{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9850746268656714,
  "eval_steps": 50,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07462686567164178,
      "grad_norm": 67.58726313106416,
      "learning_rate": 5e-07,
      "logits/chosen": -2.728022575378418,
      "logits/rejected": -2.724689245223999,
      "logps/chosen": -256.9372253417969,
      "logps/rejected": -205.8960418701172,
      "loss": 0.6906,
      "rewards/accuracies": 0.3499999940395355,
      "rewards/chosen": 0.012728104367852211,
      "rewards/margins": 0.00967978872358799,
      "rewards/rejected": 0.003048315178602934,
      "step": 5
    },
    {
      "epoch": 0.14925373134328357,
      "grad_norm": 46.21946007010114,
      "learning_rate": 1e-06,
      "logits/chosen": -2.7133026123046875,
      "logits/rejected": -2.6811420917510986,
      "logps/chosen": -248.5566864013672,
      "logps/rejected": -197.87265014648438,
      "loss": 0.648,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": 0.4624863564968109,
      "rewards/margins": 0.19924615323543549,
      "rewards/rejected": 0.26324018836021423,
      "step": 10
    },
    {
      "epoch": 0.22388059701492538,
      "grad_norm": 39.90578699736001,
      "learning_rate": 9.983100718730718e-07,
      "logits/chosen": -2.5118980407714844,
      "logits/rejected": -2.4951248168945312,
      "logps/chosen": -241.9331512451172,
      "logps/rejected": -202.6668701171875,
      "loss": 0.6322,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 1.4060888290405273,
      "rewards/margins": 0.5313084721565247,
      "rewards/rejected": 0.8747804760932922,
      "step": 15
    },
    {
      "epoch": 0.29850746268656714,
      "grad_norm": 62.96556196720542,
      "learning_rate": 9.932517109205849e-07,
      "logits/chosen": -2.3295137882232666,
      "logits/rejected": -2.323544502258301,
      "logps/chosen": -236.51119995117188,
      "logps/rejected": -223.9521026611328,
      "loss": 0.6902,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": 1.4421606063842773,
      "rewards/margins": 0.5772963762283325,
      "rewards/rejected": 0.8648644685745239,
      "step": 20
    },
    {
      "epoch": 0.373134328358209,
      "grad_norm": 52.70054919387278,
      "learning_rate": 9.848591102083375e-07,
      "logits/chosen": -2.291351795196533,
      "logits/rejected": -2.298973560333252,
      "logps/chosen": -248.5207061767578,
      "logps/rejected": -212.27572631835938,
      "loss": 0.6318,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 1.2592790126800537,
      "rewards/margins": 0.7029441595077515,
      "rewards/rejected": 0.5563349723815918,
      "step": 25
    },
    {
      "epoch": 0.44776119402985076,
      "grad_norm": 39.25239538787194,
      "learning_rate": 9.731890013043367e-07,
      "logits/chosen": -2.3153462409973145,
      "logits/rejected": -2.2776923179626465,
      "logps/chosen": -230.8772430419922,
      "logps/rejected": -191.97027587890625,
      "loss": 0.6011,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 1.068457841873169,
      "rewards/margins": 0.6923708319664001,
      "rewards/rejected": 0.3760870099067688,
      "step": 30
    },
    {
      "epoch": 0.5223880597014925,
      "grad_norm": 42.62760194044431,
      "learning_rate": 9.583202707897073e-07,
      "logits/chosen": -2.3093903064727783,
      "logits/rejected": -2.291133403778076,
      "logps/chosen": -219.6804656982422,
      "logps/rejected": -220.82974243164062,
      "loss": 0.6313,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": 0.7500138282775879,
      "rewards/margins": 0.756812572479248,
      "rewards/rejected": -0.006798672489821911,
      "step": 35
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 40.93212888100826,
      "learning_rate": 9.403534270080829e-07,
      "logits/chosen": -2.398266315460205,
