{
  "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": 284.73684625892474,
      "learning_rate": 4.386080498729062e-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": 269.4644096553184,
      "learning_rate": 4.386080498729062e-07,
      "logits/chosen": -2.5559744834899902,
      "logits/rejected": -2.5380678176879883,
      "logps/chosen": -87.90437316894531,
      "logps/rejected": -80.99159240722656,
      "loss": 0.698,
      "rewards/accuracies": 0.2013888955116272,
      "rewards/chosen": 0.004661789629608393,
      "rewards/margins": -0.01561733242124319,
      "rewards/rejected": 0.02027912251651287,
      "step": 10
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 171.18434231702736,
      "learning_rate": 8.772160997458124e-07,
      "logits/chosen": -2.604264736175537,
      "logits/rejected": -2.5549380779266357,
      "logps/chosen": -102.35679626464844,
      "logps/rejected": -89.08101654052734,
      "loss": 0.6597,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 0.4296715259552002,
      "rewards/margins": 0.08247259259223938,
      "rewards/rejected": 0.34719887375831604,
      "step": 20
    },
    {
      "epoch": 0.3125,
      "grad_norm": 307.77121119192236,
      "learning_rate": 8.320291172024512e-07,
      "logits/chosen": -2.482062578201294,
      "logits/rejected": -2.4983456134796143,
      "logps/chosen": -64.65339660644531,
      "logps/rejected": -73.08040618896484,
      "loss": 0.7129,
      "rewards/accuracies": 0.3499999940395355,
      "rewards/chosen": 1.2031724452972412,
      "rewards/margins": 0.40556639432907104,
      "rewards/rejected": 0.7976058721542358,
      "step": 30
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 247.69654790448098,
      "learning_rate": 7.868421346590901e-07,
      "logits/chosen": -2.5462698936462402,
      "logits/rejected": -2.5348854064941406,
      "logps/chosen": -68.5112075805664,
      "logps/rejected": -67.58045196533203,
      "loss": 0.7089,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": 1.5711699724197388,
      "rewards/margins": 0.21717211604118347,
      "rewards/rejected": 1.353997826576233,
      "step": 40
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 245.28947413610663,
      "learning_rate": 7.416551521157288e-07,
      "logits/chosen": -2.4909956455230713,
      "logits/rejected": -2.5039918422698975,
      "logps/chosen": -47.20299530029297,
      "logps/rejected": -55.72983932495117,
      "loss": 0.7414,
      "rewards/accuracies": 0.24375000596046448,
      "rewards/chosen": 1.5983082056045532,
      "rewards/margins": 0.22424061596393585,
      "rewards/rejected": 1.3740675449371338,
      "step": 50
    },
    {
      "epoch": 0.625,
      "grad_norm": 185.80480485868813,
      "learning_rate": 6.964681695723677e-07,
      "logits/chosen": -2.6212210655212402,
      "logits/rejected": -2.6057589054107666,
      "logps/chosen": -73.73361206054688,
      "logps/rejected": -75.0650863647461,
      "loss": 0.7278,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 2.306000232696533,
      "rewards/margins": 0.4594852924346924,
      "rewards/rejected": 1.8465149402618408,
      "step": 60
    },
    {
      "epoch": 0.7291666666666666,
      "grad_norm": 238.44222468726358,
      "learning_rate": 6.512811870290064e-07,
      "logits/chosen": -2.616278886795044,
      "logits/rejected": -2.6074726581573486,
      "logps/chosen": -91.4566421508789,
      "logps/rejected": -81.8633041381836,
      "loss": 0.7043,
      "rewards/accuracies": 0.35624998807907104,
      "rewards/chosen": 2.112656354904175,
      "rewards/margins": 0.5624496936798096,
      "rewards/rejected": 1.5502065420150757,
      "step": 70
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 359.10559072083385,
      "learning_rate": 6.060942044856453e-07,
      "logits/chosen": -2.593449115753174,
      "logits/rejected": -2.54801607131958,
      "logps/chosen": -80.93285369873047,
      "logps/rejected": -76.7210693359375,
      "loss": 0.7058,
      "rewards/accuracies": 0.41874998807907104,
      "rewards/chosen": 2.0366339683532715,
      "rewards/margins": 0.886386513710022,
      "rewards/rejected": 1.1502474546432495,
      "step": 80
    },
    {
      "epoch": 0.9375,
      "grad_norm": 283.4561861873084,
      "learning_rate": 5.609072219422841e-07,
      "logits/chosen": -2.5106282234191895,
      "logits/rejected": -2.506967306137085,
      "logps/chosen": -51.3546028137207,
      "logps/rejected": -61.58504104614258,
      "loss": 0.7583,
      "rewards/accuracies": 0.23125000298023224,
      "rewards/chosen": 1.134612798690796,
      "rewards/margins": 0.32073575258255005,
      "rewards/rejected": 0.8138772249221802,
      "step": 90
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 21.926819298392022,
      "learning_rate": 5.157202393989229e-07,
      "logits/chosen": -2.4917004108428955,
      "logits/rejected": -2.476295232772827,
      "logps/chosen": -65.5259780883789,
      "logps/rejected": -74.4820327758789,
      "loss": 0.5775,
      "rewards/accuracies": 0.40625,
      "rewards/chosen": 3.455796718597412,
      "rewards/margins": 4.517679691314697,
      "rewards/rejected": -1.0618823766708374,
      "step": 100
    },
    {
      "epoch": 1.0416666666666667,
      "eval_logits/chosen": -2.5355653762817383,
