{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9932523616734144,
  "eval_steps": 400,
  "global_step": 115,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00863697705802969,
      "grad_norm": 4.147310027334897,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -1.0488052368164062,
      "logits/rejected": -0.5750762224197388,
      "logps/chosen": -273.85137939453125,
      "logps/rejected": -288.77301025390625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.043184885290148446,
      "grad_norm": 4.588427612553101,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": -1.007105827331543,
      "logits/rejected": -0.7435885667800903,
      "logps/chosen": -253.44888305664062,
      "logps/rejected": -266.1280517578125,
      "loss": 0.6932,
      "rewards/accuracies": 0.4140625,
      "rewards/chosen": 0.0007548874709755182,
      "rewards/margins": 0.0008569365600124002,
      "rewards/rejected": -0.0001020491763483733,
      "step": 5
    },
    {
      "epoch": 0.08636977058029689,
      "grad_norm": 4.548746866725734,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -0.9015377759933472,
      "logits/rejected": -0.7045949101448059,
      "logps/chosen": -256.82440185546875,
      "logps/rejected": -265.10888671875,
      "loss": 0.693,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.0008276132866740227,
      "rewards/margins": 5.403275281423703e-05,
      "rewards/rejected": -0.0008816460031084716,
      "step": 10
    },
    {
      "epoch": 0.12955465587044535,
      "grad_norm": 4.335294831627272,
      "learning_rate": 4.989541370516523e-07,
      "logits/chosen": -0.9198816418647766,
      "logits/rejected": -0.6794015169143677,
      "logps/chosen": -249.69180297851562,
      "logps/rejected": -263.43963623046875,
      "loss": 0.6917,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.006573961116373539,
      "rewards/margins": 0.003779920982196927,
      "rewards/rejected": -0.010353881865739822,
      "step": 15
    },
    {
      "epoch": 0.17273954116059378,
      "grad_norm": 4.361096350569026,
      "learning_rate": 4.925944144036026e-07,
      "logits/chosen": -0.966783344745636,
      "logits/rejected": -0.6541803479194641,
      "logps/chosen": -256.15704345703125,
      "logps/rejected": -271.27569580078125,
      "loss": 0.6873,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.01756737381219864,
      "rewards/margins": 0.011632733047008514,
      "rewards/rejected": -0.029200103133916855,
      "step": 20
    },
    {
      "epoch": 0.21592442645074225,
      "grad_norm": 7.667484004542981,
      "learning_rate": 4.806034494187948e-07,
      "logits/chosen": -0.9687163233757019,
      "logits/rejected": -0.6905485391616821,
      "logps/chosen": -263.11126708984375,
      "logps/rejected": -281.65869140625,
      "loss": 0.6805,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.046140603721141815,
      "rewards/margins": 0.02697494998574257,
      "rewards/rejected": -0.07311554253101349,
      "step": 25
    },
    {
      "epoch": 0.2591093117408907,
      "grad_norm": 4.383791095797972,
      "learning_rate": 4.632595833354104e-07,
      "logits/chosen": -0.9939810633659363,
      "logits/rejected": -0.7504791021347046,
      "logps/chosen": -259.5863952636719,
      "logps/rejected": -277.5854187011719,
      "loss": 0.6719,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.09566532075405121,
      "rewards/margins": 0.046156905591487885,
      "rewards/rejected": -0.1418222188949585,
      "step": 30
    },
    {
      "epoch": 0.30229419703103916,
      "grad_norm": 7.675641265735793,
      "learning_rate": 4.409654120384862e-07,
      "logits/chosen": -1.0092413425445557,
      "logits/rejected": -0.7530118227005005,
      "logps/chosen": -287.18853759765625,
      "logps/rejected": -316.9033203125,
      "loss": 0.6416,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.2571231424808502,
      "rewards/margins": 0.14525508880615234,
      "rewards/rejected": -0.4023781716823578,
      "step": 35
    },
    {
      "epoch": 0.34547908232118757,
      "grad_norm": 8.57248295436779,
      "learning_rate": 4.1423844077058456e-07,
