{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 313,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003194888178913738,
      "grad_norm": 159.3779178763582,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -3.640625,
      "logits/rejected": -3.703125,
      "logps/chosen": -173.0,
      "logps/rejected": -142.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03194888178913738,
      "grad_norm": 142.0171514848489,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -3.59375,
      "logits/rejected": -3.59375,
      "logps/chosen": -146.0,
      "logps/rejected": -152.0,
      "loss": 0.6656,
      "rewards/accuracies": 0.4722222089767456,
      "rewards/chosen": 0.005889892578125,
      "rewards/margins": 0.060302734375,
      "rewards/rejected": -0.05419921875,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 41.93159336713555,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.53125,
      "logps/chosen": -151.0,
      "logps/rejected": -164.0,
      "loss": 0.3441,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1484375,
      "rewards/margins": 1.09375,
      "rewards/rejected": -0.94921875,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 9.157588889007195,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.46875,
      "logps/chosen": -157.0,
      "logps/rejected": -208.0,
      "loss": 0.0363,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.71484375,
      "rewards/margins": 4.90625,
      "rewards/rejected": -4.1875,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 0.17402017672333522,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.453125,
      "logps/chosen": -138.0,
      "logps/rejected": -237.0,
      "loss": 0.0017,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7265625,
      "rewards/margins": 9.625,
      "rewards/rejected": -8.875,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 0.5204944038679334,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.390625,
      "logps/chosen": -138.0,
      "logps/rejected": -280.0,
      "loss": 0.0004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.796875,
      "rewards/margins": 12.875,
      "rewards/rejected": -12.125,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 1.0403500251971276,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.359375,
      "logps/chosen": -145.0,
      "logps/rejected": -292.0,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.458984375,
      "rewards/margins": 13.5625,
      "rewards/rejected": -13.125,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 0.03485477392790473,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.40625,
      "logps/chosen": -144.0,
      "logps/rejected": -306.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.10791015625,
      "rewards/margins": 15.9375,
      "rewards/rejected": -15.875,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 0.5699036915216166,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.390625,
      "logps/chosen": -159.0,
      "logps/rejected": -336.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.349609375,
      "rewards/margins": 18.5,
      "rewards/rejected": -18.125,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 0.000586271534556838,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.359375,
      "logps/chosen": -141.0,
      "logps/rejected": -342.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.53125,
      "rewards/margins": 19.375,
      "rewards/rejected": -18.75,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.6226854251495038,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.375,
      "logps/chosen": -143.0,
      "logps/rejected": -338.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.451171875,
      "rewards/margins": 19.125,
      "rewards/rejected": -18.625,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.010449160008542551,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.328125,
      "logps/chosen": -152.0,
      "logps/rejected": -360.0,
      "loss": 0.0004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.10205078125,
      "rewards/margins": 20.5,
      "rewards/rejected": -20.625,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.0009548362087768612,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.375,
      "logps/chosen": -157.0,
      "logps/rejected": -362.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.119140625,
      "rewards/margins": 19.75,
      "rewards/rejected": -19.75,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.00013646586994902306,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.328125,
      "logps/chosen": -160.0,
      "logps/rejected": -380.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1240234375,
      "rewards/margins": 21.625,
      "rewards/rejected": -21.5,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.011640529634937136,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.328125,
      "logps/chosen": -133.0,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.466796875,
      "rewards/margins": 25.875,
      "rewards/rejected": -25.375,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 3.475718964256916e-05,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.265625,
      "logps/chosen": -148.0,
      "logps/rejected": -388.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.42578125,
      "rewards/margins": 23.625,
      "rewards/rejected": -23.25,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 1.9692663945470987,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.296875,
      "logps/chosen": -149.0,
      "logps/rejected": -416.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.396484375,
      "rewards/margins": 26.375,
      "rewards/rejected": -26.0,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.08170048165908884,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.34375,
      "logps/chosen": -166.0,
      "logps/rejected": -422.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.140625,
      "rewards/margins": 25.0,
      "rewards/rejected": -27.125,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.0035499679004452286,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -3.515625,
      "logits/rejected": -3.375,
      "logps/chosen": -175.0,
      "logps/rejected": -422.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.21875,
      "rewards/margins": 23.75,
      "rewards/rejected": -26.0,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.04104575399817043,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.34375,
      "logps/chosen": -176.0,
      "logps/rejected": -440.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.3125,
      "rewards/margins": 26.25,
      "rewards/rejected": -28.5,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 3.7685629191447305e-05,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.328125,
      "logps/chosen": -171.0,
      "logps/rejected": -448.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.4453125,
      "rewards/margins": 26.375,
      "rewards/rejected": -27.875,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 9.004933203527336e-06,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.359375,
      "logps/chosen": -161.0,
      "logps/rejected": -428.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.388671875,
      "rewards/margins": 27.5,
      "rewards/rejected": -28.0,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.0029222087631083824,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.328125,
      "logps/chosen": -147.0,
      "logps/rejected": -460.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.326171875,
      "rewards/margins": 30.5,
      "rewards/rejected": -30.875,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 0.00044147141232812285,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.34375,
      "logps/chosen": -146.0,
      "logps/rejected": -456.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.51953125,
      "rewards/margins": 29.5,
      "rewards/rejected": -30.125,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.010403918315756865,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -3.515625,
      "logits/rejected": -3.3125,
      "logps/chosen": -152.0,
      "logps/rejected": -436.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.2294921875,
      "rewards/margins": 28.0,
      "rewards/rejected": -28.25,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.015941514809629027,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.390625,
      "logps/chosen": -164.0,
      "logps/rejected": -452.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.5546875,
      "rewards/margins": 28.875,
      "rewards/rejected": -29.5,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 5.9135229813990824e-05,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -3.515625,
      "logits/rejected": -3.34375,
      "logps/chosen": -138.0,
      "logps/rejected": -446.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.00848388671875,
      "rewards/margins": 28.75,
      "rewards/rejected": -28.75,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 3.054828362608032e-05,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.3125,
      "logps/chosen": -164.0,
      "logps/rejected": -452.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.365234375,
      "rewards/margins": 29.0,
      "rewards/rejected": -29.375,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 0.00011745687765947919,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.359375,
      "logps/chosen": -186.0,
      "logps/rejected": -464.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.515625,
      "rewards/margins": 30.5,
      "rewards/rejected": -31.0,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 1.417074129799485e-06,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.34375,
      "logps/chosen": -149.0,
      "logps/rejected": -478.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.05810546875,
      "rewards/margins": 31.75,
      "rewards/rejected": -31.75,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 9.730034757690613e-07,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.359375,
      "logps/chosen": -174.0,
      "logps/rejected": -466.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.58203125,
      "rewards/margins": 30.5,
      "rewards/rejected": -31.0,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.0022653315379438928,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.375,
      "logps/chosen": -182.0,
      "logps/rejected": -462.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.2177734375,
      "rewards/margins": 30.0,
      "rewards/rejected": -30.125,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -3.53125,
      "eval_logits/rejected": -3.296875,
      "eval_logps/chosen": -150.0,
      "eval_logps/rejected": -476.0,
      "eval_loss": 3.5477933124639094e-07,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -0.275390625,
      "eval_rewards/margins": 33.0,
      "eval_rewards/rejected": -33.25,
      "eval_runtime": 15.4158,
      "eval_samples_per_second": 12.974,
      "eval_steps_per_second": 0.454,
      "step": 313
    }
  ],
  "logging_steps": 10,
  "max_steps": 939,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}