{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 50,
  "global_step": 352,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.028409090909090908,
      "grad_norm": 90.01349453953418,
      "learning_rate": 1.3888888888888888e-07,
      "logits/chosen": -2.8592569828033447,
      "logits/rejected": -2.642957925796509,
      "logps/chosen": -390.56573486328125,
      "logps/rejected": -607.8802490234375,
      "loss": 1.3762,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": 0.001940618036314845,
      "rewards/margins": 0.013666175305843353,
      "rewards/rejected": -0.011725558899343014,
      "step": 10
    },
    {
      "epoch": 0.056818181818181816,
      "grad_norm": 38.16702665689489,
      "learning_rate": 2.7777777777777776e-07,
      "logits/chosen": -2.842255115509033,
      "logits/rejected": -2.694774627685547,
      "logps/chosen": -328.5555419921875,
      "logps/rejected": -774.9117431640625,
      "loss": 1.0262,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 0.08589581400156021,
      "rewards/margins": 0.660616934299469,
      "rewards/rejected": -0.5747210383415222,
      "step": 20
    },
    {
      "epoch": 0.08522727272727272,
      "grad_norm": 9.474354712422207,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.857499837875366,
      "logits/rejected": -2.675124406814575,
      "logps/chosen": -295.38397216796875,
      "logps/rejected": -1112.8095703125,
      "loss": 0.3585,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.36898040771484375,
      "rewards/margins": 4.717532157897949,
      "rewards/rejected": -4.3485517501831055,
      "step": 30
    },
    {
      "epoch": 0.11363636363636363,
      "grad_norm": 2.844664039016391,
      "learning_rate": 4.998023493068254e-07,
      "logits/chosen": -2.8356635570526123,
      "logits/rejected": -2.645285129547119,
      "logps/chosen": -314.98876953125,
      "logps/rejected": -2130.987060546875,
      "loss": 0.083,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3176669478416443,
      "rewards/margins": 14.54858112335205,
      "rewards/rejected": -14.230911254882812,
      "step": 40
    },
    {
      "epoch": 0.14204545454545456,
      "grad_norm": 0.18581946577481065,
      "learning_rate": 4.975823666181255e-07,
      "logits/chosen": -2.8282175064086914,
      "logits/rejected": -2.554896831512451,
      "logps/chosen": -441.171875,
      "logps/rejected": -4335.5703125,
      "loss": 0.011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.785366415977478,
      "rewards/margins": 36.03093338012695,
      "rewards/rejected": -36.8162956237793,
      "step": 50
    },
    {
      "epoch": 0.14204545454545456,
      "eval_logits/chosen": -2.965341806411743,
      "eval_logits/rejected": -2.5409843921661377,
      "eval_logps/chosen": -459.0153503417969,
      "eval_logps/rejected": -4976.9013671875,
      "eval_loss": 0.009123104624450207,
      "eval_rewards/accuracies": 0.9959677457809448,
      "eval_rewards/chosen": -0.9203360676765442,
      "eval_rewards/margins": 42.44078063964844,
      "eval_rewards/rejected": -43.361122131347656,
      "eval_runtime": 195.4208,
      "eval_samples_per_second": 19.983,
      "eval_steps_per_second": 0.317,
      "step": 50
    },
    {
      "epoch": 0.17045454545454544,
      "grad_norm": 1.8496563592278583,
      "learning_rate": 4.929173350101024e-07,
      "logits/chosen": -3.018855571746826,
      "logits/rejected": -2.489605188369751,
      "logps/chosen": -478.2201232910156,
      "logps/rejected": -5343.02783203125,
      "loss": 0.0066,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -1.1832506656646729,
      "rewards/margins": 45.8184814453125,
      "rewards/rejected": -47.00173568725586,
      "step": 60
    },
    {
      "epoch": 0.19886363636363635,
      "grad_norm": 0.42838497413063575,
      "learning_rate": 4.858533249305336e-07,
      "logits/chosen": -2.8961918354034424,
      "logits/rejected": -1.8984692096710205,
      "logps/chosen": -537.8681640625,
      "logps/rejected": -5729.29150390625,
      "loss": 0.0061,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.587074875831604,
      "rewards/margins": 49.57848358154297,
      "rewards/rejected": -51.165557861328125,
      "step": 70
    },
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 23.183269598091908,
      "learning_rate": 4.764600984163808e-07,
      "logits/chosen": -3.015650510787964,
