{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 155,
  "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": -2.8946032524108887,
      "logits/rejected": -2.8606104850769043,
      "logps/chosen": -149.65138244628906,
      "logps/rejected": -147.93389892578125,
      "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": -2.7723653316497803,
      "logits/rejected": -2.7413721084594727,
      "logps/chosen": -127.34146881103516,
      "logps/rejected": -128.1923065185547,
      "loss": 0.693,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": 0.0019323664018884301,
      "rewards/margins": 0.0003104716306552291,
      "rewards/rejected": 0.001621894771233201,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989790503518888e-06,
      "logits/chosen": -2.78098464012146,
      "logits/rejected": -2.7467544078826904,
      "logps/chosen": -123.1203384399414,
      "logps/rejected": -123.0350341796875,
      "loss": 0.6922,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": 0.028792519122362137,
      "rewards/margins": 0.0023718727752566338,
      "rewards/rejected": 0.026420647278428078,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8758891663695165e-06,
      "logits/chosen": -2.7641711235046387,
      "logits/rejected": -2.7216341495513916,
      "logps/chosen": -122.10086822509766,
      "logps/rejected": -121.503662109375,
      "loss": 0.6897,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": 0.07782017439603806,
      "rewards/margins": 0.005080958362668753,
      "rewards/rejected": 0.07273921370506287,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.64113778721764e-06,
      "logits/chosen": -2.7458434104919434,
      "logits/rejected": -2.717912197113037,
      "logps/chosen": -109.0793228149414,
      "logps/rejected": -111.13133239746094,
      "loss": 0.6888,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.0947461873292923,
      "rewards/margins": 0.010546943172812462,
      "rewards/rejected": 0.08419923484325409,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297477038394368e-06,
      "logits/chosen": -2.7164647579193115,
      "logits/rejected": -2.6716389656066895,
      "logps/chosen": -124.75947570800781,
      "logps/rejected": -124.14982604980469,
      "loss": 0.6863,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.07808887958526611,
      "rewards/margins": 0.012092621996998787,
      "rewards/rejected": 0.06599625200033188,
      "step": 50
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.862387287468095e-06,
      "logits/chosen": -2.7368290424346924,
      "logits/rejected": -2.7067949771881104,
      "logps/chosen": -122.44651794433594,
      "logps/rejected": -123.14102935791016,
      "loss": 0.6814,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": 0.04377375915646553,
      "rewards/margins": 0.0176816638559103,
      "rewards/rejected": 0.02609209157526493,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.357999455114148e-06,
      "logits/chosen": -2.6553173065185547,
      "logits/rejected": -2.6178958415985107,
      "logps/chosen": -131.94534301757812,
      "logps/rejected": -128.55323791503906,
      "loss": 0.6783,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": 0.01449825894087553,
      "rewards/margins": 0.04684123024344444,
      "rewards/rejected": -0.03234297037124634,
      "step": 70
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.8099693217402807e-06,
      "logits/chosen": -2.680434226989746,
      "logits/rejected": -2.652169704437256,
      "logps/chosen": -135.7716522216797,
      "logps/rejected": -137.4171142578125,
      "loss": 0.6796,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -0.02753743901848793,
      "rewards/margins": 0.03442533686757088,
      "rewards/rejected": -0.06196277216076851,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.24617254146973e-06,
      "logits/chosen": -2.683811664581299,
      "logits/rejected": -2.6760880947113037,
      "logps/chosen": -109.57649230957031,
      "logps/rejected": -118.2197265625,
      "loss": 0.6791,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": 0.08844764530658722,
      "rewards/margins": 0.03314244747161865,
      "rewards/rejected": 0.055305201560258865,
      "step": 90
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6952867418370707e-06,
      "logits/chosen": -2.662198543548584,
      "logits/rejected": -2.6310229301452637,
      "logps/chosen": -123.90461730957031,
      "logps/rejected": -129.31741333007812,
      "loss": 0.6763,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.07869021594524384,
      "rewards/margins": 0.050858981907367706,
      "rewards/rejected": 0.027831237763166428,
      "step": 100
    },
    {
      "epoch": 0.65,
      "eval_logits/chosen": -2.662461042404175,
      "eval_logits/rejected": -2.5754306316375732,
      "eval_logps/chosen": -289.1737365722656,
      "eval_logps/rejected": -269.4944763183594,
      "eval_loss": 0.6652680039405823,
      "eval_rewards/accuracies": 0.6399999856948853,
      "eval_rewards/chosen": -0.02158256620168686,
      "eval_rewards/margins": 0.060695480555295944,
      "eval_rewards/rejected": -0.0822780430316925,
      "eval_runtime": 382.0664,
      "eval_samples_per_second": 5.235,
      "eval_steps_per_second": 0.654,
      "step": 100
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1853328309581139e-06,
      "logits/chosen": -2.6986496448516846,
      "logits/rejected": -2.6445116996765137,
      "logps/chosen": -126.5402603149414,
      "logps/rejected": -126.73878479003906,
      "loss": 0.6724,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": 0.008479948155581951,
      "rewards/margins": 0.04984438419342041,
      "rewards/rejected": -0.04136443883180618,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.422497088545436e-07,
      "logits/chosen": -2.6892261505126953,
      "logits/rejected": -2.61507511138916,
      "logps/chosen": -121.74040222167969,
      "logps/rejected": -122.19820404052734,
      "loss": 0.6739,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": -0.008245681412518024,
      "rewards/margins": 0.041609544306993484,
      "rewards/rejected": -0.049855224788188934,
      "step": 120
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.8857488050544903e-07,
      "logits/chosen": -2.6735100746154785,
      "logits/rejected": -2.6348578929901123,
      "logps/chosen": -125.81143951416016,
      "logps/rejected": -133.61953735351562,
      "loss": 0.6736,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.015428361482918262,
      "rewards/margins": 0.057859934866428375,
      "rewards/rejected": -0.04243157058954239,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4229808148697732e-07,
      "logits/chosen": -2.6265532970428467,
      "logits/rejected": -2.633122205734253,
      "logps/chosen": -131.5351104736328,
      "logps/rejected": -136.5831756591797,
      "loss": 0.6743,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": 0.021531684324145317,
      "rewards/margins": 0.03703152388334274,
      "rewards/rejected": -0.015499832108616829,
      "step": 140
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.5946226744029402e-08,
      "logits/chosen": -2.687284231185913,
      "logits/rejected": -2.641932249069214,
      "logps/chosen": -122.74555969238281,
      "logps/rejected": -129.6962127685547,
      "loss": 0.6756,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.03600780665874481,
      "rewards/margins": 0.05919266864657402,
      "rewards/rejected": -0.023184867575764656,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 155,
      "total_flos": 0.0,
      "train_loss": 0.6804818130308582,
      "train_runtime": 6451.0629,
      "train_samples_per_second": 3.075,
      "train_steps_per_second": 0.024
    }
  ],
  "logging_steps": 10,
  "max_steps": 155,
  "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
}