{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7960199004975125,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03980099502487562,
      "grad_norm": 9.524229049682617,
      "learning_rate": 4.980474401576887e-07,
      "logits/chosen": -0.1925463080406189,
      "logits/rejected": -0.15873177349567413,
      "logps/chosen": -239.6322021484375,
      "logps/rejected": -242.57594299316406,
      "loss": 0.8819,
      "rewards/accuracies": 0.49281251430511475,
      "rewards/chosen": -1.3072433471679688,
      "rewards/margins": 0.008615786209702492,
      "rewards/rejected": -1.315859079360962,
      "step": 100
    },
    {
      "epoch": 0.07960199004975124,
      "grad_norm": 7.333236217498779,
      "learning_rate": 4.922202605502572e-07,
      "logits/chosen": 0.046361636370420456,
      "logits/rejected": 0.08236894011497498,
      "logps/chosen": -231.10765075683594,
      "logps/rejected": -233.24984741210938,
      "loss": 0.8361,
      "rewards/accuracies": 0.48875001072883606,
      "rewards/chosen": -0.9621695876121521,
      "rewards/margins": -0.021255964413285255,
      "rewards/rejected": -0.9409136772155762,
      "step": 200
    },
    {
      "epoch": 0.11940298507462686,
      "grad_norm": 5.164106369018555,
      "learning_rate": 4.82609484512869e-07,
      "logits/chosen": 0.0250965878367424,
      "logits/rejected": 0.054141998291015625,
      "logps/chosen": -229.79965209960938,
      "logps/rejected": -231.60562133789062,
      "loss": 0.8114,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.8564431667327881,
      "rewards/margins": -0.01034404058009386,
      "rewards/rejected": -0.8460990786552429,
      "step": 300
    },
    {
      "epoch": 0.15920398009950248,
      "grad_norm": 5.328795909881592,
      "learning_rate": 4.6936523696827614e-07,
      "logits/chosen": 0.039505548775196075,
      "logits/rejected": 0.05482972040772438,
      "logps/chosen": -233.69537353515625,
      "logps/rejected": -236.25210571289062,
      "loss": 0.7907,
      "rewards/accuracies": 0.49406251311302185,
      "rewards/chosen": -0.8005841970443726,
      "rewards/margins": 0.00799934659153223,
      "rewards/rejected": -0.8085834980010986,
      "step": 400
    },
    {
      "epoch": 0.19900497512437812,
      "grad_norm": 8.521681785583496,
      "learning_rate": 4.5269439940365644e-07,
      "logits/chosen": 0.11151163280010223,
      "logits/rejected": 0.1460810899734497,
      "logps/chosen": -227.23020935058594,
      "logps/rejected": -230.19451904296875,
      "loss": 0.7849,
      "rewards/accuracies": 0.4946874976158142,
      "rewards/chosen": -0.7644023299217224,
      "rewards/margins": -0.004693666007369757,
      "rewards/rejected": -0.7597086429595947,
      "step": 500
    },
    {
      "epoch": 0.23880597014925373,
      "grad_norm": 7.143113136291504,
      "learning_rate": 4.328573782827409e-07,
      "logits/chosen": 0.09673156589269638,
      "logits/rejected": 0.14390873908996582,
      "logps/chosen": -225.45919799804688,
      "logps/rejected": -225.44052124023438,
      "loss": 0.77,
      "rewards/accuracies": 0.5045312643051147,
      "rewards/chosen": -0.7256423830986023,
      "rewards/margins": 0.004257932770997286,
      "rewards/rejected": -0.7299003005027771,
      "step": 600
    },
    {
      "epoch": 0.27860696517412936,
      "grad_norm": 6.525391101837158,
      "learning_rate": 4.1016403737218373e-07,
      "logits/chosen": 0.05486857891082764,
      "logits/rejected": 0.09620587527751923,
      "logps/chosen": -228.3573760986328,
      "logps/rejected": -223.32395935058594,
      "loss": 0.7696,