      "logits/rejected": -1.4822012186050415,
      "logps/chosen": -574.1959838867188,
      "logps/rejected": -7196.5283203125,
      "loss": 0.007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.6787515878677368,
      "rewards/margins": 63.5491943359375,
      "rewards/rejected": -65.22795104980469,
      "step": 80
    },
    {
      "epoch": 0.2556818181818182,
      "grad_norm": 0.2590854602849931,
      "learning_rate": 4.6483042014491527e-07,
      "logits/chosen": -2.880042314529419,
      "logits/rejected": -1.109546422958374,
      "logps/chosen": -574.42529296875,
      "logps/rejected": -6483.9189453125,
      "loss": 0.0194,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.123119831085205,
      "rewards/margins": 56.34221267700195,
      "rewards/rejected": -58.4653434753418,
      "step": 90
    },
    {
      "epoch": 0.2840909090909091,
      "grad_norm": 0.2975727339596996,
      "learning_rate": 4.510791413176912e-07,
      "logits/chosen": -2.7033655643463135,
      "logits/rejected": -0.46150803565979004,
      "logps/chosen": -574.6309814453125,
      "logps/rejected": -7050.77099609375,
      "loss": 0.0029,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0516088008880615,
      "rewards/margins": 61.398292541503906,
      "rewards/rejected": -63.44989776611328,
      "step": 100
    },
    {
      "epoch": 0.2840909090909091,
      "eval_logits/chosen": -2.5256500244140625,
      "eval_logits/rejected": 0.035765551030635834,
      "eval_logps/chosen": -558.7725830078125,
      "eval_logps/rejected": -6818.2158203125,
      "eval_loss": 0.0016336780972778797,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.9179083108901978,
      "eval_rewards/margins": 59.856361389160156,
      "eval_rewards/rejected": -61.77427673339844,
      "eval_runtime": 192.2507,
      "eval_samples_per_second": 20.312,
      "eval_steps_per_second": 0.322,
      "step": 100
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.04984017585029643,
      "learning_rate": 4.353420654246546e-07,
      "logits/chosen": -2.2970995903015137,
      "logits/rejected": 0.2104659080505371,
      "logps/chosen": -568.5081787109375,
      "logps/rejected": -6664.22412109375,
      "loss": 0.0043,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.930869460105896,
      "rewards/margins": 58.71869659423828,
      "rewards/rejected": -60.649559020996094,
      "step": 110
    },
    {
      "epoch": 0.3409090909090909,
      "grad_norm": 0.03797940797804543,
      "learning_rate": 4.177746070897592e-07,
      "logits/chosen": -1.7653045654296875,
      "logits/rejected": 1.1280748844146729,
      "logps/chosen": -600.4093017578125,
      "logps/rejected": -7361.03662109375,
      "loss": 0.0084,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.188924789428711,
      "rewards/margins": 65.04915618896484,
      "rewards/rejected": -67.23808288574219,
      "step": 120
    },
    {
      "epoch": 0.3693181818181818,
      "grad_norm": 0.47736957873502217,
      "learning_rate": 3.9855025724292763e-07,
      "logits/chosen": -2.10666823387146,
      "logits/rejected": 1.7087266445159912,
      "logps/chosen": -594.9124755859375,
      "logps/rejected": -7779.0634765625,
      "loss": 0.003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.2446084022521973,
      "rewards/margins": 69.14064025878906,
      "rewards/rejected": -71.38525390625,
      "step": 130
    },
    {
      "epoch": 0.3977272727272727,
      "grad_norm": 1.020605288874105,
      "learning_rate": 3.7785886977585555e-07,
      "logits/chosen": -2.2910099029541016,
      "logits/rejected": 1.9655479192733765,
      "logps/chosen": -569.71533203125,
      "logps/rejected": -8695.2900390625,
      "loss": 0.0045,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0367894172668457,
      "rewards/margins": 78.4962158203125,
      "rewards/rejected": -80.53299713134766,
      "step": 140
    },
    {
      "epoch": 0.42613636363636365,
      "grad_norm": 0.5904857854916888,
      "learning_rate": 3.5590478660213206e-07,
      "logits/chosen": -2.2953405380249023,
      "logits/rejected": 1.9216325283050537,
      "logps/chosen": -595.8374633789062,
      "logps/rejected": -8074.73974609375,
      "loss": 0.0023,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.1777424812316895,
      "rewards/margins": 71.44985961914062,
