{
  "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": 0.3283297121524811,
      "learning_rate": 5e-07,
      "logits/chosen": 0.08735769987106323,
      "logits/rejected": 0.2894211411476135,
      "logps/chosen": -131.21682739257812,
      "logps/ref_chosen": -121.79497528076172,
      "logps/ref_rejected": -161.43960571289062,
      "logps/rejected": -170.8655242919922,
      "loss": 0.6172,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.09421855211257935,
      "rewards/margins": 4.066087421961129e-05,
      "rewards/rejected": -0.09425921738147736,
      "sft_loss": 0.4654408097267151,
      "step": 1
    },
    {
      "epoch": 0.1844380403458213,
      "grad_norm": 0.3361700177192688,
      "learning_rate": 4.849231551964771e-07,
      "logits/chosen": 0.13249221444129944,
      "logits/rejected": 0.2818200886249542,
      "logps/chosen": -133.42129516601562,
      "logps/ref_chosen": -124.1944808959961,
      "logps/ref_rejected": -155.79832458496094,
      "logps/rejected": -164.98802185058594,
      "loss": 0.6221,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.092268206179142,
      "rewards/margins": -0.00037108425749465823,
      "rewards/rejected": -0.09189712256193161,
      "sft_loss": 0.4797438383102417,
      "step": 2
    },
    {
      "epoch": 0.276657060518732,
      "grad_norm": 0.3492422103881836,
      "learning_rate": 4.415111107797445e-07,
      "logits/chosen": 0.1395522505044937,
      "logits/rejected": 0.3097372353076935,
      "logps/chosen": -134.55520629882812,
      "logps/ref_chosen": -126.107666015625,
      "logps/ref_rejected": -166.2165985107422,
      "logps/rejected": -174.74267578125,
      "loss": 0.6219,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.08447550237178802,
      "rewards/margins": 0.0007849468383938074,
      "rewards/rejected": -0.08526045083999634,
      "sft_loss": 0.48026716709136963,
      "step": 3
    },
    {
      "epoch": 0.3688760806916426,
      "grad_norm": 0.40037524700164795,
      "learning_rate": 3.75e-07,
      "logits/chosen": 0.1005653589963913,
      "logits/rejected": 0.24767020344734192,
      "logps/chosen": -132.58848571777344,
      "logps/ref_chosen": -125.09526062011719,
      "logps/ref_rejected": -163.86485290527344,
      "logps/rejected": -171.28343200683594,
      "loss": 0.6207,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": -0.07493225485086441,
      "rewards/margins": -0.0007463583024218678,
      "rewards/rejected": -0.07418590039014816,
      "sft_loss": 0.47509297728538513,
      "step": 4
    },
    {
      "epoch": 0.4610951008645533,
      "grad_norm": 0.3622381389141083,
      "learning_rate": 2.934120444167326e-07,
      "logits/chosen": 0.09210827201604843,
      "logits/rejected": 0.27451103925704956,
      "logps/chosen": -127.89833068847656,
      "logps/ref_chosen": -121.33724212646484,
      "logps/ref_rejected": -157.88401794433594,
      "logps/rejected": -164.47955322265625,
      "loss": 0.6192,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.06561093032360077,
      "rewards/margins": 0.00034457340370863676,
      "rewards/rejected": -0.0659554973244667,
      "sft_loss": 0.4717320203781128,
      "step": 5
    },
    {
      "epoch": 0.553314121037464,
      "grad_norm": 0.3330410420894623,
      "learning_rate": 2.065879555832674e-07,
      "logits/chosen": 0.08373218774795532,
      "logits/rejected": 0.31112056970596313,
      "logps/chosen": -130.05555725097656,
      "logps/ref_chosen": -124.18991088867188,
      "logps/ref_rejected": -162.19686889648438,
      "logps/rejected": -168.12203979492188,
      "loss": 0.6132,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.05865645781159401,
      "rewards/margins": 0.0005952669307589531,
      "rewards/rejected": -0.059251733124256134,
      "sft_loss": 0.4540290832519531,
      "step": 6
    },
    {
      "epoch": 0.6455331412103746,
      "grad_norm": 0.3554254174232483,
      "learning_rate": 1.2500000000000005e-07,
      "logits/chosen": 0.0741545632481575,
      "logits/rejected": 0.2337346076965332,
      "logps/chosen": -128.43069458007812,
      "logps/ref_chosen": -124.93177032470703,
      "logps/ref_rejected": -159.1948699951172,
      "logps/rejected": -162.75405883789062,
      "loss": 0.6147,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.03498939424753189,
      "rewards/margins": 0.0006023755413480103,
      "rewards/rejected": -0.035591769963502884,
      "sft_loss": 0.45834434032440186,
      "step": 7
    },
    {
      "epoch": 0.7377521613832853,
      "grad_norm": 0.3500949740409851,
      "learning_rate": 5.848888922025552e-08,
      "logits/chosen": 0.14493891596794128,
      "logits/rejected": 0.311847060918808,
      "logps/chosen": -127.07202911376953,
      "logps/ref_chosen": -124.96209716796875,
      "logps/ref_rejected": -158.53147888183594,
      "logps/rejected": -160.7977752685547,
      "loss": 0.6134,
      "rewards/accuracies": 0.6328125,
      "rewards/chosen": -0.021099410951137543,
      "rewards/margins": 0.001563590602017939,
      "rewards/rejected": -0.022663000971078873,
      "sft_loss": 0.4554786682128906,
      "step": 8
    },
    {
      "epoch": 0.829971181556196,
      "grad_norm": 0.4512203335762024,
      "learning_rate": 1.507684480352292e-08,
      "logits/chosen": 0.10647760331630707,
      "logits/rejected": 0.3222857713699341,
      "logps/chosen": -131.0767059326172,
      "logps/ref_chosen": -129.31468200683594,
      "logps/ref_rejected": -173.19229125976562,
      "logps/rejected": -174.9317169189453,
      "loss": 0.6109,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": -0.01762016862630844,
      "rewards/margins": -0.00022587231069337577,
      "rewards/rejected": -0.0173942968249321,
      "sft_loss": 0.44625312089920044,
      "step": 9
    },
    {
      "epoch": 0.9221902017291066,
      "grad_norm": 0.2977113425731659,
      "learning_rate": 0.0,
      "logits/chosen": 0.1302759051322937,
      "logits/rejected": 0.2790878713130951,
      "logps/chosen": -127.37201690673828,
      "logps/ref_chosen": -125.88874816894531,
      "logps/ref_rejected": -154.29812622070312,
      "logps/rejected": -155.82936096191406,
      "loss": 0.6057,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.014832701534032822,
      "rewards/margins": 0.00047985377022996545,
      "rewards/rejected": -0.015312554314732552,
      "sft_loss": 0.4313240647315979,
      "step": 10
    },
    {
      "epoch": 0.9221902017291066,
      "step": 10,
      "total_flos": 0.0,
      "train_loss": 0.6159235239028931,
      "train_runtime": 636.1291,
      "train_samples_per_second": 2.18,
      "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
}