{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.14496692941922623,
  "eval_steps": 50,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007248346470961312,
      "grad_norm": 0.07235438376665115,
      "learning_rate": 4.998766400914329e-06,
      "logits/chosen": -1.9785633087158203,
      "logits/rejected": -2.5380234718322754,
      "logps/chosen": -0.28137442469596863,
      "logps/rejected": -0.37793582677841187,
      "loss": 7.3908,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.42206162214279175,
      "rewards/margins": 0.14484205842018127,
      "rewards/rejected": -0.5669037103652954,
      "step": 10
    },
    {
      "epoch": 0.014496692941922623,
      "grad_norm": 0.07455909997224808,
      "learning_rate": 4.9950668210706795e-06,
      "logits/chosen": -2.0576319694519043,
      "logits/rejected": -2.499069929122925,
      "logps/chosen": -0.2769497036933899,
      "logps/rejected": -0.33560773730278015,
      "loss": 7.3753,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.41542449593544006,
      "rewards/margins": 0.08798708021640778,
      "rewards/rejected": -0.503411591053009,
      "step": 20
    },
    {
      "epoch": 0.021745039412883936,
      "grad_norm": 0.09561269730329514,
      "learning_rate": 4.9889049115077e-06,
      "logits/chosen": -2.1018643379211426,
      "logits/rejected": -2.377673625946045,
      "logps/chosen": -0.2673099935054779,
      "logps/rejected": -0.3057115972042084,
      "loss": 7.4516,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.40096497535705566,
      "rewards/margins": 0.05760239437222481,
      "rewards/rejected": -0.45856744050979614,
      "step": 30
    },
    {
      "epoch": 0.028993385883845247,
      "grad_norm": 0.08169445395469666,
      "learning_rate": 4.980286753286196e-06,
      "logits/chosen": -2.1627352237701416,
      "logits/rejected": -2.3873391151428223,
      "logps/chosen": -0.27636194229125977,
      "logps/rejected": -0.3703404664993286,
      "loss": 7.3888,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.4145428538322449,
      "rewards/margins": 0.14096775650978088,
      "rewards/rejected": -0.5555106401443481,
      "step": 40
    },
    {
      "epoch": 0.03624173235480656,
      "grad_norm": 0.08877147734165192,
      "learning_rate": 4.9692208514878445e-06,
      "logits/chosen": -2.1400368213653564,
      "logits/rejected": -2.44627046585083,
      "logps/chosen": -0.25384199619293213,
      "logps/rejected": -0.309225857257843,
      "loss": 7.518,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.3807629942893982,
      "rewards/margins": 0.08307582885026932,
      "rewards/rejected": -0.4638388752937317,
      "step": 50
    },
    {
      "epoch": 0.03624173235480656,
      "eval_logits/chosen": -2.1149837970733643,
      "eval_logits/rejected": -2.4759538173675537,
      "eval_logps/chosen": -0.28288090229034424,
      "eval_logps/rejected": -0.3434317111968994,
      "eval_loss": 0.9120966196060181,
      "eval_rewards/accuracies": 0.5089285969734192,
      "eval_rewards/chosen": -0.42432135343551636,
      "eval_rewards/margins": 0.09082622081041336,
      "eval_rewards/rejected": -0.5151475667953491,
      "eval_runtime": 30.2473,
      "eval_samples_per_second": 29.49,
      "eval_steps_per_second": 3.703,
      "step": 50
    },
    {
      "epoch": 0.04349007882576787,
      "grad_norm": 0.11774340271949768,
      "learning_rate": 4.9557181268217225e-06,
      "logits/chosen": -2.0567967891693115,
      "logits/rejected": -2.545668125152588,
      "logps/chosen": -0.2460743933916092,
      "logps/rejected": -0.32390663027763367,
      "loss": 7.3929,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.3691115975379944,
      "rewards/margins": 0.1167483925819397,
      "rewards/rejected": -0.4858599603176117,
      "step": 60
    },
    {
      "epoch": 0.05073842529672919,
      "grad_norm": 0.10012238472700119,
      "learning_rate": 4.939791904846869e-06,
      "logits/chosen": -1.9286606311798096,
      "logits/rejected": -2.4620895385742188,
      "logps/chosen": -0.25992274284362793,
      "logps/rejected": -0.3665553629398346,
      "loss": 7.1863,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.3898841440677643,
      "rewards/margins": 0.15994893014431,
      "rewards/rejected": -0.5498330593109131,
