616613418531,
      "grad_norm": 2196.148346791048,
      "importance_weight": 1.0,
      "learning_rate": 5e-07,
      "loss": 4.25025749206543,
      "reverse_kl": 0.25299072265625,
      "reward": 0.25299072265625,
      "step": 41
    },
    {
      "advantage": 0.0015718936920166016,
      "epoch": 0.536741214057508,
      "grad_norm": 732.8143262835811,
      "importance_weight": 1.0,
      "learning_rate": 4.793375128755933e-07,
      "loss": 2.3447933197021484,
      "reverse_kl": -0.0015718936920166016,
      "reward": -0.0015718936920166016,
      "step": 42
    },
    {
      "advantage": -0.12131786346435547,
      "epoch": 0.549520766773163,
      "grad_norm": 2242.00593726535,
      "importance_weight": 1.0,
      "learning_rate": 4.5871032726383385e-07,
      "loss": 1.5312654972076416,
      "reverse_kl": 0.12131786346435547,
      "reward": 0.12131786346435547,
      "step": 43
    },
    {
      "advantage": -0.305877685546875,
      "epoch": 0.5623003194888179,
      "grad_norm": 1136.0092305842322,
      "importance_weight": 1.0,
      "learning_rate": 4.381536843653261e-07,
      "loss": 3.839193344116211,
      "reverse_kl": 0.305877685546875,
      "reward": 0.305877685546875,
      "step": 44
    },
    {
      "advantage": -0.16869544982910156,
      "epoch": 0.5750798722044729,
      "grad_norm": 1581.728010442836,
      "importance_weight": 1.0,
      "learning_rate": 4.1770270485963294e-07,
      "loss": 4.090610980987549,
      "reverse_kl": 0.16869544982910156,
      "reward": 0.16869544982910156,
      "step": 45
    },
    {
      "advantage": 0.12481689453125,
      "epoch": 0.5878594249201278,
      "grad_norm": 958.0526951211081,
      "importance_weight": 1.0,
      "learning_rate": 3.973923289021829e-07,
      "loss": 2.3874318599700928,
      "reverse_kl": -0.12481689453125,
      "reward": -0.12481689453125,
      "step": 46
    },
    {
      "advantage": -0.09581756591796875,
      "epoch": 0.6006389776357828,
      "grad_norm": 1340.7939718360853,
      "importance_weight": 1.0,
      "learning_rate": 3.772572564296004e-07,
      "loss": 1.3532946109771729,
      "reverse_kl": 0.09581756591796875,
      "reward": 0.09581756591796875,
      "step": 47
    },
    {
      "advantage": -0.2921600341796875,
      "epoch": 0.6134185303514377,
      "grad_norm": 2813.341942965456,
      "importance_weight": 1.0,
      "learning_rate": 3.5733188787544746e-07,
      "loss": 2.3117501735687256,
      "reverse_kl": 0.2921600341796875,
      "reward": 0.2921600341796875,
      "step": 48
    },
    {
      "advantage": 0.02086639404296875,
      "epoch": 0.6261980830670927,
      "grad_norm": 484.24089550941784,
      "importance_weight": 1.0,
      "learning_rate": 3.3765026539765827e-07,
      "loss": 1.2950079441070557,
      "reverse_kl": -0.02086639404296875,
      "reward": -0.02086639404296875,
      "step": 49
    },
    {
      "advantage": -0.0614471435546875,
      "epoch": 0.6389776357827476,
      "grad_norm": 1461.479792219086,
      "importance_weight": 1.0,
      "learning_rate": 3.1824601471808497e-07,
      "loss": 3.096958637237549,
      "reverse_kl": 0.0614471435546875,
      "reward": 0.0614471435546875,
      "step": 50
    },
    {
      "advantage": -0.06619381904602051,
      "epoch": 0.6517571884984026,
      "grad_norm": 2000.1416857576135,
      "importance_weight": 1.0,
      "learning_rate": 2.9915228767351535e-07,
      "loss": 5.684658050537109,
      "reverse_kl": 0.06619381904602051,
      "reward": 0.06619381904602051,
      "step": 51
    },
    {
      "advantage": -0.6048431396484375,
      "epoch": 0.6645367412140575,
      "grad_norm": 5582.848235693696,
      "importance_weight": 1.0,
      "learning_rate": 2.8040170557631485e-07,
      "loss": 2.930215358734131,
      "reverse_kl": 0.6048431396484375,
      "reward": 0.6048431396484375,
      "step": 52
    },
    {
      "advantage": -0.2693181037902832,
      "epoch": 0.6773162939297125,
      "grad_norm": 1321.6539287352055,
      "importance_weight": 1.0,
      "learning_rate": 2.620263034814632e-07,
      "loss": 2.3553826808929443,
      "reverse_kl": 0.2693181037902832,
      "reward": 0.2693181037902832,
      "step": 53
    },
    {
      "advantage": -0.1400146484375,
