{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9874476987447699,
  "eval_steps": 500,
  "global_step": 59,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1.6666666666666664e-08,
      "logits/chosen": -2.5598204135894775,
      "logits/rejected": -2.54148006439209,
      "logps/chosen": -231.18460083007812,
      "logps/pi_response": -153.6551971435547,
      "logps/ref_response": -153.6551971435547,
      "logps/rejected": -481.5467529296875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.860114570402053e-08,
      "logits/chosen": -2.5779542922973633,
      "logits/rejected": -2.4980809688568115,
      "logps/chosen": -278.1629333496094,
      "logps/pi_response": -136.0104217529297,
      "logps/ref_response": -135.50714111328125,
      "logps/rejected": -486.1741027832031,
      "loss": 0.6872,
      "rewards/accuracies": 0.5833333134651184,
      "rewards/chosen": -0.016039522364735603,
      "rewards/margins": 0.00932595506310463,
      "rewards/rejected": -0.025365475565195084,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.374915007591053e-08,
      "logits/chosen": -2.5219459533691406,
      "logits/rejected": -2.475327968597412,
      "logps/chosen": -305.75347900390625,
      "logps/pi_response": -152.5146026611328,
      "logps/ref_response": -147.16122436523438,
      "logps/rejected": -493.6817321777344,
      "loss": 0.6359,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.19058512151241302,
      "rewards/margins": 0.14784559607505798,
      "rewards/rejected": -0.3384307026863098,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.738232820012406e-08,
      "logits/chosen": -2.4824159145355225,
      "logits/rejected": -2.4323043823242188,
      "logps/chosen": -296.9906921386719,
      "logps/pi_response": -145.09451293945312,
      "logps/ref_response": -141.3872528076172,
      "logps/rejected": -551.1701049804688,
      "loss": 0.586,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.26490381360054016,
      "rewards/margins": 0.3826388120651245,
      "rewards/rejected": -0.6475426554679871,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.8496739886173992e-08,
      "logits/chosen": -2.471738815307617,
      "logits/rejected": -2.4239768981933594,
      "logps/chosen": -319.9653015136719,
      "logps/pi_response": -151.64266967773438,
      "logps/ref_response": -145.77267456054688,
      "logps/rejected": -595.8931884765625,
      "loss": 0.5548,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.4044647216796875,
      "rewards/margins": 0.6303603649139404,
      "rewards/rejected": -1.034825086593628,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 6.947819411632222e-09,
      "logits/chosen": -2.472991466522217,
      "logits/rejected": -2.4102094173431396,
      "logps/chosen": -351.69976806640625,
      "logps/pi_response": -155.18856811523438,
      "logps/ref_response": -148.97946166992188,
      "logps/rejected": -625.48193359375,
      "loss": 0.5394,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.5062131285667419,
      "rewards/margins": 0.6884612441062927,
      "rewards/rejected": -1.1946743726730347,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.5946870900816836,
      "train_runtime": 3503.4654,
      "train_samples_per_second": 4.362,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 59,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}