{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 1.25e-06,
      "logits/chosen": -2.400381326675415,
      "logits/rejected": -2.39224910736084,
      "logps/chosen": -298.7495422363281,
      "logps/rejected": -385.2267761230469,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.1682252883911133,
      "logits/rejected": -2.1295580863952637,
      "logps/chosen": -284.8713684082031,
      "logps/rejected": -366.9027099609375,
      "loss": 0.6526,
      "rewards/accuracies": 0.5798611044883728,
      "rewards/chosen": -0.2899032235145569,
      "rewards/margins": 0.10106495022773743,
      "rewards/rejected": -0.3909682035446167,
      "step": 10
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -1.947609305381775,
      "logits/rejected": -1.8020164966583252,
      "logps/chosen": -310.9280090332031,
      "logps/rejected": -387.20355224609375,
      "loss": 0.5644,
      "rewards/accuracies": 0.7906249761581421,
      "rewards/chosen": -0.3633348345756531,
      "rewards/margins": 0.3639495074748993,
      "rewards/rejected": -0.7272843718528748,
      "step": 20
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -1.7138926982879639,
      "logits/rejected": -1.5106457471847534,
      "logps/chosen": -347.11175537109375,
      "logps/rejected": -445.07470703125,
      "loss": 0.513,
      "rewards/accuracies": 0.765625,
      "rewards/chosen": -0.46522030234336853,
      "rewards/margins": 0.533661961555481,
      "rewards/rejected": -0.9988821148872375,
      "step": 30
    },
    {
      "epoch": 1.0,
      "step": 39,
      "total_flos": 0.0,
      "train_loss": 0.5628757874170939,
      "train_runtime": 1697.0362,
      "train_samples_per_second": 2.946,
      "train_steps_per_second": 0.023
    }
  ],
  "logging_steps": 10,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}