{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.995910949568378,
  "eval_steps": 400,
  "global_step": 137,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03634711494775102,
      "grad_norm": 42.18056900627306,
      "learning_rate": 2.857142857142857e-07,
      "logits/chosen": -10.936590194702148,
      "logits/rejected": -11.232436180114746,
      "logps/chosen": -0.6059076189994812,
      "logps/rejected": -0.5706321001052856,
      "loss": 5.4752,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": -6.05907678604126,
      "rewards/margins": -0.352755606174469,
      "rewards/rejected": -5.7063212394714355,
      "step": 5
    },
    {
      "epoch": 0.07269422989550205,
      "grad_norm": 33.128716826319426,
      "learning_rate": 5.714285714285714e-07,
      "logits/chosen": -10.911026954650879,
      "logits/rejected": -11.181764602661133,
      "logps/chosen": -0.6222352385520935,
      "logps/rejected": -0.5984331965446472,
      "loss": 5.5068,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": -6.222352504730225,
      "rewards/margins": -0.23802027106285095,
      "rewards/rejected": -5.984332084655762,
      "step": 10
    },
    {
      "epoch": 0.10904134484325306,
      "grad_norm": 50.848111590558474,
      "learning_rate": 7.998695344323425e-07,
      "logits/chosen": -10.209379196166992,
      "logits/rejected": -10.468820571899414,
      "logps/chosen": -0.6080077290534973,
      "logps/rejected": -0.5699423551559448,
      "loss": 5.3874,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": -6.080076694488525,
      "rewards/margins": -0.3806529641151428,
      "rewards/rejected": -5.699423789978027,
      "step": 15
    },
    {
      "epoch": 0.1453884597910041,
      "grad_norm": 35.145738479410355,
      "learning_rate": 7.953121695121394e-07,
      "logits/chosen": -9.792947769165039,
      "logits/rejected": -10.061820030212402,
      "logps/chosen": -0.5358205437660217,
      "logps/rejected": -0.5112504363059998,
      "loss": 5.3654,
      "rewards/accuracies": 0.3499999940395355,
      "rewards/chosen": -5.358205318450928,
      "rewards/margins": -0.24570095539093018,
      "rewards/rejected": -5.112504482269287,
      "step": 20
    },
    {
      "epoch": 0.18173557473875512,
      "grad_norm": 35.43905001965967,
      "learning_rate": 7.843163833184991e-07,
      "logits/chosen": -9.215612411499023,
      "logits/rejected": -9.5372314453125,
      "logps/chosen": -0.47783246636390686,
      "logps/rejected": -0.4458232522010803,
      "loss": 5.2677,
      "rewards/accuracies": 0.33125001192092896,
      "rewards/chosen": -4.778324604034424,
      "rewards/margins": -0.32009226083755493,
      "rewards/rejected": -4.4582319259643555,
      "step": 25
    },
    {
      "epoch": 0.21808268968650613,
      "grad_norm": 34.872469426364724,
      "learning_rate": 7.670612634414511e-07,
      "logits/chosen": -8.819177627563477,
      "logits/rejected": -8.965360641479492,
      "logps/chosen": -0.49288082122802734,
      "logps/rejected": -0.4752427041530609,
      "loss": 5.1863,
      "rewards/accuracies": 0.34375,
      "rewards/chosen": -4.928807735443115,
      "rewards/margins": -0.17638036608695984,
      "rewards/rejected": -4.752427577972412,
      "step": 30
    },
    {
      "epoch": 0.25442980463425713,
      "grad_norm": 41.683131083762184,
      "learning_rate": 7.438278427948805e-07,
      "logits/chosen": -8.983338356018066,
      "logits/rejected": -9.358332633972168,
      "logps/chosen": -0.45715832710266113,
      "logps/rejected": -0.4428199231624603,
      "loss": 5.1479,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": -4.571582794189453,
      "rewards/margins": -0.14338405430316925,
      "rewards/rejected": -4.428199291229248,
      "step": 35
    },
    {
      "epoch": 0.2907769195820082,
      "grad_norm": 35.49350724475253,
      "learning_rate": 7.149945224533862e-07,
      "logits/chosen": -8.473846435546875,
      "logits/rejected": -8.634288787841797,
      "logps/chosen": -0.49668779969215393,
      "logps/rejected": -0.4822346568107605,
      "loss": 5.1059,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": -4.9668779373168945,
      "rewards/margins": -0.14453084766864777,
      "rewards/rejected": -4.8223466873168945,
      "step": 40
    },
    {
      "epoch": 0.3271240345297592,
      "grad_norm": 35.201815027229166,
      "learning_rate": 6.810309086608129e-07,
      "logits/chosen": -9.485041618347168,
      "logits/rejected": -9.641453742980957,
      "logps/chosen": -0.4795761704444885,
      "logps/rejected": -0.4812702536582947,
      "loss": 5.0687,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": -4.795762062072754,
      "rewards/margins": 0.016940664499998093,
      "rewards/rejected": -4.812702655792236,