      "epoch": 0.6900958466453674,
      "grad_norm": 2308.030886933676,
      "importance_weight": 1.0,
      "learning_rate": 2.440574754551996e-07,
      "loss": 1.4683387279510498,
      "reverse_kl": 0.1400146484375,
      "reward": 0.1400146484375,
      "step": 54
    },
    {
      "advantage": 0.1058807373046875,
      "epoch": 0.7028753993610224,
      "grad_norm": 1394.4296745563015,
      "importance_weight": 1.0,
      "learning_rate": 2.2652592093878665e-07,
      "loss": 2.3433170318603516,
      "reverse_kl": -0.1058807373046875,
      "reward": -0.1058807373046875,
      "step": 55
    },
    {
      "advantage": -0.15239191055297852,
      "epoch": 0.7156549520766773,
      "grad_norm": 1934.9536869786018,
      "importance_weight": 1.0,
      "learning_rate": 2.0946159229903088e-07,
      "loss": 2.8346152305603027,
      "reverse_kl": 0.15239191055297852,
      "reward": 0.15239191055297852,
      "step": 56
    },
    {
      "advantage": -0.4003019332885742,
      "epoch": 0.7284345047923323,
      "grad_norm": 2455.118744097761,
      "importance_weight": 1.0,
      "learning_rate": 1.9289364365516607e-07,
      "loss": 2.331786632537842,
      "reverse_kl": 0.4003019332885742,
      "reward": 0.4003019332885742,
      "step": 57
    },
    {
      "advantage": -0.0343017578125,
      "epoch": 0.7412140575079872,
      "grad_norm": 2098.7018731096477,
      "importance_weight": 1.0,
      "learning_rate": 1.768503810695295e-07,
      "loss": 3.555081367492676,
      "reverse_kl": 0.0343017578125,
      "reward": 0.0343017578125,
      "step": 58
    },
    {
      "advantage": 0.0009593963623046875,
      "epoch": 0.7539936102236422,
      "grad_norm": 600.6726102343208,
      "importance_weight": 1.0,
      "learning_rate": 1.6135921418712955e-07,
      "loss": 1.0303959846496582,
      "reverse_kl": -0.0009593963623046875,
      "reward": -0.0009593963623046875,
      "step": 59
    },
    {
      "advantage": -0.240631103515625,
      "epoch": 0.7667731629392971,
      "grad_norm": 1945.6859172276784,
      "importance_weight": 1.0,
      "learning_rate": 1.4644660940672627e-07,
      "loss": 3.855902910232544,
      "reverse_kl": 0.240631103515625,
      "reward": 0.240631103515625,
      "step": 60
    },
    {
      "advantage": 0.11822509765625,
      "epoch": 0.7795527156549521,
      "grad_norm": 1065.389130516611,
      "importance_weight": 1.0,
      "learning_rate": 1.321380446634342e-07,
      "loss": 2.0869712829589844,
      "reverse_kl": -0.11822509765625,
      "reward": -0.11822509765625,
      "step": 61
    },
    {
      "advantage": -0.009402468800544739,
      "epoch": 0.792332268370607,
      "grad_norm": 1558.7229527128636,
      "importance_weight": 1.0,
      "learning_rate": 1.1845796590009683e-07,
      "loss": 5.005650520324707,
      "reverse_kl": 0.009402468800544739,
      "reward": 0.009402468800544739,
      "step": 62
    },
    {
      "advantage": 0.00140380859375,
      "epoch": 0.805111821086262,
      "grad_norm": 1262.339264743914,
      "importance_weight": 1.0,
      "learning_rate": 1.0542974530180327e-07,
      "loss": 1.710388422012329,
      "reverse_kl": -0.00140380859375,
      "reward": -0.00140380859375,
      "step": 63
    },
    {
      "advantage": -0.1585693359375,
      "epoch": 0.8178913738019169,
      "grad_norm": 956.6598014756286,
      "importance_weight": 1.0,
      "learning_rate": 9.307564136490254e-08,
      "loss": 2.5448691844940186,
      "reverse_kl": 0.1585693359375,
      "reward": 0.1585693359375,
      "step": 64
    },
    {
      "advantage": -0.826171875,
      "epoch": 0.8306709265175719,
      "grad_norm": 1865.801779087802,
      "importance_weight": 1.0,
      "learning_rate": 8.141676086873573e-08,
      "loss": 4.3569440841674805,
      "reverse_kl": 0.826171875,
      "reward": 0.826171875,
      "step": 65
    },
    {
      "advantage": -0.0063018798828125,
      "epoch": 0.8434504792332268,
      "grad_norm": 1027.3810566963693,
      "importance_weight": 1.0,
      "learning_rate": 7.047302281505735e-08,
      "loss": 2.0769460201263428,
      "reverse_kl": 0.0063018798828125,
      "reward": 0.0063018798828125,
      "step": 66
    },
    {
      "advantage": -0.05352973937988281,
      "epoch": 0.8562300319488818,
