{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 782,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 6.329113924050633e-09,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -403.3199157714844,
      "logps/real": -443.6107177734375,
      "loss": 4.7453,
      "rewards/accuracies": 0.5,
      "rewards/generated": -12.943833351135254,
      "rewards/margins": -2.53641414642334,
      "rewards/real": -15.48024845123291,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.329113924050633e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -487.2853088378906,
      "logps/real": -384.8033142089844,
      "loss": 3.4516,
      "rewards/accuracies": 0.6111111044883728,
      "rewards/generated": -19.51491928100586,
      "rewards/margins": 6.083561420440674,
      "rewards/real": -13.431358337402344,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.2658227848101266e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -473.446044921875,
      "logps/real": -413.171630859375,
      "loss": 3.9705,
      "rewards/accuracies": 0.637499988079071,
      "rewards/generated": -18.367900848388672,
      "rewards/margins": 3.9024269580841064,
      "rewards/real": -14.465472221374512,
      "step": 20
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.89873417721519e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -510.4585876464844,
      "logps/real": -408.74212646484375,
      "loss": 3.6955,
      "rewards/accuracies": 0.6875,
      "rewards/generated": -21.059778213500977,
      "rewards/margins": 7.762242317199707,
      "rewards/real": -13.297533988952637,
      "step": 30
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.5316455696202533e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -451.734130859375,
      "logps/real": -368.16998291015625,
      "loss": 3.9056,
      "rewards/accuracies": 0.625,
      "rewards/generated": -16.805126190185547,
      "rewards/margins": 4.4605584144592285,
      "rewards/real": -12.344568252563477,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.1645569620253163e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -552.9141845703125,
      "logps/real": -397.95318603515625,
      "loss": 2.9742,
      "rewards/accuracies": 0.737500011920929,
      "rewards/generated": -25.695674896240234,
      "rewards/margins": 10.653547286987305,
      "rewards/real": -15.042126655578613,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.79746835443038e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -509.5240783691406,
      "logps/real": -390.6533203125,
      "loss": 2.5682,
      "rewards/accuracies": 0.737500011920929,
      "rewards/generated": -22.568603515625,
      "rewards/margins": 7.904494285583496,
      "rewards/real": -14.664111137390137,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.4303797468354424e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -554.5638427734375,
      "logps/real": -382.84832763671875,
      "loss": 1.7793,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/generated": -28.215290069580078,
      "rewards/margins": 14.945714950561523,
      "rewards/real": -13.269571304321289,
      "step": 70
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.992887624466572e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -574.03759765625,
      "logps/real": -402.69232177734375,
      "loss": 1.4194,
      "rewards/accuracies": 0.800000011920929,
      "rewards/generated": -28.118602752685547,
      "rewards/margins": 12.523630142211914,
      "rewards/real": -15.594972610473633,
      "step": 80
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.92176386913229e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -577.7903442382812,
      "logps/real": -401.81561279296875,
      "loss": 1.2635,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/generated": -30.364093780517578,
      "rewards/margins": 14.755389213562012,
      "rewards/real": -15.608701705932617,
      "step": 90
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.850640113798008e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -654.4276733398438,
      "logps/real": -442.893798828125,
      "loss": 0.9213,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/generated": -35.57740020751953,
      "rewards/margins": 18.7423038482666,
      "rewards/real": -16.835100173950195,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.779516358463727e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -637.9244384765625,
      "logps/real": -426.7471618652344,
      "loss": 1.0798,
      "rewards/accuracies": 0.875,
      "rewards/generated": -35.024356842041016,
      "rewards/margins": 18.848251342773438,
      "rewards/real": -16.17610740661621,
      "step": 110
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.7083926031294454e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -647.3193969726562,
      "logps/real": -419.98095703125,
      "loss": 0.7428,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -35.827938079833984,
      "rewards/margins": 21.43227767944336,
      "rewards/real": -14.395665168762207,
      "step": 120
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.6372688477951633e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -647.6121826171875,
      "logps/real": -413.42401123046875,
      "loss": 0.6566,
      "rewards/accuracies": 0.875,
      "rewards/generated": -35.408538818359375,
      "rewards/margins": 20.003904342651367,
      "rewards/real": -15.404635429382324,
      "step": 130
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.5661450924608817e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -697.9991455078125,
      "logps/real": -401.5552673339844,
      "loss": 0.9421,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -38.38452911376953,
      "rewards/margins": 23.595943450927734,
      "rewards/real": -14.788581848144531,
      "step": 140
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.4950213371266e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -672.8363037109375,
      "logps/real": -401.61590576171875,
      "loss": 0.6587,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/generated": -38.686241149902344,
      "rewards/margins": 23.072139739990234,
      "rewards/real": -15.614102363586426,
      "step": 150
