{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 620,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16227180527383367,
      "grad_norm": 84.16447448730469,
      "kl": 0.7359374761581421,
      "learning_rate": 7.258064516129032e-08,
      "logits/chosen": -36952473.6,
      "logits/rejected": -292326604.8,
      "logps/chosen": -899.2125,
      "logps/rejected": -619.225,
      "loss": 0.494,
      "rewards/chosen": 0.0840850830078125,
      "rewards/margins": 0.06186370849609375,
      "rewards/rejected": 0.02222137451171875,
      "step": 10
    },
    {
      "epoch": 0.32454361054766734,
      "grad_norm": 60.70732498168945,
      "kl": 0.30781251192092896,
      "learning_rate": 1.532258064516129e-07,
      "logits/chosen": -46751744.0,
      "logits/rejected": -247562649.6,
      "logps/chosen": -1051.1125,
      "logps/rejected": -544.9375,
      "loss": 0.4834,
      "rewards/chosen": 0.0416412353515625,
      "rewards/margins": 0.09996337890625,
      "rewards/rejected": -0.0583221435546875,
      "step": 20
    },
    {
      "epoch": 0.486815415821501,
      "grad_norm": 43.18465042114258,
      "kl": 0.4087890684604645,
      "learning_rate": 2.3387096774193546e-07,
      "logits/chosen": -55412326.4,
      "logits/rejected": -207881830.4,
      "logps/chosen": -937.7375,
      "logps/rejected": -543.55,
      "loss": 0.4162,
      "rewards/chosen": 0.17140350341796876,
      "rewards/margins": 1861222.9714035036,
      "rewards/rejected": -1861222.8,
      "step": 30
    },
    {
      "epoch": 0.6490872210953347,
      "grad_norm": 48.49949264526367,
      "kl": 6.178906440734863,
      "learning_rate": 3.1451612903225806e-07,
      "logits/chosen": -53701836.8,
      "logits/rejected": -276274380.8,
      "logps/chosen": -980.5875,
      "logps/rejected": -471.225,
      "loss": 0.327,
      "rewards/chosen": 1.36524658203125,
      "rewards/margins": 2.0919921875,
      "rewards/rejected": -0.72674560546875,
      "step": 40
    },
    {
      "epoch": 0.8113590263691683,
      "grad_norm": 28.66044807434082,
      "kl": 28.497461318969727,
      "learning_rate": 3.951612903225806e-07,
      "logits/chosen": -39727513.6,
      "logits/rejected": -272444620.8,
      "logps/chosen": -1023.975,
      "logps/rejected": -532.5125,
      "loss": 0.2801,
      "rewards/chosen": 3.29287109375,
      "rewards/margins": -253437.48212890624,
      "rewards/rejected": 253440.775,
      "step": 50
    },
    {
      "epoch": 0.973630831643002,
      "grad_norm": 17.18927764892578,
      "kl": 73.3125,
      "learning_rate": 4.758064516129032e-07,
      "logits/chosen": -54368870.4,
      "logits/rejected": -200237465.6,
      "logps/chosen": -853.9375,
      "logps/rejected": -491.55,
      "loss": 0.2727,
      "rewards/chosen": 8.0046875,
      "rewards/margins": 2261238.6046875003,
      "rewards/rejected": -2261230.6,
      "step": 60
    },
    {
      "epoch": 1.129817444219067,
      "grad_norm": 9.472125053405762,
      "kl": 98.43506622314453,
      "learning_rate": 4.998058751077704e-07,
      "logits/chosen": -20553196.051948052,
      "logits/rejected": -251854422.44155845,
      "logps/chosen": -842.7792207792207,
      "logps/rejected": -573.1753246753246,
      "loss": 0.2091,
      "rewards/chosen": 9.786525974025974,
      "rewards/margins": 5.313108766233766,
      "rewards/rejected": 4.473417207792208,
      "step": 70
    },
    {
      "epoch": 1.2920892494929006,
      "grad_norm": 8.546073913574219,
      "kl": 122.63749694824219,
      "learning_rate": 4.98855784989581e-07,
      "logits/chosen": -29129113.6,
      "logits/rejected": -209353113.6,
