{
  "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": -2.696837902069092,
      "logits/rejected": -2.596374273300171,
      "logps/chosen": -215.154052734375,
      "logps/pi_response": -176.18202209472656,
      "logps/ref_response": -176.18202209472656,
      "logps/rejected": -289.2363586425781,
      "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": -2.621392250061035,
      "logits/rejected": -2.592118978500366,
      "logps/chosen": -235.2295684814453,
      "logps/pi_response": -177.52685546875,
      "logps/ref_response": -178.44647216796875,
      "logps/rejected": -303.8392333984375,
      "loss": 0.6804,
      "rewards/accuracies": 0.6076388955116272,
      "rewards/chosen": -0.016610508784651756,
      "rewards/margins": 0.039140839129686356,
      "rewards/rejected": -0.05575134977698326,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.187457503795526e-07,
      "logits/chosen": -2.651366710662842,
      "logits/rejected": -2.6005194187164307,
      "logps/chosen": -272.60650634765625,
      "logps/pi_response": -166.6388397216797,
      "logps/ref_response": -174.33192443847656,
      "logps/rejected": -367.46075439453125,
      "loss": 0.6014,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.24459533393383026,
      "rewards/margins": 0.45078325271606445,
      "rewards/rejected": -0.6953786015510559,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.8691164100062034e-07,
      "logits/chosen": -2.671945333480835,
      "logits/rejected": -2.617596387863159,
      "logps/chosen": -301.997802734375,
      "logps/pi_response": -184.88941955566406,
      "logps/ref_response": -172.4146270751953,
      "logps/rejected": -417.10992431640625,
      "loss": 0.5494,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.5642356872558594,
      "rewards/margins": 0.7100368738174438,
      "rewards/rejected": -1.2742725610733032,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4248369943086995e-07,
      "logits/chosen": -2.594676971435547,
      "logits/rejected": -2.551079750061035,
      "logps/chosen": -301.282470703125,
      "logps/pi_response": -201.1599884033203,
      "logps/ref_response": -164.72154235839844,
      "logps/rejected": -445.9717712402344,
      "loss": 0.5072,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.7301871180534363,
      "rewards/margins": 0.8498421907424927,
      "rewards/rejected": -1.5800292491912842,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.473909705816111e-08,
      "logits/chosen": -2.5517385005950928,
      "logits/rejected": -2.5155932903289795,
      "logps/chosen": -337.0827331542969,
      "logps/pi_response": -220.1709442138672,
      "logps/ref_response": -174.74102783203125,
      "logps/rejected": -440.63677978515625,
      "loss": 0.4993,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.9003208875656128,
      "rewards/margins": 0.711205780506134,
      "rewards/rejected": -1.6115267276763916,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.5582792960991294,
      "train_runtime": 3533.8117,
      "train_samples_per_second": 4.325,
      "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
}