{
  "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": 0.48872771859169006,
      "learning_rate": 5e-07,
      "logits/chosen": 0.12278681248426437,
      "logits/rejected": 0.26080453395843506,
      "logps/chosen": -132.40695190429688,
      "logps/ref_chosen": -121.57450103759766,
      "logps/ref_rejected": -166.17074584960938,
      "logps/rejected": -177.05752563476562,
      "loss": 0.5381,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -0.10832449048757553,
      "rewards/margins": 0.000543450063560158,
      "rewards/rejected": -0.10886794328689575,
      "sft_loss": 0.4607769250869751,
      "step": 1
    },
    {
      "epoch": 0.1844380403458213,
      "grad_norm": 0.7124770283699036,
      "learning_rate": 4.849231551964771e-07,
      "logits/chosen": 0.1629990041255951,
      "logits/rejected": 0.3135248124599457,
      "logps/chosen": -134.13861083984375,
      "logps/ref_chosen": -123.470703125,
      "logps/ref_rejected": -159.41018676757812,
      "logps/rejected": -170.13693237304688,
      "loss": 0.5337,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.10667898505926132,
      "rewards/margins": 0.000588476425036788,
      "rewards/rejected": -0.10726746171712875,
      "sft_loss": 0.4541751444339752,
      "step": 2
    },
    {
      "epoch": 0.276657060518732,
      "grad_norm": 0.5838692784309387,
      "learning_rate": 4.415111107797445e-07,
      "logits/chosen": 0.09741576015949249,
      "logits/rejected": 0.2005864381790161,
      "logps/chosen": -134.69981384277344,
      "logps/ref_chosen": -124.5863037109375,
      "logps/ref_rejected": -152.92868041992188,
      "logps/rejected": -163.17599487304688,
      "loss": 0.5444,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.10113508999347687,
      "rewards/margins": 0.0013378520961850882,
      "rewards/rejected": -0.1024729460477829,
      "sft_loss": 0.4703802764415741,
      "step": 3
    },
    {
      "epoch": 0.3688760806916426,
      "grad_norm": 0.5232725739479065,
      "learning_rate": 3.75e-07,
      "logits/chosen": 0.11469627916812897,
      "logits/rejected": 0.34779858589172363,
      "logps/chosen": -131.96142578125,
      "logps/ref_chosen": -123.3237533569336,
      "logps/ref_rejected": -160.29180908203125,
      "logps/rejected": -169.1221466064453,
      "loss": 0.5541,
      "rewards/accuracies": 0.6015625,
      "rewards/chosen": -0.08637689054012299,
      "rewards/margins": 0.00192636763677001,
      "rewards/rejected": -0.08830326050519943,
      "sft_loss": 0.48508843779563904,
      "step": 4
    },
    {
      "epoch": 0.4610951008645533,
      "grad_norm": 0.4373687505722046,
      "learning_rate": 2.934120444167326e-07,
      "logits/chosen": 0.13797219097614288,
      "logits/rejected": 0.2943338453769684,
      "logps/chosen": -129.5037841796875,
      "logps/ref_chosen": -122.18938446044922,
      "logps/ref_rejected": -166.5259246826172,
      "logps/rejected": -173.7913055419922,
      "loss": 0.5453,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.07314412295818329,
      "rewards/margins": -0.0004904267261736095,
      "rewards/rejected": -0.07265370339155197,
      "sft_loss": 0.4711988866329193,
      "step": 5
    },
    {
      "epoch": 0.553314121037464,
      "grad_norm": 0.41881853342056274,
      "learning_rate": 2.065879555832674e-07,
      "logits/chosen": 0.1323821097612381,
      "logits/rejected": 0.31871122121810913,
      "logps/chosen": -132.2285614013672,
      "logps/ref_chosen": -125.81028747558594,
      "logps/ref_rejected": -156.31320190429688,
      "logps/rejected": -162.79730224609375,
      "loss": 0.5281,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.06418267637491226,
      "rewards/margins": 0.0006581920897588134,
      "rewards/rejected": -0.06484086811542511,
      "sft_loss": 0.44570034742355347,
      "step": 6
    },
    {
      "epoch": 0.6455331412103746,
      "grad_norm": 0.3571808934211731,
      "learning_rate": 1.2500000000000005e-07,
      "logits/chosen": 0.18207934498786926,
      "logits/rejected": 0.28170838952064514,
      "logps/chosen": -130.76219177246094,
      "logps/ref_chosen": -128.7080535888672,
      "logps/ref_rejected": -158.98272705078125,
      "logps/rejected": -161.1389923095703,
      "loss": 0.5175,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.02054133079946041,
      "rewards/margins": 0.0010213707573711872,
      "rewards/rejected": -0.02156270295381546,
      "sft_loss": 0.4298672676086426,
      "step": 7
    },
    {
      "epoch": 0.7377521613832853,
      "grad_norm": 0.4075394570827484,
      "learning_rate": 5.848888922025552e-08,
      "logits/chosen": 0.10367591679096222,
      "logits/rejected": 0.28194165229797363,
      "logps/chosen": -124.07286834716797,
      "logps/ref_chosen": -123.2734375,
      "logps/ref_rejected": -156.3369903564453,
      "logps/rejected": -157.23104858398438,
      "loss": 0.5234,
      "rewards/accuracies": 0.6015625,
      "rewards/chosen": -0.007994325831532478,
      "rewards/margins": 0.0009461576119065285,
      "rewards/rejected": -0.008940483443439007,
      "sft_loss": 0.4388274848461151,
      "step": 8
    },
    {
      "epoch": 0.829971181556196,
      "grad_norm": 0.230843648314476,
      "learning_rate": 1.507684480352292e-08,
      "logits/chosen": 0.06345280259847641,
      "logits/rejected": 0.206377774477005,
      "logps/chosen": -127.7571792602539,
      "logps/ref_chosen": -127.38499450683594,
      "logps/ref_rejected": -157.14312744140625,
      "logps/rejected": -157.56546020507812,
      "loss": 0.5129,
      "rewards/accuracies": 0.5859375,
      "rewards/chosen": -0.00372200645506382,
      "rewards/margins": 0.0005012964247725904,
      "rewards/rejected": -0.004223302938044071,
      "sft_loss": 0.42283880710601807,
      "step": 9
    },
    {
      "epoch": 0.9221902017291066,
      "grad_norm": 0.43585899472236633,
      "learning_rate": 0.0,
      "logits/chosen": 0.08928333222866058,
      "logits/rejected": 0.238207146525383,
      "logps/chosen": -124.22206115722656,
      "logps/ref_chosen": -123.99081420898438,
      "logps/ref_rejected": -154.0594940185547,
      "logps/rejected": -154.32521057128906,
      "loss": 0.5223,
      "rewards/accuracies": 0.5234375,
      "rewards/chosen": -0.0023124003782868385,
      "rewards/margins": 0.00034469965612515807,
      "rewards/rejected": -0.002657099859789014,
      "sft_loss": 0.4369175434112549,
      "step": 10
    },
    {
      "epoch": 0.9221902017291066,
      "step": 10,
      "total_flos": 0.0,
      "train_loss": 0.5319805860519409,
      "train_runtime": 668.7287,
      "train_samples_per_second": 2.074,
      "train_steps_per_second": 0.015
    }
  ],
  "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
}