      "logps/chosen": -882.325,
      "logps/rejected": -479.55,
      "loss": 0.1937,
      "rewards/chosen": 11.08515625,
      "rewards/margins": 4.6421875,
      "rewards/rejected": 6.44296875,
      "step": 80
    },
    {
      "epoch": 1.4543610547667343,
      "grad_norm": 8.500472068786621,
      "kl": 121.0999984741211,
      "learning_rate": 4.971170810820278e-07,
      "logits/chosen": -40557363.2,
      "logits/rejected": -195141632.0,
      "logps/chosen": -795.875,
      "logps/rejected": -480.8625,
      "loss": 0.1752,
      "rewards/chosen": 11.70703125,
      "rewards/margins": 1535225.60703125,
      "rewards/rejected": -1535213.9,
      "step": 90
    },
    {
      "epoch": 1.616632860040568,
      "grad_norm": 12.312355995178223,
      "kl": 109.1312484741211,
      "learning_rate": 4.945952732665755e-07,
      "logits/chosen": -8305868.8,
      "logits/rejected": -211968409.6,
      "logps/chosen": -808.3,
      "logps/rejected": -387.16875,
      "loss": 0.1271,
      "rewards/chosen": 11.8703125,
      "rewards/margins": 8.137158203125,
      "rewards/rejected": 3.733154296875,
      "step": 100
    },
    {
      "epoch": 1.7789046653144016,
      "grad_norm": 7.883743762969971,
      "kl": 109.5562515258789,
      "learning_rate": 4.912983530488966e-07,
      "logits/chosen": -37431603.2,
      "logits/rejected": -249792102.4,
      "logps/chosen": -916.45625,
      "logps/rejected": -489.2125,
      "loss": 0.147,
      "rewards/chosen": 11.49296875,
      "rewards/margins": -253333.48203125,
      "rewards/rejected": 253344.975,
      "step": 110
    },
    {
      "epoch": 1.9411764705882353,
      "grad_norm": 3.3018791675567627,
      "kl": 114.1500015258789,
      "learning_rate": 4.872367682341172e-07,
      "logits/chosen": -64476160.0,
      "logits/rejected": -196329881.6,
      "logps/chosen": -797.3875,
      "logps/rejected": -489.9875,
      "loss": 0.1421,
      "rewards/chosen": 12.90703125,
      "rewards/margins": 9.448437499999999,
      "rewards/rejected": 3.45859375,
      "step": 120
    },
    {
      "epoch": 2.0973630831643,
      "grad_norm": 13.774515151977539,
      "kl": 92.26298522949219,
      "learning_rate": 4.824233898181509e-07,
      "logits/chosen": -42439241.14285714,
      "logits/rejected": -251607172.98701298,
      "logps/chosen": -861.1233766233767,
      "logps/rejected": -612.8311688311688,
      "loss": 0.1248,
      "rewards/chosen": 11.994318181818182,
      "rewards/margins": 2648996.7215909087,
      "rewards/rejected": -2648984.727272727,
      "step": 130
    },
    {
      "epoch": 2.259634888438134,
      "grad_norm": 11.28150463104248,
      "kl": 78.5999984741211,
      "learning_rate": 4.76873471200039e-07,
      "logits/chosen": -43557273.6,
      "logits/rejected": -245417984.0,
      "logps/chosen": -808.3875,
      "logps/rejected": -577.90625,
      "loss": 0.1185,
      "rewards/chosen": 9.87734375,
      "rewards/margins": 9.4081298828125,
      "rewards/rejected": 0.4692138671875,
      "step": 140
    },
    {
      "epoch": 2.4219066937119678,
      "grad_norm": 6.963021755218506,
      "kl": 72.09375,
      "learning_rate": 4.7060459984455474e-07,
      "logits/chosen": -70427648.0,
      "logits/rejected": -230709657.6,
      "logps/chosen": -924.20625,
      "logps/rejected": -505.775,
      "loss": 0.1058,
      "rewards/chosen": 10.9953125,
      "rewards/margins": 1436882.7953125,
      "rewards/rejected": -1436871.8,
      "step": 150
    },
    {
      "epoch": 2.584178498985801,
      "grad_norm": 11.5512056350708,
      "kl": 59.21875,
      "learning_rate": 4.6363664154824735e-07,
      "logits/chosen": -75585536.0,
