{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.5625e-07,
      "logits/chosen": -0.23665529489517212,
      "logits/rejected": 0.00744676636531949,
      "logps/chosen": -473.1935729980469,
      "logps/rejected": -432.5630798339844,
      "loss": 2.0507,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5625e-06,
      "logits/chosen": -0.1829964816570282,
      "logits/rejected": -0.12090256810188293,
      "logps/chosen": -383.1433410644531,
      "logps/rejected": -396.7732238769531,
      "loss": 2.1635,
      "rewards/accuracies": 0.3611111044883728,
      "rewards/chosen": 0.00011064788850490004,
      "rewards/margins": -0.0001980961678782478,
      "rewards/rejected": 0.00030874405638314784,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": -0.16048185527324677,
      "logits/rejected": -0.10916981846094131,
      "logps/chosen": -385.62884521484375,
      "logps/rejected": -405.12811279296875,
      "loss": 2.1708,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.0015514277620241046,
      "rewards/margins": 0.0006486388738267124,
      "rewards/rejected": -0.002200066577643156,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.6875000000000004e-06,
      "logits/chosen": -0.13327451050281525,
      "logits/rejected": -0.09481740742921829,
      "logps/chosen": -383.2904968261719,
      "logps/rejected": -386.16241455078125,
      "loss": 2.1136,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.007082388736307621,
      "rewards/margins": 0.004239413887262344,
      "rewards/rejected": -0.011321802623569965,
      "step": 30
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -0.20482906699180603,
      "logits/rejected": -0.11631456762552261,
      "logps/chosen": -449.19073486328125,
      "logps/rejected": -432.53546142578125,
      "loss": 2.0563,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.04025329276919365,
      "rewards/margins": 0.009497672319412231,
      "rewards/rejected": -0.04975096508860588,
      "step": 40
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.949188496058089e-06,
      "logits/chosen": -0.18474937975406647,
      "logits/rejected": -0.15179498493671417,
      "logps/chosen": -466.480224609375,
      "logps/rejected": -564.7294311523438,
      "loss": 2.0206,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.1005839928984642,
      "rewards/margins": 0.05746855586767197,
      "rewards/rejected": -0.15805256366729736,
      "step": 50
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -0.2444579154253006,
      "logits/rejected": -0.17071941494941711,
      "logps/chosen": -678.161865234375,
      "logps/rejected": -776.6124877929688,
      "loss": 1.9129,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.25659433007240295,
      "rewards/margins": 0.11048071086406708,
      "rewards/rejected": -0.3670750558376312,
      "step": 60
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.7761938666470405e-06,
      "logits/chosen": -0.2789674401283264,
      "logits/rejected": -0.15984465181827545,
      "logps/chosen": -767.4395751953125,
      "logps/rejected": -960.1740112304688,
      "loss": 1.9228,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.37694883346557617,
      "rewards/margins": 0.17607252299785614,
      "rewards/rejected": -0.5530213117599487,
      "step": 70
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -0.1913307160139084,
      "logits/rejected": -0.14110524952411652,
      "logps/chosen": -764.4078979492188,
      "logps/rejected": -998.7781982421875,
      "loss": 1.9313,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.3331916630268097,
      "rewards/margins": 0.25039905309677124,
      "rewards/rejected": -0.5835906863212585,
      "step": 80
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.4890613722044526e-06,
      "logits/chosen": -0.23566904664039612,
      "logits/rejected": -0.15600116550922394,
      "logps/chosen": -777.1472778320312,
      "logps/rejected": -871.2056884765625,
      "loss": 1.7804,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.37899401783943176,
      "rewards/margins": 0.1299220621585846,
      "rewards/rejected": -0.5089160203933716,
      "step": 90
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -0.25944238901138306,
      "logits/rejected": -0.1812916249036789,
      "logps/chosen": -820.12255859375,
      "logps/rejected": -1078.459228515625,
      "loss": 1.7734,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.4638814926147461,
      "rewards/margins": 0.24260278046131134,
      "rewards/rejected": -0.7064841985702515,
      "step": 100
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.102189034962561e-06,
