{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9921671018276762,
  "eval_steps": 100,
  "global_step": 95,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "logits/chosen": 0.8826487064361572,
      "logits/rejected": 0.921362042427063,
      "logps/chosen": -36.58121871948242,
      "logps/rejected": -54.902320861816406,
      "loss": 0.0001,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-06,
      "logits/chosen": 0.8916183114051819,
      "logits/rejected": 0.873969316482544,
      "logps/chosen": -87.77237701416016,
      "logps/rejected": -96.3612289428711,
      "loss": 0.0001,
      "rewards/accuracies": 0.2222222238779068,
      "rewards/chosen": 0.0002965294988825917,
      "rewards/margins": 0.00019970430003013462,
      "rewards/rejected": 9.682523523224518e-05,
      "step": 10
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.83118057351089e-06,
      "logits/chosen": 0.7805377244949341,
      "logits/rejected": 0.8473492860794067,
      "logps/chosen": -91.75300598144531,
      "logps/rejected": -85.11421966552734,
      "loss": 0.0001,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 4.452450230019167e-05,
      "rewards/margins": -0.00017617340199649334,
      "rewards/rejected": 0.00022069788246881217,
      "step": 20
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3475222930516484e-06,
      "logits/chosen": 0.8648212552070618,
      "logits/rejected": 0.8478900194168091,
      "logps/chosen": -85.289306640625,
      "logps/rejected": -78.0145492553711,
      "loss": 0.0001,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": 0.00029949203599244356,
      "rewards/margins": 6.509931699838489e-05,
      "rewards/rejected": 0.00023439270444214344,
      "step": 30
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.6143458894413463e-06,
      "logits/chosen": 0.7649093270301819,
      "logits/rejected": 0.8426358103752136,
      "logps/chosen": -122.76078033447266,
      "logps/rejected": -108.79180908203125,
      "loss": 0.0001,
      "rewards/accuracies": 0.26249998807907104,
      "rewards/chosen": -0.0002823365502990782,
      "rewards/margins": -9.583422070136294e-05,
      "rewards/rejected": -0.00018650232232175767,
      "step": 40
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.730670898658255e-06,
      "logits/chosen": 0.8404033780097961,
      "logits/rejected": 0.8685251474380493,
      "logps/chosen": -65.78496551513672,
      "logps/rejected": -71.65745544433594,
      "loss": 0.0001,
      "rewards/accuracies": 0.23749999701976776,
      "rewards/chosen": -0.0002987021580338478,
      "rewards/margins": -0.0003552257257979363,
      "rewards/rejected": 5.65235604881309e-05,
      "step": 50
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8158425248197931e-06,
      "logits/chosen": 0.8147390484809875,
      "logits/rejected": 0.8857711553573608,
      "logps/chosen": -117.61279296875,
      "logps/rejected": -110.9428482055664,
      "loss": 0.0001,
      "rewards/accuracies": 0.30000001192092896,
      "rewards/chosen": 0.00020218415011186153,
      "rewards/margins": 0.0003176866448484361,
      "rewards/rejected": -0.00011550241470104083,
      "step": 60
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.934134090518593e-07,
      "logits/chosen": 0.8052595257759094,
      "logits/rejected": 0.8626030087471008,
      "logps/chosen": -82.66513061523438,
      "logps/rejected": -83.91613006591797,
      "loss": 0.0001,
      "rewards/accuracies": 0.2750000059604645,
      "rewards/chosen": -1.9253784557804465e-05,
      "rewards/margins": 6.199729250511155e-05,
      "rewards/rejected": -8.125107706291601e-05,
      "step": 70
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7445716067596506e-07,
      "logits/chosen": 0.7661033272743225,
      "logits/rejected": 0.829667866230011,
      "logps/chosen": -105.8118667602539,
      "logps/rejected": -104.9473876953125,
      "loss": 0.0001,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": 2.1945637854514644e-05,
      "rewards/margins": -0.00017051820759661496,
      "rewards/rejected": 0.00019246386364102364,
      "step": 80
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.256725079024554e-08,
      "logits/chosen": 0.8347122073173523,
      "logits/rejected": 0.8553287386894226,
      "logps/chosen": -77.658203125,
      "logps/rejected": -89.99678039550781,
      "loss": 0.0001,
      "rewards/accuracies": 0.34375,
      "rewards/chosen": 0.0003329566097818315,
      "rewards/margins": 0.0006740919779986143,
      "rewards/rejected": -0.0003411353682167828,
      "step": 90
    },
    {
      "epoch": 0.99,
      "step": 95,
      "total_flos": 0.0,
      "train_loss": 0.0001254068662283795,
      "train_runtime": 1151.127,
      "train_samples_per_second": 5.31,
      "train_steps_per_second": 0.083
    }
  ],
  "logging_steps": 10,
  "max_steps": 95,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}