      "step": 70
    },
    {
      "epoch": 0.057986771767690494,
      "grad_norm": 0.07387609034776688,
      "learning_rate": 4.921457902821578e-06,
      "logits/chosen": -2.0778262615203857,
      "logits/rejected": -2.4614813327789307,
      "logps/chosen": -0.24521782994270325,
      "logps/rejected": -0.3450419306755066,
      "loss": 7.3662,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.3678267002105713,
      "rewards/margins": 0.1497361809015274,
      "rewards/rejected": -0.5175628662109375,
      "step": 80
    },
    {
      "epoch": 0.0652351182386518,
      "grad_norm": 0.09718403220176697,
      "learning_rate": 4.900734214192358e-06,
      "logits/chosen": -1.998186707496643,
      "logits/rejected": -2.4382071495056152,
      "logps/chosen": -0.24080593883991241,
      "logps/rejected": -0.3280099332332611,
      "loss": 7.268,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.36120888590812683,
      "rewards/margins": 0.13080602884292603,
      "rewards/rejected": -0.49201488494873047,
      "step": 90
    },
    {
      "epoch": 0.07248346470961312,
      "grad_norm": 0.08059138059616089,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -1.894728422164917,
      "logits/rejected": -2.475407838821411,
      "logps/chosen": -0.2019994705915451,
      "logps/rejected": -0.29733040928840637,
      "loss": 7.2544,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.30299919843673706,
      "rewards/margins": 0.14299637079238892,
      "rewards/rejected": -0.445995569229126,
      "step": 100
    },
    {
      "epoch": 0.07248346470961312,
      "eval_logits/chosen": -2.1233773231506348,
      "eval_logits/rejected": -2.4810428619384766,
      "eval_logps/chosen": -0.24400465190410614,
      "eval_logps/rejected": -0.3260345160961151,
      "eval_loss": 0.8920583724975586,
      "eval_rewards/accuracies": 0.5625,
      "eval_rewards/chosen": -0.3660070598125458,
      "eval_rewards/margins": 0.12304472178220749,
      "eval_rewards/rejected": -0.4890517592430115,
      "eval_runtime": 30.28,
      "eval_samples_per_second": 29.458,
      "eval_steps_per_second": 3.699,
      "step": 100
    },
    {
      "epoch": 0.07973181118057443,
      "grad_norm": 0.060006480664014816,
      "learning_rate": 4.852201922385564e-06,
      "logits/chosen": -2.1128883361816406,
      "logits/rejected": -2.5161118507385254,
      "logps/chosen": -0.23731942474842072,
      "logps/rejected": -0.31481316685676575,
      "loss": 7.1762,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.3559790849685669,
      "rewards/margins": 0.11624068021774292,
      "rewards/rejected": -0.4722197651863098,
      "step": 110
    },
    {
      "epoch": 0.08698015765153574,
      "grad_norm": 0.0602690726518631,
      "learning_rate": 4.824441214720629e-06,
      "logits/chosen": -2.0212602615356445,
      "logits/rejected": -2.4894890785217285,
      "logps/chosen": -0.18956038355827332,
      "logps/rejected": -0.31935763359069824,
      "loss": 7.1442,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2843405604362488,
      "rewards/margins": 0.19469590485095978,
      "rewards/rejected": -0.47903648018836975,
      "step": 120
    },
    {
      "epoch": 0.09422850412249706,
      "grad_norm": 0.06425223499536514,
      "learning_rate": 4.794386564209953e-06,
      "logits/chosen": -2.06717848777771,
      "logits/rejected": -2.5086405277252197,
      "logps/chosen": -0.2075866460800171,
      "logps/rejected": -0.3124083876609802,
      "loss": 7.1449,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.31137990951538086,
      "rewards/margins": 0.15723267197608948,
      "rewards/rejected": -0.46861258149147034,
      "step": 130
    },
    {
      "epoch": 0.10147685059345837,
      "grad_norm": 0.08793757855892181,
      "learning_rate": 4.762067631165049e-06,
      "logits/chosen": -2.0242953300476074,
      "logits/rejected": -2.4943737983703613,
      "logps/chosen": -0.19989363849163055,
      "logps/rejected": -0.32057636976242065,
      "loss": 7.122,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.29984045028686523,
      "rewards/margins": 0.18102414906024933,
      "rewards/rejected": -0.48086461424827576,
      "step": 140
    },
    {
      "epoch": 0.10872519706441967,
      "grad_norm": 0.09698841720819473,