      "rewards/accuracies": 0.5034375190734863,
      "rewards/chosen": -0.7243590354919434,
      "rewards/margins": 0.0008349200943484902,
      "rewards/rejected": -0.7251940369606018,
      "step": 700
    },
    {
      "epoch": 0.31840796019900497,
      "grad_norm": 8.071100234985352,
      "learning_rate": 3.849688575211836e-07,
      "logits/chosen": 0.004134657327085733,
      "logits/rejected": 0.02180260606110096,
      "logps/chosen": -234.44007873535156,
      "logps/rejected": -232.92913818359375,
      "loss": 0.7682,
      "rewards/accuracies": 0.5073437690734863,
      "rewards/chosen": -0.7022367119789124,
      "rewards/margins": 0.0076367598958313465,
      "rewards/rejected": -0.7098734974861145,
      "step": 800
    },
    {
      "epoch": 0.3582089552238806,
      "grad_norm": 9.765380859375,
      "learning_rate": 3.576653995009154e-07,
      "logits/chosen": 0.018308693543076515,
      "logits/rejected": 0.04190211370587349,
      "logps/chosen": -224.4905242919922,
      "logps/rejected": -228.26222229003906,
      "loss": 0.7467,
      "rewards/accuracies": 0.5228124856948853,
      "rewards/chosen": -0.7309367656707764,
      "rewards/margins": 0.04116936773061752,
      "rewards/rejected": -0.7721061110496521,
      "step": 900
    },
    {
      "epoch": 0.39800995024875624,
      "grad_norm": 6.442622184753418,
      "learning_rate": 3.286801563968721e-07,
      "logits/chosen": -0.018274417147040367,
      "logits/rejected": -0.015018883161246777,
      "logps/chosen": -230.45724487304688,
      "logps/rejected": -229.37107849121094,
      "loss": 0.7517,
      "rewards/accuracies": 0.5123437643051147,
      "rewards/chosen": -0.738405704498291,
      "rewards/margins": 0.026958582922816277,
      "rewards/rejected": -0.765364408493042,
      "step": 1000
    },
    {
      "epoch": 0.43781094527363185,
      "grad_norm": 3.437434434890747,
      "learning_rate": 2.9846589158269034e-07,
      "logits/chosen": -0.03212662786245346,
      "logits/rejected": -0.016695672646164894,
      "logps/chosen": -233.62451171875,
      "logps/rejected": -230.5555419921875,
      "loss": 0.7495,
      "rewards/accuracies": 0.5182812213897705,
      "rewards/chosen": -0.6797711253166199,
      "rewards/margins": 0.029431035742163658,
      "rewards/rejected": -0.7092021107673645,
      "step": 1100
    },
    {
      "epoch": 0.47761194029850745,
      "grad_norm": 9.21783447265625,
      "learning_rate": 2.674945663394993e-07,
      "logits/chosen": -0.022503716871142387,
      "logits/rejected": -0.0002497506211511791,
      "logps/chosen": -224.26544189453125,
      "logps/rejected": -220.6869354248047,
      "loss": 0.7452,
      "rewards/accuracies": 0.5198437571525574,
      "rewards/chosen": -0.6790177822113037,
      "rewards/margins": 0.027901820838451385,
      "rewards/rejected": -0.7069195508956909,
      "step": 1200
    },
    {
      "epoch": 0.5174129353233831,
      "grad_norm": 5.184940814971924,
      "learning_rate": 2.3624996759476285e-07,
      "logits/chosen": -0.02832232229411602,
      "logits/rejected": 0.001360323396511376,
      "logps/chosen": -225.9873504638672,
      "logps/rejected": -224.61253356933594,
      "loss": 0.7448,
      "rewards/accuracies": 0.5210937261581421,
      "rewards/chosen": -0.6495236158370972,
      "rewards/margins": 0.03301383554935455,
      "rewards/rejected": -0.6825373768806458,
      "step": 1300
    },
    {
      "epoch": 0.5572139303482587,
      "grad_norm": 4.051429271697998,
      "learning_rate": 2.0522015093886614e-07,
      "logits/chosen": -0.07805901765823364,
      "logits/rejected": -0.054497309029102325,
