{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 287,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003484320557491289,
      "grad_norm": 225.57155111737745,
      "learning_rate": 6.896551724137931e-09,
      "logits/chosen": -2.5345611572265625,
      "logits/rejected": -2.581700563430786,
      "logps/chosen": -60.002105712890625,
      "logps/rejected": -99.98374938964844,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03484320557491289,
      "grad_norm": 216.6261369698797,
      "learning_rate": 6.896551724137931e-08,
      "logits/chosen": -2.563714027404785,
      "logits/rejected": -2.562321901321411,
      "logps/chosen": -59.66276550292969,
      "logps/rejected": -73.39944458007812,
      "loss": 0.6916,
      "rewards/accuracies": 0.2291666716337204,
      "rewards/chosen": -0.001211778144352138,
      "rewards/margins": 0.00601451750844717,
      "rewards/rejected": -0.007226296700537205,
      "step": 10
    },
    {
      "epoch": 0.06968641114982578,
      "grad_norm": 262.5210498959883,
      "learning_rate": 1.3793103448275863e-07,
      "logits/chosen": -2.6048154830932617,
      "logits/rejected": -2.5637078285217285,
      "logps/chosen": -104.09293365478516,
      "logps/rejected": -94.90653991699219,
      "loss": 0.6903,
      "rewards/accuracies": 0.33125001192092896,
      "rewards/chosen": 0.013311183080077171,
      "rewards/margins": 0.018149277195334435,
      "rewards/rejected": -0.004838095046579838,
      "step": 20
    },
    {
      "epoch": 0.10452961672473868,
      "grad_norm": 268.8517801211917,
      "learning_rate": 1.9961240310077517e-07,
      "logits/chosen": -2.5926074981689453,
      "logits/rejected": -2.572615623474121,
      "logps/chosen": -82.38002014160156,
      "logps/rejected": -91.52497863769531,
      "loss": 0.6791,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.04545076563954353,
      "rewards/margins": 0.039520200341939926,
      "rewards/rejected": 0.005930562969297171,
      "step": 30
    },
    {
      "epoch": 0.13937282229965156,
      "grad_norm": 209.92212859559183,
      "learning_rate": 1.9573643410852711e-07,
      "logits/chosen": -2.4985804557800293,
      "logits/rejected": -2.496473550796509,
      "logps/chosen": -77.8263931274414,
      "logps/rejected": -73.18889617919922,
      "loss": 0.6608,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": 0.03588957339525223,
      "rewards/margins": 0.10904321819543839,
      "rewards/rejected": -0.07315365225076675,
      "step": 40
    },
    {
      "epoch": 0.17421602787456447,
      "grad_norm": 197.5547212101371,
      "learning_rate": 1.9186046511627906e-07,
      "logits/chosen": -2.53145170211792,
      "logits/rejected": -2.5351381301879883,
      "logps/chosen": -63.558860778808594,
      "logps/rejected": -75.93846130371094,
      "loss": 0.6646,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": 0.1483130156993866,
      "rewards/margins": 0.08396634459495544,
      "rewards/rejected": 0.06434665620326996,
      "step": 50
    },
    {
      "epoch": 0.20905923344947736,
      "grad_norm": 191.33111792340372,
      "learning_rate": 1.87984496124031e-07,
      "logits/chosen": -2.4998061656951904,
      "logits/rejected": -2.493762254714966,
      "logps/chosen": -71.91557312011719,
      "logps/rejected": -67.55859375,
      "loss": 0.6549,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 0.5034655928611755,
      "rewards/margins": 0.17197737097740173,
      "rewards/rejected": 0.3314882814884186,
      "step": 60
    },
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 204.75139928311083,
      "learning_rate": 1.8410852713178295e-07,
      "logits/chosen": -2.5222249031066895,
      "logits/rejected": -2.5176424980163574,
      "logps/chosen": -62.58455276489258,
      "logps/rejected": -67.23956298828125,
      "loss": 0.6491,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": 0.644144594669342,
      "rewards/margins": 0.18119266629219055,
      "rewards/rejected": 0.4629518389701843,
      "step": 70
    },
    {
      "epoch": 0.2787456445993031,
      "grad_norm": 237.92667425371948,