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.4238975817923186e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -641.0380859375,
      "logps/real": -377.9221496582031,
      "loss": 0.6261,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/generated": -35.72499084472656,
      "rewards/margins": 21.804719924926758,
      "rewards/real": -13.920272827148438,
      "step": 160
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.3527738264580364e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -676.1116943359375,
      "logps/real": -410.857421875,
      "loss": 0.6189,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/generated": -37.884979248046875,
      "rewards/margins": 23.008445739746094,
      "rewards/real": -14.876535415649414,
      "step": 170
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.2816500711237554e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -654.6505737304688,
      "logps/real": -370.24224853515625,
      "loss": 0.3247,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -36.68430709838867,
      "rewards/margins": 23.027376174926758,
      "rewards/real": -13.656933784484863,
      "step": 180
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.2105263157894733e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -665.4993286132812,
      "logps/real": -364.58880615234375,
      "loss": 0.5709,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -37.38275146484375,
      "rewards/margins": 25.380136489868164,
      "rewards/real": -12.002609252929688,
      "step": 190
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.1394025604551917e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -684.380615234375,
      "logps/real": -392.02288818359375,
      "loss": 0.6823,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/generated": -38.82280349731445,
      "rewards/margins": 24.800228118896484,
      "rewards/real": -14.02257251739502,
      "step": 200
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.06827880512091e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -741.0132446289062,
      "logps/real": -383.6390075683594,
      "loss": 0.5369,
      "rewards/accuracies": 0.925000011920929,
      "rewards/generated": -43.17304229736328,
      "rewards/margins": 28.80326271057129,
      "rewards/real": -14.369776725769043,
      "step": 210
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.9971550497866285e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -665.7184448242188,
      "logps/real": -401.6400146484375,
      "loss": 0.3696,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -38.721866607666016,
      "rewards/margins": 24.154865264892578,
      "rewards/real": -14.567001342773438,
      "step": 220
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.926031294452347e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -681.5037231445312,
      "logps/real": -394.03509521484375,
      "loss": 0.2532,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -38.438758850097656,
      "rewards/margins": 23.877817153930664,
      "rewards/real": -14.560938835144043,
      "step": 230
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.8549075391180653e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -705.4946899414062,
      "logps/real": -414.25146484375,
      "loss": 0.4719,
      "rewards/accuracies": 0.925000011920929,
      "rewards/generated": -40.05225372314453,
      "rewards/margins": 24.665658950805664,
      "rewards/real": -15.386594772338867,
      "step": 240
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.783783783783784e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -718.469970703125,
      "logps/real": -393.9696350097656,
      "loss": 0.326,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -43.078765869140625,
      "rewards/margins": 28.980609893798828,
      "rewards/real": -14.09815788269043,
      "step": 250
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.7126600284495016e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -721.9268798828125,
      "logps/real": -403.2998046875,
      "loss": 0.4804,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -41.29940414428711,
      "rewards/margins": 26.352609634399414,
      "rewards/real": -14.946797370910645,
      "step": 260
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.6415362731152206e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -690.9553833007812,
      "logps/real": -374.5037536621094,
      "loss": 0.3827,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -40.69682693481445,
      "rewards/margins": 26.715112686157227,
      "rewards/real": -13.981710433959961,
      "step": 270
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.5704125177809385e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -725.0104370117188,
      "logps/real": -406.488037109375,
      "loss": 0.4101,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -42.66962432861328,
      "rewards/margins": 27.242467880249023,
      "rewards/real": -15.427154541015625,
      "step": 280
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.4992887624466574e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -697.3206787109375,
      "logps/real": -400.65753173828125,
      "loss": 0.3629,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -39.595970153808594,
      "rewards/margins": 24.689983367919922,
      "rewards/real": -14.90599250793457,
      "step": 290
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.4281650071123753e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -711.7612915039062,
      "logps/real": -388.0057678222656,
      "loss": 0.2122,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -42.39947509765625,
      "rewards/margins": 27.109222412109375,
      "rewards/real": -15.290254592895508,
      "step": 300
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.3570412517780937e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -713.0751953125,
      "logps/real": -423.7640686035156,
      "loss": 0.3657,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/generated": -41.861183166503906,
      "rewards/margins": 26.213199615478516,
      "rewards/real": -15.647982597351074,
      "step": 310
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.285917496443812e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -695.4622192382812,
