{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.888888888888889,
  "eval_steps": 50,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 61.74356859047619,
      "learning_rate": 5e-07,
      "logits/chosen": -2.7662081718444824,
      "logits/rejected": -2.7523112297058105,
      "logps/chosen": -261.1184387207031,
      "logps/rejected": -234.32962036132812,
      "loss": 0.6896,
      "rewards/accuracies": 0.39375001192092896,
      "rewards/chosen": 0.02241230569779873,
      "rewards/margins": 0.010422506369650364,
      "rewards/rejected": 0.011989799328148365,
      "step": 5
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 63.88312506249413,
      "learning_rate": 1e-06,
      "logits/chosen": -2.707808494567871,
      "logits/rejected": -2.6938366889953613,
      "logps/chosen": -284.2864685058594,
      "logps/rejected": -212.7003631591797,
      "loss": 0.6396,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 0.7215784788131714,
      "rewards/margins": 0.3653421998023987,
      "rewards/rejected": 0.35623636841773987,
      "step": 10
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 28.759118326736992,
      "learning_rate": 9.284285880837946e-07,
      "logits/chosen": -2.544132947921753,
      "logits/rejected": -2.5163872241973877,
      "logps/chosen": -259.2568359375,
      "logps/rejected": -224.86819458007812,
      "loss": 0.5276,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 1.8950694799423218,
      "rewards/margins": 1.3206126689910889,
      "rewards/rejected": 0.5744568109512329,
      "step": 15
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 44.99419433783415,
      "learning_rate": 7.342042203498951e-07,
      "logits/chosen": -2.4921061992645264,
      "logits/rejected": -2.452922821044922,
      "logps/chosen": -254.24813842773438,
      "logps/rejected": -218.63784790039062,
      "loss": 0.3624,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 2.1607468128204346,
      "rewards/margins": 2.315674304962158,
      "rewards/rejected": -0.15492752194404602,
      "step": 20
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 37.984688059234884,
      "learning_rate": 4.7293054570729126e-07,
      "logits/chosen": -2.4480080604553223,
      "logits/rejected": -2.469356060028076,
      "logps/chosen": -235.30078125,
      "logps/rejected": -223.0630340576172,
      "loss": 0.3487,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": 2.4314210414886475,
      "rewards/margins": 2.905514717102051,
      "rewards/rejected": -0.47409337759017944,
      "step": 25
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 14.063153978656265,
      "learning_rate": 2.1940646731880885e-07,
      "logits/chosen": -2.446575403213501,
      "logits/rejected": -2.457470417022705,
      "logps/chosen": -252.03466796875,
      "logps/rejected": -218.13442993164062,
      "loss": 0.2276,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 3.03205943107605,
      "rewards/margins": 3.497868299484253,
      "rewards/rejected": -0.46580901741981506,
      "step": 30
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 14.44411483401936,
      "learning_rate": 4.621229016452155e-08,
      "logits/chosen": -2.4535937309265137,
      "logits/rejected": -2.4528815746307373,
      "logps/chosen": -264.5810546875,
      "logps/rejected": -230.9309844970703,
      "loss": 0.1866,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 3.1574349403381348,
      "rewards/margins": 3.6558170318603516,
      "rewards/rejected": -0.4983822703361511,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 459455468666880.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}