{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9905956112852664,
  "eval_steps": 500,
  "global_step": 79,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 6.25e-08,
      "logits/chosen": -2.9087584018707275,
      "logits/rejected": -2.8338208198547363,
      "logps/chosen": -352.96319580078125,
      "logps/pi_response": -77.43819427490234,
      "logps/ref_response": -77.43819427490234,
      "logps/rejected": -170.26690673828125,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.990217055187362e-07,
      "logits/chosen": -2.7869324684143066,
      "logits/rejected": -2.7639520168304443,
      "logps/chosen": -237.83615112304688,
      "logps/pi_response": -73.73171997070312,
      "logps/ref_response": -73.52912139892578,
      "logps/rejected": -170.54678344726562,
      "loss": 0.6905,
      "rewards/accuracies": 0.5486111044883728,
      "rewards/chosen": 0.005888951942324638,
      "rewards/margins": 0.004442126490175724,
      "rewards/rejected": 0.0014468259178102016,
      "step": 10
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.655786431300069e-07,
      "logits/chosen": -2.7454209327697754,
      "logits/rejected": -2.6917433738708496,
      "logps/chosen": -248.8080596923828,
      "logps/pi_response": -79.94108581542969,
      "logps/ref_response": -71.4201431274414,
      "logps/rejected": -174.76712036132812,
      "loss": 0.6604,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.04190480336546898,
      "rewards/margins": 0.08115339279174805,
      "rewards/rejected": -0.03924858942627907,
      "step": 20
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.9061232191019517e-07,
      "logits/chosen": -2.658690929412842,
      "logits/rejected": -2.623389959335327,
      "logps/chosen": -234.90585327148438,
      "logps/pi_response": -105.19673156738281,
      "logps/ref_response": -68.3199691772461,
      "logps/rejected": -190.49130249023438,
      "loss": 0.6266,
      "rewards/accuracies": 0.7406250238418579,
      "rewards/chosen": -0.0959092304110527,
      "rewards/margins": 0.18841782212257385,
      "rewards/rejected": -0.28432708978652954,
      "step": 30
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.8856223324132555e-07,
      "logits/chosen": -2.665231943130493,
      "logits/rejected": -2.6431384086608887,
      "logps/chosen": -259.9793395996094,
      "logps/pi_response": -145.41070556640625,
      "logps/ref_response": -75.2870864868164,
      "logps/rejected": -230.671142578125,
      "loss": 0.5798,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.22865232825279236,
      "rewards/margins": 0.35143420100212097,
      "rewards/rejected": -0.5800865292549133,
      "step": 40
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.7908455541642582e-07,
      "logits/chosen": -2.667163133621216,
      "logits/rejected": -2.633802890777588,
      "logps/chosen": -287.7627258300781,
      "logps/pi_response": -154.95919799804688,
      "logps/ref_response": -74.21197509765625,
      "logps/rejected": -249.1497039794922,
      "loss": 0.5501,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.32577013969421387,
      "rewards/margins": 0.43573999404907227,
      "rewards/rejected": -0.7615101933479309,
      "step": 50
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.32661172908373e-08,
      "logits/chosen": -2.664795398712158,
      "logits/rejected": -2.6296298503875732,
      "logps/chosen": -257.57293701171875,
      "logps/pi_response": -161.33213806152344,
      "logps/ref_response": -66.34608459472656,
      "logps/rejected": -266.1256408691406,
      "loss": 0.5298,
      "rewards/accuracies": 0.796875,
      "rewards/chosen": -0.4251323342323303,
      "rewards/margins": 0.5614072680473328,
      "rewards/rejected": -0.9865396618843079,
      "step": 60
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.956279997278043e-08,
      "logits/chosen": -2.6611087322235107,
      "logits/rejected": -2.6245808601379395,
      "logps/chosen": -307.0917053222656,
      "logps/pi_response": -187.67933654785156,
      "logps/ref_response": -77.6395034790039,
      "logps/rejected": -278.4061279296875,
      "loss": 0.5189,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.49217742681503296,
      "rewards/margins": 0.6050472259521484,
      "rewards/rejected": -1.097224473953247,
      "step": 70
    },
    {
      "epoch": 0.99,
      "step": 79,
      "total_flos": 0.0,
      "train_loss": 0.5867953602271744,
      "train_runtime": 4674.1967,
      "train_samples_per_second": 4.36,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 79,
  "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
}