      "logits/rejected": -2.383267641067505,
      "logps/chosen": -220.89498901367188,
      "logps/rejected": -231.33663940429688,
      "loss": 0.6053,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.6151864528656006,
      "rewards/margins": 0.6149100065231323,
      "rewards/rejected": 0.00027652084827423096,
      "step": 40
    },
    {
      "epoch": 0.6716417910447762,
      "grad_norm": 44.441524435463975,
      "learning_rate": 9.19409920658098e-07,
      "logits/chosen": -2.38905668258667,
      "logits/rejected": -2.418109893798828,
      "logps/chosen": -243.25039672851562,
      "logps/rejected": -227.03500366210938,
      "loss": 0.6158,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": 0.7249364256858826,
      "rewards/margins": 0.903680682182312,
      "rewards/rejected": -0.17874428629875183,
      "step": 45
    },
    {
      "epoch": 0.746268656716418,
      "grad_norm": 36.109012224798654,
      "learning_rate": 8.956313238215823e-07,
      "logits/chosen": -2.4329724311828613,
      "logits/rejected": -2.393233060836792,
      "logps/chosen": -235.9673614501953,
      "logps/rejected": -207.4291229248047,
      "loss": 0.5788,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": 0.7963231801986694,
      "rewards/margins": 1.1296331882476807,
      "rewards/rejected": -0.3333100378513336,
      "step": 50
    },
    {
      "epoch": 0.746268656716418,
      "eval_logits/chosen": -2.3947904109954834,
      "eval_logits/rejected": -2.381561756134033,
      "eval_logps/chosen": -248.11183166503906,
      "eval_logps/rejected": -209.1226348876953,
      "eval_loss": 0.5886067152023315,
      "eval_rewards/accuracies": 0.7666666507720947,
      "eval_rewards/chosen": 1.3111679553985596,
      "eval_rewards/margins": 1.323790431022644,
      "eval_rewards/rejected": -0.01262261439114809,
      "eval_runtime": 126.2751,
      "eval_samples_per_second": 15.047,
      "eval_steps_per_second": 0.238,
      "step": 50
    },
    {
      "epoch": 0.8208955223880597,
      "grad_norm": 37.42293105255099,
      "learning_rate": 8.691783729769873e-07,
      "logits/chosen": -2.3496580123901367,
      "logits/rejected": -2.370349407196045,
      "logps/chosen": -237.4831085205078,
      "logps/rejected": -204.3516082763672,
      "loss": 0.5805,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 0.9665477871894836,
      "rewards/margins": 1.0714900493621826,
      "rewards/rejected": -0.10494247823953629,
      "step": 55
    },
    {
      "epoch": 0.8955223880597015,
      "grad_norm": 36.18527140538969,
      "learning_rate": 8.402298824670029e-07,
      "logits/chosen": -2.3262667655944824,
      "logits/rejected": -2.3069987297058105,
      "logps/chosen": -252.2436065673828,
      "logps/rejected": -218.7754364013672,
      "loss": 0.5423,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 1.2422854900360107,
      "rewards/margins": 1.209867238998413,
      "rewards/rejected": 0.032418303191661835,
      "step": 60
    },
    {
      "epoch": 0.9701492537313433,
      "grad_norm": 33.097833441355846,
      "learning_rate": 8.089815357650089e-07,
      "logits/chosen": -2.33039927482605,
      "logits/rejected": -2.325122833251953,
      "logps/chosen": -254.16928100585938,
      "logps/rejected": -207.00668334960938,
      "loss": 0.5462,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 1.058384656906128,
      "rewards/margins": 1.3959627151489258,
      "rewards/rejected": -0.33757805824279785,
      "step": 65
    },
    {
      "epoch": 1.044776119402985,
      "grad_norm": 20.023976264712793,
      "learning_rate": 7.756445627110522e-07,
      "logits/chosen": -2.3418185710906982,
      "logits/rejected": -2.3507180213928223,
      "logps/chosen": -239.88333129882812,
      "logps/rejected": -218.7294464111328,