      "logits/chosen": -0.22316499054431915,
      "logits/rejected": -0.19235864281654358,
      "logps/chosen": -885.9295654296875,
      "logps/rejected": -1113.445068359375,
      "loss": 1.8059,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.4847044050693512,
      "rewards/margins": 0.23728513717651367,
      "rewards/rejected": -0.7219894528388977,
      "step": 110
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -0.20930325984954834,
      "logits/rejected": -0.1453815996646881,
      "logps/chosen": -880.3770751953125,
      "logps/rejected": -1030.3275146484375,
      "loss": 1.8441,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.4620415270328522,
      "rewards/margins": 0.17096856236457825,
      "rewards/rejected": -0.6330100893974304,
      "step": 120
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.634976249348867e-06,
      "logits/chosen": -0.15696187317371368,
      "logits/rejected": -0.1074260026216507,
      "logps/chosen": -885.1849365234375,
      "logps/rejected": -1044.3388671875,
      "loss": 1.9112,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.4623374938964844,
      "rewards/margins": 0.17165368795394897,
      "rewards/rejected": -0.6339911222457886,
      "step": 130
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -0.2678197920322418,
      "logits/rejected": -0.23175963759422302,
      "logps/chosen": -806.1285400390625,
      "logps/rejected": -1082.8525390625,
      "loss": 1.8023,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.42001548409461975,
      "rewards/margins": 0.21150164306163788,
      "rewards/rejected": -0.6315171122550964,
      "step": 140
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.1108510153447352e-06,
      "logits/chosen": -0.21387240290641785,
      "logits/rejected": -0.1143934354186058,
      "logps/chosen": -815.8262939453125,
      "logps/rejected": -906.0357666015625,
      "loss": 1.808,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.37800681591033936,
      "rewards/margins": 0.16527661681175232,
      "rewards/rejected": -0.5432834029197693,
      "step": 150
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -0.14601436257362366,
      "logits/rejected": -0.1241162046790123,
      "logps/chosen": -840.3963623046875,
      "logps/rejected": -1006.9793090820312,
      "loss": 1.9465,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.38944554328918457,
      "rewards/margins": 0.19957077503204346,
      "rewards/rejected": -0.589016318321228,
      "step": 160
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.556095160739513e-06,
      "logits/chosen": -0.1653786599636078,
      "logits/rejected": -0.12808741629123688,
      "logps/chosen": -730.0451049804688,
      "logps/rejected": -965.7391357421875,
      "loss": 1.769,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.35579684376716614,
      "rewards/margins": 0.20604142546653748,
      "rewards/rejected": -0.5618382692337036,
      "step": 170
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -0.19225947558879852,
      "logits/rejected": -0.17646418511867523,
      "logps/chosen": -787.419677734375,
      "logps/rejected": -1047.8543701171875,
      "loss": 1.747,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.4115837514400482,
      "rewards/margins": 0.2577529549598694,
      "rewards/rejected": -0.6693366765975952,
      "step": 180
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9985264605418185e-06,
      "logits/chosen": -0.21607378125190735,
      "logits/rejected": -0.18134012818336487,
      "logps/chosen": -850.4358520507812,
      "logps/rejected": -1074.816162109375,
      "loss": 1.7703,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.42398494482040405,
      "rewards/margins": 0.27196767926216125,
      "rewards/rejected": -0.6959526538848877,
      "step": 190
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -0.18237197399139404,
      "logits/rejected": -0.15335983037948608,
      "logps/chosen": -753.6973266601562,
      "logps/rejected": -888.6695556640625,
      "loss": 1.8577,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.3945154547691345,
      "rewards/margins": 0.15583784878253937,
      "rewards/rejected": -0.5503532886505127,
      "step": 200
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.466103737583699e-06,
      "logits/chosen": -0.14478984475135803,
      "logits/rejected": -0.1423361599445343,
      "logps/chosen": -786.3085327148438,
      "logps/rejected": -1014.6956787109375,
      "loss": 1.8453,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.39204883575439453,
      "rewards/margins": 0.2218962460756302,
      "rewards/rejected": -0.6139450073242188,