      "rewards/rejected": -73.62760925292969,
      "step": 150
    },
    {
      "epoch": 0.42613636363636365,
      "eval_logits/chosen": -2.4309346675872803,
      "eval_logits/rejected": 1.8614341020584106,
      "eval_logps/chosen": -558.0396728515625,
      "eval_logps/rejected": -7546.6005859375,
      "eval_loss": 0.001163232373073697,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.9105795621871948,
      "eval_rewards/margins": 67.14754486083984,
      "eval_rewards/rejected": -69.05812072753906,
      "eval_runtime": 193.8607,
      "eval_samples_per_second": 20.143,
      "eval_steps_per_second": 0.32,
      "step": 150
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 5.391869442805235,
      "learning_rate": 3.3290481963801696e-07,
      "logits/chosen": -2.435987949371338,
      "logits/rejected": 2.048819065093994,
      "logps/chosen": -532.9867553710938,
      "logps/rejected": -7208.0419921875,
      "loss": 0.0007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.628488540649414,
      "rewards/margins": 64.28703308105469,
      "rewards/rejected": -65.91552734375,
      "step": 160
    },
    {
      "epoch": 0.48295454545454547,
      "grad_norm": 0.031936510635288025,
      "learning_rate": 3.0908610963322626e-07,
      "logits/chosen": -2.173886775970459,
      "logits/rejected": 2.0431816577911377,
      "logps/chosen": -577.22314453125,
      "logps/rejected": -7725.6796875,
      "loss": 0.0074,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.896048903465271,
      "rewards/margins": 68.07765197753906,
      "rewards/rejected": -69.97369384765625,
      "step": 170
    },
    {
      "epoch": 0.5113636363636364,
      "grad_norm": 0.018094668159337083,
      "learning_rate": 2.846838829972671e-07,
      "logits/chosen": -1.9897123575210571,
      "logits/rejected": 2.31811785697937,
      "logps/chosen": -567.0270385742188,
      "logps/rejected": -7191.4697265625,
      "loss": 0.0021,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.061227798461914,
      "rewards/margins": 63.72364044189453,
      "rewards/rejected": -65.78486633300781,
      "step": 180
    },
    {
      "epoch": 0.5397727272727273,
      "grad_norm": 0.3615128328570267,
      "learning_rate": 2.5993912877423147e-07,
      "logits/chosen": -1.5569745302200317,
      "logits/rejected": 1.8910541534423828,
      "logps/chosen": -519.6112670898438,
      "logps/rejected": -8060.75390625,
      "loss": 0.0013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.7399555444717407,
      "rewards/margins": 71.95148468017578,
      "rewards/rejected": -73.69144439697266,
      "step": 190
    },
    {
      "epoch": 0.5681818181818182,
      "grad_norm": 1.460001191569848,
      "learning_rate": 2.3509621870754504e-07,
      "logits/chosen": -1.1832640171051025,
      "logits/rejected": 2.8146424293518066,
      "logps/chosen": -540.25,
      "logps/rejected": -7328.6376953125,
      "loss": 0.0037,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.751201868057251,
      "rewards/margins": 65.22674560546875,
      "rewards/rejected": -66.97795104980469,
      "step": 200
    },
    {
      "epoch": 0.5681818181818182,
      "eval_logits/chosen": -1.0991721153259277,
      "eval_logits/rejected": 2.4905478954315186,
      "eval_logps/chosen": -560.5505981445312,
      "eval_logps/rejected": -7817.95947265625,
      "eval_loss": 0.0009270149166695774,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.9356889724731445,
      "eval_rewards/margins": 69.83601379394531,
      "eval_rewards/rejected": -71.7717056274414,
      "eval_runtime": 195.666,
      "eval_samples_per_second": 19.957,
      "eval_steps_per_second": 0.317,
      "step": 200
    },
    {
      "epoch": 0.5965909090909091,
      "grad_norm": 0.6173080616307016,
      "learning_rate": 2.1040049389819624e-07,
      "logits/chosen": -1.0591375827789307,
      "logits/rejected": 2.6843762397766113,
      "logps/chosen": -585.5335083007812,
      "logps/rejected": -7571.32666015625,
      "loss": 0.0065,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9489023685455322,
      "rewards/margins": 67.44532775878906,
      "rewards/rejected": -69.39422607421875,
      "step": 210
    },
    {
      "epoch": 0.625,
      "grad_norm": 3.3870532652701733,
      "learning_rate": 1.8609584188988133e-07,
      "logits/chosen": -0.8726997375488281,