      "step": 45
    },
    {
      "epoch": 0.36347114947751025,
      "grad_norm": 41.62812201333068,
      "learning_rate": 6.424901643866552e-07,
      "logits/chosen": -9.312422752380371,
      "logits/rejected": -9.527371406555176,
      "logps/chosen": -0.5089246034622192,
      "logps/rejected": -0.5194448232650757,
      "loss": 5.0107,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": -5.0892462730407715,
      "rewards/margins": 0.10520193725824356,
      "rewards/rejected": -5.194448471069336,
      "step": 50
    },
    {
      "epoch": 0.39981826442526125,
      "grad_norm": 34.69469413711179,
      "learning_rate": 6e-07,
      "logits/chosen": -9.82011604309082,
      "logits/rejected": -9.918914794921875,
      "logps/chosen": -0.5091612935066223,
      "logps/rejected": -0.5071789026260376,
      "loss": 4.9611,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -5.091612815856934,
      "rewards/margins": -0.019824281334877014,
      "rewards/rejected": -5.071788311004639,
      "step": 55
    },
    {
      "epoch": 0.43616537937301225,
      "grad_norm": 39.20092819974629,
      "learning_rate": 5.542524497952543e-07,
      "logits/chosen": -10.392576217651367,
      "logits/rejected": -10.515856742858887,
      "logps/chosen": -0.49750223755836487,
      "logps/rejected": -0.5498560070991516,
      "loss": 4.7725,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": -4.975022315979004,
      "rewards/margins": 0.5235382318496704,
      "rewards/rejected": -5.498560428619385,
      "step": 60
    },
    {
      "epoch": 0.4725124943207633,
      "grad_norm": 41.76932417294435,
      "learning_rate": 5.059926008786647e-07,
      "logits/chosen": -9.944269180297852,
      "logits/rejected": -10.23642635345459,
      "logps/chosen": -0.5527420043945312,
      "logps/rejected": -0.6039702296257019,
      "loss": 4.9196,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -5.5274200439453125,
      "rewards/margins": 0.5122829675674438,
      "rewards/rejected": -6.039702415466309,
      "step": 65
    },
    {
      "epoch": 0.5088596092685143,
      "grad_norm": 42.78642756929462,
      "learning_rate": 4.5600645798745166e-07,
      "logits/chosen": -10.92486572265625,
      "logits/rejected": -11.130925178527832,
      "logps/chosen": -0.5507891774177551,
      "logps/rejected": -0.5964936017990112,
      "loss": 4.7685,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -5.50789213180542,
      "rewards/margins": 0.4570443034172058,
      "rewards/rejected": -5.964935779571533,
      "step": 70
    },
    {
      "epoch": 0.5452067242162654,
      "grad_norm": 43.736655032116566,
      "learning_rate": 4.051081418863895e-07,
      "logits/chosen": -10.675444602966309,
      "logits/rejected": -10.829668045043945,
      "logps/chosen": -0.6147274374961853,
      "logps/rejected": -0.6446117758750916,
      "loss": 4.8607,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -6.147274494171143,
      "rewards/margins": 0.29884377121925354,
      "rewards/rejected": -6.446118354797363,
      "step": 75
    },
    {
      "epoch": 0.5815538391640164,
      "grad_norm": 50.47780593376421,
      "learning_rate": 3.541266298406398e-07,
      "logits/chosen": -11.719624519348145,
      "logits/rejected": -11.85960578918457,
      "logps/chosen": -0.6017196774482727,
      "logps/rejected": -0.6400548815727234,
      "loss": 4.7166,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -6.017197608947754,
      "rewards/margins": 0.38335245847702026,
      "rewards/rejected": -6.400549411773682,
      "step": 80
    },
    {
      "epoch": 0.6179009541117674,
      "grad_norm": 49.58320001105474,
      "learning_rate": 3.0389225412181565e-07,
      "logits/chosen": -11.57774543762207,
      "logits/rejected": -11.787096977233887,
      "logps/chosen": -0.639508843421936,
      "logps/rejected": -0.7087122797966003,
      "loss": 4.6635,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -6.395088195800781,
      "rewards/margins": 0.6920341849327087,
      "rewards/rejected": -7.087122917175293,
      "step": 85
    },
    {
      "epoch": 0.6542480690595184,
      "grad_norm": 60.72521586951035,
      "learning_rate": 2.5522317844515273e-07,
      "logits/chosen": -11.539664268493652,
      "logits/rejected": -11.73334789276123,
      "logps/chosen": -0.6763519644737244,
      "logps/rejected": -0.719671905040741,
      "loss": 4.6877,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -6.763520240783691,
      "rewards/margins": 0.4331996440887451,
      "rewards/rejected": -7.196720123291016,
      "step": 90
    },
    {
      "epoch": 0.6905951840072694,
      "grad_norm": 73.18876407608641,
      "learning_rate": 2.0891207259509476e-07,
      "logits/chosen": -11.891935348510742,
