{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6017191977077365,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08595988538681948,
      "grad_norm": 93.6100539565976,
      "learning_rate": 1e-06,
      "logits/chosen": -1.03125,
      "logits/rejected": -0.84765625,
      "logps/chosen": -194.0,
      "logps/rejected": -228.0,
      "loss": 0.6921,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.17191977077363896,
      "grad_norm": 81.32351057726953,
      "learning_rate": 1e-06,
      "logits/chosen": -1.03125,
      "logits/rejected": -0.83203125,
      "logps/chosen": -217.0,
      "logps/rejected": -276.0,
      "loss": 0.5897,
      "rewards/accuracies": 0.8833333253860474,
      "rewards/chosen": -0.005401611328125,
      "rewards/margins": 0.22265625,
      "rewards/rejected": -0.2275390625,
      "step": 2
    },
    {
      "epoch": 0.25787965616045844,
      "grad_norm": 54.70178903970555,
      "learning_rate": 1e-06,
      "logits/chosen": -1.03125,
      "logits/rejected": -0.80859375,
      "logps/chosen": -194.0,
      "logps/rejected": -220.0,
      "loss": 0.3507,
      "rewards/accuracies": 0.9666666388511658,
      "rewards/chosen": 0.02587890625,
      "rewards/margins": 0.83203125,
      "rewards/rejected": -0.8046875,
      "step": 3
    },
    {
      "epoch": 0.3438395415472779,
      "grad_norm": 43.28884532989534,
      "learning_rate": 1e-06,
      "logits/chosen": -1.0390625,
      "logits/rejected": -0.8203125,
      "logps/chosen": -161.0,
      "logps/rejected": -222.0,
      "loss": 0.3005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.08056640625,
      "rewards/margins": 1.1953125,
      "rewards/rejected": -1.1171875,
      "step": 4
    },
    {
      "epoch": 0.4297994269340974,
      "grad_norm": 11.27681555628311,
      "learning_rate": 1e-06,
      "logits/chosen": -0.9453125,
      "logits/rejected": -0.8203125,
      "logps/chosen": -138.0,
      "logps/rejected": -242.0,
      "loss": 0.0703,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2158203125,
      "rewards/margins": 3.328125,
      "rewards/rejected": -3.109375,
      "step": 5
    },
    {
      "epoch": 0.5157593123209169,
      "grad_norm": 7.154503627923726,
      "learning_rate": 1e-06,
      "logits/chosen": -1.046875,
      "logits/rejected": -0.84765625,
      "logps/chosen": -185.0,
      "logps/rejected": -262.0,
      "loss": 0.0443,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.197265625,
      "rewards/margins": 3.625,
      "rewards/rejected": -3.421875,
      "step": 6
    },
    {
      "epoch": 0.6017191977077364,
      "grad_norm": 4.580182139528718,
      "learning_rate": 1e-06,
      "logits/chosen": -1.0625,
      "logits/rejected": -0.828125,
      "logps/chosen": -177.0,
      "logps/rejected": -249.0,
      "loss": 0.0283,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.21484375,
      "rewards/margins": 4.5,
      "rewards/rejected": -4.28125,
      "step": 7
    },
    {
      "epoch": 0.6876790830945558,
      "grad_norm": 3.6258438418031407,
      "learning_rate": 1e-06,
      "logits/chosen": -0.96484375,
      "logits/rejected": -0.890625,
      "logps/chosen": -180.0,
      "logps/rejected": -278.0,
      "loss": 0.023,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1865234375,
      "rewards/margins": 4.5625,
      "rewards/rejected": -4.375,
      "step": 8
    },
    {
      "epoch": 0.7736389684813754,
      "grad_norm": 0.19960970927261662,
      "learning_rate": 1e-06,
      "logits/chosen": -0.99609375,
      "logits/rejected": -0.8671875,
      "logps/chosen": -181.0,
      "logps/rejected": -338.0,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.37890625,
      "rewards/margins": 9.4375,