      "logits/rejected": 2.3560879230499268,
      "logps/chosen": -635.2283935546875,
      "logps/rejected": -6522.76708984375,
      "loss": 0.0054,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.706278085708618,
      "rewards/margins": 56.02503204345703,
      "rewards/rejected": -58.73130416870117,
      "step": 220
    },
    {
      "epoch": 0.6534090909090909,
      "grad_norm": 84.74544834753664,
      "learning_rate": 1.624222881090439e-07,
      "logits/chosen": -1.068495512008667,
      "logits/rejected": 2.1958107948303223,
      "logps/chosen": -655.9327392578125,
      "logps/rejected": -7632.9365234375,
      "loss": 0.0224,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.6126813888549805,
      "rewards/margins": 66.86088562011719,
      "rewards/rejected": -69.47355651855469,
      "step": 230
    },
    {
      "epoch": 0.6818181818181818,
      "grad_norm": 0.655326329468716,
      "learning_rate": 1.3961362544602212e-07,
      "logits/chosen": -0.8508334159851074,
      "logits/rejected": 2.6375107765197754,
      "logps/chosen": -604.7547607421875,
      "logps/rejected": -6603.0595703125,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.287423849105835,
      "rewards/margins": 57.80244827270508,
      "rewards/rejected": -60.08986282348633,
      "step": 240
    },
    {
      "epoch": 0.7102272727272727,
      "grad_norm": 0.8566182395838629,
      "learning_rate": 1.1789510538684522e-07,
      "logits/chosen": -0.9583337903022766,
      "logits/rejected": 2.481625556945801,
      "logps/chosen": -564.140869140625,
      "logps/rejected": -8373.404296875,
      "loss": 0.0015,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.12107253074646,
      "rewards/margins": 74.31715393066406,
      "rewards/rejected": -76.43821716308594,
      "step": 250
    },
    {
      "epoch": 0.7102272727272727,
      "eval_logits/chosen": -0.9364973902702332,
      "eval_logits/rejected": 2.6537041664123535,
      "eval_logps/chosen": -579.6390991210938,
      "eval_logps/rejected": -7301.04736328125,
      "eval_loss": 0.0008862165850587189,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.126574754714966,
      "eval_rewards/margins": 64.47601318359375,
      "eval_rewards/rejected": -66.60258483886719,
      "eval_runtime": 193.2705,
      "eval_samples_per_second": 20.205,
      "eval_steps_per_second": 0.321,
      "step": 250
    },
    {
      "epoch": 0.7386363636363636,
      "grad_norm": 0.3669759262567033,
      "learning_rate": 9.748121349736891e-08,
      "logits/chosen": -0.815521240234375,
      "logits/rejected": 2.5474696159362793,
      "logps/chosen": -614.1104736328125,
      "logps/rejected": -7755.7451171875,
      "loss": 0.0009,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.302823305130005,
      "rewards/margins": 68.47334289550781,
      "rewards/rejected": -70.77616882324219,
      "step": 260
    },
    {
      "epoch": 0.7670454545454546,
      "grad_norm": 0.14376906522497312,
      "learning_rate": 7.857355122839673e-08,
      "logits/chosen": -0.8737711906433105,
      "logits/rejected": 2.419881582260132,
      "logps/chosen": -592.2354125976562,
      "logps/rejected": -7475.37109375,
      "loss": 0.0009,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.113858461380005,
      "rewards/margins": 65.95025634765625,
      "rewards/rejected": -68.0641098022461,
      "step": 270
    },
    {
      "epoch": 0.7954545454545454,
      "grad_norm": 0.17275320368872452,
      "learning_rate": 6.135884496044244e-08,
      "logits/chosen": -0.8250367045402527,
      "logits/rejected": 2.680527687072754,
      "logps/chosen": -574.1306762695312,
      "logps/rejected": -7469.40087890625,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9495769739151,
      "rewards/margins": 66.2955322265625,
      "rewards/rejected": -68.24510955810547,
      "step": 280
    },
    {
      "epoch": 0.8238636363636364,
      "grad_norm": 0.036939229346128824,
      "learning_rate": 4.600710195020982e-08,
      "logits/chosen": -0.6854007244110107,
      "logits/rejected": 2.711479902267456,
      "logps/chosen": -595.4219970703125,
      "logps/rejected": -7571.99853515625,
      "loss": 0.0054,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0627870559692383,
      "rewards/margins": 67.24461364746094,
      "rewards/rejected": -69.30741119384766,
      "step": 290
    },
    {
