{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.8099547511312217,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36199095022624433,
      "grad_norm": 0.5034113526344299,
      "learning_rate": 9.259259259259259e-07,
      "logits/chosen": 1.8096959590911865,
      "logits/rejected": 1.7158682346343994,
      "logps/chosen": -106.03636169433594,
      "logps/rejected": -84.16211700439453,
      "loss": 0.695,
      "rewards/accuracies": 0.3499999940395355,
      "rewards/chosen": -0.00430720392614603,
      "rewards/margins": 0.00014101080887485296,
      "rewards/rejected": -0.004448213614523411,
      "step": 10
    },
    {
      "epoch": 0.7239819004524887,
      "grad_norm": 0.4689481854438782,
      "learning_rate": 1.8518518518518519e-06,
      "logits/chosen": 1.6904691457748413,
      "logits/rejected": 1.6791242361068726,
      "logps/chosen": -82.1568603515625,
      "logps/rejected": -81.06039428710938,
      "loss": 0.6942,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.002815738320350647,
      "rewards/margins": 0.005291810259222984,
      "rewards/rejected": -0.008107547648251057,
      "step": 20
    },
    {
      "epoch": 1.085972850678733,
      "grad_norm": 0.5041429996490479,
      "learning_rate": 2.7777777777777783e-06,
      "logits/chosen": 1.877396583557129,
      "logits/rejected": 1.9337213039398193,
      "logps/chosen": -76.86979675292969,
      "logps/rejected": -86.98431396484375,
      "loss": 0.6937,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.002632059855386615,
      "rewards/margins": -0.00260960147716105,
      "rewards/rejected": -2.245856376248412e-05,
      "step": 30
    },
    {
      "epoch": 1.4479638009049773,
      "grad_norm": 0.49786925315856934,
      "learning_rate": 3.7037037037037037e-06,
      "logits/chosen": 1.622865915298462,
      "logits/rejected": 1.649457573890686,
      "logps/chosen": -77.10165405273438,
      "logps/rejected": -75.56939697265625,
      "loss": 0.6927,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.005956621374934912,
      "rewards/margins": -0.0038306519854813814,
      "rewards/rejected": -0.0021259686909615993,
      "step": 40
    },
    {
      "epoch": 1.8099547511312217,
      "grad_norm": 0.6152529120445251,
      "learning_rate": 4.62962962962963e-06,
      "logits/chosen": 1.7763454914093018,
      "logits/rejected": 1.8556989431381226,
      "logps/chosen": -84.42935943603516,
      "logps/rejected": -97.08312225341797,
      "loss": 0.6937,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.008006314747035503,
      "rewards/margins": 0.005445700138807297,
      "rewards/rejected": 0.0025606155395507812,
      "step": 50
    }
  ],
  "logging_steps": 10,
  "max_steps": 540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.715295541568471e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}