      "logps/chosen": -227.91299438476562,
      "logps/rejected": -231.43540954589844,
      "loss": 0.7352,
      "rewards/accuracies": 0.5206249952316284,
      "rewards/chosen": -0.6832866072654724,
      "rewards/margins": 0.05871019884943962,
      "rewards/rejected": -0.7419967651367188,
      "step": 1400
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 4.9932379722595215,
      "learning_rate": 1.7488981696314154e-07,
      "logits/chosen": -0.07948292791843414,
      "logits/rejected": -0.04982204735279083,
      "logps/chosen": -226.94752502441406,
      "logps/rejected": -225.03204345703125,
      "loss": 0.7349,
      "rewards/accuracies": 0.5214062333106995,
      "rewards/chosen": -0.6902438998222351,
      "rewards/margins": 0.03936518728733063,
      "rewards/rejected": -0.7296090722084045,
      "step": 1500
    },
    {
      "epoch": 0.6368159203980099,
      "grad_norm": 8.183513641357422,
      "learning_rate": 1.4573274000458839e-07,
      "logits/chosen": -0.08282151073217392,
      "logits/rejected": -0.06661933660507202,
      "logps/chosen": -227.28172302246094,
      "logps/rejected": -227.94952392578125,
      "loss": 0.7374,
      "rewards/accuracies": 0.5310937762260437,
      "rewards/chosen": -0.6844578385353088,
      "rewards/margins": 0.03974698856472969,
      "rewards/rejected": -0.7242047190666199,
      "step": 1600
    },
    {
      "epoch": 0.6766169154228856,
      "grad_norm": 5.1135573387146,
      "learning_rate": 1.1820436756391414e-07,
      "logits/chosen": -0.1615159809589386,
      "logits/rejected": -0.13768981397151947,
      "logps/chosen": -230.94302368164062,
      "logps/rejected": -229.08021545410156,
      "loss": 0.7359,
      "rewards/accuracies": 0.5257812738418579,
      "rewards/chosen": -0.654849648475647,
      "rewards/margins": 0.04666764661669731,
      "rewards/rejected": -0.7015172839164734,
      "step": 1700
    },
    {
      "epoch": 0.7164179104477612,
      "grad_norm": 4.888497829437256,
      "learning_rate": 9.273470599753375e-08,
      "logits/chosen": -0.07336300611495972,
      "logits/rejected": -0.0592733770608902,
      "logps/chosen": -223.34539794921875,
      "logps/rejected": -226.36752319335938,
      "loss": 0.7341,
      "rewards/accuracies": 0.5315625071525574,
      "rewards/chosen": -0.6613838076591492,
      "rewards/margins": 0.04804745316505432,
      "rewards/rejected": -0.7094312310218811,
      "step": 1800
    },
    {
      "epoch": 0.7562189054726368,
      "grad_norm": 6.096445083618164,
      "learning_rate": 6.972160361242119e-08,
      "logits/chosen": -0.11273051053285599,
      "logits/rejected": -0.09009736031293869,
      "logps/chosen": -223.08956909179688,
      "logps/rejected": -226.54151916503906,
      "loss": 0.7362,
      "rewards/accuracies": 0.5253124833106995,
      "rewards/chosen": -0.6732974052429199,
      "rewards/margins": 0.03796848654747009,
      "rewards/rejected": -0.7112659215927124,
      "step": 1900
    },
    {
      "epoch": 0.7960199004975125,
      "grad_norm": 4.404871940612793,
      "learning_rate": 4.952453608509e-08,
      "logits/chosen": -0.12635847926139832,
      "logits/rejected": -0.09713621437549591,
      "logps/chosen": -230.83767700195312,
      "logps/rejected": -230.70140075683594,
      "loss": 0.7348,
      "rewards/accuracies": 0.5314062237739563,
      "rewards/chosen": -0.6768426299095154,
      "rewards/margins": 0.04782358556985855,
      "rewards/rejected": -0.7246662378311157,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 2512,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}