      "loss": 0.3251,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": 1.4150011539459229,
      "rewards/margins": 1.9995105266571045,
      "rewards/rejected": -0.584509551525116,
      "step": 70
    },
    {
      "epoch": 1.1194029850746268,
      "grad_norm": 23.675520442711395,
      "learning_rate": 7.404443116588547e-07,
      "logits/chosen": -2.398703098297119,
      "logits/rejected": -2.3791286945343018,
      "logps/chosen": -230.90005493164062,
      "logps/rejected": -201.66864013671875,
      "loss": 0.221,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 1.7823137044906616,
      "rewards/margins": 2.5341625213623047,
      "rewards/rejected": -0.7518486380577087,
      "step": 75
    },
    {
      "epoch": 1.1940298507462686,
      "grad_norm": 15.225016299890036,
      "learning_rate": 7.036187261857288e-07,
      "logits/chosen": -2.3974556922912598,
      "logits/rejected": -2.388159990310669,
      "logps/chosen": -242.7141571044922,
      "logps/rejected": -225.48001098632812,
      "loss": 0.2114,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 2.029745578765869,
      "rewards/margins": 3.112726926803589,
      "rewards/rejected": -1.0829811096191406,
      "step": 80
    },
    {
      "epoch": 1.2686567164179103,
      "grad_norm": 20.47443225957059,
      "learning_rate": 6.654167366624008e-07,
      "logits/chosen": -2.3772902488708496,
      "logits/rejected": -2.351728916168213,
      "logps/chosen": -233.95236206054688,
      "logps/rejected": -230.03994750976562,
      "loss": 0.2168,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 1.4827768802642822,
      "rewards/margins": 2.783128499984741,
      "rewards/rejected": -1.3003517389297485,
      "step": 85
    },
    {
      "epoch": 1.3432835820895521,
      "grad_norm": 23.797530563309465,
      "learning_rate": 6.260965775552713e-07,
      "logits/chosen": -2.3302738666534424,
      "logits/rejected": -2.3255810737609863,
      "logps/chosen": -242.00732421875,
      "logps/rejected": -240.259521484375,
      "loss": 0.2087,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 1.8802661895751953,
      "rewards/margins": 3.69026517868042,
      "rewards/rejected": -1.8099991083145142,
      "step": 90
    },
    {
      "epoch": 1.417910447761194,
      "grad_norm": 22.2827308670365,
      "learning_rate": 5.859240418356614e-07,
      "logits/chosen": -2.346041679382324,
      "logits/rejected": -2.3258674144744873,
      "logps/chosen": -241.3457794189453,
      "logps/rejected": -232.9997100830078,
      "loss": 0.2454,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 2.422487497329712,
      "rewards/margins": 3.9014511108398438,
      "rewards/rejected": -1.4789636135101318,
      "step": 95
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 23.32738252611807,
      "learning_rate": 5.451706842957421e-07,
      "logits/chosen": -2.3272500038146973,
      "logits/rejected": -2.290189266204834,
      "logps/chosen": -237.8894805908203,
      "logps/rejected": -224.0364227294922,
      "loss": 0.2623,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 2.44073748588562,
      "rewards/margins": 3.8086400032043457,
      "rewards/rejected": -1.3679022789001465,
      "step": 100
    },
    {
      "epoch": 1.4925373134328357,
      "eval_logits/chosen": -2.272735118865967,
      "eval_logits/rejected": -2.24358868598938,
      "eval_logps/chosen": -245.39662170410156,
      "eval_logps/rejected": -213.9952850341797,
      "eval_loss": 0.6056690216064453,
      "eval_rewards/accuracies": 0.8041666746139526,
      "eval_rewards/chosen": 1.5826876163482666,
      "eval_rewards/margins": 2.0825743675231934,
      "eval_rewards/rejected": -0.49988678097724915,
      "eval_runtime": 126.0708,
      "eval_samples_per_second": 15.071,
      "eval_steps_per_second": 0.238,
      "step": 100
    },
    {
      "epoch": 1.5671641791044775,
