{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 100,
  "global_step": 192,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010416666666666666,
      "grad_norm": 24.66960228608589,
      "learning_rate": 2.0229548209232687e-08,
      "logits/chosen": -2.590585231781006,
      "logits/rejected": -2.5664222240448,
      "logps/chosen": -80.29847717285156,
      "logps/rejected": -53.10200881958008,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.10416666666666667,
      "grad_norm": 22.00841805745166,
      "learning_rate": 2.0229548209232686e-07,
      "logits/chosen": -2.5565450191497803,
      "logits/rejected": -2.5385091304779053,
      "logps/chosen": -87.94749450683594,
      "logps/rejected": -81.03849792480469,
      "loss": 0.6933,
      "rewards/accuracies": 0.1666666716337204,
      "rewards/chosen": -0.0012115496210753918,
      "rewards/margins": -0.001211885130032897,
      "rewards/rejected": 3.3554130141055794e-07,
      "step": 10
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 19.440282490720026,
      "learning_rate": 4.045909641846537e-07,
      "logits/chosen": -2.612945318222046,
      "logits/rejected": -2.5642082691192627,
      "logps/chosen": -103.17449951171875,
      "logps/rejected": -89.7829360961914,
      "loss": 0.6911,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.006612158380448818,
      "rewards/margins": 0.0028112882282584906,
      "rewards/rejected": 0.0038008708506822586,
      "step": 20
    },
    {
      "epoch": 0.3125,
      "grad_norm": 25.805649105098475,
      "learning_rate": 3.8688218719717384e-07,
      "logits/chosen": -2.508690357208252,
      "logits/rejected": -2.5234837532043457,
      "logps/chosen": -66.37418365478516,
      "logps/rejected": -74.30455017089844,
      "loss": 0.6851,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.03983448073267937,
      "rewards/margins": 0.016553020104765892,
      "rewards/rejected": 0.023281460627913475,
      "step": 30
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 17.133204904463383,
      "learning_rate": 3.6917341020969396e-07,
      "logits/chosen": -2.555058479309082,
      "logits/rejected": -2.5424702167510986,
      "logps/chosen": -69.98268127441406,
      "logps/rejected": -69.5423583984375,
      "loss": 0.6777,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": 0.08108103275299072,
      "rewards/margins": 0.03720525652170181,
      "rewards/rejected": 0.04387578368186951,
      "step": 40
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 22.625402130027865,
      "learning_rate": 3.5146463322221414e-07,
      "logits/chosen": -2.4543604850769043,
      "logits/rejected": -2.467179775238037,
      "logps/chosen": -49.026615142822266,
      "logps/rejected": -58.0882568359375,
      "loss": 0.6703,
      "rewards/accuracies": 0.23125000298023224,
      "rewards/chosen": 0.07023879140615463,
      "rewards/margins": 0.03948055952787399,
      "rewards/rejected": 0.030758222565054893,
      "step": 50
    },
    {
      "epoch": 0.625,
      "grad_norm": 23.61323531343505,
      "learning_rate": 3.3375585623473426e-07,
      "logits/chosen": -2.530721664428711,
      "logits/rejected": -2.5115602016448975,
      "logps/chosen": -78.50758361816406,
      "logps/rejected": -81.587158203125,
      "loss": 0.6625,
      "rewards/accuracies": 0.32499998807907104,
      "rewards/chosen": 0.02104257419705391,
      "rewards/margins": 0.1053396612405777,
      "rewards/rejected": -0.08429709076881409,
      "step": 60
    },
    {
      "epoch": 0.7291666666666666,
      "grad_norm": 21.448302962597314,
      "learning_rate": 3.1604707924725434e-07,
      "logits/chosen": -2.4924867153167725,
      "logits/rejected": -2.4797446727752686,
      "logps/chosen": -92.79881286621094,
      "logps/rejected": -86.05406188964844,
      "loss": 0.6497,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 0.13287237286567688,
      "rewards/margins": 0.15550200641155243,
      "rewards/rejected": -0.02262965776026249,
      "step": 70
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 34.07385900953939,
      "learning_rate": 2.983383022597745e-07,
      "logits/chosen": -2.470043182373047,
      "logits/rejected": -2.412937879562378,
      "logps/chosen": -82.65133666992188,
      "logps/rejected": -82.9260482788086,
      "loss": 0.636,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": 0.11218090355396271,
      "rewards/margins": 0.24496188759803772,
      "rewards/rejected": -0.13278096914291382,
      "step": 80
    },
    {
      "epoch": 0.9375,
      "grad_norm": 29.698805422230546,
      "learning_rate": 2.806295252722946e-07,
      "logits/chosen": -2.402101755142212,
      "logits/rejected": -2.3976664543151855,
      "logps/chosen": -55.17719650268555,
      "logps/rejected": -69.18394470214844,
      "loss": 0.6461,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": -0.04486609995365143,
      "rewards/margins": 0.1693090796470642,
      "rewards/rejected": -0.21417517960071564,
      "step": 90
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 22.229206377048136,
      "learning_rate": 2.6292074828481476e-07,
      "logits/chosen": -2.388129711151123,
      "logits/rejected": -2.3694980144500732,
      "logps/chosen": -71.2069320678711,
      "logps/rejected": -80.87518310546875,
      "loss": 0.5735,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": 0.08674298971891403,
      "rewards/margins": 0.4183635711669922,
      "rewards/rejected": -0.33162060379981995,
      "step": 100
    },
    {
