{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984301412872841,
  "eval_steps": 100,
  "global_step": 477,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.0416666666666666e-08,
      "logits/chosen": -2.6597137451171875,
      "logits/rejected": -2.5902962684631348,
      "logps/chosen": -296.01092529296875,
      "logps/rejected": -290.09039306640625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0416666666666667e-07,
      "logits/chosen": -2.624401807785034,
      "logits/rejected": -2.58382511138916,
      "logps/chosen": -278.5157775878906,
      "logps/rejected": -242.1708984375,
      "loss": 0.6931,
      "rewards/accuracies": 0.4618055522441864,
      "rewards/chosen": 0.00044631207128986716,
      "rewards/margins": 0.0005090843187645078,
      "rewards/rejected": -6.277219654293731e-05,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": -2.6496779918670654,
      "logits/rejected": -2.5882985591888428,
      "logps/chosen": -276.2167663574219,
      "logps/rejected": -246.029052734375,
      "loss": 0.6924,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.0011385921388864517,
      "rewards/margins": 0.001740106614306569,
      "rewards/rejected": -0.0006015143590047956,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -2.6365559101104736,
      "logits/rejected": -2.5856800079345703,
      "logps/chosen": -291.7817077636719,
      "logps/rejected": -274.3924255371094,
      "loss": 0.6887,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": 0.0060002789832651615,
      "rewards/margins": 0.009898515418171883,
      "rewards/rejected": -0.0038982369005680084,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.5962817668914795,
      "logits/rejected": -2.533491611480713,
      "logps/chosen": -304.540283203125,
      "logps/rejected": -277.9544677734375,
      "loss": 0.6757,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": 0.02992216870188713,
      "rewards/margins": 0.03784631937742233,
      "rewards/rejected": -0.007924148812890053,
      "step": 40
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.999731868769026e-07,
      "logits/chosen": -2.550187110900879,
      "logits/rejected": -2.5181498527526855,
      "logps/chosen": -298.8063049316406,
      "logps/rejected": -287.06878662109375,
      "loss": 0.6604,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.0016534685855731368,
      "rewards/margins": 0.07625629007816315,
      "rewards/rejected": -0.07790975272655487,
      "step": 50
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.990353313429303e-07,
      "logits/chosen": -2.5104401111602783,
      "logits/rejected": -2.409574031829834,
      "logps/chosen": -286.75982666015625,
      "logps/rejected": -255.2505340576172,
      "loss": 0.6393,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.038099173456430435,
      "rewards/margins": 0.16247370839118958,
      "rewards/rejected": -0.2005728781223297,
      "step": 60
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.967625656594781e-07,
      "logits/chosen": -2.5491867065429688,
      "logits/rejected": -2.4775824546813965,
      "logps/chosen": -305.6695556640625,
      "logps/rejected": -294.116455078125,
      "loss": 0.6222,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.25629425048828125,
      "rewards/margins": 0.2100234478712082,
      "rewards/rejected": -0.466317743062973,
      "step": 70
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.93167072587771e-07,
      "logits/chosen": -2.512619972229004,
      "logits/rejected": -2.459233522415161,
      "logps/chosen": -313.1577453613281,
      "logps/rejected": -333.73199462890625,
      "loss": 0.6115,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.26731860637664795,
      "rewards/margins": 0.19400887191295624,
      "rewards/rejected": -0.4613274931907654,
      "step": 80
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.882681251368548e-07,
      "logits/chosen": -2.453090190887451,
      "logits/rejected": -2.3753929138183594,
      "logps/chosen": -286.51220703125,
      "logps/rejected": -299.60174560546875,
      "loss": 0.5834,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.2537994682788849,
      "rewards/margins": 0.3461161255836487,
      "rewards/rejected": -0.5999155044555664,
      "step": 90
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.820919832540181e-07,
      "logits/chosen": -2.481619358062744,
      "logits/rejected": -2.3836333751678467,
      "logps/chosen": -316.19061279296875,
      "logps/rejected": -341.30596923828125,
      "loss": 0.5736,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.3343900740146637,
      "rewards/margins": 0.4206913113594055,
      "rewards/rejected": -0.755081295967102,
      "step": 100
    },
    {
      "epoch": 0.21,
      "eval_logits/chosen": -2.432966470718384,
      "eval_logits/rejected": -2.3509771823883057,