      "grad_norm": 18.955563276330228,
      "learning_rate": 5.041119859162068e-07,
      "logits/chosen": -2.266010046005249,
      "logits/rejected": -2.224538564682007,
      "logps/chosen": -229.598876953125,
      "logps/rejected": -229.48312377929688,
      "loss": 0.2328,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 2.0126357078552246,
      "rewards/margins": 3.625946044921875,
      "rewards/rejected": -1.6133100986480713,
      "step": 105
    },
    {
      "epoch": 1.6417910447761193,
      "grad_norm": 22.19988117931717,
      "learning_rate": 4.630254916940423e-07,
      "logits/chosen": -2.241581678390503,
      "logits/rejected": -2.228926658630371,
      "logps/chosen": -234.0880889892578,
      "logps/rejected": -211.94546508789062,
      "loss": 0.2614,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 1.8189502954483032,
      "rewards/margins": 3.2742087841033936,
      "rewards/rejected": -1.4552589654922485,
      "step": 110
    },
    {
      "epoch": 1.716417910447761,
      "grad_norm": 25.211183388734273,
      "learning_rate": 4.2218893451814e-07,
      "logits/chosen": -2.240575075149536,
      "logits/rejected": -2.2353932857513428,
      "logps/chosen": -234.106201171875,
      "logps/rejected": -225.87057495117188,
      "loss": 0.2607,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 1.9395558834075928,
      "rewards/margins": 3.2138640880584717,
      "rewards/rejected": -1.2743080854415894,
      "step": 115
    },
    {
      "epoch": 1.7910447761194028,
      "grad_norm": 20.408403893097674,
      "learning_rate": 3.8187835777481375e-07,
      "logits/chosen": -2.2678146362304688,
      "logits/rejected": -2.248746395111084,
      "logps/chosen": -234.76974487304688,
      "logps/rejected": -229.4392852783203,
      "loss": 0.2434,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 2.1745102405548096,
      "rewards/margins": 3.79044771194458,
      "rewards/rejected": -1.6159374713897705,
      "step": 120
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 23.108934441307838,
      "learning_rate": 3.423662493738687e-07,
      "logits/chosen": -2.2544753551483154,
      "logits/rejected": -2.226003408432007,
      "logps/chosen": -215.0713348388672,
      "logps/rejected": -198.3508758544922,
      "loss": 0.2838,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": 1.776116967201233,
      "rewards/margins": 3.0843865871429443,
      "rewards/rejected": -1.3082692623138428,
      "step": 125
    },
    {
      "epoch": 1.9402985074626866,
      "grad_norm": 24.748752156744665,
      "learning_rate": 3.039196998086687e-07,
      "logits/chosen": -2.273759365081787,
      "logits/rejected": -2.2541890144348145,
      "logps/chosen": -239.5052947998047,
      "logps/rejected": -248.71157836914062,
      "loss": 0.2544,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": 2.4912400245666504,
      "rewards/margins": 3.995276927947998,
      "rewards/rejected": -1.5040366649627686,
      "step": 130
    },
    {
      "epoch": 2.014925373134328,
      "grad_norm": 17.84439016806703,
      "learning_rate": 2.667985967011878e-07,
      "logits/chosen": -2.287829875946045,
      "logits/rejected": -2.2854392528533936,
      "logps/chosen": -243.9482879638672,
      "logps/rejected": -236.39785766601562,
      "loss": 0.2205,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 2.2010834217071533,
      "rewards/margins": 3.6561388969421387,
      "rewards/rejected": -1.455055832862854,
      "step": 135
    },
    {
      "epoch": 2.08955223880597,
      "grad_norm": 17.681759766337574,
      "learning_rate": 2.3125386803640183e-07,
      "logits/chosen": -2.303140163421631,
      "logits/rejected": -2.2794697284698486,
      "logps/chosen": -229.8805389404297,
      "logps/rejected": -261.2996520996094,
