{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 46,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10869565217391304,
      "grad_norm": 1.5240029096603394,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -3.1438066959381104,
      "logits/rejected": -2.998044490814209,
      "logps/chosen": -24.425586700439453,
      "logps/rejected": -109.31417083740234,
      "loss": 0.2266,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 1.709136962890625,
      "rewards/margins": 1.7067525386810303,
      "rewards/rejected": 0.0023843557573854923,
      "step": 5
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 0.8518787622451782,
      "learning_rate": 4.883490980137327e-06,
      "logits/chosen": -3.2125651836395264,
      "logits/rejected": -3.0023326873779297,
      "logps/chosen": -24.60028648376465,
      "logps/rejected": -109.94728088378906,
      "loss": 0.1743,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 1.6953380107879639,
      "rewards/margins": 2.0240752696990967,
      "rewards/rejected": -0.32873719930648804,
      "step": 10
    },
    {
      "epoch": 0.32608695652173914,
      "grad_norm": 0.5248653888702393,
      "learning_rate": 4.428722949554858e-06,
      "logits/chosen": -3.3195624351501465,
      "logits/rejected": -3.0724987983703613,
      "logps/chosen": -23.181148529052734,
      "logps/rejected": -118.40291595458984,
      "loss": 0.0617,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5336229801177979,
      "rewards/margins": 3.1812326908111572,
      "rewards/rejected": -1.6476097106933594,
      "step": 15
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 0.5495370626449585,
      "learning_rate": 3.6942995462806574e-06,
      "logits/chosen": -3.4612669944763184,
      "logits/rejected": -3.146296501159668,
      "logps/chosen": -25.121320724487305,
      "logps/rejected": -125.0635986328125,
      "loss": 0.0497,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.540027141571045,
      "rewards/margins": 3.769411563873291,
      "rewards/rejected": -2.229383945465088,
      "step": 20
    },
    {
      "epoch": 0.5434782608695652,
      "grad_norm": 0.3498240113258362,
      "learning_rate": 2.786708563496002e-06,
      "logits/chosen": -3.4969794750213623,
      "logits/rejected": -3.1035914421081543,
      "logps/chosen": -26.37749671936035,
      "logps/rejected": -134.34664916992188,
      "loss": 0.0391,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.542913794517517,
      "rewards/margins": 4.45734167098999,
      "rewards/rejected": -2.914428234100342,
      "step": 25
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.21621277928352356,
      "learning_rate": 1.8375462445083464e-06,
      "logits/chosen": -3.5592663288116455,
      "logits/rejected": -3.240027666091919,
      "logps/chosen": -27.510631561279297,
      "logps/rejected": -126.76861572265625,
      "loss": 0.0459,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4070671796798706,
      "rewards/margins": 4.541848659515381,
      "rewards/rejected": -3.1347813606262207,
      "step": 30
    },
    {
      "epoch": 0.7608695652173914,
      "grad_norm": 0.2057134360074997,
      "learning_rate": 9.844364725834058e-07,
      "logits/chosen": -3.556133270263672,
      "logits/rejected": -3.2353644371032715,
      "logps/chosen": -26.806324005126953,
      "logps/rejected": -128.81968688964844,
      "loss": 0.056,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 1.3792188167572021,
      "rewards/margins": 4.967148780822754,
      "rewards/rejected": -3.5879299640655518,
      "step": 35
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.21634213626384735,
      "learning_rate": 3.510759825319976e-07,
      "logits/chosen": -3.608375072479248,
      "logits/rejected": -3.2185797691345215,
      "logps/chosen": -26.37796974182129,
      "logps/rejected": -136.79074096679688,
      "loss": 0.036,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4289178848266602,
      "rewards/margins": 4.961454391479492,
      "rewards/rejected": -3.532536745071411,
      "step": 40
    },
    {
      "epoch": 0.9782608695652174,
      "grad_norm": 0.40187039971351624,
      "learning_rate": 2.9298940549128962e-08,
      "logits/chosen": -3.6041157245635986,
      "logits/rejected": -3.2674224376678467,
      "logps/chosen": -27.230480194091797,
      "logps/rejected": -134.53451538085938,
      "loss": 0.0598,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.397599697113037,
      "rewards/margins": 4.8428955078125,
      "rewards/rejected": -3.4452953338623047,
      "step": 45
    }
  ],
  "logging_steps": 5,
  "max_steps": 46,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}