      "step": 210
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -0.1539977490901947,
      "logits/rejected": -0.10850413888692856,
      "logps/chosen": -802.6807861328125,
      "logps/rejected": -1012.9494018554688,
      "loss": 1.679,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.37582096457481384,
      "rewards/margins": 0.24124114215373993,
      "rewards/rejected": -0.6170621514320374,
      "step": 220
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.855248903979505e-07,
      "logits/chosen": -0.2050359696149826,
      "logits/rejected": -0.11318047344684601,
      "logps/chosen": -806.9402465820312,
      "logps/rejected": -1053.7979736328125,
      "loss": 1.7812,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.3787270784378052,
      "rewards/margins": 0.25472280383110046,
      "rewards/rejected": -0.633449912071228,
      "step": 230
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -0.18793925642967224,
      "logits/rejected": -0.13918839395046234,
      "logps/chosen": -818.8219604492188,
      "logps/rejected": -1038.0279541015625,
      "loss": 1.7375,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.40735897421836853,
      "rewards/margins": 0.23022215068340302,
      "rewards/rejected": -0.6375811696052551,
      "step": 240
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.808881491049723e-07,
      "logits/chosen": -0.13271953165531158,
      "logits/rejected": -0.10349054634571075,
      "logps/chosen": -846.697265625,
      "logps/rejected": -1090.2091064453125,
      "loss": 1.7763,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.39346030354499817,
      "rewards/margins": 0.25347214937210083,
      "rewards/rejected": -0.6469324827194214,
      "step": 250
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -0.10151909291744232,
      "logits/rejected": -0.09530433267354965,
      "logps/chosen": -817.2677612304688,
      "logps/rejected": -1003.5374755859375,
      "loss": 1.7718,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.4086136817932129,
      "rewards/margins": 0.1990220844745636,
      "rewards/rejected": -0.6076357960700989,
      "step": 260
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.7248368952908055e-07,
      "logits/chosen": -0.1809629648923874,
      "logits/rejected": -0.14152123034000397,
      "logps/chosen": -830.4895629882812,
      "logps/rejected": -1007.2437744140625,
      "loss": 1.789,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.4253358840942383,
      "rewards/margins": 0.21133287250995636,
      "rewards/rejected": -0.6366687417030334,
      "step": 270
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -0.1590525209903717,
      "logits/rejected": -0.10606285184621811,
      "logps/chosen": -817.2872924804688,
      "logps/rejected": -1020.6212158203125,
      "loss": 1.8107,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.3928419053554535,
      "rewards/margins": 0.2158069908618927,
      "rewards/rejected": -0.6086488962173462,
      "step": 280
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.577619905828281e-08,
      "logits/chosen": -0.1407514065504074,
      "logits/rejected": -0.14306677877902985,
      "logps/chosen": -782.491455078125,
      "logps/rejected": -1054.53564453125,
      "loss": 1.7329,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.41321054100990295,
      "rewards/margins": 0.27371519804000854,
      "rewards/rejected": -0.6869257092475891,
      "step": 290
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -0.13833969831466675,
      "logits/rejected": -0.14348629117012024,
      "logps/chosen": -808.810791015625,
      "logps/rejected": -1050.609619140625,
      "loss": 1.6973,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.40211352705955505,
      "rewards/margins": 0.259845107793808,
      "rewards/rejected": -0.6619586944580078,
      "step": 300
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.294126437336734e-10,
      "logits/chosen": -0.18316778540611267,
      "logits/rejected": -0.11073551326990128,
      "logps/chosen": -916.8952026367188,
      "logps/rejected": -1118.8402099609375,
      "loss": 1.7781,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.47190985083580017,
      "rewards/margins": 0.21381942927837372,
      "rewards/rejected": -0.6857293248176575,
      "step": 310
    },
    {
      "epoch": 1.0,
      "step": 312,
      "total_flos": 0.0,
      "train_loss": 1.8545298270690136,
      "train_runtime": 4214.6598,
      "train_samples_per_second": 3.559,
      "train_steps_per_second": 0.074
    }
  ],
  "logging_steps": 10,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}