{
  "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": 8.333333333333333e-08,
      "logits/chosen": -2.3157501220703125,
      "logits/rejected": -2.2659952640533447,
      "logps/chosen": -302.7703857421875,
      "logps/pi_response": -161.13186645507812,
      "logps/ref_response": -161.13186645507812,
      "logps/rejected": -577.6012573242188,
      "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": 4.930057285201027e-07,
      "logits/chosen": -2.270664691925049,
      "logits/rejected": -2.160485029220581,
      "logps/chosen": -325.5273132324219,
      "logps/pi_response": -161.80162048339844,
      "logps/ref_response": -144.94566345214844,
      "logps/rejected": -661.7462158203125,
      "loss": 0.6232,
      "rewards/accuracies": 0.6458333134651184,
      "rewards/chosen": -0.3242569863796234,
      "rewards/margins": 0.34182536602020264,
      "rewards/rejected": -0.6660823822021484,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.187457503795526e-07,
      "logits/chosen": -2.028958320617676,
      "logits/rejected": -1.9086393117904663,
      "logps/chosen": -435.76434326171875,
      "logps/pi_response": -224.1667022705078,
      "logps/ref_response": -153.24801635742188,
      "logps/rejected": -934.2482299804688,
      "loss": 0.6154,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -1.4426922798156738,
      "rewards/margins": 1.9100526571273804,
      "rewards/rejected": -3.3527450561523438,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.8691164100062034e-07,
      "logits/chosen": -2.0624918937683105,
      "logits/rejected": -1.951886773109436,
      "logps/chosen": -382.1322326660156,
      "logps/pi_response": -174.34063720703125,
      "logps/ref_response": -144.47926330566406,
      "logps/rejected": -816.6314697265625,
      "loss": 0.4817,
      "rewards/accuracies": 0.753125011920929,
      "rewards/chosen": -0.9019044637680054,
      "rewards/margins": 1.3625807762145996,
      "rewards/rejected": -2.2644851207733154,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4248369943086995e-07,
      "logits/chosen": -2.110487699508667,
      "logits/rejected": -1.97927987575531,
      "logps/chosen": -389.966064453125,
      "logps/pi_response": -180.74072265625,
      "logps/ref_response": -157.57518005371094,
      "logps/rejected": -769.4873046875,
      "loss": 0.4754,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.7284509539604187,
      "rewards/margins": 0.9590038061141968,
      "rewards/rejected": -1.6874548196792603,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.473909705816111e-08,
      "logits/chosen": -2.0629687309265137,
      "logits/rejected": -1.9410479068756104,
      "logps/chosen": -383.071044921875,
      "logps/pi_response": -181.2548065185547,
      "logps/ref_response": -156.56509399414062,
      "logps/rejected": -840.9393310546875,
      "loss": 0.4496,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -0.7260923385620117,
      "rewards/margins": 1.2465245723724365,
      "rewards/rejected": -1.9726167917251587,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.5167979992042153,
      "train_runtime": 3407.459,
      "train_samples_per_second": 4.485,
      "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
}