{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.7241379310344827,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 0.5030671954154968,
      "learning_rate": 8.620689655172415e-07,
      "logits/chosen": 1.8564815521240234,
      "logits/rejected": 1.8255866765975952,
      "logps/chosen": -95.54290008544922,
      "logps/rejected": -79.79582214355469,
      "loss": 0.693,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.0028573228046298027,
      "rewards/margins": 0.0005112116923555732,
      "rewards/rejected": -0.0033685355447232723,
      "step": 10
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 0.45081979036331177,
      "learning_rate": 1.724137931034483e-06,
      "logits/chosen": 1.781947374343872,
      "logits/rejected": 1.6911979913711548,
      "logps/chosen": -104.20463562011719,
      "logps/rejected": -78.91150665283203,
      "loss": 0.6929,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": 0.00018558502779342234,
      "rewards/margins": -0.002152198925614357,
      "rewards/rejected": 0.0023377849720418453,
      "step": 20
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 0.41456305980682373,
      "learning_rate": 2.5862068965517246e-06,
      "logits/chosen": 1.7937275171279907,
      "logits/rejected": 1.729128122329712,
      "logps/chosen": -90.21867370605469,
      "logps/rejected": -71.69265747070312,
      "loss": 0.6935,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.003012509550899267,
      "rewards/margins": 0.009945740923285484,
      "rewards/rejected": -0.006933231838047504,
      "step": 30
    },
    {
      "epoch": 1.3793103448275863,
      "grad_norm": 0.45128729939460754,
      "learning_rate": 3.448275862068966e-06,
      "logits/chosen": 1.864363670349121,
      "logits/rejected": 1.9002879858016968,
      "logps/chosen": -87.1126480102539,
      "logps/rejected": -77.18392181396484,
      "loss": 0.6923,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.0013037443859502673,
      "rewards/margins": 0.006342612206935883,
      "rewards/rejected": -0.005038867238909006,
      "step": 40
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 0.4960842728614807,
      "learning_rate": 4.310344827586207e-06,
      "logits/chosen": 1.7825076580047607,
      "logits/rejected": 1.829602837562561,
      "logps/chosen": -78.63069915771484,
      "logps/rejected": -91.36685180664062,
      "loss": 0.6954,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.009792634285986423,
      "rewards/margins": 0.014146638102829456,
      "rewards/rejected": -0.0043540047481656075,
      "step": 50
    }
  ],
  "logging_steps": 10,
  "max_steps": 580,
  "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": 1.683084854511534e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}