{
  "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": 77.31810180796927,
      "learning_rate": 1.3888888888888888e-07,
      "logits/chosen": -2.8591513633728027,
      "logits/rejected": -2.6428322792053223,
      "logps/chosen": -390.54095458984375,
      "logps/rejected": -607.8343505859375,
      "loss": 1.3778,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.002188617829233408,
      "rewards/margins": 0.013455559499561787,
      "rewards/rejected": -0.011266940273344517,
      "step": 10
    },
    {
      "epoch": 0.056818181818181816,
      "grad_norm": 38.318260170576636,
      "learning_rate": 2.7777777777777776e-07,
      "logits/chosen": -2.842257261276245,
      "logits/rejected": -2.6946189403533936,
      "logps/chosen": -328.62799072265625,
      "logps/rejected": -775.0516357421875,
      "loss": 1.0648,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.08517134189605713,
      "rewards/margins": 0.6612905859947205,
      "rewards/rejected": -0.5761191844940186,
      "step": 20
    },
    {
      "epoch": 0.08522727272727272,
      "grad_norm": 10.337672358301827,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.856661319732666,
      "logits/rejected": -2.6734800338745117,
      "logps/chosen": -295.8897399902344,
      "logps/rejected": -1121.2652587890625,
      "loss": 0.4072,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.36392277479171753,
      "rewards/margins": 4.797031402587891,
      "rewards/rejected": -4.433108329772949,
      "step": 30
    },
    {
      "epoch": 0.11363636363636363,
      "grad_norm": 3.166840866383729,
      "learning_rate": 4.998023493068254e-07,
      "logits/chosen": -2.8248767852783203,
      "logits/rejected": -2.627469539642334,
      "logps/chosen": -318.3464050292969,
      "logps/rejected": -2185.42578125,
      "loss": 0.1013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.28409063816070557,
      "rewards/margins": 15.0593900680542,
      "rewards/rejected": -14.775299072265625,
      "step": 40
    },
    {
      "epoch": 0.14204545454545456,
      "grad_norm": 1.0472115502346335,
      "learning_rate": 4.975823666181255e-07,
      "logits/chosen": -2.8039865493774414,
      "logits/rejected": -2.5030531883239746,
      "logps/chosen": -466.212890625,
      "logps/rejected": -4635.23095703125,
      "loss": 0.0135,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0357766151428223,
      "rewards/margins": 38.77712631225586,
      "rewards/rejected": -39.81290054321289,
      "step": 50
    },
    {
      "epoch": 0.14204545454545456,
      "eval_logits/chosen": -2.9291088581085205,
      "eval_logits/rejected": -2.438577175140381,
      "eval_logps/chosen": -499.10443115234375,
      "eval_logps/rejected": -5381.90283203125,
      "eval_loss": 0.00992405042052269,
      "eval_rewards/accuracies": 0.9959677457809448,
      "eval_rewards/chosen": -1.321226954460144,
      "eval_rewards/margins": 46.089908599853516,
      "eval_rewards/rejected": -47.41114044189453,
      "eval_runtime": 194.7089,
      "eval_samples_per_second": 20.056,
      "eval_steps_per_second": 0.318,
      "step": 50
    },
    {
      "epoch": 0.17045454545454544,
      "grad_norm": 0.9673005612652019,
      "learning_rate": 4.929173350101024e-07,
      "logits/chosen": -2.9523978233337402,
      "logits/rejected": -2.3117947578430176,
      "logps/chosen": -503.43365478515625,
      "logps/rejected": -5926.720703125,
      "loss": 0.0114,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -1.4353859424591064,
      "rewards/margins": 51.40327835083008,
      "rewards/rejected": -52.83866500854492,
      "step": 60
    },
    {
      "epoch": 0.19886363636363635,
      "grad_norm": 0.3988175169368107,
      "learning_rate": 4.858533249305336e-07,
      "logits/chosen": -2.878004550933838,
      "logits/rejected": -1.9669125080108643,
      "logps/chosen": -545.6673583984375,
      "logps/rejected": -6498.20263671875,
      "loss": 0.0065,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.6650670766830444,
      "rewards/margins": 57.189605712890625,
      "rewards/rejected": -58.85467529296875,
      "step": 70
    },
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 13.315517602902379,
      "learning_rate": 4.764600984163808e-07,
      "logits/chosen": -2.780555248260498,
      "logits/rejected": -1.0796103477478027,
      "logps/chosen": -584.189697265625,
      "logps/rejected": -7447.61572265625,
      "loss": 0.0056,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.7786884307861328,
      "rewards/margins": 65.96015167236328,
      "rewards/rejected": -67.73884582519531,