      "logits/rejected": -257813708.8,
      "logps/chosen": -759.025,
      "logps/rejected": -496.0625,
      "loss": 0.095,
      "rewards/chosen": 10.133203125,
      "rewards/margins": 11.517364501953125,
      "rewards/rejected": -1.384161376953125,
      "step": 160
    },
    {
      "epoch": 2.7464503042596347,
      "grad_norm": 16.846378326416016,
      "kl": 42.296875,
      "learning_rate": 4.5599167748554633e-07,
      "logits/chosen": -99328000.0,
      "logits/rejected": -318544281.6,
      "logps/chosen": -964.01875,
      "logps/rejected": -570.0125,
      "loss": 0.1034,
      "rewards/chosen": 8.240234375,
      "rewards/margins": 304753.265234375,
      "rewards/rejected": -304745.025,
      "step": 170
    },
    {
      "epoch": 2.9087221095334685,
      "grad_norm": 7.493341445922852,
      "kl": 31.861719131469727,
      "learning_rate": 4.476939342344246e-07,
      "logits/chosen": -141419724.8,
      "logits/rejected": -281131417.6,
      "logps/chosen": -841.39375,
      "logps/rejected": -558.1875,
      "loss": 0.0987,
      "rewards/chosen": 7.83671875,
      "rewards/margins": -186562.66328125,
      "rewards/rejected": 186570.5,
      "step": 180
    },
    {
      "epoch": 3.0649087221095335,
      "grad_norm": 7.3059892654418945,
      "kl": 38.8863639831543,
      "learning_rate": 4.387697070033649e-07,
      "logits/chosen": -149665499.42857143,
      "logits/rejected": -332706696.3116883,
      "logps/chosen": -899.012987012987,
      "logps/rejected": -628.8181818181819,
      "loss": 0.0836,
      "rewards/chosen": 10.13271103896104,
      "rewards/margins": 2618391.2236201297,
      "rewards/rejected": -2618381.090909091,
      "step": 190
    },
    {
      "epoch": 3.227180527383367,
      "grad_norm": 3.7938599586486816,
      "kl": 54.193748474121094,
      "learning_rate": 4.2924727630292124e-07,
      "logits/chosen": -99653632.0,
      "logits/rejected": -326130892.8,
      "logps/chosen": -845.28125,
      "logps/rejected": -579.2875,
      "loss": 0.0948,
      "rewards/chosen": 9.30859375,
      "rewards/margins": 11.3248046875,
      "rewards/rejected": -2.0162109375,
      "step": 200
    },
    {
      "epoch": 3.389452332657201,
      "grad_norm": 6.6560492515563965,
      "kl": 59.662498474121094,
      "learning_rate": 4.1915681832593934e-07,
      "logits/chosen": -138940006.4,
      "logits/rejected": -283273625.6,
      "logps/chosen": -865.6625,
      "logps/rejected": -578.7625,
      "loss": 0.0752,
      "rewards/chosen": 10.1765625,
      "rewards/margins": 1564678.5765624999,
      "rewards/rejected": -1564668.4,
      "step": 210
    },
    {
      "epoch": 3.5517241379310347,
      "grad_norm": 7.188765525817871,
      "kl": 39.96953201293945,
      "learning_rate": 4.085303093204377e-07,
      "logits/chosen": -145563238.4,
      "logits/rejected": -305337139.2,
      "logps/chosen": -854.525,
      "logps/rejected": -579.0125,
      "loss": 0.0652,
      "rewards/chosen": 9.2734375,
      "rewards/margins": 12.3808837890625,
      "rewards/rejected": -3.1074462890625,
      "step": 220
    },
    {
      "epoch": 3.713995943204868,
      "grad_norm": 9.175395011901855,
      "kl": 26.307811737060547,
      "learning_rate": 3.974014242581871e-07,
      "logits/chosen": -173807206.4,
      "logits/rejected": -361424486.4,
      "logps/chosen": -956.9375,
      "logps/rejected": -474.825,
      "loss": 0.0615,
      "rewards/chosen": 7.969140625,
      "rewards/margins": 536591.069140625,
      "rewards/rejected": -536583.1,
      "step": 230
    },
    {
      "epoch": 3.8762677484787016,
      "grad_norm": 7.2789306640625,