      "logps/real": -386.4324645996094,
      "loss": 0.3005,
      "rewards/accuracies": 0.925000011920929,
      "rewards/generated": -39.45234298706055,
      "rewards/margins": 25.696590423583984,
      "rewards/real": -13.755752563476562,
      "step": 320
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.2147937411095305e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -774.9369506835938,
      "logps/real": -426.2724609375,
      "loss": 0.2325,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -48.17813491821289,
      "rewards/margins": 31.29937171936035,
      "rewards/real": -16.87876319885254,
      "step": 330
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.1436699857752484e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -714.9234008789062,
      "logps/real": -395.0069885253906,
      "loss": 0.2981,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -41.547645568847656,
      "rewards/margins": 27.00480079650879,
      "rewards/real": -14.54284381866455,
      "step": 340
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.0725462304409674e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -721.8656005859375,
      "logps/real": -400.1606750488281,
      "loss": 0.2019,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -44.031803131103516,
      "rewards/margins": 28.76715087890625,
      "rewards/real": -15.264646530151367,
      "step": 350
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.001422475106685e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -728.89599609375,
      "logps/real": -470.1971130371094,
      "loss": 0.5931,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -43.573814392089844,
      "rewards/margins": 24.9804630279541,
      "rewards/real": -18.593351364135742,
      "step": 360
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.9302987197724037e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -667.4353637695312,
      "logps/real": -369.68328857421875,
      "loss": 0.4031,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -38.71464920043945,
      "rewards/margins": 24.013246536254883,
      "rewards/real": -14.701400756835938,
      "step": 370
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.8591749644381226e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -708.6209716796875,
      "logps/real": -395.7604675292969,
      "loss": 0.3691,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -42.240318298339844,
      "rewards/margins": 26.32097816467285,
      "rewards/real": -15.919347763061523,
      "step": 380
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.7880512091038405e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -766.4454345703125,
      "logps/real": -423.30755615234375,
      "loss": 0.2758,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -47.435791015625,
      "rewards/margins": 30.477214813232422,
      "rewards/real": -16.958572387695312,
      "step": 390
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.716927453769559e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -768.6455078125,
      "logps/real": -449.0997009277344,
      "loss": 0.3455,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -47.01398849487305,
      "rewards/margins": 28.302906036376953,
      "rewards/real": -18.711084365844727,
      "step": 400
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.6458036984352773e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -734.2765502929688,
      "logps/real": -447.5357971191406,
      "loss": 0.375,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -46.2476921081543,
      "rewards/margins": 27.413455963134766,
      "rewards/real": -18.834239959716797,
      "step": 410
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.574679943100996e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -730.5106201171875,
      "logps/real": -453.9371643066406,
      "loss": 0.0816,
      "rewards/accuracies": 0.925000011920929,
      "rewards/generated": -45.583648681640625,
      "rewards/margins": 27.334802627563477,
      "rewards/real": -18.248844146728516,
      "step": 420
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.5035561877667136e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -766.5603637695312,
      "logps/real": -438.89788818359375,
      "loss": 0.1735,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -46.9077262878418,
      "rewards/margins": 29.52130126953125,
      "rewards/real": -17.386432647705078,
      "step": 430
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.4324324324324326e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -781.421630859375,
      "logps/real": -405.1501770019531,
      "loss": 0.3965,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -49.6665153503418,
      "rewards/margins": 33.62762451171875,
      "rewards/real": -16.03889274597168,
      "step": 440
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.3613086770981507e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -760.2469482421875,
      "logps/real": -404.4407653808594,
      "loss": 0.4128,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -47.4620361328125,
      "rewards/margins": 30.4351863861084,
      "rewards/real": -17.026851654052734,
      "step": 450
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.290184921763869e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -820.1843872070312,
      "logps/real": -420.02587890625,
      "loss": 0.1306,
      "rewards/accuracies": 1.0,
      "rewards/generated": -51.192138671875,
      "rewards/margins": 35.38946533203125,
      "rewards/real": -15.8026704788208,
      "step": 460
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.2190611664295875e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -759.2930908203125,
      "logps/real": -438.888916015625,
      "loss": 0.2487,
      "rewards/accuracies": 0.925000011920929,
      "rewards/generated": -45.420188903808594,
      "rewards/margins": 27.34377670288086,
      "rewards/real": -18.0764102935791,
      "step": 470
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.1479374110953057e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -777.7584838867188,
      "logps/real": -388.6899108886719,
      "loss": 0.3922,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -48.69025421142578,