      "step": 80
    },
    {
      "epoch": 0.2556818181818182,
      "grad_norm": 0.060351047390217884,
      "learning_rate": 4.6483042014491527e-07,
      "logits/chosen": -2.8976826667785645,
      "logits/rejected": -1.835821509361267,
      "logps/chosen": -552.1006469726562,
      "logps/rejected": -6640.58984375,
      "loss": 0.0148,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -1.8998725414276123,
      "rewards/margins": 58.132171630859375,
      "rewards/rejected": -60.03204345703125,
      "step": 90
    },
    {
      "epoch": 0.2840909090909091,
      "grad_norm": 1.627678131457394,
      "learning_rate": 4.510791413176912e-07,
      "logits/chosen": -2.7810730934143066,
      "logits/rejected": -0.7030321359634399,
      "logps/chosen": -615.0423583984375,
      "logps/rejected": -8701.7958984375,
      "loss": 0.0109,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.455723285675049,
      "rewards/margins": 77.5044174194336,
      "rewards/rejected": -79.96015167236328,
      "step": 100
    },
    {
      "epoch": 0.2840909090909091,
      "eval_logits/chosen": -2.5427753925323486,
      "eval_logits/rejected": -0.03400120511651039,
      "eval_logps/chosen": -626.6861572265625,
      "eval_logps/rejected": -8739.6259765625,
      "eval_loss": 0.0025400689337402582,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.5970447063446045,
      "eval_rewards/margins": 78.39134216308594,
      "eval_rewards/rejected": -80.9883804321289,
      "eval_runtime": 194.2924,
      "eval_samples_per_second": 20.099,
      "eval_steps_per_second": 0.319,
      "step": 100
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.04122796046055686,
      "learning_rate": 4.353420654246546e-07,
      "logits/chosen": -2.4864227771759033,
      "logits/rejected": -0.10299022495746613,
      "logps/chosen": -605.4061279296875,
      "logps/rejected": -7677.7255859375,
      "loss": 0.0083,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.29984974861145,
      "rewards/margins": 68.48472595214844,
      "rewards/rejected": -70.78457641601562,
      "step": 110
    },
    {
      "epoch": 0.3409090909090909,
      "grad_norm": 0.09070738570484652,
      "learning_rate": 4.177746070897592e-07,
      "logits/chosen": -2.3363277912139893,
      "logits/rejected": 0.5452026724815369,
      "logps/chosen": -594.1629638671875,
      "logps/rejected": -7349.20068359375,
      "loss": 0.0063,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.1264612674713135,
      "rewards/margins": 64.9932632446289,
      "rewards/rejected": -67.1197280883789,
      "step": 120
    },
    {
      "epoch": 0.3693181818181818,
      "grad_norm": 0.4807602533089145,
      "learning_rate": 3.9855025724292763e-07,
      "logits/chosen": -2.243814468383789,
      "logits/rejected": 1.6017320156097412,
      "logps/chosen": -609.6871337890625,
      "logps/rejected": -7587.91552734375,
      "loss": 0.0019,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.392354965209961,
      "rewards/margins": 67.08141326904297,
      "rewards/rejected": -69.47377014160156,
      "step": 130
    },
    {
      "epoch": 0.3977272727272727,
      "grad_norm": 1.856213293772278,
      "learning_rate": 3.7785886977585555e-07,
      "logits/chosen": -1.8541723489761353,
      "logits/rejected": 2.44315505027771,
      "logps/chosen": -582.9074096679688,
      "logps/rejected": -9157.189453125,
      "loss": 0.0035,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.168710231781006,
      "rewards/margins": 82.98329162597656,
      "rewards/rejected": -85.1520004272461,
      "step": 140
    },
    {
      "epoch": 0.42613636363636365,
      "grad_norm": 1.4955330936107978,
      "learning_rate": 3.5590478660213206e-07,
      "logits/chosen": -1.5623462200164795,
      "logits/rejected": 2.7078518867492676,
      "logps/chosen": -628.0105590820312,
      "logps/rejected": -8812.60546875,
      "loss": 0.0017,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.499473810195923,
      "rewards/margins": 78.50679779052734,
      "rewards/rejected": -81.00627136230469,
      "step": 150
    },
    {
      "epoch": 0.42613636363636365,
      "eval_logits/chosen": -1.6475961208343506,
      "eval_logits/rejected": 2.8304052352905273,
      "eval_logps/chosen": -566.4089965820312,
      "eval_logps/rejected": -8356.697265625,
      "eval_loss": 0.0010887953685596585,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.9942734241485596,
      "eval_rewards/margins": 75.1648178100586,
      "eval_rewards/rejected": -77.15908813476562,
      "eval_runtime": 194.5753,
