{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9221902017291066,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09221902017291066,
      "grad_norm": 0.7050405740737915,
      "learning_rate": 5e-07,
      "logits/chosen": 0.1478731483221054,
      "logits/rejected": 0.28315749764442444,
      "logps/chosen": -134.61978149414062,
      "logps/ref_chosen": -134.61978149414062,
      "logps/ref_rejected": -173.30970764160156,
      "logps/rejected": -173.30970764160156,
      "loss": 0.4709,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "sft_loss": 0.47093769907951355,
      "step": 1
    },
    {
      "epoch": 0.1844380403458213,
      "grad_norm": 0.6069543957710266,
      "learning_rate": 4.849231551964771e-07,
      "logits/chosen": 0.14105559885501862,
      "logits/rejected": 0.2469622939825058,
      "logps/chosen": -130.12448120117188,
      "logps/ref_chosen": -130.12448120117188,
      "logps/ref_rejected": -163.5547637939453,
      "logps/rejected": -163.5547637939453,
      "loss": 0.4654,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "sft_loss": 0.4653787612915039,
      "step": 2
    },
    {
      "epoch": 0.276657060518732,
      "grad_norm": 0.6174837946891785,
      "learning_rate": 4.415111107797445e-07,
      "logits/chosen": 0.06450571119785309,
      "logits/rejected": 0.24848753213882446,
      "logps/chosen": -129.13465881347656,
      "logps/ref_chosen": -129.6482391357422,
      "logps/ref_rejected": -167.99598693847656,
      "logps/rejected": -167.31204223632812,
      "loss": 0.4657,
      "rewards/accuracies": 0.3671875,
      "rewards/chosen": 0.005135728046298027,
      "rewards/margins": -0.0017038242658600211,
      "rewards/rejected": 0.00683955242857337,
      "sft_loss": 0.4656541049480438,
      "step": 3
    },
    {
      "epoch": 0.3688760806916426,
      "grad_norm": 0.5388476848602295,
      "learning_rate": 3.75e-07,
      "logits/chosen": 0.1188218742609024,
      "logits/rejected": 0.26108819246292114,
      "logps/chosen": -130.52902221679688,
      "logps/ref_chosen": -132.77403259277344,
      "logps/ref_rejected": -168.05410766601562,
      "logps/rejected": -165.7576904296875,
      "loss": 0.4781,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": 0.022450026124715805,
      "rewards/margins": -0.0005142414593137801,
      "rewards/rejected": 0.022964270785450935,
      "sft_loss": 0.47812312841415405,
      "step": 4
    },
    {
      "epoch": 0.4610951008645533,
      "grad_norm": 0.9245567917823792,
      "learning_rate": 2.934120444167326e-07,
      "logits/chosen": 0.12649387121200562,
      "logits/rejected": 0.28299272060394287,
      "logps/chosen": -132.39537048339844,
      "logps/ref_chosen": -135.82839965820312,
      "logps/ref_rejected": -161.7675323486328,
      "logps/rejected": -158.30758666992188,
      "loss": 0.4737,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": 0.03433013707399368,
      "rewards/margins": -0.0002691660774871707,
      "rewards/rejected": 0.03459930792450905,
      "sft_loss": 0.47371014952659607,
      "step": 5
    },
    {
      "epoch": 0.553314121037464,
      "grad_norm": 0.6128867268562317,
      "learning_rate": 2.065879555832674e-07,
      "logits/chosen": 0.10092329978942871,
      "logits/rejected": 0.20842014253139496,
      "logps/chosen": -133.4036407470703,
      "logps/ref_chosen": -137.62083435058594,
      "logps/ref_rejected": -165.1855926513672,
      "logps/rejected": -160.89614868164062,
      "loss": 0.4566,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": 0.04217197746038437,
      "rewards/margins": -0.000722518190741539,
      "rewards/rejected": 0.04289449751377106,
      "sft_loss": 0.45664921402931213,
      "step": 6
    },
    {
      "epoch": 0.6455331412103746,
      "grad_norm": 0.5220282077789307,
      "learning_rate": 1.2500000000000005e-07,
      "logits/chosen": 0.12199896574020386,
      "logits/rejected": 0.26001349091529846,
      "logps/chosen": -130.01318359375,
      "logps/ref_chosen": -140.14544677734375,
      "logps/ref_rejected": -173.2746124267578,
      "logps/rejected": -163.0173797607422,
      "loss": 0.439,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": 0.10132250934839249,
      "rewards/margins": -0.0012498770374804735,
      "rewards/rejected": 0.10257238149642944,
      "sft_loss": 0.43903031945228577,
      "step": 7
    },
    {
      "epoch": 0.7377521613832853,
      "grad_norm": 0.6048033833503723,
      "learning_rate": 5.848888922025552e-08,
      "logits/chosen": 0.08684300631284714,
      "logits/rejected": 0.25959163904190063,
      "logps/chosen": -120.96797943115234,
      "logps/ref_chosen": -132.29315185546875,
      "logps/ref_rejected": -175.901611328125,
      "logps/rejected": -164.5523223876953,
      "loss": 0.435,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": 0.11325177550315857,
      "rewards/margins": -0.00024128076620399952,
      "rewards/rejected": 0.11349306255578995,
      "sft_loss": 0.43499475717544556,
      "step": 8
    },
    {
      "epoch": 0.829971181556196,
      "grad_norm": 0.5645197629928589,
      "learning_rate": 1.507684480352292e-08,
      "logits/chosen": 0.10725325345993042,
      "logits/rejected": 0.2104402333498001,
      "logps/chosen": -120.40372467041016,
      "logps/ref_chosen": -132.26475524902344,
      "logps/ref_rejected": -158.7421112060547,
      "logps/rejected": -146.69927978515625,
      "loss": 0.4335,
      "rewards/accuracies": 0.3984375,
      "rewards/chosen": 0.11861036717891693,
      "rewards/margins": -0.0018178187310695648,
      "rewards/rejected": 0.1204281896352768,
      "sft_loss": 0.4335402548313141,
      "step": 9
    },
    {
      "epoch": 0.9221902017291066,
      "grad_norm": 0.6507696509361267,
      "learning_rate": 0.0,
      "logits/chosen": 0.13997340202331543,
      "logits/rejected": 0.26554781198501587,
      "logps/chosen": -127.0248031616211,
      "logps/ref_chosen": -138.58920288085938,
      "logps/ref_rejected": -173.31874084472656,
      "logps/rejected": -161.5537872314453,
      "loss": 0.4145,
      "rewards/accuracies": 0.421875,
      "rewards/chosen": 0.11564403027296066,
      "rewards/margins": -0.0020055007189512253,
      "rewards/rejected": 0.11764952540397644,
      "sft_loss": 0.41452255845069885,
      "step": 10
    },
    {
      "epoch": 0.9221902017291066,
      "step": 10,
      "total_flos": 0.0,
      "train_loss": 0.45325408279895785,
      "train_runtime": 617.7917,
      "train_samples_per_second": 2.245,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}