      "rewards/margins": 32.126991271972656,
      "rewards/real": -16.563264846801758,
      "step": 480
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.076813655761024e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -804.2822875976562,
      "logps/real": -438.558837890625,
      "loss": 0.2558,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -50.539390563964844,
      "rewards/margins": 31.528533935546875,
      "rewards/real": -19.0108585357666,
      "step": 490
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.0056899004267425e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -799.6202392578125,
      "logps/real": -483.55096435546875,
      "loss": 0.2945,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -50.81739807128906,
      "rewards/margins": 30.668746948242188,
      "rewards/real": -20.14865493774414,
      "step": 500
    },
    {
      "epoch": 0.64,
      "eval_logits/generated": -Infinity,
      "eval_logits/real": -Infinity,
      "eval_logps/generated": -509.32861328125,
      "eval_logps/real": -313.0279541015625,
      "eval_loss": 0.1748354285955429,
      "eval_rewards/accuracies": 0.9442675113677979,
      "eval_rewards/generated": -21.789350509643555,
      "eval_rewards/margins": 15.725247383117676,
      "eval_rewards/real": -6.0641021728515625,
      "eval_runtime": 590.026,
      "eval_samples_per_second": 8.474,
      "eval_steps_per_second": 0.266,
      "step": 500
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.9345661450924607e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -757.3658447265625,
      "logps/real": -412.8812561035156,
      "loss": 0.2523,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -45.53099822998047,
      "rewards/margins": 29.076038360595703,
      "rewards/real": -16.454959869384766,
      "step": 510
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.863442389758179e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -780.4212036132812,
      "logps/real": -437.5557556152344,
      "loss": 0.2147,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -49.51309585571289,
      "rewards/margins": 32.93183517456055,
      "rewards/real": -16.581256866455078,
      "step": 520
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.7923186344238975e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -735.9564208984375,
      "logps/real": -407.6651306152344,
      "loss": 0.2975,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -44.98466491699219,
      "rewards/margins": 29.423254013061523,
      "rewards/real": -15.561413764953613,
      "step": 530
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.721194879089616e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -807.4033203125,
      "logps/real": -429.4000549316406,
      "loss": 0.2254,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -51.46739959716797,
      "rewards/margins": 32.75578689575195,
      "rewards/real": -18.711612701416016,
      "step": 540
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.650071123755334e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -811.77099609375,
      "logps/real": -425.84649658203125,
      "loss": 0.3105,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -52.33929443359375,
      "rewards/margins": 34.459083557128906,
      "rewards/real": -17.88020896911621,
      "step": 550
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5789473684210525e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -788.242919921875,
      "logps/real": -440.695556640625,
      "loss": 0.1702,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -49.456298828125,
      "rewards/margins": 31.41439437866211,
      "rewards/real": -18.041906356811523,
      "step": 560
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.507823613086771e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -752.26611328125,
      "logps/real": -427.10113525390625,
      "loss": 0.3281,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -47.79071807861328,
      "rewards/margins": 30.950246810913086,
      "rewards/real": -16.84047508239746,
      "step": 570
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.436699857752489e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -826.0186767578125,
      "logps/real": -424.66680908203125,
      "loss": 0.3042,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -51.909454345703125,
      "rewards/margins": 34.6719856262207,
      "rewards/real": -17.237468719482422,
      "step": 580
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.3655761024182077e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -816.61962890625,
      "logps/real": -420.100830078125,
      "loss": 0.2635,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -52.91407012939453,
      "rewards/margins": 36.203983306884766,
      "rewards/real": -16.710086822509766,
      "step": 590
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.2944523470839261e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -820.4493408203125,
      "logps/real": -428.4378967285156,
      "loss": 0.347,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -51.51692581176758,
      "rewards/margins": 34.30039596557617,
      "rewards/real": -17.216527938842773,
      "step": 600
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.2233285917496443e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -768.8092041015625,
      "logps/real": -420.62200927734375,
      "loss": 0.0588,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -47.871437072753906,
      "rewards/margins": 32.33845901489258,
      "rewards/real": -15.532976150512695,
      "step": 610
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.1522048364153626e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -786.0013427734375,
      "logps/real": -442.2757873535156,
      "loss": 0.2483,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -50.278663635253906,
      "rewards/margins": 33.101016998291016,
      "rewards/real": -17.17764663696289,
      "step": 620
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.0810810810810811e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -787.63818359375,
      "logps/real": -404.19122314453125,
      "loss": 0.1336,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -50.15456008911133,