      "loss": 0.1265,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 2.40771222114563,
      "rewards/margins": 4.214150428771973,
      "rewards/rejected": -1.8064378499984741,
      "step": 140
    },
    {
      "epoch": 2.1641791044776117,
      "grad_norm": 14.355965133474355,
      "learning_rate": 1.9752578596124952e-07,
      "logits/chosen": -2.2925033569335938,
      "logits/rejected": -2.2765016555786133,
      "logps/chosen": -239.452392578125,
      "logps/rejected": -258.3964538574219,
      "loss": 0.1418,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 2.2769477367401123,
      "rewards/margins": 4.572505474090576,
      "rewards/rejected": -2.295557975769043,
      "step": 145
    },
    {
      "epoch": 2.2388059701492535,
      "grad_norm": 13.075018639785867,
      "learning_rate": 1.6584234261399532e-07,
      "logits/chosen": -2.2472779750823975,
      "logits/rejected": -2.2457058429718018,
      "logps/chosen": -215.5850372314453,
      "logps/rejected": -245.4586181640625,
      "loss": 0.1176,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 2.197263479232788,
      "rewards/margins": 4.124143123626709,
      "rewards/rejected": -1.9268791675567627,
      "step": 150
    },
    {
      "epoch": 2.2388059701492535,
      "eval_logits/chosen": -2.2484564781188965,
      "eval_logits/rejected": -2.223518133163452,
      "eval_logps/chosen": -246.53546142578125,
      "eval_logps/rejected": -216.13336181640625,
      "eval_loss": 0.5922633409500122,
      "eval_rewards/accuracies": 0.7875000238418579,
      "eval_rewards/chosen": 1.4688013792037964,
      "eval_rewards/margins": 2.1824986934661865,
      "eval_rewards/rejected": -0.7136973738670349,
      "eval_runtime": 125.9912,
      "eval_samples_per_second": 15.08,
      "eval_steps_per_second": 0.238,
      "step": 150
    },
    {
      "epoch": 2.3134328358208958,
      "grad_norm": 13.851661957735093,
      "learning_rate": 1.3641770896292082e-07,
      "logits/chosen": -2.2474873065948486,
      "logits/rejected": -2.247633457183838,
      "logps/chosen": -244.7652587890625,
      "logps/rejected": -235.2009735107422,
      "loss": 0.1112,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 2.8528218269348145,
      "rewards/margins": 4.657607555389404,
      "rewards/rejected": -1.804785966873169,
      "step": 155
    },
    {
      "epoch": 2.388059701492537,
      "grad_norm": 21.198096536107492,
      "learning_rate": 1.0945078707215221e-07,
      "logits/chosen": -2.260723829269409,
      "logits/rejected": -2.250204563140869,
      "logps/chosen": -234.06942749023438,
      "logps/rejected": -261.3254699707031,
      "loss": 0.1333,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 2.2776079177856445,
      "rewards/margins": 4.555299282073975,
      "rewards/rejected": -2.27769136428833,
      "step": 160
    },
    {
      "epoch": 2.4626865671641793,
      "grad_norm": 17.709308743546742,
      "learning_rate": 8.512386558088919e-08,
      "logits/chosen": -2.2542097568511963,
      "logits/rejected": -2.2354049682617188,
      "logps/chosen": -227.0293426513672,
      "logps/rejected": -220.221435546875,
      "loss": 0.1199,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 2.320430278778076,
      "rewards/margins": 4.206162452697754,
      "rewards/rejected": -1.8857324123382568,
      "step": 165
    },
    {
      "epoch": 2.5373134328358207,
      "grad_norm": 10.709332675693139,
      "learning_rate": 6.360138748461013e-08,
      "logits/chosen": -2.2639975547790527,
      "logits/rejected": -2.234525203704834,
      "logps/chosen": -229.2438507080078,
      "logps/rejected": -223.31326293945312,
      "loss": 0.1403,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 2.291680335998535,
      "rewards/margins": 4.302089691162109,
      "rewards/rejected": -2.010409116744995,
      "step": 170