      "eval_logps/chosen": -322.4753723144531,
      "eval_logps/rejected": -346.4595947265625,
      "eval_loss": 0.584247350692749,
      "eval_rewards/accuracies": 0.7242063283920288,
      "eval_rewards/chosen": -0.3836560547351837,
      "eval_rewards/margins": 0.46315449476242065,
      "eval_rewards/rejected": -0.846810519695282,
      "eval_runtime": 210.4195,
      "eval_samples_per_second": 9.505,
      "eval_steps_per_second": 0.299,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.7467175306295647e-07,
      "logits/chosen": -2.392451763153076,
      "logits/rejected": -2.2609496116638184,
      "logps/chosen": -329.49127197265625,
      "logps/rejected": -338.2959289550781,
      "loss": 0.5785,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.32086434960365295,
      "rewards/margins": 0.4109874665737152,
      "rewards/rejected": -0.7318518161773682,
      "step": 110
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.6604720940421207e-07,
      "logits/chosen": -1.5522091388702393,
      "logits/rejected": -1.2740141153335571,
      "logps/chosen": -365.6708984375,
      "logps/rejected": -378.64117431640625,
      "loss": 0.5441,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.8018746376037598,
      "rewards/margins": 0.6111066341400146,
      "rewards/rejected": -1.4129812717437744,
      "step": 120
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.5626458262912735e-07,
      "logits/chosen": -1.4644163846969604,
      "logits/rejected": -1.1872146129608154,
      "logps/chosen": -324.7268371582031,
      "logps/rejected": -374.5579833984375,
      "loss": 0.549,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.4890199601650238,
      "rewards/margins": 0.5601330995559692,
      "rewards/rejected": -1.0491530895233154,
      "step": 130
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.453763107901675e-07,
      "logits/chosen": -0.9067952036857605,
      "logits/rejected": -0.45260438323020935,
      "logps/chosen": -365.0186462402344,
      "logps/rejected": -414.08526611328125,
      "loss": 0.5265,
      "rewards/accuracies": 0.784375011920929,
      "rewards/chosen": -0.6849763989448547,
      "rewards/margins": 0.7878872752189636,
      "rewards/rejected": -1.4728636741638184,
      "step": 140
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3344075855595097e-07,
      "logits/chosen": -0.6669697761535645,
      "logits/rejected": -0.12073500454425812,
      "logps/chosen": -363.00494384765625,
      "logps/rejected": -401.21533203125,
      "loss": 0.5248,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.7877362966537476,
      "rewards/margins": 0.7080036401748657,
      "rewards/rejected": -1.4957398176193237,
      "step": 150
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.2052190435769554e-07,
      "logits/chosen": -0.41461697220802307,
      "logits/rejected": 0.06870967149734497,
      "logps/chosen": -365.81103515625,
      "logps/rejected": -428.4544982910156,
      "loss": 0.5073,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.8387645483016968,
      "rewards/margins": 0.7532976269721985,
      "rewards/rejected": -1.59206223487854,
      "step": 160
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.0668899744407567e-07,
      "logits/chosen": -0.7587008476257324,
      "logits/rejected": -0.3191295862197876,
      "logps/chosen": -342.67425537109375,
      "logps/rejected": -367.0728759765625,
      "loss": 0.5653,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.8000960350036621,
      "rewards/margins": 0.5834600329399109,
      "rewards/rejected": -1.3835561275482178,
      "step": 170
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.920161866827889e-07,
      "logits/chosen": -0.17186331748962402,
      "logits/rejected": 0.25430601835250854,
      "logps/chosen": -365.8907165527344,
      "logps/rejected": -429.11688232421875,
      "loss": 0.5225,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.884574294090271,
      "rewards/margins": 0.7979339361190796,
      "rewards/rejected": -1.6825082302093506,
      "step": 180
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.765821230985757e-07,
      "logits/chosen": 0.03502330183982849,
      "logits/rejected": 0.6186638474464417,
      "logps/chosen": -348.1022644042969,
      "logps/rejected": -428.8955993652344,
      "loss": 0.5105,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.7972752451896667,
      "rewards/margins": 0.9095140695571899,
      "rewards/rejected": -1.706789255142212,
      "step": 190
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.604695382782159e-07,
      "logits/chosen": -0.39503103494644165,
      "logits/rejected": 0.1367359161376953,
      "logps/chosen": -365.77545166015625,
      "logps/rejected": -400.91998291015625,