      "grad_norm": 321.61226654551314,
      "importance_weight": 1.0,
      "learning_rate": 6.026312439675551e-08,
      "loss": 0.9115631580352783,
      "reverse_kl": 0.05352973937988281,
      "reward": 0.05352973937988281,
      "step": 67
    },
    {
      "advantage": -0.037384033203125,
      "epoch": 0.8690095846645367,
      "grad_norm": 565.8598867705407,
      "importance_weight": 1.0,
      "learning_rate": 5.080450905401057e-08,
      "loss": 2.5955116748809814,
      "reverse_kl": 0.037384033203125,
      "reward": 0.037384033203125,
      "step": 68
    },
    {
      "advantage": 0.01715666800737381,
      "epoch": 0.8817891373801917,
      "grad_norm": 1548.5297190041179,
      "importance_weight": 1.0,
      "learning_rate": 4.2113336672471245e-08,
      "loss": 2.921511650085449,
      "reverse_kl": -0.01715666800737381,
      "reward": -0.01715666800737381,
      "step": 69
    },
    {
      "advantage": -0.48827362060546875,
      "epoch": 0.8945686900958466,
      "grad_norm": 1129.5806146465313,
      "importance_weight": 1.0,
      "learning_rate": 3.420445597436056e-08,
      "loss": 3.740647554397583,
      "reverse_kl": 0.48827362060546875,
      "reward": 0.48827362060546875,
      "step": 70
    },
    {
      "advantage": -0.06817626953125,
      "epoch": 0.9073482428115016,
      "grad_norm": 690.8545848293572,
      "importance_weight": 1.0,
      "learning_rate": 2.7091379149682682e-08,
      "loss": 1.2684175968170166,
      "reverse_kl": 0.06817626953125,
      "reward": 0.06817626953125,
      "step": 71
    },
    {
      "advantage": -0.15771491592749953,
      "epoch": 0.9201277955271565,
      "grad_norm": 2301.1837059655113,
      "importance_weight": 1.0,
      "learning_rate": 2.0786258770873645e-08,
      "loss": 1.5807130336761475,
      "reverse_kl": 0.15771491592749953,
      "reward": 0.15771491592749953,
      "step": 72
    },
    {
      "advantage": -0.2563939094543457,
      "epoch": 0.9329073482428115,
      "grad_norm": 606.8915564417606,
      "importance_weight": 1.0,
      "learning_rate": 1.5299867030334813e-08,
      "loss": 1.5268313884735107,
      "reverse_kl": 0.2563939094543457,
      "reward": 0.2563939094543457,
      "step": 73
    },
    {
      "advantage": -0.1568593755364418,
      "epoch": 0.9456869009584664,
      "grad_norm": 877.2732697019068,
      "importance_weight": 1.0,
      "learning_rate": 1.0641577336322761e-08,
      "loss": 2.0240747928619385,
      "reverse_kl": 0.1568593755364418,
      "reward": 0.1568593755364418,
      "step": 74
    },
    {
      "advantage": -0.21515464782714844,
      "epoch": 0.9584664536741214,
      "grad_norm": 1394.657901482745,
      "importance_weight": 1.0,
      "learning_rate": 6.819348298638839e-09,
      "loss": 2.1968772411346436,
      "reverse_kl": 0.21515464782714844,
      "reward": 0.21515464782714844,
      "step": 75
    },
    {
      "advantage": -0.4272308349609375,
      "epoch": 0.9712460063897763,
      "grad_norm": 664.7715685534853,
      "importance_weight": 1.0,
      "learning_rate": 3.8397101314774915e-09,
      "loss": 3.3966917991638184,
      "reverse_kl": 0.4272308349609375,
      "reward": 0.4272308349609375,
      "step": 76
    },
    {
      "advantage": -0.285400390625,
      "epoch": 0.9840255591054313,
      "grad_norm": 1453.0775811658589,
      "importance_weight": 1.0,
      "learning_rate": 1.7077534966650765e-09,
      "loss": 2.557816982269287,
      "reverse_kl": 0.285400390625,
      "reward": 0.285400390625,
      "step": 77
    },
    {
      "advantage": 0.060047149658203125,
      "epoch": 0.9968051118210862,
      "grad_norm": 1197.0430107864017,
      "importance_weight": 1.0,
      "learning_rate": 4.271208063494902e-10,
      "loss": 2.7366576194763184,
      "reverse_kl": -0.060047149658203125,
      "reward": -0.060047149658203125,
      "step": 78
    },
    {
      "advantage": 0.044921875,
      "epoch": 1.0,
      "grad_norm": 1197.0430107864017,
      "importance_weight": 1.0,
      "learning_rate": 0.0,
      "loss": -1.370888113975525,
      "reverse_kl": -0.044921875,
      "reward": -0.044921875,
      "step": 79
    }
  ],
  "logging_steps": 1,
  "max_steps": 79,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}