{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 63.1178072888425,
      "learning_rate": 3.125e-08,
      "logits/chosen": -1.482092022895813,
      "logits/rejected": 0.7751112580299377,
      "logps/chosen": -254.50701904296875,
      "logps/rejected": -67.8974609375,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "grad_norm": 21.62022121482692,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -1.8638765811920166,
      "logits/rejected": 0.6638774275779724,
      "logps/chosen": -329.24737548828125,
      "logps/rejected": -65.78995513916016,
      "loss": 0.6723,
      "rewards/accuracies": 0.7256944179534912,
      "rewards/chosen": 0.039733994752168655,
      "rewards/margins": 0.0414588563144207,
      "rewards/rejected": -0.0017248644726350904,
      "step": 10
    },
    {
      "epoch": 0.13,
      "grad_norm": 9.802449200015984,
      "learning_rate": 4.989935734988097e-07,
      "logits/chosen": -2.5045132637023926,
      "logits/rejected": -0.7919284701347351,
      "logps/chosen": -281.2373962402344,
      "logps/rejected": -69.94824981689453,
      "loss": 0.5442,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": 0.3721862733364105,
      "rewards/margins": 0.40567564964294434,
      "rewards/rejected": -0.03348935395479202,
      "step": 20
    },
    {
      "epoch": 0.19,
      "grad_norm": 6.344038188899024,
      "learning_rate": 4.877641290737883e-07,
      "logits/chosen": -2.8063747882843018,
      "logits/rejected": -2.0033249855041504,
      "logps/chosen": -273.3968505859375,
      "logps/rejected": -83.54493713378906,
      "loss": 0.43,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.5773380398750305,
      "rewards/margins": 0.7373520135879517,
      "rewards/rejected": -0.16001388430595398,
      "step": 30
    },
    {
      "epoch": 0.26,
      "grad_norm": 4.802832110120652,
      "learning_rate": 4.646121984004665e-07,
      "logits/chosen": -2.8680758476257324,
      "logits/rejected": -2.3258016109466553,
      "logps/chosen": -263.2470397949219,
      "logps/rejected": -92.8746109008789,
      "loss": 0.3813,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.5783931612968445,
      "rewards/margins": 0.836522102355957,
      "rewards/rejected": -0.25812894105911255,
      "step": 40
    },
    {
      "epoch": 0.32,
      "grad_norm": 3.9555813235365616,
      "learning_rate": 4.3069871595684787e-07,
      "logits/chosen": -2.8306097984313965,
      "logits/rejected": -2.3546886444091797,
      "logps/chosen": -258.2146301269531,
      "logps/rejected": -98.2398452758789,
      "loss": 0.3589,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6665330529212952,
      "rewards/margins": 0.9788433313369751,
      "rewards/rejected": -0.31231021881103516,
      "step": 50
    },
    {
      "epoch": 0.38,
      "grad_norm": 4.001436828366304,
      "learning_rate": 3.877242453630256e-07,
      "logits/chosen": -2.819143772125244,
      "logits/rejected": -2.3954625129699707,
      "logps/chosen": -278.05682373046875,
      "logps/rejected": -101.1539077758789,
      "loss": 0.3482,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6159041523933411,
      "rewards/margins": 0.9530438184738159,
      "rewards/rejected": -0.33713972568511963,
      "step": 60
    },
    {
      "epoch": 0.45,
      "grad_norm": 3.7138660870592304,
      "learning_rate": 3.378437060203357e-07,
      "logits/chosen": -2.806363344192505,
      "logits/rejected": -2.4681384563446045,
      "logps/chosen": -264.37713623046875,
      "logps/rejected": -102.52363586425781,
      "loss": 0.3343,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6885555386543274,
      "rewards/margins": 1.0429760217666626,
      "rewards/rejected": -0.3544204831123352,
      "step": 70
    },
    {
      "epoch": 0.51,
      "grad_norm": 3.603973508900059,
      "learning_rate": 2.8355831645441387e-07,
      "logits/chosen": -2.8176445960998535,