      "logits/chosen": -1.3185703754425049,
      "logits/rejected": -0.9748894572257996,
      "logps/chosen": -328.22552490234375,
      "logps/rejected": -359.8807678222656,
      "loss": 0.6414,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.6841151714324951,
      "rewards/margins": 0.19913408160209656,
      "rewards/rejected": -0.8832491636276245,
      "step": 40
    },
    {
      "epoch": 0.38866396761133604,
      "grad_norm": 12.061629079887432,
      "learning_rate": 3.8369907149976854e-07,
      "logits/chosen": -1.420593500137329,
      "logits/rejected": -1.1807670593261719,
      "logps/chosen": -342.5406188964844,
      "logps/rejected": -385.1309509277344,
      "loss": 0.5983,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9875448346138,
      "rewards/margins": 0.29662927985191345,
      "rewards/rejected": -1.2841740846633911,
      "step": 45
    },
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 14.093087795015201,
      "learning_rate": 3.5005620178906946e-07,
      "logits/chosen": -1.455536127090454,
      "logits/rejected": -1.192246675491333,
      "logps/chosen": -368.83203125,
      "logps/rejected": -424.71746826171875,
      "loss": 0.6055,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.1400392055511475,
      "rewards/margins": 0.5143567323684692,
      "rewards/rejected": -1.6543958187103271,
      "step": 50
    },
    {
      "epoch": 0.4750337381916329,
      "grad_norm": 15.447721210624003,
      "learning_rate": 3.1409076945484506e-07,
      "logits/chosen": -1.4618675708770752,
      "logits/rejected": -1.0830767154693604,
      "logps/chosen": -355.78460693359375,
      "logps/rejected": -439.4186096191406,
      "loss": 0.5889,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.087812900543213,
      "rewards/margins": 0.5787354707717896,
      "rewards/rejected": -1.666548490524292,
      "step": 55
    },
    {
      "epoch": 0.5182186234817814,
      "grad_norm": 81.08313417012548,
      "learning_rate": 2.76637624984969e-07,
      "logits/chosen": -1.527682900428772,
      "logits/rejected": -1.2854654788970947,
      "logps/chosen": -379.35748291015625,
      "logps/rejected": -489.2784118652344,
      "loss": 0.6214,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.3896359205245972,
      "rewards/margins": 0.8340662121772766,
      "rewards/rejected": -2.2237021923065186,
      "step": 60
    },
    {
      "epoch": 0.5614035087719298,
      "grad_norm": 17.702377673251867,
      "learning_rate": 2.3856615250480137e-07,
      "logits/chosen": -1.4518334865570068,
      "logits/rejected": -1.2453023195266724,
      "logps/chosen": -411.2533264160156,
      "logps/rejected": -539.5147705078125,
      "loss": 0.5552,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -1.6073110103607178,
      "rewards/margins": 1.1000343561172485,
      "rewards/rejected": -2.707345485687256,
      "step": 65
    },
    {
      "epoch": 0.6045883940620783,
      "grad_norm": 29.87177949584331,
      "learning_rate": 2.0076008912832354e-07,
      "logits/chosen": -1.4434335231781006,
      "logits/rejected": -1.3378995656967163,
      "logps/chosen": -387.21954345703125,
      "logps/rejected": -468.7499084472656,
      "loss": 0.5621,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -1.3989627361297607,
      "rewards/margins": 0.7221517562866211,
      "rewards/rejected": -2.121114492416382,
      "step": 70
    },
    {
      "epoch": 0.6477732793522267,
      "grad_norm": 12.350143755505647,
      "learning_rate": 1.640970111393718e-07,
      "logits/chosen": -1.4322903156280518,
      "logits/rejected": -1.3162221908569336,
      "logps/chosen": -376.51715087890625,
      "logps/rejected": -457.11688232421875,
      "loss": 0.5743,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.2510515451431274,
      "rewards/margins": 0.697724461555481,
      "rewards/rejected": -1.9487760066986084,
      "step": 75
    },
    {
      "epoch": 0.6909581646423751,
      "grad_norm": 13.161201224229147,
      "learning_rate": 1.2942796318163593e-07,
      "logits/chosen": -1.448541283607483,
      "logits/rejected": -1.256659984588623,