      "epoch": 1.0416666666666667,
      "eval_logits/chosen": -2.4165902137756348,
      "eval_logits/rejected": -2.3982186317443848,
      "eval_logps/chosen": -79.08678436279297,
      "eval_logps/rejected": -91.30322265625,
      "eval_loss": 0.6444148421287537,
      "eval_rewards/accuracies": 0.32341268658638,
      "eval_rewards/chosen": -0.1721993237733841,
      "eval_rewards/margins": 0.2090616375207901,
      "eval_rewards/rejected": -0.3812609612941742,
      "eval_runtime": 113.7367,
      "eval_samples_per_second": 17.584,
      "eval_steps_per_second": 0.554,
      "step": 100
    },
    {
      "epoch": 1.1458333333333333,
      "grad_norm": 18.970483409354923,
      "learning_rate": 2.452119712973349e-07,
      "logits/chosen": -2.379981279373169,
      "logits/rejected": -2.420714855194092,
      "logps/chosen": -61.6137809753418,
      "logps/rejected": -94.87492370605469,
      "loss": 0.5017,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": 0.1406974196434021,
      "rewards/margins": 0.6424610018730164,
      "rewards/rejected": -0.5017635822296143,
      "step": 110
    },
    {
      "epoch": 1.25,
      "grad_norm": 25.179155759555773,
      "learning_rate": 2.27503194309855e-07,
      "logits/chosen": -2.4509730339050293,
      "logits/rejected": -2.412595510482788,
      "logps/chosen": -103.92436218261719,
      "logps/rejected": -112.41646575927734,
      "loss": 0.5125,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": 0.1529318243265152,
      "rewards/margins": 0.8678812980651855,
      "rewards/rejected": -0.7149494886398315,
      "step": 120
    },
    {
      "epoch": 1.3541666666666667,
      "grad_norm": 24.005615665444957,
      "learning_rate": 2.0979441732237514e-07,
      "logits/chosen": -2.365518093109131,
      "logits/rejected": -2.3705790042877197,
      "logps/chosen": -82.37178802490234,
      "logps/rejected": -104.2152328491211,
      "loss": 0.4926,
      "rewards/accuracies": 0.4312500059604645,
      "rewards/chosen": 0.15715977549552917,
      "rewards/margins": 0.7146759033203125,
      "rewards/rejected": -0.5575161576271057,
      "step": 130
    },
    {
      "epoch": 1.4583333333333333,
      "grad_norm": 30.460528822298023,
      "learning_rate": 1.9208564033489532e-07,
      "logits/chosen": -2.3874213695526123,
      "logits/rejected": -2.410029888153076,
      "logps/chosen": -52.55329513549805,
      "logps/rejected": -84.75013732910156,
      "loss": 0.5088,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.12236322462558746,
      "rewards/margins": 0.6131333708763123,
      "rewards/rejected": -0.7354966402053833,
      "step": 140
    },
    {
      "epoch": 1.5625,
      "grad_norm": 23.593513513479692,
      "learning_rate": 1.7437686334741544e-07,
      "logits/chosen": -2.447330951690674,
      "logits/rejected": -2.4216654300689697,
      "logps/chosen": -73.29743957519531,
      "logps/rejected": -96.33416748046875,
      "loss": 0.4942,
      "rewards/accuracies": 0.4312500059604645,
      "rewards/chosen": -0.04307596758008003,
      "rewards/margins": 0.809691309928894,
      "rewards/rejected": -0.852767288684845,
      "step": 150
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 30.463548320289735,
      "learning_rate": 1.5666808635993556e-07,
      "logits/chosen": -2.4006917476654053,
      "logits/rejected": -2.410930871963501,
      "logps/chosen": -77.80570983886719,
      "logps/rejected": -112.13191223144531,
      "loss": 0.4833,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 0.116888128221035,
      "rewards/margins": 1.0234750509262085,
      "rewards/rejected": -0.9065868258476257,
      "step": 160
    },
    {
      "epoch": 1.7708333333333335,
      "grad_norm": 21.813576335226717,
      "learning_rate": 1.389593093724557e-07,
      "logits/chosen": -2.349498748779297,
      "logits/rejected": -2.345773458480835,
      "logps/chosen": -75.15486145019531,
      "logps/rejected": -106.83766174316406,
      "loss": 0.4879,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": 0.00244431896135211,
      "rewards/margins": 0.8961756825447083,
      "rewards/rejected": -0.893731415271759,
      "step": 170
    },
    {
      "epoch": 1.875,
      "grad_norm": 24.30833424266402,
      "learning_rate": 1.2125053238497584e-07,
      "logits/chosen": -2.3641536235809326,
      "logits/rejected": -2.3408045768737793,
      "logps/chosen": -69.39147186279297,
      "logps/rejected": -93.45851135253906,
      "loss": 0.467,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": 0.09233461320400238,
      "rewards/margins": 0.8359875679016113,
      "rewards/rejected": -0.7436529397964478,
      "step": 180
    },
    {
      "epoch": 1.9791666666666665,
      "grad_norm": 22.898056429808634,
      "learning_rate": 1.0354175539749597e-07,
      "logits/chosen": -2.3930718898773193,
      "logits/rejected": -2.3902525901794434,
      "logps/chosen": -86.51902770996094,
      "logps/rejected": -118.46821594238281,
      "loss": 0.4742,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": 0.07279382646083832,
      "rewards/margins": 1.116233229637146,
      "rewards/rejected": -1.0434393882751465,
      "step": 190
    },
    {
      "epoch": 2.0,
      "step": 192,
      "total_flos": 0.0,
      "train_loss": 0.0,
      "train_runtime": 0.0564,
      "train_samples_per_second": 216706.859,
      "train_steps_per_second": 3403.216
    }
  ],
  "logging_steps": 10,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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
}