    },
    {
      "epoch": 2.611940298507463,
      "grad_norm": 15.275450815772352,
      "learning_rate": 4.5028838547699346e-08,
      "logits/chosen": -2.238157272338867,
      "logits/rejected": -2.2448458671569824,
      "logps/chosen": -223.23828125,
      "logps/rejected": -251.35513305664062,
      "loss": 0.1258,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 2.0068905353546143,
      "rewards/margins": 4.182096481323242,
      "rewards/rejected": -2.175205945968628,
      "step": 175
    },
    {
      "epoch": 2.6865671641791042,
      "grad_norm": 10.670496827369625,
      "learning_rate": 2.9531763861505964e-08,
      "logits/chosen": -2.246856689453125,
      "logits/rejected": -2.2432849407196045,
      "logps/chosen": -230.0186309814453,
      "logps/rejected": -235.7653045654297,
      "loss": 0.1521,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 2.0715701580047607,
      "rewards/margins": 4.315110683441162,
      "rewards/rejected": -2.2435405254364014,
      "step": 180
    },
    {
      "epoch": 2.7611940298507465,
      "grad_norm": 13.296818618991479,
      "learning_rate": 1.7214919195619125e-08,
      "logits/chosen": -2.2300615310668945,
      "logits/rejected": -2.224147319793701,
      "logps/chosen": -233.1608428955078,
      "logps/rejected": -218.90585327148438,
      "loss": 0.1109,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 2.1774356365203857,
      "rewards/margins": 4.205111503601074,
      "rewards/rejected": -2.0276761054992676,
      "step": 185
    },
    {
      "epoch": 2.835820895522388,
      "grad_norm": 17.367538516485407,
      "learning_rate": 8.161562878982398e-09,
      "logits/chosen": -2.2490463256835938,
      "logits/rejected": -2.2461249828338623,
      "logps/chosen": -229.1264190673828,
      "logps/rejected": -234.72836303710938,
      "loss": 0.1378,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 2.0733642578125,
      "rewards/margins": 4.36142635345459,
      "rewards/rejected": -2.288062334060669,
      "step": 190
    },
    {
      "epoch": 2.91044776119403,
      "grad_norm": 19.281670591260532,
      "learning_rate": 2.432892997526026e-09,
      "logits/chosen": -2.247451066970825,
      "logits/rejected": -2.228789806365967,
      "logps/chosen": -236.5612030029297,
      "logps/rejected": -241.4217987060547,
      "loss": 0.1481,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 2.031038761138916,
      "rewards/margins": 4.2105302810668945,
      "rewards/rejected": -2.1794912815093994,
      "step": 195
    },
    {
      "epoch": 2.9850746268656714,
      "grad_norm": 16.314964119933162,
      "learning_rate": 6.763371270035457e-11,
      "logits/chosen": -2.2567026615142822,
      "logits/rejected": -2.2431235313415527,
      "logps/chosen": -244.52407836914062,
      "logps/rejected": -221.75595092773438,
      "loss": 0.1387,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 2.3067245483398438,
      "rewards/margins": 4.139992713928223,
      "rewards/rejected": -1.8332679271697998,
      "step": 200
    },
    {
      "epoch": 2.9850746268656714,
      "eval_logits/chosen": -2.243741035461426,
      "eval_logits/rejected": -2.2175543308258057,
      "eval_logps/chosen": -248.4080047607422,
      "eval_logps/rejected": -218.735107421875,
      "eval_loss": 0.6027006506919861,
      "eval_rewards/accuracies": 0.7875000238418579,
      "eval_rewards/chosen": 1.2815468311309814,
      "eval_rewards/margins": 2.2554140090942383,
      "eval_rewards/rejected": -0.9738672375679016,
      "eval_runtime": 125.8747,
      "eval_samples_per_second": 15.094,
      "eval_steps_per_second": 0.238,
      "step": 200
    }
  ],
  "logging_steps": 5,
  "max_steps": 201,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2358113407598592.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}