      "kl": 35.146873474121094,
      "learning_rate": 3.858054301201047e-07,
      "logits/chosen": -169469542.4,
      "logits/rejected": -343206297.6,
      "logps/chosen": -873.7625,
      "logps/rejected": -612.525,
      "loss": 0.0585,
      "rewards/chosen": 8.37109375,
      "rewards/margins": -114471.84140625,
      "rewards/rejected": 114480.2125,
      "step": 240
    },
    {
      "epoch": 4.032454361054767,
      "grad_norm": 7.370634078979492,
      "kl": 37.38230514526367,
      "learning_rate": 3.7377907413663576e-07,
      "logits/chosen": -175320077.2987013,
      "logits/rejected": -325758178.0779221,
      "logps/chosen": -908.3181818181819,
      "logps/rejected": -593.3116883116883,
      "loss": 0.0588,
      "rewards/chosen": 9.853896103896103,
      "rewards/margins": 2647334.529220779,
      "rewards/rejected": -2647324.675324675,
      "step": 250
    },
    {
      "epoch": 4.1947261663286,
      "grad_norm": 20.8977108001709,
      "kl": 22.143749237060547,
      "learning_rate": 3.613604673372861e-07,
      "logits/chosen": -160774553.6,
      "logits/rejected": -361404825.6,
      "logps/chosen": -796.53125,
      "logps/rejected": -669.5625,
      "loss": 0.0639,
      "rewards/chosen": 8.620703125,
      "rewards/margins": 12.99384765625,
      "rewards/rejected": -4.37314453125,
      "step": 260
    },
    {
      "epoch": 4.356997971602434,
      "grad_norm": 8.267151832580566,
      "kl": 14.619531631469727,
      "learning_rate": 3.4858896377832965e-07,
      "logits/chosen": -194864332.8,
      "logits/rejected": -343074406.4,
      "logps/chosen": -909.8375,
      "logps/rejected": -613.9625,
      "loss": 0.0528,
      "rewards/chosen": 8.049609375,
      "rewards/margins": 1472961.049609375,
      "rewards/rejected": -1472953.0,
      "step": 270
    },
    {
      "epoch": 4.519269776876268,
      "grad_norm": 6.327075004577637,
      "kl": 20.830469131469727,
      "learning_rate": 3.355050358314172e-07,
      "logits/chosen": -216122982.4,
      "logits/rejected": -317764403.2,
      "logps/chosen": -820.175,
      "logps/rejected": -619.325,
      "loss": 0.0507,
      "rewards/chosen": 8.112890625,
      "rewards/margins": 13.183007812500001,
      "rewards/rejected": -5.0701171875,
      "step": 280
    },
    {
      "epoch": 4.681541582150102,
      "grad_norm": 3.9251317977905273,
      "kl": 18.582813262939453,
      "learning_rate": 3.2215014592828766e-07,
      "logits/chosen": -225445478.4,
      "logits/rejected": -377572556.8,
      "logps/chosen": -922.80625,
      "logps/rejected": -511.4125,
      "loss": 0.0355,
      "rewards/chosen": 8.7359375,
      "rewards/margins": 684053.0359375001,
      "rewards/rejected": -684044.3,
      "step": 290
    },
    {
      "epoch": 4.8438133874239355,
      "grad_norm": 2.7528088092803955,
      "kl": 32.806251525878906,
      "learning_rate": 3.085666151680205e-07,
      "logits/chosen": -194420736.0,
      "logits/rejected": -402663014.4,
      "logps/chosen": -906.09375,
      "logps/rejected": -579.35,
      "loss": 0.0304,
      "rewards/chosen": 9.00625,
      "rewards/margins": 22823.6671875,
      "rewards/rejected": -22814.6609375,
      "step": 300
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.467460632324219,
      "kl": 34.149349212646484,
      "learning_rate": 2.947974892032063e-07,
      "logits/chosen": -231085042.7012987,
      "logits/rejected": -335724756.77922076,
      "logps/chosen": -854.9805194805194,
      "logps/rejected": -645.4155844155844,
      "loss": 0.0361,
      "rewards/chosen": 10.591720779220779,
      "rewards/margins": 2898001.8644480524,