      "logits/rejected": -11.978799819946289,
      "logps/chosen": -0.6321767568588257,
      "logps/rejected": -0.7210510969161987,
      "loss": 4.6661,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -6.321768283843994,
      "rewards/margins": 0.8887429237365723,
      "rewards/rejected": -7.210511207580566,
      "step": 95
    },
    {
      "epoch": 0.7269422989550205,
      "grad_norm": 58.29758563166122,
      "learning_rate": 1.6571320226872206e-07,
      "logits/chosen": -12.378867149353027,
      "logits/rejected": -12.450654983520508,
      "logps/chosen": -0.7091072797775269,
      "logps/rejected": -0.7949134111404419,
      "loss": 4.5318,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -7.091072082519531,
      "rewards/margins": 0.8580614924430847,
      "rewards/rejected": -7.949133396148682,
      "step": 100
    },
    {
      "epoch": 0.7632894139027715,
      "grad_norm": 65.14186021974389,
      "learning_rate": 1.2633014440382787e-07,
      "logits/chosen": -12.120596885681152,
      "logits/rejected": -12.198526382446289,
      "logps/chosen": -0.7462261319160461,
      "logps/rejected": -0.8084705471992493,
      "loss": 4.5075,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -7.462261199951172,
      "rewards/margins": 0.622443437576294,
      "rewards/rejected": -8.084705352783203,
      "step": 105
    },
    {
      "epoch": 0.7996365288505225,
      "grad_norm": 66.21038830050585,
      "learning_rate": 9.14043280712228e-08,
      "logits/chosen": -12.642437934875488,
      "logits/rejected": -12.874890327453613,
      "logps/chosen": -0.7560291290283203,
      "logps/rejected": -0.8245932459831238,
      "loss": 4.6235,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -7.560290336608887,
      "rewards/margins": 0.6856421828269958,
      "rewards/rejected": -8.245932579040527,
      "step": 110
    },
    {
      "epoch": 0.8359836437982735,
      "grad_norm": 58.78661492817544,
      "learning_rate": 6.150458756494239e-08,
      "logits/chosen": -12.5349760055542,
      "logits/rejected": -12.633232116699219,
      "logps/chosen": -0.7407893538475037,
      "logps/rejected": -0.8003439903259277,
      "loss": 4.537,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -7.407894134521484,
      "rewards/margins": 0.5955458879470825,
      "rewards/rejected": -8.003438949584961,
      "step": 115
    },
    {
      "epoch": 0.8723307587460245,
      "grad_norm": 72.85901824273444,
      "learning_rate": 3.711789783843522e-08,
      "logits/chosen": -12.179682731628418,
      "logits/rejected": -12.421493530273438,
      "logps/chosen": -0.7077151536941528,
      "logps/rejected": -0.8436015248298645,
      "loss": 4.41,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -7.077151298522949,
      "rewards/margins": 1.358863115310669,
      "rewards/rejected": -8.436014175415039,
      "step": 120
    },
    {
      "epoch": 0.9086778736937755,
      "grad_norm": 81.44112780212329,
      "learning_rate": 1.8641443178027784e-08,
      "logits/chosen": -12.844386100769043,
      "logits/rejected": -13.204521179199219,
      "logps/chosen": -0.7584520578384399,
      "logps/rejected": -0.8550151586532593,
      "loss": 4.4737,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -7.584519863128662,
      "rewards/margins": 0.965630829334259,
      "rewards/rejected": -8.550151824951172,
      "step": 125
    },
    {
      "epoch": 0.9450249886415266,
      "grad_norm": 63.15208705033616,
      "learning_rate": 6.376148290617145e-09,
      "logits/chosen": -13.006174087524414,
      "logits/rejected": -13.145883560180664,
      "logps/chosen": -0.7715423703193665,
      "logps/rejected": -0.8375405073165894,
      "loss": 4.6746,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -7.715423583984375,
      "rewards/margins": 0.6599816679954529,
      "rewards/rejected": -8.375406265258789,
      "step": 130
    },
    {
      "epoch": 0.9813721035892776,
      "grad_norm": 69.30502594164692,
      "learning_rate": 5.217771643080127e-10,
      "logits/chosen": -12.759674072265625,
      "logits/rejected": -12.864163398742676,
      "logps/chosen": -0.7980765700340271,
      "logps/rejected": -0.8657040596008301,
      "loss": 4.3661,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -7.980767250061035,
      "rewards/margins": 0.6762741804122925,
      "rewards/rejected": -8.6570405960083,
      "step": 135
    },
    {
      "epoch": 0.995910949568378,
      "step": 137,
      "total_flos": 0.0,
      "train_loss": 4.865393144370866,
      "train_runtime": 3024.5765,
      "train_samples_per_second": 5.821,
      "train_steps_per_second": 0.045
    }
  ],
  "logging_steps": 5,
  "max_steps": 137,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}