      "rewards/margins": 33.99864959716797,
      "rewards/real": -16.15591049194336,
      "step": 630
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.0099573257467994e-07,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -830.99462890625,
      "logps/real": -457.3229064941406,
      "loss": 0.2925,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -53.06385040283203,
      "rewards/margins": 34.269439697265625,
      "rewards/real": -18.794404983520508,
      "step": 640
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.388335704125178e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -777.55517578125,
      "logps/real": -405.3326110839844,
      "loss": 0.137,
      "rewards/accuracies": 1.0,
      "rewards/generated": -49.94211196899414,
      "rewards/margins": 32.77922821044922,
      "rewards/real": -17.16288948059082,
      "step": 650
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.677098150782361e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -775.7921142578125,
      "logps/real": -408.35302734375,
      "loss": 0.0992,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -48.498905181884766,
      "rewards/margins": 32.695289611816406,
      "rewards/real": -15.803617477416992,
      "step": 660
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.965860597439544e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -764.4813232421875,
      "logps/real": -427.140625,
      "loss": 0.2396,
      "rewards/accuracies": 0.925000011920929,
      "rewards/generated": -48.443748474121094,
      "rewards/margins": 31.320148468017578,
      "rewards/real": -17.123600006103516,
      "step": 670
    },
    {
      "epoch": 0.87,
      "learning_rate": 7.254623044096728e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -803.4662475585938,
      "logps/real": -435.402099609375,
      "loss": 0.2362,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -51.644203186035156,
      "rewards/margins": 33.23621368408203,
      "rewards/real": -18.40799331665039,
      "step": 680
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.543385490753911e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -863.4392700195312,
      "logps/real": -439.1971130371094,
      "loss": 0.1949,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -55.93394088745117,
      "rewards/margins": 36.83943176269531,
      "rewards/real": -19.094507217407227,
      "step": 690
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.832147937411095e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -779.9309692382812,
      "logps/real": -411.91455078125,
      "loss": 0.2666,
      "rewards/accuracies": 0.9375,
      "rewards/generated": -47.416011810302734,
      "rewards/margins": 29.500951766967773,
      "rewards/real": -17.915063858032227,
      "step": 700
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.120910384068278e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -826.6921997070312,
      "logps/real": -447.87762451171875,
      "loss": 0.258,
      "rewards/accuracies": 0.949999988079071,
      "rewards/generated": -53.74137496948242,
      "rewards/margins": 35.463531494140625,
      "rewards/real": -18.277841567993164,
      "step": 710
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.4096728307254624e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -834.2398681640625,
      "logps/real": -427.90966796875,
      "loss": 0.1272,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -53.82038497924805,
      "rewards/margins": 36.249298095703125,
      "rewards/real": -17.571086883544922,
      "step": 720
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.698435277382646e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -838.7642822265625,
      "logps/real": -436.14642333984375,
      "loss": 0.2078,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -55.14348602294922,
      "rewards/margins": 37.05797576904297,
      "rewards/real": -18.08551597595215,
      "step": 730
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.9871977240398294e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -798.5349731445312,
      "logps/real": -427.6378479003906,
      "loss": 0.2621,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -52.086219787597656,
      "rewards/margins": 34.135223388671875,
      "rewards/real": -17.95099449157715,
      "step": 740
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.275960170697013e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -835.6724853515625,
      "logps/real": -414.00152587890625,
      "loss": 0.1564,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -52.597930908203125,
      "rewards/margins": 36.33824920654297,
      "rewards/real": -16.259681701660156,
      "step": 750
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.564722617354196e-08,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -850.904296875,
      "logps/real": -392.68310546875,
      "loss": 0.1515,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -55.202125549316406,
      "rewards/margins": 38.95097351074219,
      "rewards/real": -16.25115394592285,
      "step": 760
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.534850640113798e-09,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -814.2432250976562,
      "logps/real": -415.373779296875,
      "loss": 0.1867,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/generated": -53.38776397705078,
      "rewards/margins": 37.246055603027344,
      "rewards/real": -16.141704559326172,
      "step": 770
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.422475106685633e-09,
      "logits/generated": -Infinity,
      "logits/real": -Infinity,
      "logps/generated": -811.1026611328125,
      "logps/real": -388.99835205078125,
      "loss": 0.2403,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/generated": -51.92308807373047,
      "rewards/margins": 35.41986846923828,
      "rewards/real": -16.503215789794922,
      "step": 780
    },
    {
      "epoch": 1.0,
      "step": 782,
      "total_flos": 0.0,
      "train_loss": 0.6284351689954493,
      "train_runtime": 6560.8804,
      "train_samples_per_second": 3.81,
      "train_steps_per_second": 0.119
    }
  ],
  "logging_steps": 10,
  "max_steps": 782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}