      "loss": 0.5116,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.7570338249206543,
      "rewards/margins": 0.7489927411079407,
      "rewards/rejected": -1.5060265064239502,
      "step": 200
    },
    {
      "epoch": 0.42,
      "eval_logits/chosen": 0.04591968283057213,
      "eval_logits/rejected": 0.5012260675430298,
      "eval_logps/chosen": -374.5287780761719,
      "eval_logps/rejected": -435.0859375,
      "eval_loss": 0.5307875871658325,
      "eval_rewards/accuracies": 0.7519841194152832,
      "eval_rewards/chosen": -0.904190182685852,
      "eval_rewards/margins": 0.8288835883140564,
      "eval_rewards/rejected": -1.7330738306045532,
      "eval_runtime": 211.7151,
      "eval_samples_per_second": 9.447,
      "eval_steps_per_second": 0.298,
      "step": 200
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4376480090239047e-07,
      "logits/chosen": 0.21950086951255798,
      "logits/rejected": 0.7312016487121582,
      "logps/chosen": -394.24078369140625,
      "logps/rejected": -439.376220703125,
      "loss": 0.5279,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -1.0525823831558228,
      "rewards/margins": 0.7364203929901123,
      "rewards/rejected": -1.789002776145935,
      "step": 210
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.265574537815398e-07,
      "logits/chosen": 0.034196797758340836,
      "logits/rejected": 0.35187411308288574,
      "logps/chosen": -332.6381530761719,
      "logps/rejected": -382.64434814453125,
      "loss": 0.5384,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.7204285860061646,
      "rewards/margins": 0.6646216511726379,
      "rewards/rejected": -1.3850500583648682,
      "step": 220
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.0893973387735683e-07,
      "logits/chosen": 0.3270489573478699,
      "logits/rejected": 0.8940129280090332,
      "logps/chosen": -341.1247253417969,
      "logps/rejected": -408.52337646484375,
      "loss": 0.5218,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.765308141708374,
      "rewards/margins": 0.8177844882011414,
      "rewards/rejected": -1.583092451095581,
      "step": 230
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.910060778827554e-07,
      "logits/chosen": 0.8036887049674988,
      "logits/rejected": 1.329053521156311,
      "logps/chosen": -384.66473388671875,
      "logps/rejected": -448.1686096191406,
      "loss": 0.5285,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -1.1919479370117188,
      "rewards/margins": 0.8430492281913757,
      "rewards/rejected": -2.0349972248077393,
      "step": 240
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7285261601056697e-07,
      "logits/chosen": 0.51897132396698,
      "logits/rejected": 1.1455881595611572,
      "logps/chosen": -373.64044189453125,
      "logps/rejected": -425.75494384765625,
      "loss": 0.519,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.9999877214431763,
      "rewards/margins": 0.7631025314331055,
      "rewards/rejected": -1.7630901336669922,
      "step": 250
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.5457665670441937e-07,
      "logits/chosen": 0.5749747157096863,
      "logits/rejected": 1.220221757888794,
      "logps/chosen": -371.4618225097656,
      "logps/rejected": -424.69439697265625,
      "loss": 0.4969,
      "rewards/accuracies": 0.765625,
      "rewards/chosen": -0.9122276306152344,
      "rewards/margins": 0.8183482885360718,
      "rewards/rejected": -1.7305759191513062,
      "step": 260
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3627616503391812e-07,
      "logits/chosen": 0.8650910258293152,
      "logits/rejected": 1.3838642835617065,
      "logps/chosen": -398.350341796875,
      "logps/rejected": -484.79547119140625,
      "loss": 0.5051,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -1.091962456703186,
      "rewards/margins": 0.9698148965835571,
      "rewards/rejected": -2.061777353286743,
      "step": 270
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1804923757009882e-07,
      "logits/chosen": 0.9545858502388,
      "logits/rejected": 1.6210914850234985,
      "logps/chosen": -365.74053955078125,
      "logps/rejected": -422.39300537109375,
      "loss": 0.5248,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.9749383926391602,
      "rewards/margins": 0.841883659362793,
      "rewards/rejected": -1.8168220520019531,
      "step": 280
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9999357655598891e-07,
      "logits/chosen": 1.363952875137329,
      "logits/rejected": 2.0388360023498535,
      "logps/chosen": -385.5817565917969,
      "logps/rejected": -436.33209228515625,
      "loss": 0.5155,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -1.1055189371109009,
      "rewards/margins": 0.77390056848526,
      "rewards/rejected": -1.8794195652008057,
      "step": 290
