{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-07,
      "logits/chosen": -2.89351749420166,
      "logits/rejected": -2.7752203941345215,
      "logps/chosen": -345.7324523925781,
      "logps/rejected": -319.42047119140625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": -2.8028833866119385,
      "logits/rejected": -2.7471988201141357,
      "logps/chosen": -255.036865234375,
      "logps/rejected": -252.82679748535156,
      "loss": 0.6928,
      "rewards/accuracies": 0.4583333432674408,
      "rewards/chosen": 0.0003066221543122083,
      "rewards/margins": 0.0006635435856878757,
      "rewards/rejected": -0.0003569214604794979,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -2.7710824012756348,
      "logits/rejected": -2.7166454792022705,
      "logps/chosen": -277.1798400878906,
      "logps/rejected": -256.997802734375,
      "loss": 0.6889,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 0.012738336808979511,
      "rewards/margins": 0.008854442276060581,
      "rewards/rejected": 0.00388389453291893,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -2.7218027114868164,
      "logits/rejected": -2.658277750015259,
      "logps/chosen": -274.3503112792969,
      "logps/rejected": -246.41128540039062,
      "loss": 0.6795,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.03419749438762665,
      "rewards/margins": 0.027278240770101547,
      "rewards/rejected": 0.006919251289218664,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.7631821632385254,
      "logits/rejected": -2.6550638675689697,
      "logps/chosen": -268.45220947265625,
      "logps/rejected": -251.58743286132812,
      "loss": 0.6693,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": 0.03486743941903114,
      "rewards/margins": 0.05773182958364487,
      "rewards/rejected": -0.022864393889904022,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -2.676445960998535,
      "logits/rejected": -2.62298321723938,
      "logps/chosen": -274.287841796875,
      "logps/rejected": -281.6821594238281,
      "loss": 0.6605,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.003023784141987562,
      "rewards/margins": 0.07856948673725128,
      "rewards/rejected": -0.08159326761960983,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -2.687849521636963,
      "logits/rejected": -2.6154909133911133,
      "logps/chosen": -287.0888366699219,
      "logps/rejected": -274.48822021484375,
      "loss": 0.6507,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.03279999643564224,
      "rewards/margins": 0.11418493092060089,
      "rewards/rejected": -0.14698493480682373,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -2.6265475749969482,
      "logits/rejected": -2.570312976837158,
      "logps/chosen": -304.10125732421875,
      "logps/rejected": -292.9505310058594,
      "loss": 0.6431,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.033603884279727936,
      "rewards/margins": 0.12847770750522614,
      "rewards/rejected": -0.16208159923553467,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.6333694458007812,
      "logits/rejected": -2.5370612144470215,
      "logps/chosen": -302.35736083984375,
      "logps/rejected": -276.11175537109375,
      "loss": 0.634,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.04339645802974701,
      "rewards/margins": 0.14248773455619812,
      "rewards/rejected": -0.18588420748710632,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -2.584850311279297,
      "logits/rejected": -2.5956850051879883,
      "logps/chosen": -292.6184997558594,
      "logps/rejected": -311.091796875,
      "loss": 0.6249,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.09801232814788818,
      "rewards/margins": 0.17943526804447174,
      "rewards/rejected": -0.27744758129119873,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -2.6319050788879395,
      "logits/rejected": -2.5922951698303223,
      "logps/chosen": -290.19976806640625,
      "logps/rejected": -273.79815673828125,
      "loss": 0.6297,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.1279720962047577,
      "rewards/margins": 0.16270975768566132,
      "rewards/rejected": -0.2906818985939026,
      "step": 100
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.6261305809020996,
      "eval_logits/rejected": -2.541619062423706,
      "eval_logps/chosen": -295.3544006347656,
      "eval_logps/rejected": -286.08172607421875,
      "eval_loss": 0.6284892559051514,
      "eval_rewards/accuracies": 0.699999988079071,
      "eval_rewards/chosen": -0.11314628273248672,
      "eval_rewards/margins": 0.1725120097398758,
      "eval_rewards/rejected": -0.2856582701206207,
      "eval_runtime": 384.1378,
      "eval_samples_per_second": 5.206,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -2.6338038444519043,
      "logits/rejected": -2.5288445949554443,
      "logps/chosen": -289.0359802246094,
      "logps/rejected": -275.2894287109375,
      "loss": 0.6177,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.06765095144510269,
      "rewards/margins": 0.20255950093269348,
      "rewards/rejected": -0.2702104449272156,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -2.552917003631592,
      "logits/rejected": -2.550574779510498,
      "logps/chosen": -274.2412109375,
      "logps/rejected": -288.6865539550781,
      "loss": 0.6251,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.10379727929830551,
      "rewards/margins": 0.1611437350511551,
      "rewards/rejected": -0.2649410367012024,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -2.611356735229492,
      "logits/rejected": -2.5477137565612793,
      "logps/chosen": -291.66290283203125,
      "logps/rejected": -301.75537109375,
      "loss": 0.6233,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.11176357418298721,
      "rewards/margins": 0.20609335601329803,
      "rewards/rejected": -0.31785690784454346,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -2.6064445972442627,
      "logits/rejected": -2.479429244995117,
      "logps/chosen": -313.6204833984375,
      "logps/rejected": -287.5814514160156,
      "loss": 0.6146,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.09765791893005371,
      "rewards/margins": 0.21785131096839905,
      "rewards/rejected": -0.31550922989845276,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -2.58495831489563,
      "logits/rejected": -2.4780099391937256,
      "logps/chosen": -300.6700134277344,
      "logps/rejected": -281.14935302734375,
      "loss": 0.6193,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": -0.11531106382608414,
      "rewards/margins": 0.22770515084266663,
      "rewards/rejected": -0.34301620721817017,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.6446946973984058,
      "train_runtime": 7192.2445,
      "train_samples_per_second": 2.781,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}