      "learning_rate": 1.8023255813953487e-07,
      "logits/chosen": -2.4706761837005615,
      "logits/rejected": -2.4605259895324707,
      "logps/chosen": -73.8516845703125,
      "logps/rejected": -76.4228286743164,
      "loss": 0.6602,
      "rewards/accuracies": 0.3062500059604645,
      "rewards/chosen": 0.7071425318717957,
      "rewards/margins": 0.2812144160270691,
      "rewards/rejected": 0.4259282052516937,
      "step": 80
    },
    {
      "epoch": 0.313588850174216,
      "grad_norm": 188.49653222846555,
      "learning_rate": 1.7635658914728681e-07,
      "logits/chosen": -2.48022723197937,
      "logits/rejected": -2.494345188140869,
      "logps/chosen": -64.58412170410156,
      "logps/rejected": -69.22706604003906,
      "loss": 0.6454,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 0.5452820062637329,
      "rewards/margins": 0.1993606686592102,
      "rewards/rejected": 0.34592127799987793,
      "step": 90
    },
    {
      "epoch": 0.34843205574912894,
      "grad_norm": 230.52946384799608,
      "learning_rate": 1.7248062015503873e-07,
      "logits/chosen": -2.460999011993408,
      "logits/rejected": -2.462146282196045,
      "logps/chosen": -73.88617706298828,
      "logps/rejected": -80.31554412841797,
      "loss": 0.6538,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 0.5401462316513062,
      "rewards/margins": 0.2530631422996521,
      "rewards/rejected": 0.28708305954933167,
      "step": 100
    },
    {
      "epoch": 0.34843205574912894,
      "eval_logits/chosen": -2.5468788146972656,
      "eval_logits/rejected": -2.530789613723755,
      "eval_logps/chosen": -73.00140380859375,
      "eval_logps/rejected": -80.36827850341797,
      "eval_loss": 0.6319671273231506,
      "eval_rewards/accuracies": 0.3472222089767456,
      "eval_rewards/chosen": 0.5810332298278809,
      "eval_rewards/margins": 0.2846057713031769,
      "eval_rewards/rejected": 0.2964273989200592,
      "eval_runtime": 114.037,
      "eval_samples_per_second": 17.538,
      "eval_steps_per_second": 0.552,
      "step": 100
    },
    {
      "epoch": 0.3832752613240418,
      "grad_norm": 251.4434825633798,
      "learning_rate": 1.6860465116279068e-07,
      "logits/chosen": -2.4884157180786133,
      "logits/rejected": -2.44923734664917,
      "logps/chosen": -72.09432220458984,
      "logps/rejected": -62.603981018066406,
      "loss": 0.6295,
      "rewards/accuracies": 0.30000001192092896,
      "rewards/chosen": 0.41952481865882874,
      "rewards/margins": 0.2609493136405945,
      "rewards/rejected": 0.15857550501823425,
      "step": 110
    },
    {
      "epoch": 0.4181184668989547,
      "grad_norm": 183.32886181861778,
      "learning_rate": 1.6472868217054262e-07,
      "logits/chosen": -2.5084633827209473,
      "logits/rejected": -2.4761087894439697,
      "logps/chosen": -76.3115234375,
      "logps/rejected": -66.27922058105469,
      "loss": 0.6149,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.44851747155189514,
      "rewards/margins": 0.29549649357795715,
      "rewards/rejected": 0.15302105247974396,
      "step": 120
    },
    {
      "epoch": 0.4529616724738676,
      "grad_norm": 260.8387325270783,
      "learning_rate": 1.6085271317829457e-07,
      "logits/chosen": -2.5476431846618652,
      "logits/rejected": -2.5282294750213623,
      "logps/chosen": -82.90985870361328,
      "logps/rejected": -87.62440490722656,
      "loss": 0.6315,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": 0.37865743041038513,
      "rewards/margins": 0.42385053634643555,
      "rewards/rejected": -0.045193083584308624,
      "step": 130
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 234.8662840093771,
      "learning_rate": 1.569767441860465e-07,
      "logits/chosen": -2.445432186126709,
      "logits/rejected": -2.4326701164245605,
      "logps/chosen": -80.27764892578125,
      "logps/rejected": -70.95781707763672,
      "loss": 0.6136,
      "rewards/accuracies": 0.34375,
      "rewards/chosen": 0.3384358584880829,
      "rewards/margins": 0.4139274060726166,
      "rewards/rejected": -0.07549159228801727,