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8220596619089573e-07,
      "logits/chosen": 1.3142567873001099,
      "logits/rejected": 1.9236018657684326,
      "logps/chosen": -394.31964111328125,
      "logps/rejected": -467.43756103515625,
      "loss": 0.5027,
      "rewards/accuracies": 0.734375,
      "rewards/chosen": -0.9480918049812317,
      "rewards/margins": 0.7692841291427612,
      "rewards/rejected": -1.7173759937286377,
      "step": 300
    },
    {
      "epoch": 0.63,
      "eval_logits/chosen": 1.1385366916656494,
      "eval_logits/rejected": 1.822389006614685,
      "eval_logps/chosen": -372.8834228515625,
      "eval_logps/rejected": -436.4477844238281,
      "eval_loss": 0.508425235748291,
      "eval_rewards/accuracies": 0.7638888955116272,
      "eval_rewards/chosen": -0.887736976146698,
      "eval_rewards/margins": 0.8589555621147156,
      "eval_rewards/rejected": -1.746692419052124,
      "eval_runtime": 211.1789,
      "eval_samples_per_second": 9.471,
      "eval_steps_per_second": 0.298,
      "step": 300
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.647817538357072e-07,
      "logits/chosen": 1.0226597785949707,
      "logits/rejected": 1.96248459815979,
      "logps/chosen": -367.84130859375,
      "logps/rejected": -404.55963134765625,
      "loss": 0.4985,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.9431735277175903,
      "rewards/margins": 0.8295876383781433,
      "rewards/rejected": -1.7727611064910889,
      "step": 310
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.478143389201113e-07,
      "logits/chosen": 1.6019757986068726,
      "logits/rejected": 2.321258068084717,
      "logps/chosen": -374.2918701171875,
      "logps/rejected": -456.11602783203125,
      "loss": 0.5035,
      "rewards/accuracies": 0.7718750238418579,
      "rewards/chosen": -1.0117809772491455,
      "rewards/margins": 0.9862319231033325,
      "rewards/rejected": -1.998012900352478,
      "step": 320
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3139467229135998e-07,
      "logits/chosen": 1.089303970336914,
      "logits/rejected": 1.724713921546936,
      "logps/chosen": -422.41314697265625,
      "logps/rejected": -462.42999267578125,
      "loss": 0.5162,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -1.184761881828308,
      "rewards/margins": 0.7901886701583862,
      "rewards/rejected": -1.9749505519866943,
      "step": 330
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1561076868822755e-07,
      "logits/chosen": 1.2003899812698364,
      "logits/rejected": 1.7042887210845947,
      "logps/chosen": -374.2175598144531,
      "logps/rejected": -423.61114501953125,
      "loss": 0.5141,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.0190467834472656,
      "rewards/margins": 0.6825306415557861,
      "rewards/rejected": -1.7015774250030518,
      "step": 340
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0054723495346482e-07,
      "logits/chosen": 1.183173418045044,
      "logits/rejected": 2.0356898307800293,
      "logps/chosen": -359.0115661621094,
      "logps/rejected": -438.88336181640625,
      "loss": 0.4766,
      "rewards/accuracies": 0.7718750238418579,
      "rewards/chosen": -1.0118539333343506,
      "rewards/margins": 0.9436109662055969,
      "rewards/rejected": -1.9554650783538818,
      "step": 350
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.628481651367875e-08,
      "logits/chosen": 1.1831741333007812,
      "logits/rejected": 1.9693758487701416,
      "logps/chosen": -408.16705322265625,
      "logps/rejected": -451.48370361328125,
      "loss": 0.4836,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": -1.0999749898910522,
      "rewards/margins": 0.8795393705368042,
      "rewards/rejected": -1.9795143604278564,
      "step": 360
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.289996455765748e-08,
      "logits/chosen": 1.1307913064956665,
      "logits/rejected": 2.096768856048584,
      "logps/chosen": -413.69805908203125,
      "logps/rejected": -458.0303649902344,
      "loss": 0.499,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -1.0460116863250732,
      "rewards/margins": 0.9790540933609009,
      "rewards/rejected": -2.0250658988952637,
      "step": 370
    },
    {
      "epoch": 0.8,
      "learning_rate": 6.046442623320145e-08,
      "logits/chosen": 0.9909412264823914,
      "logits/rejected": 1.6467933654785156,
      "logps/chosen": -381.5520324707031,
      "logps/rejected": -453.1272888183594,
      "loss": 0.4727,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -1.0082350969314575,
      "rewards/margins": 0.8770313262939453,
      "rewards/rejected": -1.8852663040161133,
      "step": 380
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.904486005914027e-08,
      "logits/chosen": 1.356707215309143,
      "logits/rejected": 1.9564104080200195,
