{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.15808974886784685,
  "eval_steps": 500,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013174145738987238,
      "grad_norm": 1.0125823020935059,
      "learning_rate": 6.25e-08,
      "logits/chosen": 9.990612030029297,
      "logits/rejected": 10.698101997375488,
      "logps/chosen": -102.88545989990234,
      "logps/ref_chosen": -102.88545989990234,
      "logps/ref_rejected": -121.84871673583984,
      "logps/rejected": -121.84871673583984,
      "loss": 0.4327,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "sft_loss": 0.36753880977630615,
      "step": 1
    },
    {
      "epoch": 0.026348291477974475,
      "grad_norm": 0.3579196035861969,
      "learning_rate": 1.25e-07,
      "logits/chosen": 10.211905479431152,
      "logits/rejected": 11.06594467163086,
      "logps/chosen": -107.70349884033203,
      "logps/ref_chosen": -107.70349884033203,
      "logps/ref_rejected": -121.89966583251953,
      "logps/rejected": -121.89966583251953,
      "loss": 0.4667,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "sft_loss": 0.41013145446777344,
      "step": 2
    },
    {
      "epoch": 0.03952243721696171,
      "grad_norm": 0.49040451645851135,
      "learning_rate": 1.875e-07,
      "logits/chosen": 10.035531044006348,
      "logits/rejected": 11.027185440063477,
      "logps/chosen": -108.23310852050781,
      "logps/ref_chosen": -107.98188781738281,
      "logps/ref_rejected": -124.51527404785156,
      "logps/rejected": -124.64785766601562,
      "loss": 0.4683,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.0025122263468801975,
      "rewards/margins": -0.0011863748077303171,
      "rewards/rejected": -0.0013258515391498804,
      "sft_loss": 0.41194257140159607,
      "step": 3
    },
    {
      "epoch": 0.05269658295594895,
      "grad_norm": 0.8740162253379822,
      "learning_rate": 2.5e-07,
      "logits/chosen": 9.860024452209473,
      "logits/rejected": 10.876106262207031,
      "logps/chosen": -109.94369506835938,
      "logps/ref_chosen": -109.20836639404297,
      "logps/ref_rejected": -119.23908996582031,
      "logps/rejected": -119.73454284667969,
      "loss": 0.4633,
      "rewards/accuracies": 0.3828125,
      "rewards/chosen": -0.007353362161666155,
      "rewards/margins": -0.002398767275735736,
      "rewards/rejected": -0.004954595118761063,
      "sft_loss": 0.40552011132240295,
      "step": 4
    },
    {
      "epoch": 0.06587072869493618,
      "grad_norm": 1.1980141401290894,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": 10.19467830657959,
      "logits/rejected": 10.95050048828125,
      "logps/chosen": -104.02793884277344,
      "logps/ref_chosen": -103.87680053710938,
      "logps/ref_rejected": -118.41618347167969,
      "logps/rejected": -118.46170806884766,
      "loss": 0.4351,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.001511452835984528,
      "rewards/margins": -0.001056289067491889,
      "rewards/rejected": -0.0004551640013232827,
      "sft_loss": 0.3704559803009033,
      "step": 5
    },
    {
      "epoch": 0.07904487443392343,
      "grad_norm": 0.928102433681488,
      "learning_rate": 3.75e-07,
      "logits/chosen": 10.701957702636719,
      "logits/rejected": 11.477033615112305,
      "logps/chosen": -107.61714935302734,
      "logps/ref_chosen": -107.58968353271484,
      "logps/ref_rejected": -122.07303619384766,
      "logps/rejected": -122.0443115234375,
      "loss": 0.4515,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": -0.0002746534300968051,
      "rewards/margins": -0.0005618570139631629,
      "rewards/rejected": 0.0002872035256586969,
      "sft_loss": 0.3909577429294586,
      "step": 6
    },
    {
      "epoch": 0.09221902017291066,
      "grad_norm": 1.0250380039215088,
      "learning_rate": 4.375e-07,
      "logits/chosen": 10.025421142578125,