      "step": 140
    },
    {
      "epoch": 0.5226480836236934,
      "grad_norm": 234.49493704296844,
      "learning_rate": 1.5310077519379846e-07,
      "logits/chosen": -2.5063323974609375,
      "logits/rejected": -2.4608452320098877,
      "logps/chosen": -78.8966064453125,
      "logps/rejected": -79.84175872802734,
      "loss": 0.6478,
      "rewards/accuracies": 0.3062500059604645,
      "rewards/chosen": 0.22921499609947205,
      "rewards/margins": 0.34269630908966064,
      "rewards/rejected": -0.1134813204407692,
      "step": 150
    },
    {
      "epoch": 0.5574912891986062,
      "grad_norm": 208.63525471171238,
      "learning_rate": 1.4922480620155038e-07,
      "logits/chosen": -2.462036609649658,
      "logits/rejected": -2.483565092086792,
      "logps/chosen": -64.0514907836914,
      "logps/rejected": -72.12866973876953,
      "loss": 0.6534,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": 0.18461140990257263,
      "rewards/margins": 0.24535641074180603,
      "rewards/rejected": -0.060745008289813995,
      "step": 160
    },
    {
      "epoch": 0.5923344947735192,
      "grad_norm": 215.9542751971873,
      "learning_rate": 1.4534883720930232e-07,
      "logits/chosen": -2.459441900253296,
      "logits/rejected": -2.444371461868286,
      "logps/chosen": -68.60945129394531,
      "logps/rejected": -77.02355194091797,
      "loss": 0.6049,
      "rewards/accuracies": 0.3375000059604645,
      "rewards/chosen": 0.1395992934703827,
      "rewards/margins": 0.416309118270874,
      "rewards/rejected": -0.27670982480049133,
      "step": 170
    },
    {
      "epoch": 0.627177700348432,
      "grad_norm": 220.77017684857242,
      "learning_rate": 1.4147286821705424e-07,
      "logits/chosen": -2.5126798152923584,
      "logits/rejected": -2.5011754035949707,
      "logps/chosen": -91.29668426513672,
      "logps/rejected": -86.69538879394531,
      "loss": 0.6539,
      "rewards/accuracies": 0.32499998807907104,
      "rewards/chosen": 0.21858568489551544,
      "rewards/margins": 0.29070785641670227,
      "rewards/rejected": -0.07212214171886444,
      "step": 180
    },
    {
      "epoch": 0.662020905923345,
      "grad_norm": 162.410148810253,
      "learning_rate": 1.3759689922480619e-07,
      "logits/chosen": -2.5120596885681152,
      "logits/rejected": -2.502159595489502,
      "logps/chosen": -70.31501770019531,
      "logps/rejected": -81.38325500488281,
      "loss": 0.6211,
      "rewards/accuracies": 0.3375000059604645,
      "rewards/chosen": 0.28721699118614197,
      "rewards/margins": 0.23069223761558533,
      "rewards/rejected": 0.05652473121881485,
      "step": 190
    },
    {
      "epoch": 0.6968641114982579,
      "grad_norm": 291.03325951738674,
      "learning_rate": 1.3372093023255813e-07,
      "logits/chosen": -2.545940399169922,
      "logits/rejected": -2.5534462928771973,
      "logps/chosen": -89.51722717285156,
      "logps/rejected": -91.98088073730469,
      "loss": 0.6462,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 0.4442625641822815,
      "rewards/margins": 0.4170360565185547,
      "rewards/rejected": 0.02722644805908203,
      "step": 200
    },
    {
      "epoch": 0.6968641114982579,
      "eval_logits/chosen": -2.522193670272827,
      "eval_logits/rejected": -2.505502939224243,
      "eval_logps/chosen": -73.84490966796875,
      "eval_logps/rejected": -81.41748046875,
      "eval_loss": 0.6323519349098206,
      "eval_rewards/accuracies": 0.3492063581943512,
      "eval_rewards/chosen": 0.2520641088485718,
      "eval_rewards/margins": 0.3648246228694916,
      "eval_rewards/rejected": -0.1127605140209198,
      "eval_runtime": 113.9702,
      "eval_samples_per_second": 17.548,
      "eval_steps_per_second": 0.553,
      "step": 200
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 333.8956625644709,
      "learning_rate": 1.2984496124031008e-07,
      "logits/chosen": -2.528879404067993,
      "logits/rejected": -2.501743793487549,
      "logps/chosen": -68.72173309326172,
      "logps/rejected": -64.01664733886719,