      "rewards/rejected": -2897991.272727273,
      "step": 310
    },
    {
      "epoch": 5.162271805273834,
      "grad_norm": 15.31023120880127,
      "kl": 17.678905487060547,
      "learning_rate": 2.8088640183003664e-07,
      "logits/chosen": -220064153.6,
      "logits/rejected": -414279270.4,
      "logps/chosen": -786.05,
      "logps/rejected": -687.275,
      "loss": 0.0426,
      "rewards/chosen": 8.684375,
      "rewards/margins": 13.99609375,
      "rewards/rejected": -5.31171875,
      "step": 320
    },
    {
      "epoch": 5.324543610547668,
      "grad_norm": 12.307633399963379,
      "kl": 4.917187690734863,
      "learning_rate": 2.668774367145913e-07,
      "logits/chosen": -235716608.0,
      "logits/rejected": -369033216.0,
      "logps/chosen": -952.45,
      "logps/rejected": -636.8875,
      "loss": 0.0409,
      "rewards/chosen": 7.369140625,
      "rewards/margins": 14.8421875,
      "rewards/rejected": -7.473046875,
      "step": 330
    },
    {
      "epoch": 5.486815415821501,
      "grad_norm": 5.51188850402832,
      "kl": 5.789843559265137,
      "learning_rate": 2.5281498769350646e-07,
      "logits/chosen": -236973260.8,
      "logits/rejected": -331428659.2,
      "logps/chosen": -838.6375,
      "logps/rejected": -630.125,
      "loss": 0.0264,
      "rewards/chosen": 7.7671875,
      "rewards/margins": 1459860.0671875,
      "rewards/rejected": -1459852.3,
      "step": 340
    },
    {
      "epoch": 5.649087221095335,
      "grad_norm": 14.79988956451416,
      "kl": 5.79296875,
      "learning_rate": 2.3874361809172427e-07,
      "logits/chosen": -255219302.4,
      "logits/rejected": -402902220.8,
      "logps/chosen": -896.85,
      "logps/rejected": -556.4,
      "loss": 0.0206,
      "rewards/chosen": 8.0015625,
      "rewards/margins": 15.924218750000001,
      "rewards/rejected": -7.92265625,
      "step": 350
    },
    {
      "epoch": 5.811359026369169,
      "grad_norm": 1.1963369846343994,
      "kl": 5.693749904632568,
      "learning_rate": 2.2470791950314196e-07,
      "logits/chosen": -237377945.6,
      "logits/rejected": -411195801.6,
      "logps/chosen": -967.6,
      "logps/rejected": -634.9125,
      "loss": 0.0207,
      "rewards/chosen": 7.40546875,
      "rewards/margins": 891114.00546875,
      "rewards/rejected": -891106.6,
      "step": 360
    },
    {
      "epoch": 5.973630831643002,
      "grad_norm": 0.8492076396942139,
      "kl": 14.582812309265137,
      "learning_rate": 2.1075237048167677e-07,
      "logits/chosen": -262532300.8,
      "logits/rejected": -366804992.0,
      "logps/chosen": -832.14375,
      "logps/rejected": -621.4625,
      "loss": 0.0228,
      "rewards/chosen": 9.843359375,
      "rewards/margins": 3079304.6433593747,
      "rewards/rejected": -3079294.8,
      "step": 370
    },
    {
      "epoch": 6.129817444219067,
      "grad_norm": 14.829185485839844,
      "kl": 13.705357551574707,
      "learning_rate": 1.96921195590551e-07,
      "logits/chosen": -247062208.83116883,
      "logits/rejected": -430358740.77922076,
      "logps/chosen": -851.1038961038961,
      "logps/rejected": -713.3376623376623,
      "loss": 0.0316,
      "rewards/chosen": 9.212256493506494,
      "rewards/margins": 15.109577922077923,
      "rewards/rejected": -5.897321428571429,
      "step": 380
    },
    {
      "epoch": 6.292089249492901,
      "grad_norm": 13.538140296936035,
      "kl": 4.005468845367432,
      "learning_rate": 1.8325822525646205e-07,
      "logits/chosen": -268140544.0,
      "logits/rejected": -391171276.8,
      "logps/chosen": -932.825,
      "logps/rejected": -651.425,
      "loss": 0.0245,
      "rewards/chosen": 7.41875,