      "learning_rate": 4.72751631047092e-06,
      "logits/chosen": -2.098658800125122,
      "logits/rejected": -2.582111120223999,
      "logps/chosen": -0.18897351622581482,
      "logps/rejected": -0.363709419965744,
      "loss": 7.1253,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.28346025943756104,
      "rewards/margins": 0.2621038556098938,
      "rewards/rejected": -0.5455641150474548,
      "step": 150
    },
    {
      "epoch": 0.10872519706441967,
      "eval_logits/chosen": -2.1921768188476562,
      "eval_logits/rejected": -2.556370973587036,
      "eval_logps/chosen": -0.22214026749134064,
      "eval_logps/rejected": -0.3240993022918701,
      "eval_loss": 0.8764163851737976,
      "eval_rewards/accuracies": 0.5803571343421936,
      "eval_rewards/chosen": -0.33321040868759155,
      "eval_rewards/margins": 0.15293852984905243,
      "eval_rewards/rejected": -0.48614898324012756,
      "eval_runtime": 30.273,
      "eval_samples_per_second": 29.465,
      "eval_steps_per_second": 3.7,
      "step": 150
    },
    {
      "epoch": 0.11597354353538099,
      "grad_norm": 0.09119638800621033,
      "learning_rate": 4.690766700109659e-06,
      "logits/chosen": -2.2321066856384277,
      "logits/rejected": -2.5475690364837646,
      "logps/chosen": -0.19602122902870178,
      "logps/rejected": -0.27682799100875854,
      "loss": 7.0307,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.2940318286418915,
      "rewards/margins": 0.12121014297008514,
      "rewards/rejected": -0.4152420163154602,
      "step": 160
    },
    {
      "epoch": 0.1232218900063423,
      "grad_norm": 0.11496366560459137,
      "learning_rate": 4.65185506750986e-06,
      "logits/chosen": -2.1506271362304688,
      "logits/rejected": -2.506520986557007,
      "logps/chosen": -0.18018664419651031,
      "logps/rejected": -0.3084454834461212,
      "loss": 7.1652,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.27028003334999084,
      "rewards/margins": 0.19238826632499695,
      "rewards/rejected": -0.462668240070343,
      "step": 170
    },
    {
      "epoch": 0.1304702364773036,
      "grad_norm": 0.09201692789793015,
      "learning_rate": 4.610819813755038e-06,
      "logits/chosen": -2.285245895385742,
      "logits/rejected": -2.573941707611084,
      "logps/chosen": -0.1844937801361084,
      "logps/rejected": -0.3112415075302124,
      "loss": 7.0022,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2767406404018402,
      "rewards/margins": 0.1901216208934784,
      "rewards/rejected": -0.4668622612953186,
      "step": 180
    },
    {
      "epoch": 0.13771858294826492,
      "grad_norm": 0.13191230595111847,
      "learning_rate": 4.567701435686405e-06,
      "logits/chosen": -2.265538454055786,
      "logits/rejected": -2.569638252258301,
      "logps/chosen": -0.21122264862060547,
      "logps/rejected": -0.3469196856021881,
      "loss": 7.004,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.3168340027332306,
      "rewards/margins": 0.20354552567005157,
      "rewards/rejected": -0.5203795433044434,
      "step": 190
    },
    {
      "epoch": 0.14496692941922623,
      "grad_norm": 0.1503789722919464,
      "learning_rate": 4.522542485937369e-06,
      "logits/chosen": -2.2021026611328125,
      "logits/rejected": -2.655961513519287,
      "logps/chosen": -0.19664816558361053,
      "logps/rejected": -0.34679359197616577,
      "loss": 7.0539,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.2949722409248352,
      "rewards/margins": 0.22521813213825226,
      "rewards/rejected": -0.5201903581619263,
      "step": 200
    },
    {
      "epoch": 0.14496692941922623,
      "eval_logits/chosen": -2.325611114501953,
      "eval_logits/rejected": -2.715676784515381,
      "eval_logps/chosen": -0.2261900007724762,
      "eval_logps/rejected": -0.3570065200328827,
      "eval_loss": 0.8565592169761658,
      "eval_rewards/accuracies": 0.5714285969734192,
      "eval_rewards/chosen": -0.3392849862575531,
      "eval_rewards/margins": 0.19622473418712616,
      "eval_rewards/rejected": -0.5355097055435181,
      "eval_runtime": 30.2681,
      "eval_samples_per_second": 29.47,
      "eval_steps_per_second": 3.7,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.387465726191206e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}