      "logits/rejected": -2.5400137901306152,
      "logps/chosen": -258.44287109375,
      "logps/rejected": -103.3671875,
      "loss": 0.329,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7031112909317017,
      "rewards/margins": 1.0713307857513428,
      "rewards/rejected": -0.3682195544242859,
      "step": 80
    },
    {
      "epoch": 0.58,
      "grad_norm": 3.227257986644031,
      "learning_rate": 2.2759017277414164e-07,
      "logits/chosen": -2.844313859939575,
      "logits/rejected": -2.568953514099121,
      "logps/chosen": -261.044189453125,
      "logps/rejected": -103.6786880493164,
      "loss": 0.3265,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6580973863601685,
      "rewards/margins": 1.024906873703003,
      "rewards/rejected": -0.36680951714515686,
      "step": 90
    },
    {
      "epoch": 0.64,
      "grad_norm": 3.783887447902626,
      "learning_rate": 1.7274575140626315e-07,
      "logits/chosen": -2.835691452026367,
      "logits/rejected": -2.5570685863494873,
      "logps/chosen": -263.1366271972656,
      "logps/rejected": -104.17109680175781,
      "loss": 0.3324,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.716522216796875,
      "rewards/margins": 1.0835347175598145,
      "rewards/rejected": -0.36701256036758423,
      "step": 100
    },
    {
      "epoch": 0.7,
      "grad_norm": 3.730463342852827,
      "learning_rate": 1.2177518064852348e-07,
      "logits/chosen": -2.8178272247314453,
      "logits/rejected": -2.5992181301116943,
      "logps/chosen": -265.06048583984375,
      "logps/rejected": -104.28125,
      "loss": 0.33,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6356825232505798,
      "rewards/margins": 1.0085790157318115,
      "rewards/rejected": -0.3728964924812317,
      "step": 110
    },
    {
      "epoch": 0.77,
      "grad_norm": 3.3151932122118555,
      "learning_rate": 7.723433775328384e-08,
      "logits/chosen": -2.8309969902038574,
      "logits/rejected": -2.600473403930664,
      "logps/chosen": -269.7681884765625,
      "logps/rejected": -104.23124694824219,
      "loss": 0.327,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6922739148139954,
      "rewards/margins": 1.0601446628570557,
      "rewards/rejected": -0.36787083745002747,
      "step": 120
    },
    {
      "epoch": 0.83,
      "grad_norm": 3.337566282002191,
      "learning_rate": 4.1356686569674335e-08,
      "logits/chosen": -2.793952465057373,
      "logits/rejected": -2.5505855083465576,
      "logps/chosen": -240.2477264404297,
      "logps/rejected": -104.47578430175781,
      "loss": 0.3267,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6651207208633423,
      "rewards/margins": 1.041259527206421,
      "rewards/rejected": -0.37613874673843384,
      "step": 130
    },
    {
      "epoch": 0.9,
      "grad_norm": 3.379211511743711,
      "learning_rate": 1.5941282340065697e-08,
      "logits/chosen": -2.860114097595215,
      "logits/rejected": -2.6306686401367188,
      "logps/chosen": -263.01873779296875,
      "logps/rejected": -104.28123474121094,
      "loss": 0.324,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6528992652893066,
      "rewards/margins": 1.0218257904052734,
      "rewards/rejected": -0.3689264953136444,
      "step": 140
    },
    {
      "epoch": 0.96,
      "grad_norm": 3.2760679534529755,
      "learning_rate": 2.2625595580163247e-09,
      "logits/chosen": -2.7721505165100098,
      "logits/rejected": -2.5658388137817383,
      "logps/chosen": -233.78543090820312,
      "logps/rejected": -104.24223327636719,
      "loss": 0.3254,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6700480580329895,
      "rewards/margins": 1.0418356657028198,
      "rewards/rejected": -0.3717876970767975,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.37798528717114377,
      "train_runtime": 17606.5611,
      "train_samples_per_second": 1.136,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}