      "logps/chosen": -365.4916076660156,
      "logps/rejected": -444.9703674316406,
      "loss": 0.552,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -1.1402069330215454,
      "rewards/margins": 0.6963046789169312,
      "rewards/rejected": -1.8365116119384766,
      "step": 80
    },
    {
      "epoch": 0.7341430499325237,
      "grad_norm": 16.80049441603505,
      "learning_rate": 9.755770331648641e-08,
      "logits/chosen": -1.5232871770858765,
      "logits/rejected": -1.2771618366241455,
      "logps/chosen": -405.6431884765625,
      "logps/rejected": -508.34942626953125,
      "loss": 0.5654,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -1.496346116065979,
      "rewards/margins": 0.8162988424301147,
      "rewards/rejected": -2.3126449584960938,
      "step": 85
    },
    {
      "epoch": 0.7773279352226721,
      "grad_norm": 14.621658954691451,
      "learning_rate": 6.92260225118122e-08,
      "logits/chosen": -1.5320792198181152,
      "logits/rejected": -1.258817195892334,
      "logps/chosen": -430.893310546875,
      "logps/rejected": -581.5165405273438,
      "loss": 0.5326,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.643710732460022,
      "rewards/margins": 1.3027892112731934,
      "rewards/rejected": -2.946500062942505,
      "step": 90
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 14.642434820372547,
      "learning_rate": 4.509057218473686e-08,
      "logits/chosen": -1.4839651584625244,
      "logits/rejected": -1.4334385395050049,
      "logps/chosen": -416.0023498535156,
      "logps/rejected": -521.580810546875,
      "loss": 0.5546,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.589343786239624,
      "rewards/margins": 0.9084415435791016,
      "rewards/rejected": -2.4977850914001465,
      "step": 95
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 19.36387744586808,
      "learning_rate": 2.5711598415256496e-08,
      "logits/chosen": -1.5397363901138306,
      "logits/rejected": -1.4642789363861084,
      "logps/chosen": -442.88238525390625,
      "logps/rejected": -596.0775146484375,
      "loss": 0.5146,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.7794828414916992,
      "rewards/margins": 1.3754386901855469,
      "rewards/rejected": -3.154921770095825,
      "step": 100
    },
    {
      "epoch": 0.9068825910931174,
      "grad_norm": 15.226538097583589,
      "learning_rate": 1.1538937189091823e-08,
      "logits/chosen": -1.541649580001831,
      "logits/rejected": -1.3221733570098877,
      "logps/chosen": -427.8173828125,
      "logps/rejected": -551.5006103515625,
      "loss": 0.527,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.7289737462997437,
      "rewards/margins": 1.1109263896942139,
      "rewards/rejected": -2.839900255203247,
      "step": 105
    },
    {
      "epoch": 0.9500674763832658,
      "grad_norm": 19.202027293438395,
      "learning_rate": 2.901572543725972e-09,
      "logits/chosen": -1.5520999431610107,
      "logits/rejected": -1.2729403972625732,
      "logps/chosen": -413.5086364746094,
      "logps/rejected": -557.3928833007812,
      "loss": 0.5247,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.6233694553375244,
      "rewards/margins": 1.2928354740142822,
      "rewards/rejected": -2.9162049293518066,
      "step": 110
    },
    {
      "epoch": 0.9932523616734144,
      "grad_norm": 12.814179367231938,
      "learning_rate": 0.0,
      "logits/chosen": -1.5744760036468506,
      "logits/rejected": -1.363406777381897,
      "logps/chosen": -421.62530517578125,
      "logps/rejected": -535.97216796875,
      "loss": 0.5234,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.6327297687530518,
      "rewards/margins": 1.0214827060699463,
      "rewards/rejected": -2.654212236404419,
      "step": 115
    },
    {
      "epoch": 0.9932523616734144,
      "step": 115,
      "total_flos": 0.0,
      "train_loss": 0.6000239335972329,
      "train_runtime": 5043.4745,
      "train_samples_per_second": 2.938,
      "train_steps_per_second": 0.023
    }
  ],
  "logging_steps": 5,
  "max_steps": 115,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}