      "rewards/rejected": -9.0625,
      "step": 9
    },
    {
      "epoch": 0.8595988538681948,
      "grad_norm": 0.06059000889431359,
      "learning_rate": 1e-06,
      "logits/chosen": -0.9609375,
      "logits/rejected": -0.92578125,
      "logps/chosen": -240.0,
      "logps/rejected": -366.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.65625,
      "rewards/margins": 11.6875,
      "rewards/rejected": -11.0625,
      "step": 10
    },
    {
      "epoch": 0.9455587392550143,
      "grad_norm": 0.1091934088172832,
      "learning_rate": 1e-06,
      "logits/chosen": -1.0390625,
      "logits/rejected": -0.921875,
      "logps/chosen": -190.0,
      "logps/rejected": -340.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.384765625,
      "rewards/margins": 12.25,
      "rewards/rejected": -11.875,
      "step": 11
    },
    {
      "epoch": 1.0859598853868195,
      "grad_norm": 0.028289652274062375,
      "learning_rate": 1e-06,
      "logits/chosen": -1.0078125,
      "logits/rejected": -0.890625,
      "logps/chosen": -157.0,
      "logps/rejected": -390.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.55859375,
      "rewards/margins": 14.0,
      "rewards/rejected": -13.4375,
      "step": 12
    },
    {
      "epoch": 1.171919770773639,
      "grad_norm": 0.06565194416158708,
      "learning_rate": 1e-06,
      "logits/chosen": -0.87109375,
      "logits/rejected": -0.8359375,
      "logps/chosen": -190.0,
      "logps/rejected": -344.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.59375,
      "rewards/margins": 12.4375,
      "rewards/rejected": -11.875,
      "step": 13
    },
    {
      "epoch": 1.2578796561604584,
      "grad_norm": 0.01625319987755828,
      "learning_rate": 1e-06,
      "logits/chosen": -1.0078125,
      "logits/rejected": -0.9765625,
      "logps/chosen": -194.0,
      "logps/rejected": -374.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.33203125,
      "rewards/margins": 13.3125,
      "rewards/rejected": -13.0,
      "step": 14
    },
    {
      "epoch": 1.343839541547278,
      "grad_norm": 0.029292463071972353,
      "learning_rate": 1e-06,
      "logits/chosen": -1.0078125,
      "logits/rejected": -0.9609375,
      "logps/chosen": -178.0,
      "logps/rejected": -368.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.61328125,
      "rewards/margins": 14.1875,
      "rewards/rejected": -13.625,
      "step": 15
    },
    {
      "epoch": 1.4297994269340975,
      "grad_norm": 0.019950557788354577,
      "learning_rate": 1e-06,
      "logits/chosen": -0.9765625,
      "logits/rejected": -0.9296875,
      "logps/chosen": -212.0,
      "logps/rejected": -376.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.51171875,
      "rewards/margins": 14.75,
      "rewards/rejected": -14.25,
      "step": 16
    },
    {
      "epoch": 1.5157593123209168,
      "grad_norm": 0.021773657770542035,
      "learning_rate": 1e-06,
      "logits/chosen": -1.046875,
      "logits/rejected": -1.0078125,
      "logps/chosen": -233.0,
      "logps/rejected": -450.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.177734375,
      "rewards/margins": 15.5,
      "rewards/rejected": -15.375,
      "step": 17
    },
    {
      "epoch": 1.6017191977077365,
      "grad_norm": 0.009409310185294985,
      "learning_rate": 1e-06,
      "logits/chosen": -0.8984375,
      "logits/rejected": -0.90625,
      "logps/chosen": -203.0,
      "logps/rejected": -366.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.65625,
      "rewards/margins": 14.75,
      "rewards/rejected": -14.0625,
      "step": 18
    }
  ],
  "logging_steps": 1,
  "max_steps": 35,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 6,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}