      "eval_samples_per_second": 20.069,
      "eval_steps_per_second": 0.319,
      "step": 150
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 8.762005226870338,
      "learning_rate": 3.3290481963801696e-07,
      "logits/chosen": -2.1097240447998047,
      "logits/rejected": 2.4051706790924072,
      "logps/chosen": -540.6964111328125,
      "logps/rejected": -7950.2998046875,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.705583930015564,
      "rewards/margins": 71.63252258300781,
      "rewards/rejected": -73.33810424804688,
      "step": 160
    },
    {
      "epoch": 0.48295454545454547,
      "grad_norm": 0.030626366308689413,
      "learning_rate": 3.0908610963322626e-07,
      "logits/chosen": -2.4956324100494385,
      "logits/rejected": 1.2054059505462646,
      "logps/chosen": -597.5648803710938,
      "logps/rejected": -8411.56640625,
      "loss": 0.006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.099465847015381,
      "rewards/margins": 74.73308563232422,
      "rewards/rejected": -76.83256530761719,
      "step": 170
    },
    {
      "epoch": 0.5113636363636364,
      "grad_norm": 0.014685643153554214,
      "learning_rate": 2.846838829972671e-07,
      "logits/chosen": -2.5342214107513428,
      "logits/rejected": 1.157962441444397,
      "logps/chosen": -590.8978271484375,
      "logps/rejected": -7980.2666015625,
      "loss": 0.0116,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.2999348640441895,
      "rewards/margins": 71.37288665771484,
      "rewards/rejected": -73.67282104492188,
      "step": 180
    },
    {
      "epoch": 0.5397727272727273,
      "grad_norm": 0.1745056386100847,
      "learning_rate": 2.5993912877423147e-07,
      "logits/chosen": -2.2881453037261963,
      "logits/rejected": 1.435770869255066,
      "logps/chosen": -543.35205078125,
      "logps/rejected": -8919.642578125,
      "loss": 0.0011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9773623943328857,
      "rewards/margins": 80.30296325683594,
      "rewards/rejected": -82.28031921386719,
      "step": 190
    },
    {
      "epoch": 0.5681818181818182,
      "grad_norm": 0.7368778028750317,
      "learning_rate": 2.3509621870754504e-07,
      "logits/chosen": -1.6877973079681396,
      "logits/rejected": 3.008781671524048,
      "logps/chosen": -565.8089599609375,
      "logps/rejected": -8412.712890625,
      "loss": 0.002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0067920684814453,
      "rewards/margins": 75.81190490722656,
      "rewards/rejected": -77.8187026977539,
      "step": 200
    },
    {
      "epoch": 0.5681818181818182,
      "eval_logits/chosen": -1.3436212539672852,
      "eval_logits/rejected": 2.8839986324310303,
      "eval_logps/chosen": -579.902099609375,
      "eval_logps/rejected": -8925.5107421875,
      "eval_loss": 0.000815804407466203,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.1292035579681396,
      "eval_rewards/margins": 80.71800994873047,
      "eval_rewards/rejected": -82.84722137451172,
      "eval_runtime": 194.6381,
      "eval_samples_per_second": 20.063,
      "eval_steps_per_second": 0.319,
      "step": 200
    },
    {
      "epoch": 0.5965909090909091,
      "grad_norm": 0.9148796537082613,
      "learning_rate": 2.1040049389819624e-07,
      "logits/chosen": -1.2771052122116089,
      "logits/rejected": 3.0268094539642334,
      "logps/chosen": -591.0138549804688,
      "logps/rejected": -8609.7705078125,
      "loss": 0.0067,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0037057399749756,
      "rewards/margins": 77.77495574951172,
      "rewards/rejected": -79.77867126464844,
      "step": 210
    },
    {
      "epoch": 0.625,
      "grad_norm": 4.610253822471946,
      "learning_rate": 1.8609584188988133e-07,
      "logits/chosen": -0.9021345376968384,
      "logits/rejected": 2.644864559173584,
      "logps/chosen": -637.8895263671875,
      "logps/rejected": -7473.07666015625,
      "loss": 0.016,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.732889175415039,
      "rewards/margins": 65.50151824951172,
      "rewards/rejected": -68.2343978881836,
      "step": 220
    },
    {
      "epoch": 0.6534090909090909,
      "grad_norm": 80.2393605437428,
      "learning_rate": 1.624222881090439e-07,
      "logits/chosen": -1.08247971534729,
      "logits/rejected": 2.224297046661377,
      "logps/chosen": -652.9135131835938,
      "logps/rejected": -8397.490234375,
      "loss": 0.0294,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.5824902057647705,
      "rewards/margins": 74.53660583496094,
      "rewards/rejected": -77.11909484863281,
