{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9047619047619047,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 115.44032311397486,
      "learning_rate": 9.931806517013612e-06,
      "logits/chosen": -2.1214675903320312,
      "logits/rejected": -2.09596586227417,
      "logps/chosen": -107.49108123779297,
      "logps/rejected": -15.508064270019531,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 2
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 313.3498585768591,
      "learning_rate": 9.397368756032445e-06,
      "logits/chosen": -2.0437583923339844,
      "logits/rejected": -2.048534393310547,
      "logps/chosen": -121.86276245117188,
      "logps/rejected": -16.88058090209961,
      "loss": 0.9007,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.31137895584106445,
      "rewards/margins": -0.054485566914081573,
      "rewards/rejected": -0.25689342617988586,
      "step": 4
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 455.64541668703777,
      "learning_rate": 8.386407858128707e-06,
      "logits/chosen": -1.9633965492248535,
      "logits/rejected": -2.0469534397125244,
      "logps/chosen": -152.4108428955078,
      "logps/rejected": -19.865943908691406,
      "loss": 2.869,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -3.2351627349853516,
      "rewards/margins": -2.399106502532959,
      "rewards/rejected": -0.8360563516616821,
      "step": 6
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 249.13606400963317,
      "learning_rate": 7.008477123264849e-06,
      "logits/chosen": -2.021455764770508,
      "logits/rejected": -2.0236542224884033,
      "logps/chosen": -156.01495361328125,
      "logps/rejected": -31.431161880493164,
      "loss": 3.0671,
      "rewards/accuracies": 0.21875,
      "rewards/chosen": -4.389909267425537,
      "rewards/margins": -2.582636594772339,
      "rewards/rejected": -1.8072729110717773,
      "step": 8
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 189.45328579107624,
      "learning_rate": 5.412896727361663e-06,
      "logits/chosen": -2.025608539581299,
      "logits/rejected": -2.0653796195983887,
      "logps/chosen": -149.14700317382812,
      "logps/rejected": -42.3158073425293,
      "loss": 1.3723,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -3.427455425262451,
      "rewards/margins": -0.4787706732749939,
      "rewards/rejected": -2.9486846923828125,
      "step": 10
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 244.97288427388463,
      "learning_rate": 3.7725725642960047e-06,
      "logits/chosen": -2.0566327571868896,
      "logits/rejected": -2.057169198989868,
      "logps/chosen": -126.93424987792969,
      "logps/rejected": -56.91143035888672,
      "loss": 1.9476,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -1.397817850112915,
      "rewards/margins": 3.1280388832092285,
      "rewards/rejected": -4.525856971740723,
      "step": 12
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 183.45909354907522,
      "learning_rate": 2.265259209387867e-06,
      "logits/chosen": -2.049287796020508,
      "logits/rejected": -2.0284643173217773,
      "logps/chosen": -130.31076049804688,
      "logps/rejected": -76.90144348144531,
      "loss": 0.5622,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 1.7893621921539307,
      "rewards/margins": 8.1104154586792,
      "rewards/rejected": -6.3210530281066895,
      "step": 14
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 123.873213446449,
      "learning_rate": 1.0542974530180327e-06,
      "logits/chosen": -2.060037136077881,
      "logits/rejected": -2.0083541870117188,
      "logps/chosen": -103.64411926269531,
      "logps/rejected": -78.95755004882812,
      "loss": 0.2122,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 2.1228506565093994,
      "rewards/margins": 8.57053279876709,
      "rewards/rejected": -6.447681903839111,
      "step": 16
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 23.274773765505252,
      "learning_rate": 2.7091379149682683e-07,
      "logits/chosen": -2.054797649383545,
      "logits/rejected": -1.9887886047363281,
      "logps/chosen": -112.17431640625,
      "logps/rejected": -83.83834838867188,
      "loss": 0.3109,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 0.16159486770629883,
      "rewards/margins": 7.233631134033203,
      "rewards/rejected": -7.0720367431640625,
      "step": 18
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 155.65960541381844,
      "learning_rate": 0.0,
      "logits/chosen": -2.08943510055542,
      "logits/rejected": -2.03725528717041,
      "logps/chosen": -96.60086822509766,
      "logps/rejected": -91.33654022216797,
      "loss": 0.8478,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 0.384034126996994,
      "rewards/margins": 8.193408966064453,
      "rewards/rejected": -7.809375286102295,
      "step": 20
    },
    {
      "epoch": 1.9047619047619047,
      "step": 20,
      "total_flos": 474810286080.0,
      "train_loss": 1.2783152908086777,
      "train_runtime": 432.4363,
      "train_samples_per_second": 0.777,
      "train_steps_per_second": 0.046
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 474810286080.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}