      "rewards/margins": 14.955078125,
      "rewards/rejected": -7.536328125,
      "step": 390
    },
    {
      "epoch": 6.454361054766734,
      "grad_norm": 3.61026930809021,
      "kl": 2.526562452316284,
      "learning_rate": 1.6980675687275614e-07,
      "logits/chosen": -282066944.0,
      "logits/rejected": -389788467.2,
      "logps/chosen": -851.1,
      "logps/rejected": -648.5375,
      "loss": 0.02,
      "rewards/chosen": 7.39765625,
      "rewards/margins": 1656471.09765625,
      "rewards/rejected": -1656463.7,
      "step": 400
    },
    {
      "epoch": 6.616632860040568,
      "grad_norm": 0.35545381903648376,
      "kl": 6.8046875,
      "learning_rate": 1.5660941759176157e-07,
      "logits/chosen": -249454592.0,
      "logits/rejected": -395798118.4,
      "logps/chosen": -850.96875,
      "logps/rejected": -529.025,
      "loss": 0.0107,
      "rewards/chosen": 8.4625,
      "rewards/margins": 16.0609375,
      "rewards/rejected": -7.5984375,
      "step": 410
    },
    {
      "epoch": 6.778904665314402,
      "grad_norm": 0.6669896245002747,
      "kl": 8.045312881469727,
      "learning_rate": 1.4370802924108988e-07,
      "logits/chosen": -258266726.4,
      "logits/rejected": -430876262.4,
      "logps/chosen": -961.0375,
      "logps/rejected": -622.3625,
      "loss": 0.0184,
      "rewards/chosen": 8.0828125,
      "rewards/margins": 1007446.3828125,
      "rewards/rejected": -1007438.3,
      "step": 420
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 0.8302009701728821,
      "kl": 6.962500095367432,
      "learning_rate": 1.3114347579197618e-07,
      "logits/chosen": -283987148.8,
      "logits/rejected": -371962675.2,
      "logps/chosen": -852.0,
      "logps/rejected": -631.225,
      "loss": 0.0145,
      "rewards/chosen": 8.53125,
      "rewards/margins": 16.691015625,
      "rewards/rejected": -8.159765625,
      "step": 430
    },
    {
      "epoch": 7.0973630831643,
      "grad_norm": 0.6516138911247253,
      "kl": 10.787337303161621,
      "learning_rate": 1.1895557379965004e-07,
      "logits/chosen": -267412413.5064935,
      "logits/rejected": -424823076.5714286,
      "logps/chosen": -890.7727272727273,
      "logps/rejected": -724.6883116883117,
      "loss": 0.0223,
      "rewards/chosen": 9.612012987012987,
      "rewards/margins": 3236774.702922078,
      "rewards/rejected": -3236765.090909091,
      "step": 440
    },
    {
      "epoch": 7.259634888438134,
      "grad_norm": 21.117584228515625,
      "kl": 12.122655868530273,
      "learning_rate": 1.0718294622630186e-07,
      "logits/chosen": -270696448.0,
      "logits/rejected": -412673638.4,
      "logps/chosen": -829.05,
      "logps/rejected": -671.0625,
      "loss": 0.0188,
      "rewards/chosen": 8.44765625,
      "rewards/margins": 14.951171875,
      "rewards/rejected": -6.503515625,
      "step": 450
    },
    {
      "epoch": 7.421906693711968,
      "grad_norm": 3.9178707599639893,
      "kl": 4.1484375,
      "learning_rate": 9.586290004649867e-08,
      "logits/chosen": -286366105.6,
      "logits/rejected": -389460787.2,
      "logps/chosen": -958.95625,
      "logps/rejected": -601.0625,
      "loss": 0.0151,
      "rewards/chosen": 8.26640625,
      "rewards/margins": 1740058.6664062499,
      "rewards/rejected": -1740050.4,
      "step": 460
    },
    {
      "epoch": 7.584178498985802,
      "grad_norm": 2.008082151412964,
      "kl": 2.745312452316284,
      "learning_rate": 8.503130802290861e-08,
      "logits/chosen": -276156416.0,
      "logits/rejected": -394657792.0,
      "logps/chosen": -786.85,
      "logps/rejected": -588.85,
      "loss": 0.0099,