      "loss": 0.6317,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": 0.2329626977443695,
      "rewards/margins": 0.471733957529068,
      "rewards/rejected": -0.2387712299823761,
      "step": 210
    },
    {
      "epoch": 0.7665505226480837,
      "grad_norm": 176.716286583645,
      "learning_rate": 1.2596899224806202e-07,
      "logits/chosen": -2.5579559803009033,
      "logits/rejected": -2.535482883453369,
      "logps/chosen": -72.86189270019531,
      "logps/rejected": -71.62294006347656,
      "loss": 0.6392,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": 0.23254899680614471,
      "rewards/margins": 0.21685466170310974,
      "rewards/rejected": 0.015694305300712585,
      "step": 220
    },
    {
      "epoch": 0.8013937282229965,
      "grad_norm": 226.51144499862002,
      "learning_rate": 1.2209302325581394e-07,
      "logits/chosen": -2.5561137199401855,
      "logits/rejected": -2.5325522422790527,
      "logps/chosen": -89.1368637084961,
      "logps/rejected": -88.74254608154297,
      "loss": 0.6316,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": 0.3368709683418274,
      "rewards/margins": 0.5684143900871277,
      "rewards/rejected": -0.23154346644878387,
      "step": 230
    },
    {
      "epoch": 0.8362369337979094,
      "grad_norm": 200.36993244283045,
      "learning_rate": 1.1821705426356587e-07,
      "logits/chosen": -2.5511903762817383,
      "logits/rejected": -2.5139944553375244,
      "logps/chosen": -86.218994140625,
      "logps/rejected": -80.59002685546875,
      "loss": 0.6428,
      "rewards/accuracies": 0.34375,
      "rewards/chosen": 0.2038179188966751,
      "rewards/margins": 0.3808489143848419,
      "rewards/rejected": -0.17703098058700562,
      "step": 240
    },
    {
      "epoch": 0.8710801393728222,
      "grad_norm": 220.61998054567871,
      "learning_rate": 1.1434108527131783e-07,
      "logits/chosen": -2.566251516342163,
      "logits/rejected": -2.52921199798584,
      "logps/chosen": -95.03993225097656,
      "logps/rejected": -90.61654663085938,
      "loss": 0.5848,
      "rewards/accuracies": 0.39375001192092896,
      "rewards/chosen": 0.15963879227638245,
      "rewards/margins": 0.42108219861984253,
      "rewards/rejected": -0.26144343614578247,
      "step": 250
    },
    {
      "epoch": 0.9059233449477352,
      "grad_norm": 172.2124811551973,
      "learning_rate": 1.1046511627906976e-07,
      "logits/chosen": -2.4705214500427246,
      "logits/rejected": -2.48685622215271,
      "logps/chosen": -58.55452346801758,
      "logps/rejected": -65.9000015258789,
      "loss": 0.6216,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.19298657774925232,
      "rewards/margins": 0.29895681142807007,
      "rewards/rejected": -0.10597021877765656,
      "step": 260
    },
    {
      "epoch": 0.9407665505226481,
      "grad_norm": 221.01197823196617,
      "learning_rate": 1.0658914728682171e-07,
      "logits/chosen": -2.5705666542053223,
      "logits/rejected": -2.5713424682617188,
      "logps/chosen": -68.17798614501953,
      "logps/rejected": -83.17510223388672,
      "loss": 0.6039,
      "rewards/accuracies": 0.34375,
      "rewards/chosen": 0.2738538086414337,
      "rewards/margins": 0.5247575640678406,
      "rewards/rejected": -0.2509037256240845,
      "step": 270
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 201.10787443979413,
      "learning_rate": 1.0271317829457364e-07,
      "logits/chosen": -2.4812636375427246,
      "logits/rejected": -2.4561610221862793,
      "logps/chosen": -67.36260986328125,
      "logps/rejected": -71.26799011230469,
      "loss": 0.6039,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.18398845195770264,
      "rewards/margins": 0.3530852496623993,
      "rewards/rejected": -0.16909679770469666,
      "step": 280
    },
    {
      "epoch": 1.0,
      "step": 287,
      "total_flos": 0.0,
      "train_loss": 0.6390685353960309,
      "train_runtime": 3341.9408,
      "train_samples_per_second": 5.488,
      "train_steps_per_second": 0.086
    }
  ],
  "logging_steps": 10,
  "max_steps": 287,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}