      "logps/chosen": -383.09393310546875,
      "logps/rejected": -455.9044494628906,
      "loss": 0.5061,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.0799481868743896,
      "rewards/margins": 0.7995551228523254,
      "rewards/rejected": -1.879503607749939,
      "step": 390
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.8702478614051345e-08,
      "logits/chosen": 1.2665964365005493,
      "logits/rejected": 1.9937832355499268,
      "logps/chosen": -441.0684509277344,
      "logps/rejected": -508.45269775390625,
      "loss": 0.4823,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.202941656112671,
      "rewards/margins": 1.0496307611465454,
      "rewards/rejected": -2.2525722980499268,
      "step": 400
    },
    {
      "epoch": 0.84,
      "eval_logits/chosen": 1.3074696063995361,
      "eval_logits/rejected": 1.9977563619613647,
      "eval_logps/chosen": -403.6374816894531,
      "eval_logps/rejected": -476.98516845703125,
      "eval_loss": 0.5037237405776978,
      "eval_rewards/accuracies": 0.761904776096344,
      "eval_rewards/chosen": -1.195277214050293,
      "eval_rewards/margins": 0.9567888975143433,
      "eval_rewards/rejected": -2.152066230773926,
      "eval_runtime": 212.3326,
      "eval_samples_per_second": 9.419,
      "eval_steps_per_second": 0.297,
      "step": 400
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.9492720416985e-08,
      "logits/chosen": 1.6189842224121094,
      "logits/rejected": 2.477853775024414,
      "logps/chosen": -386.3858947753906,
      "logps/rejected": -442.0225524902344,
      "loss": 0.5095,
      "rewards/accuracies": 0.7593749761581421,
      "rewards/chosen": -1.2031550407409668,
      "rewards/margins": 0.845988929271698,
      "rewards/rejected": -2.0491440296173096,
      "step": 410
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1464952759020856e-08,
      "logits/chosen": 1.4740936756134033,
      "logits/rejected": 1.9605319499969482,
      "logps/chosen": -398.62274169921875,
      "logps/rejected": -460.81201171875,
      "loss": 0.5112,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -1.1762058734893799,
      "rewards/margins": 0.8007314801216125,
      "rewards/rejected": -1.9769372940063477,
      "step": 420
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4662207078575684e-08,
      "logits/chosen": 1.1320947408676147,
      "logits/rejected": 1.8564393520355225,
      "logps/chosen": -397.1152648925781,
      "logps/rejected": -444.3277282714844,
      "loss": 0.5032,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.07761549949646,
      "rewards/margins": 0.886116623878479,
      "rewards/rejected": -1.963732361793518,
      "step": 430
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.12094829893642e-09,
      "logits/chosen": 1.1670544147491455,
      "logits/rejected": 1.9005542993545532,
      "logps/chosen": -362.5390319824219,
      "logps/rejected": -448.8984375,
      "loss": 0.4901,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -1.019899606704712,
      "rewards/margins": 0.9710233807563782,
      "rewards/rejected": -1.9909226894378662,
      "step": 440
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.8708793644441086e-09,
      "logits/chosen": 1.165825605392456,
      "logits/rejected": 1.7851331233978271,
      "logps/chosen": -393.0291442871094,
      "logps/rejected": -466.2897033691406,
      "loss": 0.5112,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.1060134172439575,
      "rewards/margins": 0.8082348704338074,
      "rewards/rejected": -1.9142482280731201,
      "step": 450
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.9347820230782295e-09,
      "logits/chosen": 1.1381809711456299,
      "logits/rejected": 2.048168897628784,
      "logps/chosen": -380.725830078125,
      "logps/rejected": -454.0580139160156,
      "loss": 0.4878,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -1.0911556482315063,
      "rewards/margins": 0.9514607191085815,
      "rewards/rejected": -2.042616367340088,
      "step": 460
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.2839470889836627e-10,
      "logits/chosen": 0.9774567484855652,
      "logits/rejected": 1.967919111251831,
      "logps/chosen": -411.60504150390625,
      "logps/rejected": -474.35650634765625,
      "loss": 0.4835,
      "rewards/accuracies": 0.796875,
      "rewards/chosen": -1.1035670042037964,
      "rewards/margins": 0.9632620811462402,
      "rewards/rejected": -2.066829204559326,
      "step": 470
    },
    {
      "epoch": 1.0,
      "step": 477,
      "total_flos": 0.0,
      "train_loss": 0.540120158305458,
      "train_runtime": 18392.9814,
      "train_samples_per_second": 3.324,
      "train_steps_per_second": 0.026
    }
  ],
  "logging_steps": 10,
  "max_steps": 477,
  "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
}