      "eval_logits/rejected": -2.5199732780456543,
      "eval_logps/chosen": -72.7430191040039,
      "eval_logps/rejected": -80.81999206542969,
      "eval_loss": 0.7711760401725769,
      "eval_rewards/accuracies": 0.3313491940498352,
      "eval_rewards/chosen": 0.7555689811706543,
      "eval_rewards/margins": 0.6222997903823853,
      "eval_rewards/rejected": 0.1332692801952362,
      "eval_runtime": 113.7538,
      "eval_samples_per_second": 17.582,
      "eval_steps_per_second": 0.554,
      "step": 100
    },
    {
      "epoch": 1.1458333333333333,
      "grad_norm": 11.001098983606164,
      "learning_rate": 4.7053325685556164e-07,
      "logits/chosen": -2.5152523517608643,
      "logits/rejected": -2.542959690093994,
      "logps/chosen": -55.308494567871094,
      "logps/rejected": -93.90345001220703,
      "loss": 0.3605,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": 4.347947120666504,
      "rewards/margins": 9.715516090393066,
      "rewards/rejected": -5.367568492889404,
      "step": 110
    },
    {
      "epoch": 1.25,
      "grad_norm": 5.506771482244887,
      "learning_rate": 4.2534627431220053e-07,
      "logits/chosen": -2.5518078804016113,
      "logits/rejected": -2.5186076164245605,
      "logps/chosen": -95.32207489013672,
      "logps/rejected": -115.25175476074219,
      "loss": 0.3819,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 5.481812477111816,
      "rewards/margins": 14.953575134277344,
      "rewards/rejected": -9.471762657165527,
      "step": 120
    },
    {
      "epoch": 1.3541666666666667,
      "grad_norm": 80.99035607988066,
      "learning_rate": 3.801592917688393e-07,
      "logits/chosen": -2.4438960552215576,
      "logits/rejected": -2.44427490234375,
      "logps/chosen": -80.55229187011719,
      "logps/rejected": -110.3847885131836,
      "loss": 0.3763,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": 2.599090814590454,
      "rewards/margins": 11.696044921875,
      "rewards/rejected": -9.096954345703125,
      "step": 130
    },
    {
      "epoch": 1.4583333333333333,
      "grad_norm": 81.79049224504786,
      "learning_rate": 3.3497230922547816e-07,
      "logits/chosen": -2.442150831222534,
      "logits/rejected": -2.4587113857269287,
      "logps/chosen": -47.28990936279297,
      "logps/rejected": -80.22529602050781,
      "loss": 0.3766,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": 0.8634545207023621,
      "rewards/margins": 7.3911895751953125,
      "rewards/rejected": -6.527735233306885,
      "step": 140
    },
    {
      "epoch": 1.5625,
      "grad_norm": 2.5199873819499934,
      "learning_rate": 2.89785326682117e-07,
      "logits/chosen": -2.4821057319641113,
      "logits/rejected": -2.4555060863494873,
      "logps/chosen": -67.4425277709961,
      "logps/rejected": -90.55575561523438,
      "loss": 0.3829,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": 2.033623456954956,
      "rewards/margins": 9.372200012207031,
      "rewards/rejected": -7.338575839996338,
      "step": 150
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 1.8477188844357901,
      "learning_rate": 2.445983441387558e-07,
      "logits/chosen": -2.4257683753967285,
      "logits/rejected": -2.4348573684692383,
      "logps/chosen": -72.42179870605469,
      "logps/rejected": -111.45719146728516,
      "loss": 0.3664,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": 3.675116777420044,
      "rewards/margins": 13.840252876281738,
      "rewards/rejected": -10.165136337280273,
      "step": 160
    },
    {
      "epoch": 1.7708333333333335,
      "grad_norm": 27.363913250696022,
      "learning_rate": 1.9941136159539462e-07,
      "logits/chosen": -2.3719429969787598,
      "logits/rejected": -2.3617918491363525,
      "logps/chosen": -68.02982330322266,
      "logps/rejected": -105.62237548828125,
      "loss": 0.394,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": 3.107600212097168,
      "rewards/margins": 12.890829086303711,
      "rewards/rejected": -9.783227920532227,
      "step": 170
    },
    {
      "epoch": 1.875,
      "grad_norm": 43.422774421532836,
      "learning_rate": 1.5422437905203344e-07,
      "logits/chosen": -2.4057819843292236,
      "logits/rejected": -2.389813184738159,
      "logps/chosen": -64.71813201904297,
      "logps/rejected": -92.5549545288086,
      "loss": 0.3757,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": 3.084803581237793,
      "rewards/margins": 11.271722793579102,
      "rewards/rejected": -8.186918258666992,
      "step": 180
    },
    {
      "epoch": 1.9791666666666665,
      "grad_norm": 2.3019013228739755,
      "learning_rate": 1.0903739650867225e-07,
      "logits/chosen": -2.454073429107666,
      "logits/rejected": -2.4500691890716553,
      "logps/chosen": -76.17430877685547,
      "logps/rejected": -113.40323638916016,
      "loss": 0.3748,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 5.310562610626221,
      "rewards/margins": 15.156728744506836,
      "rewards/rejected": -9.846165657043457,
      "step": 190
    },
    {
      "epoch": 2.0,
      "step": 192,
      "total_flos": 0.0,
      "train_loss": 0.5438582648833593,
      "train_runtime": 2038.8345,
      "train_samples_per_second": 5.997,
      "train_steps_per_second": 0.094
    }
  ],
  "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
}