{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.26586905948820205,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.026586905948820207,
      "grad_norm": 235.8455810546875,
      "learning_rate": 9e-06,
      "logits/chosen": 1.961168646812439,
      "logits/rejected": 2.3471131324768066,
      "logps/chosen": -180.52096557617188,
      "logps/rejected": -306.41729736328125,
      "loss": 41.09808349609375,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 127.35591888427734,
      "rewards/margins": 56.6036376953125,
      "rewards/rejected": 70.75228118896484,
      "step": 10
    },
    {
      "epoch": 0.053173811897640415,
      "grad_norm": 132.22105407714844,
      "learning_rate": 9e-06,
      "logits/chosen": 2.862368106842041,
      "logits/rejected": 3.0271706581115723,
      "logps/chosen": -199.9718017578125,
      "logps/rejected": -383.2444763183594,
      "loss": 30.093466186523436,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 132.59613037109375,
      "rewards/margins": 104.64494323730469,
      "rewards/rejected": 27.951187133789062,
      "step": 20
    },
    {
      "epoch": 0.07976071784646062,
      "grad_norm": 234.4345703125,
      "learning_rate": 7.88888888888889e-06,
      "logits/chosen": 3.7322630882263184,
      "logits/rejected": 3.8905537128448486,
      "logps/chosen": -219.1993865966797,
      "logps/rejected": -487.88043212890625,
      "loss": 23.500108337402345,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 141.04718017578125,
      "rewards/margins": 154.50144958496094,
      "rewards/rejected": -13.454266548156738,
      "step": 30
    },
    {
      "epoch": 0.10634762379528083,
      "grad_norm": 110.76012420654297,
      "learning_rate": 6.777777777777779e-06,
      "logits/chosen": 3.839090347290039,
      "logits/rejected": 4.133347988128662,
      "logps/chosen": -210.9350128173828,
      "logps/rejected": -591.01513671875,
      "loss": 23.24909210205078,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": 109.74186706542969,
      "rewards/margins": 179.082275390625,
      "rewards/rejected": -69.34040832519531,
      "step": 40
    },
    {
      "epoch": 0.13293452974410103,
      "grad_norm": 221.89686584472656,
      "learning_rate": 5.666666666666667e-06,
      "logits/chosen": 5.0413312911987305,
      "logits/rejected": 5.170095920562744,
      "logps/chosen": -311.14190673828125,
      "logps/rejected": -735.1285400390625,
      "loss": 20.513507080078124,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": 93.5618667602539,
      "rewards/margins": 231.3604736328125,
      "rewards/rejected": -137.79861450195312,
      "step": 50
    },
    {
      "epoch": 0.15952143569292124,
      "grad_norm": 190.1571044921875,
      "learning_rate": 4.555555555555556e-06,
      "logits/chosen": 5.5310869216918945,
      "logits/rejected": 5.918412208557129,
      "logps/chosen": -291.5699157714844,
      "logps/rejected": -943.5242919921875,
      "loss": 4.122843551635742,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 46.140045166015625,
      "rewards/margins": 318.23345947265625,
      "rewards/rejected": -272.0934143066406,
      "step": 60
    },
    {
      "epoch": 0.18610834164174145,
      "grad_norm": 20.02957534790039,
      "learning_rate": 3.444444444444445e-06,
      "logits/chosen": 6.543715476989746,
      "logits/rejected": 6.924801826477051,
      "logps/chosen": -418.068359375,
      "logps/rejected": -989.2503662109375,
      "loss": 5.07203254699707,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 34.52830123901367,
      "rewards/margins": 294.2066345214844,
      "rewards/rejected": -259.6783447265625,
      "step": 70
    },
    {
      "epoch": 0.21269524759056166,
      "grad_norm": 365.81671142578125,
      "learning_rate": 2.3333333333333336e-06,
      "logits/chosen": 7.004462242126465,
      "logits/rejected": 7.306379795074463,
      "logps/chosen": -456.623291015625,
      "logps/rejected": -1127.0947265625,
      "loss": 4.6394092559814455,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 18.725841522216797,
      "rewards/margins": 349.7669372558594,
      "rewards/rejected": -331.0411071777344,
      "step": 80
    },
    {
      "epoch": 0.23928215353938184,
      "grad_norm": 442.5915222167969,
      "learning_rate": 1.2222222222222223e-06,
      "logits/chosen": 7.063316345214844,
      "logits/rejected": 7.30740213394165,
      "logps/chosen": -479.52117919921875,
      "logps/rejected": -1115.75146484375,
      "loss": 3.6795909881591795,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 14.30505084991455,
      "rewards/margins": 339.4307556152344,
      "rewards/rejected": -325.1257019042969,
      "step": 90
    },
    {
      "epoch": 0.26586905948820205,
      "grad_norm": 170.1773223876953,
      "learning_rate": 1.1111111111111112e-07,
      "logits/chosen": 6.971957206726074,
      "logits/rejected": 7.46600866317749,
      "logps/chosen": -429.0472106933594,
      "logps/rejected": -1123.8218994140625,
      "loss": 3.1557437896728517,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 5.929503440856934,
      "rewards/margins": 339.28887939453125,
      "rewards/rejected": -333.359375,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}