      "rewards/chosen": 7.89765625,
      "rewards/margins": 16.633984375,
      "rewards/rejected": -8.736328125,
      "step": 470
    },
    {
      "epoch": 7.746450304259635,
      "grad_norm": 0.24131633341312408,
      "kl": 5.1953125,
      "learning_rate": 7.472249502698685e-08,
      "logits/chosen": -269829734.4,
      "logits/rejected": -436561510.4,
      "logps/chosen": -967.49375,
      "logps/rejected": -638.0625,
      "loss": 0.0157,
      "rewards/chosen": 7.919921875,
      "rewards/margins": 883133.219921875,
      "rewards/rejected": -883125.3,
      "step": 480
    },
    {
      "epoch": 7.9087221095334685,
      "grad_norm": 2.947861909866333,
      "kl": 6.618750095367432,
      "learning_rate": 6.496912926486278e-08,
      "logits/chosen": -284753920.0,
      "logits/rejected": -377408716.8,
      "logps/chosen": -833.10625,
      "logps/rejected": -611.9375,
      "loss": 0.0083,
      "rewards/chosen": 8.5046875,
      "rewards/margins": 183315.91718750002,
      "rewards/rejected": -183307.4125,
      "step": 490
    },
    {
      "epoch": 8.064908722109534,
      "grad_norm": 5.739405155181885,
      "kl": 9.308441162109375,
      "learning_rate": 5.580211875313345e-08,
      "logits/chosen": -287616226.0779221,
      "logits/rejected": -422971046.23376626,
      "logps/chosen": -899.9415584415584,
      "logps/rejected": -680.6233766233767,
      "loss": 0.0178,
      "rewards/chosen": 10.097402597402597,
      "rewards/margins": 3274245.9415584416,
      "rewards/rejected": -3274235.8441558443,
      "step": 500
    },
    {
      "epoch": 8.227180527383368,
      "grad_norm": 6.1757097244262695,
      "kl": 10.469531059265137,
      "learning_rate": 4.725051337262475e-08,
      "logits/chosen": -251002880.0,
      "logits/rejected": -420275814.4,
      "logps/chosen": -854.85625,
      "logps/rejected": -645.3125,
      "loss": 0.0152,
      "rewards/chosen": 8.628125,
      "rewards/margins": 15.3453125,
      "rewards/rejected": -6.7171875,
      "step": 510
    },
    {
      "epoch": 8.3894523326572,
      "grad_norm": 5.071248531341553,
      "kl": 7.877343654632568,
      "learning_rate": 3.934141281050909e-08,
      "logits/chosen": -291821977.6,
      "logits/rejected": -385135411.2,
      "logps/chosen": -886.45,
      "logps/rejected": -652.85,
      "loss": 0.0155,
      "rewards/chosen": 8.568359375,
      "rewards/margins": 1805589.568359375,
      "rewards/rejected": -1805581.0,
      "step": 520
    },
    {
      "epoch": 8.551724137931034,
      "grad_norm": 0.6050207614898682,
      "kl": 5.170312404632568,
      "learning_rate": 3.209988068250688e-08,
      "logits/chosen": -285185638.4,
      "logits/rejected": -395824332.8,
      "logps/chosen": -863.9625,
      "logps/rejected": -640.1625,
      "loss": 0.0084,
      "rewards/chosen": 8.50703125,
      "rewards/margins": 16.423046875,
      "rewards/rejected": -7.916015625,
      "step": 530
    },
    {
      "epoch": 8.713995943204868,
      "grad_norm": 2.260059118270874,
      "kl": 3.5218749046325684,
      "learning_rate": 2.5548865107314604e-08,
      "logits/chosen": -286359552.0,
      "logits/rejected": -428421939.2,
      "logps/chosen": -954.11875,
      "logps/rejected": -522.375,
      "loss": 0.0129,
      "rewards/chosen": 8.103125,
      "rewards/margins": 902798.203125,
      "rewards/rejected": -902790.1,
      "step": 540
    },
    {
      "epoch": 8.876267748478702,
      "grad_norm": 1.0818209648132324,
      "kl": 4.739062309265137,
      "learning_rate": 1.970912598495689e-08,
      "logits/chosen": -280064819.2,
      "logits/rejected": -416330547.2,