      "logits/rejected": 10.72871208190918,
      "logps/chosen": -107.13175201416016,
      "logps/ref_chosen": -107.42727661132812,
      "logps/ref_rejected": -116.87063598632812,
      "logps/rejected": -116.28421020507812,
      "loss": 0.4392,
      "rewards/accuracies": 0.3984375,
      "rewards/chosen": 0.002955180360004306,
      "rewards/margins": -0.0029091311153024435,
      "rewards/rejected": 0.005864311475306749,
      "sft_loss": 0.3753029406070709,
      "step": 7
    },
    {
      "epoch": 0.1053931659118979,
      "grad_norm": 0.5661666393280029,
      "learning_rate": 5e-07,
      "logits/chosen": 10.203546524047852,
      "logits/rejected": 11.103278160095215,
      "logps/chosen": -104.93194580078125,
      "logps/ref_chosen": -105.60282135009766,
      "logps/ref_rejected": -119.53916931152344,
      "logps/rejected": -118.93331909179688,
      "loss": 0.4416,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": 0.006708861328661442,
      "rewards/margins": 0.0006504050688818097,
      "rewards/rejected": 0.006058456376194954,
      "sft_loss": 0.3787955939769745,
      "step": 8
    },
    {
      "epoch": 0.11856731165088513,
      "grad_norm": 0.820360541343689,
      "learning_rate": 4.997252228714278e-07,
      "logits/chosen": 10.184520721435547,
      "logits/rejected": 11.154094696044922,
      "logps/chosen": -104.26238250732422,
      "logps/ref_chosen": -105.46086120605469,
      "logps/ref_rejected": -119.00373840332031,
      "logps/rejected": -117.88744354248047,
      "loss": 0.4437,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": 0.011985024437308311,
      "rewards/margins": 0.0008220230811275542,
      "rewards/rejected": 0.011163001880049706,
      "sft_loss": 0.38146448135375977,
      "step": 9
    },
    {
      "epoch": 0.13174145738987236,
      "grad_norm": 0.4781506061553955,
      "learning_rate": 4.989014955054745e-07,
      "logits/chosen": 10.042634963989258,
      "logits/rejected": 10.866905212402344,
      "logps/chosen": -101.11405944824219,
      "logps/ref_chosen": -104.21009826660156,
      "logps/ref_rejected": -118.9209213256836,
      "logps/rejected": -115.99314880371094,
      "loss": 0.4088,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": 0.030960241332650185,
      "rewards/margins": 0.0016824830090627074,
      "rewards/rejected": 0.029277760535478592,
      "sft_loss": 0.3378788232803345,
      "step": 10
    },
    {
      "epoch": 0.14491560312885962,
      "grad_norm": 0.8178320527076721,
      "learning_rate": 4.975306286336627e-07,
      "logits/chosen": 9.987105369567871,
      "logits/rejected": 11.181533813476562,
      "logps/chosen": -101.77717590332031,
      "logps/ref_chosen": -105.94319152832031,
      "logps/ref_rejected": -122.76007843017578,
      "logps/rejected": -119.00365447998047,
      "loss": 0.4478,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": 0.04166024178266525,
      "rewards/margins": 0.0040960111655294895,
      "rewards/rejected": 0.0375642292201519,
      "sft_loss": 0.3869646191596985,
      "step": 11
    },
    {
      "epoch": 0.15808974886784685,
      "grad_norm": 0.7931886315345764,
      "learning_rate": 4.956156357188939e-07,
      "logits/chosen": 9.913724899291992,
      "logits/rejected": 10.605714797973633,
      "logps/chosen": -104.08253479003906,
      "logps/ref_chosen": -109.08442687988281,
      "logps/ref_rejected": -121.41947174072266,
      "logps/rejected": -116.61964416503906,
      "loss": 0.4233,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": 0.05001899227499962,
      "rewards/margins": 0.002020882908254862,
      "rewards/rejected": 0.04799811542034149,
      "sft_loss": 0.3560585379600525,
      "step": 12
    }
  ],
  "logging_steps": 1,
  "max_steps": 75,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 12,
  "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
}