      "epoch": 0.8522727272727273,
      "grad_norm": 1.0005598289334474,
      "learning_rate": 3.2669931390104374e-08,
      "logits/chosen": -0.7712670564651489,
      "logits/rejected": 2.537261724472046,
      "logps/chosen": -546.4788818359375,
      "logps/rejected": -8221.009765625,
      "loss": 0.0017,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9022438526153564,
      "rewards/margins": 74.09959411621094,
      "rewards/rejected": -76.00184631347656,
      "step": 300
    },
    {
      "epoch": 0.8522727272727273,
      "eval_logits/chosen": -0.6079820394515991,
      "eval_logits/rejected": 2.6897590160369873,
      "eval_logps/chosen": -575.6599731445312,
      "eval_logps/rejected": -7868.724609375,
      "eval_loss": 0.0006528676021844149,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.086782693862915,
      "eval_rewards/margins": 70.19257354736328,
      "eval_rewards/rejected": -72.27935791015625,
      "eval_runtime": 193.3061,
      "eval_samples_per_second": 20.201,
      "eval_steps_per_second": 0.321,
      "step": 300
    },
    {
      "epoch": 0.8806818181818182,
      "grad_norm": 0.0660494270632579,
      "learning_rate": 2.147904716149135e-08,
      "logits/chosen": -0.5389717817306519,
      "logits/rejected": 2.602092981338501,
      "logps/chosen": -555.8554077148438,
      "logps/rejected": -7988.98974609375,
      "loss": 0.0032,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9378883838653564,
      "rewards/margins": 70.9469985961914,
      "rewards/rejected": -72.8848876953125,
      "step": 310
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.0488754281750263,
      "learning_rate": 1.254496706805433e-08,
      "logits/chosen": -0.7335542440414429,
      "logits/rejected": 2.5863664150238037,
      "logps/chosen": -590.0091552734375,
      "logps/rejected": -8039.3125,
      "loss": 0.0007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0733609199523926,
      "rewards/margins": 71.79994201660156,
      "rewards/rejected": -73.87330627441406,
      "step": 320
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.2519430705585286,
      "learning_rate": 5.955921395237318e-09,
      "logits/chosen": -0.6468337178230286,
      "logits/rejected": 2.546152353286743,
      "logps/chosen": -548.1500244140625,
      "logps/rejected": -8072.88037109375,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.928818941116333,
      "rewards/margins": 72.32246398925781,
      "rewards/rejected": -74.25126647949219,
      "step": 330
    },
    {
      "epoch": 0.9659090909090909,
      "grad_norm": 0.8725515919272617,
      "learning_rate": 1.7769815745066474e-09,
      "logits/chosen": -0.8968937993049622,
      "logits/rejected": 2.3741860389709473,
      "logps/chosen": -561.4879760742188,
      "logps/rejected": -7435.28125,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9250719547271729,
      "rewards/margins": 66.34008026123047,
      "rewards/rejected": -68.26514434814453,
      "step": 340
    },
    {
      "epoch": 0.9943181818181818,
      "grad_norm": 3.1933280173345713,
      "learning_rate": 4.9417557483610875e-11,
      "logits/chosen": -0.6227324604988098,
      "logits/rejected": 2.5037405490875244,
      "logps/chosen": -589.6453247070312,
      "logps/rejected": -7948.50927734375,
      "loss": 0.0036,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.153775215148926,
      "rewards/margins": 70.8557357788086,
      "rewards/rejected": -73.00951385498047,
      "step": 350
    },
    {
      "epoch": 0.9943181818181818,
      "eval_logits/chosen": -0.5257502198219299,
      "eval_logits/rejected": 2.7151966094970703,
      "eval_logps/chosen": -577.6112670898438,
      "eval_logps/rejected": -7961.02392578125,
      "eval_loss": 0.000620901002548635,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.106295585632324,
      "eval_rewards/margins": 71.09603881835938,
      "eval_rewards/rejected": -73.20234680175781,
      "eval_runtime": 193.4431,
      "eval_samples_per_second": 20.187,
      "eval_steps_per_second": 0.321,
      "step": 350
    },
    {
      "epoch": 1.0,
      "step": 352,
      "total_flos": 0.0,
      "train_loss": 0.08502711300074721,
      "train_runtime": 9968.7204,
      "train_samples_per_second": 4.514,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 10,
  "max_steps": 352,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}