      "logps/chosen": -878.925,
      "logps/rejected": -671.425,
      "loss": 0.008,
      "rewards/chosen": 7.967578125,
      "rewards/margins": 222435.842578125,
      "rewards/rejected": -222427.875,
      "step": 550
    },
    {
      "epoch": 9.032454361054766,
      "grad_norm": 2.1863319873809814,
      "kl": 5.1055192947387695,
      "learning_rate": 1.4599169209513568e-08,
      "logits/chosen": -289883601.45454544,
      "logits/rejected": -403381740.0519481,
      "logps/chosen": -919.012987012987,
      "logps/rejected": -651.961038961039,
      "loss": 0.0142,
      "rewards/chosen": 8.921266233766234,
      "rewards/margins": 3287868.6615259736,
      "rewards/rejected": -3287859.74025974,
      "step": 560
    },
    {
      "epoch": 9.1947261663286,
      "grad_norm": 2.410034656524658,
      "kl": 4.432812690734863,
      "learning_rate": 1.023518802469947e-08,
      "logits/chosen": -269074432.0,
      "logits/rejected": -430492876.8,
      "logps/chosen": -797.04375,
      "logps/rejected": -712.25,
      "loss": 0.0147,
      "rewards/chosen": 8.553125,
      "rewards/margins": 15.996875,
      "rewards/rejected": -7.44375,
      "step": 570
    },
    {
      "epoch": 9.356997971602434,
      "grad_norm": 1.298397421836853,
      "kl": 5.707812309265137,
      "learning_rate": 6.631011708138207e-09,
      "logits/chosen": -292247961.6,
      "logits/rejected": -404599603.2,
      "logps/chosen": -903.8875,
      "logps/rejected": -643.45,
      "loss": 0.0152,
      "rewards/chosen": 8.49375,
      "rewards/margins": 1818702.09375,
      "rewards/rejected": -1818693.6,
      "step": 580
    },
    {
      "epoch": 9.519269776876268,
      "grad_norm": 0.7316223382949829,
      "kl": 5.643750190734863,
      "learning_rate": 3.798061746947995e-09,
      "logits/chosen": -298290380.8,
      "logits/rejected": -368705536.0,
      "logps/chosen": -820.0125,
      "logps/rejected": -653.4375,
      "loss": 0.009,
      "rewards/chosen": 8.12421875,
      "rewards/margins": 15.853515625,
      "rewards/rejected": -7.729296875,
      "step": 590
    },
    {
      "epoch": 9.681541582150102,
      "grad_norm": 0.2766137421131134,
      "kl": 4.55859375,
      "learning_rate": 1.7453156435165983e-09,
      "logits/chosen": -298279731.2,
      "logits/rejected": -418748825.6,
      "logps/chosen": -924.475,
      "logps/rejected": -543.5875,
      "loss": 0.0058,
      "rewards/chosen": 8.64140625,
      "rewards/margins": 902797.3414062499,
      "rewards/rejected": -902788.7,
      "step": 600
    },
    {
      "epoch": 9.843813387423936,
      "grad_norm": 0.1658095270395279,
      "kl": 5.104687690734863,
      "learning_rate": 4.792784661642458e-10,
      "logits/chosen": -260448256.0,
      "logits/rejected": -439471308.8,
      "logps/chosen": -921.39375,
      "logps/rejected": -623.55,
      "loss": 0.0136,
      "rewards/chosen": 7.892578125,
      "rewards/margins": 219158.467578125,
      "rewards/rejected": -219150.575,
      "step": 610
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.4314670264720917,
      "kl": 6.086039066314697,
      "learning_rate": 3.962234848359225e-12,
      "logits/chosen": -294411543.27272725,
      "logits/rejected": -375213175.6883117,
      "logps/chosen": -868.5064935064935,
      "logps/rejected": -691.6623376623377,
      "loss": 0.0102,
      "rewards/chosen": 9.39163961038961,
      "rewards/margins": 3321915.885146104,
      "rewards/rejected": -3321906.4935064935,
      "step": 620
    }
  ],
  "logging_steps": 10,
  "max_steps": 620,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 200,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}