{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.99581589958159,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -2.305778980255127,
      "logits/rejected": -2.224325656890869,
      "logps/chosen": -270.25244140625,
      "logps/rejected": -383.8268127441406,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.394684314727783,
      "logits/rejected": -2.4029245376586914,
      "logps/chosen": -217.4110565185547,
      "logps/rejected": -323.976318359375,
      "loss": 0.6897,
      "rewards/accuracies": 0.5416666865348816,
      "rewards/chosen": -0.010575544089078903,
      "rewards/margins": 0.014531731605529785,
      "rewards/rejected": -0.02510727569460869,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.293762683868408,
      "logits/rejected": -2.2932822704315186,
      "logps/chosen": -260.174560546875,
      "logps/rejected": -345.7673034667969,
      "loss": 0.6705,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.30990877747535706,
      "rewards/margins": 0.16011206805706024,
      "rewards/rejected": -0.47002077102661133,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.2889437675476074,
      "logits/rejected": -2.2858946323394775,
      "logps/chosen": -283.4850158691406,
      "logps/rejected": -348.92822265625,
      "loss": 0.6862,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3383755683898926,
      "rewards/margins": 0.19907911121845245,
      "rewards/rejected": -0.5374546647071838,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.295836925506592,
      "logits/rejected": -2.288512706756592,
      "logps/chosen": -269.2564392089844,
      "logps/rejected": -314.68768310546875,
      "loss": 0.6686,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.06676869094371796,
      "rewards/margins": 0.07462646812200546,
      "rewards/rejected": -0.14139513671398163,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.2943179607391357,
      "logits/rejected": -2.2712674140930176,
      "logps/chosen": -272.5740051269531,
      "logps/rejected": -336.09210205078125,
      "loss": 0.6653,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.16639836132526398,
      "rewards/margins": 0.1392066478729248,
      "rewards/rejected": -0.30560502409935,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.3181185722351074,
      "logits/rejected": -2.3116087913513184,
      "logps/chosen": -279.1436462402344,
      "logps/rejected": -351.61798095703125,
      "loss": 0.6473,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.30815738439559937,
      "rewards/margins": 0.14108426868915558,
      "rewards/rejected": -0.44924163818359375,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.3251430988311768,
      "logits/rejected": -2.2963356971740723,
      "logps/chosen": -280.5970458984375,
      "logps/rejected": -364.6551208496094,
      "loss": 0.651,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.3270416855812073,
      "rewards/margins": 0.1887568086385727,
      "rewards/rejected": -0.5157985091209412,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.308258056640625,
      "logits/rejected": -2.2962486743927,
      "logps/chosen": -282.85528564453125,
      "logps/rejected": -327.85491943359375,
      "loss": 0.6311,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.2814248204231262,
      "rewards/margins": 0.1915736347436905,
      "rewards/rejected": -0.4729984700679779,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.2949376106262207,
      "logits/rejected": -2.3052544593811035,
      "logps/chosen": -250.7218017578125,
      "logps/rejected": -336.38140869140625,
      "loss": 0.6149,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.2139570415019989,
      "rewards/margins": 0.24394384026527405,
      "rewards/rejected": -0.45790091156959534,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.319019317626953,
      "logits/rejected": -2.298459529876709,
      "logps/chosen": -291.30572509765625,
      "logps/rejected": -360.62750244140625,
      "loss": 0.6427,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.31846240162849426,
      "rewards/margins": 0.2448347508907318,
      "rewards/rejected": -0.5632971525192261,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -2.3022727966308594,
      "logits/rejected": -2.3168272972106934,
      "logps/chosen": -281.0862731933594,
      "logps/rejected": -347.0366516113281,
      "loss": 0.6439,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.29117727279663086,
      "rewards/margins": 0.17553743720054626,
      "rewards/rejected": -0.4667147099971771,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.6554931792892328,
      "train_runtime": 1987.3183,
      "train_samples_per_second": 7.69,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 119,
  "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
}