{
  "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": 0.8317140340805054,
      "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": 0.4978,
      "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": 0.2926611006259918,
      "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": 0.5233,
      "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": 0.3214672803878784,
      "learning_rate": 1.875e-07,
      "logits/chosen": 10.036933898925781,
      "logits/rejected": 11.024795532226562,
      "logps/chosen": -108.28660583496094,
      "logps/ref_chosen": -107.98188781738281,
      "logps/ref_rejected": -124.51527404785156,
      "logps/rejected": -124.7075424194336,
      "loss": 0.5249,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.0030471612699329853,
      "rewards/margins": -0.001124453847296536,
      "rewards/rejected": -0.0019227075390517712,
      "sft_loss": 0.4123927652835846,
      "step": 3
    },
    {
      "epoch": 0.05269658295594895,
      "grad_norm": 0.7239967584609985,
      "learning_rate": 2.5e-07,
      "logits/chosen": 9.835367202758789,
      "logits/rejected": 10.854362487792969,
      "logps/chosen": -109.89208221435547,
      "logps/ref_chosen": -109.20836639404297,
      "logps/ref_rejected": -119.23908996582031,
      "logps/rejected": -119.69357299804688,
      "loss": 0.5208,
      "rewards/accuracies": 0.453125,
      "rewards/chosen": -0.006837163120508194,
      "rewards/margins": -0.0022922407370060682,
      "rewards/rejected": -0.004544922150671482,
      "sft_loss": 0.4051341712474823,
      "step": 4
    },
    {
      "epoch": 0.06587072869493618,
      "grad_norm": 0.9211171269416809,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": 10.207865715026855,
      "logits/rejected": 10.963621139526367,
      "logps/chosen": -104.02088928222656,
      "logps/ref_chosen": -103.87680053710938,
      "logps/ref_rejected": -118.41618347167969,
      "logps/rejected": -118.34123992919922,
      "loss": 0.5001,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0014409449649974704,
      "rewards/margins": -0.002190487692132592,
      "rewards/rejected": 0.0007495426689274609,
      "sft_loss": 0.3706103563308716,
      "step": 5
    },
    {
      "epoch": 0.07904487443392343,
      "grad_norm": 0.7005686163902283,
      "learning_rate": 3.75e-07,
      "logits/chosen": 10.686239242553711,
      "logits/rejected": 11.462547302246094,
      "logps/chosen": -108.1670913696289,
      "logps/ref_chosen": -107.58968353271484,
      "logps/ref_rejected": -122.07303619384766,
      "logps/rejected": -122.58065032958984,
      "loss": 0.5134,
      "rewards/accuracies": 0.484375,
      "rewards/chosen": -0.005774094723165035,
      "rewards/margins": -0.0006979470490477979,
      "rewards/rejected": -0.005076148081570864,
      "sft_loss": 0.39328432083129883,
      "step": 6
    },
    {
      "epoch": 0.09221902017291066,
      "grad_norm": 0.8737779259681702,
      "learning_rate": 4.375e-07,
      "logits/chosen": 10.040006637573242,
      "logits/rejected": 10.747206687927246,
      "logps/chosen": -107.037353515625,
      "logps/ref_chosen": -107.42727661132812,
      "logps/ref_rejected": -116.87063598632812,
      "logps/rejected": -116.42378997802734,
      "loss": 0.5022,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": 0.0038992553018033504,
      "rewards/margins": -0.0005691752885468304,
      "rewards/rejected": 0.004468431230634451,
      "sft_loss": 0.37478095293045044,
      "step": 7
    },
    {
      "epoch": 0.1053931659118979,
      "grad_norm": 0.46314188838005066,
      "learning_rate": 5e-07,
      "logits/chosen": 10.196634292602539,
      "logits/rejected": 11.089151382446289,
      "logps/chosen": -104.9737548828125,
      "logps/ref_chosen": -105.60282135009766,
      "logps/ref_rejected": -119.53916931152344,
      "logps/rejected": -118.9640884399414,
      "loss": 0.5047,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": 0.006290654186159372,
      "rewards/margins": 0.0005399012006819248,
      "rewards/rejected": 0.0057507529854774475,
      "sft_loss": 0.37916287779808044,
      "step": 8
    },
    {
      "epoch": 0.11856731165088513,
      "grad_norm": 0.6739105582237244,
      "learning_rate": 4.997252228714278e-07,
      "logits/chosen": 10.174893379211426,
      "logits/rejected": 11.141225814819336,
      "logps/chosen": -104.830078125,
      "logps/ref_chosen": -105.46086120605469,
      "logps/ref_rejected": -119.00373840332031,
      "logps/rejected": -118.41084289550781,
      "loss": 0.5073,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": 0.006307927425950766,
      "rewards/margins": 0.00037892413092777133,
      "rewards/rejected": 0.005929003469645977,
      "sft_loss": 0.3834277391433716,
      "step": 9
    },
    {
      "epoch": 0.13174145738987236,
      "grad_norm": 0.32560959458351135,
      "learning_rate": 4.989014955054745e-07,
      "logits/chosen": 10.061932563781738,
      "logits/rejected": 10.884778022766113,
      "logps/chosen": -102.10572052001953,
      "logps/ref_chosen": -104.21009826660156,
      "logps/ref_rejected": -118.9209213256836,
      "logps/rejected": -117.1087646484375,
      "loss": 0.4814,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.02104363776743412,
      "rewards/margins": 0.00292214541696012,
      "rewards/rejected": 0.018121493980288506,
      "sft_loss": 0.34126415848731995,
      "step": 10
    },
    {
      "epoch": 0.14491560312885962,
      "grad_norm": 0.8538271188735962,
      "learning_rate": 4.975306286336627e-07,
      "logits/chosen": 9.989534378051758,
      "logits/rejected": 11.177312850952148,
      "logps/chosen": -102.71998596191406,
      "logps/ref_chosen": -105.94319152832031,
      "logps/ref_rejected": -122.76007843017578,
      "logps/rejected": -119.8812026977539,
      "loss": 0.5112,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": 0.03223201259970665,
      "rewards/margins": 0.0034431489184498787,
      "rewards/rejected": 0.028788862749934196,
      "sft_loss": 0.3909299075603485,
      "step": 11
    },
    {
      "epoch": 0.15808974886784685,
      "grad_norm": 0.599084734916687,
      "learning_rate": 4.956156357188939e-07,
      "logits/chosen": 9.907474517822266,
      "logits/rejected": 10.599059104919434,
      "logps/chosen": -105.25791931152344,
      "logps/ref_chosen": -109.08442687988281,
      "logps/ref_rejected": -121.41947174072266,
      "logps/rejected": -117.7214584350586,
      "loss": 0.4934,
      "rewards/accuracies": 0.5546875,
      "rewards/chosen": 0.03826504573225975,
      "rewards/margins": 0.00128496577963233,
      "rewards/rejected": 0.036980077624320984,
      "sft_loss": 0.3606122136116028,
      "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
}