{
  "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": -2.4273462295532227,
      "logits/rejected": -2.3812289237976074,
      "logps/chosen": -259.1644592285156,
      "logps/pi_response": -133.5627899169922,
      "logps/ref_response": -133.5627899169922,
      "logps/rejected": -533.5703735351562,
      "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": -2.4278411865234375,
      "logits/rejected": -2.3395986557006836,
      "logps/chosen": -301.99102783203125,
      "logps/pi_response": -131.2730255126953,
      "logps/ref_response": -130.72630310058594,
      "logps/rejected": -550.7373046875,
      "loss": 0.6856,
      "rewards/accuracies": 0.5659722089767456,
      "rewards/chosen": -0.015166527591645718,
      "rewards/margins": 0.0135720856487751,
      "rewards/rejected": -0.028738614171743393,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.374915007591053e-08,
      "logits/chosen": -2.374455213546753,
      "logits/rejected": -2.3172311782836914,
      "logps/chosen": -314.00921630859375,
      "logps/pi_response": -147.02317810058594,
      "logps/ref_response": -140.91171264648438,
      "logps/rejected": -590.6365966796875,
      "loss": 0.6173,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.21441540122032166,
      "rewards/margins": 0.23125095665454865,
      "rewards/rejected": -0.4456663727760315,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.738232820012406e-08,
      "logits/chosen": -2.3242835998535156,
      "logits/rejected": -2.2661571502685547,
      "logps/chosen": -332.5072326660156,
      "logps/pi_response": -142.86166381835938,
      "logps/ref_response": -136.68536376953125,
      "logps/rejected": -620.6819458007812,
      "loss": 0.5712,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.3769785761833191,
      "rewards/margins": 0.45380154252052307,
      "rewards/rejected": -0.8307801485061646,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.8496739886173992e-08,
      "logits/chosen": -2.316802740097046,
      "logits/rejected": -2.2453653812408447,
      "logps/chosen": -333.75244140625,
      "logps/pi_response": -144.06423950195312,
      "logps/ref_response": -137.30410766601562,
      "logps/rejected": -665.38232421875,
      "loss": 0.536,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.479484885931015,
      "rewards/margins": 0.7068859338760376,
      "rewards/rejected": -1.1863707304000854,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 6.947819411632222e-09,
      "logits/chosen": -2.30759334564209,
      "logits/rejected": -2.2497036457061768,
      "logps/chosen": -345.99603271484375,
      "logps/pi_response": -148.74740600585938,
      "logps/ref_response": -140.4210968017578,
      "logps/rejected": -725.7435302734375,
      "loss": 0.5314,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.531948447227478,
      "rewards/margins": 0.8935807347297668,
      "rewards/rejected": -1.4255291223526,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.5845334731926353,
      "train_runtime": 3408.0859,
      "train_samples_per_second": 4.484,
      "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
}