      "step": 230
    },
    {
      "epoch": 0.6818181818181818,
      "grad_norm": 0.967663894443262,
      "learning_rate": 1.3961362544602212e-07,
      "logits/chosen": -1.0073425769805908,
      "logits/rejected": 2.469111442565918,
      "logps/chosen": -597.539794921875,
      "logps/rejected": -7158.27490234375,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.2152743339538574,
      "rewards/margins": 63.4267463684082,
      "rewards/rejected": -65.64202117919922,
      "step": 240
    },
    {
      "epoch": 0.7102272727272727,
      "grad_norm": 0.5670683312609751,
      "learning_rate": 1.1789510538684522e-07,
      "logits/chosen": -1.1133817434310913,
      "logits/rejected": 2.4663312435150146,
      "logps/chosen": -563.9259643554688,
      "logps/rejected": -9071.0546875,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.1189236640930176,
      "rewards/margins": 81.29580688476562,
      "rewards/rejected": -83.41471862792969,
      "step": 250
    },
    {
      "epoch": 0.7102272727272727,
      "eval_logits/chosen": -1.153204321861267,
      "eval_logits/rejected": 2.5446887016296387,
      "eval_logps/chosen": -581.1539916992188,
      "eval_logps/rejected": -7875.69921875,
      "eval_loss": 0.0009098726441152394,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.1417224407196045,
      "eval_rewards/margins": 70.20738220214844,
      "eval_rewards/rejected": -72.34910583496094,
      "eval_runtime": 194.9128,
      "eval_samples_per_second": 20.035,
      "eval_steps_per_second": 0.318,
      "step": 250
    },
    {
      "epoch": 0.7386363636363636,
      "grad_norm": 1.0164128695484267,
      "learning_rate": 9.748121349736891e-08,
      "logits/chosen": -1.0999664068222046,
      "logits/rejected": 2.403217315673828,
      "logps/chosen": -622.03173828125,
      "logps/rejected": -8243.462890625,
      "loss": 0.0014,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.3820364475250244,
      "rewards/margins": 73.27131652832031,
      "rewards/rejected": -75.65335845947266,
      "step": 260
    },
    {
      "epoch": 0.7670454545454546,
      "grad_norm": 0.08769848802929946,
      "learning_rate": 7.857355122839673e-08,
      "logits/chosen": -1.1708943843841553,
      "logits/rejected": 2.2550511360168457,
      "logps/chosen": -594.3570556640625,
      "logps/rejected": -7704.54541015625,
      "loss": 0.0016,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.135075092315674,
      "rewards/margins": 68.22078704833984,
      "rewards/rejected": -70.3558578491211,
      "step": 270
    },
    {
      "epoch": 0.7954545454545454,
      "grad_norm": 0.21922033699532475,
      "learning_rate": 6.135884496044244e-08,
      "logits/chosen": -0.8851491212844849,
      "logits/rejected": 2.587214708328247,
      "logps/chosen": -582.6546630859375,
      "logps/rejected": -7667.8701171875,
      "loss": 0.0019,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.034817934036255,
      "rewards/margins": 68.19500732421875,
      "rewards/rejected": -70.22981262207031,
      "step": 280
    },
    {
      "epoch": 0.8238636363636364,
      "grad_norm": 0.1479090204803531,
      "learning_rate": 4.600710195020982e-08,
      "logits/chosen": -0.6345096826553345,
      "logits/rejected": 2.7067768573760986,
      "logps/chosen": -606.1185913085938,
      "logps/rejected": -7632.70068359375,
      "loss": 0.006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.1697540283203125,
      "rewards/margins": 67.74466705322266,
      "rewards/rejected": -69.91442108154297,
      "step": 290
    },
    {
      "epoch": 0.8522727272727273,
      "grad_norm": 1.431809898038739,
      "learning_rate": 3.2669931390104374e-08,
      "logits/chosen": -0.6104838252067566,
      "logits/rejected": 2.62327241897583,
      "logps/chosen": -560.2278442382812,
      "logps/rejected": -8326.173828125,
      "loss": 0.0013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0397324562072754,
      "rewards/margins": 75.01374816894531,
      "rewards/rejected": -77.05347442626953,
      "step": 300
    },
    {
      "epoch": 0.8522727272727273,
      "eval_logits/chosen": -0.43478134274482727,
      "eval_logits/rejected": 2.7312941551208496,
      "eval_logps/chosen": -587.4812622070312,
      "eval_logps/rejected": -8038.0322265625,
      "eval_loss": 0.0010067835683003068,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.2049953937530518,
      "eval_rewards/margins": 71.7674331665039,
      "eval_rewards/rejected": -73.9724349975586,
      "eval_runtime": 193.3426,
      "eval_samples_per_second": 20.197,
      "eval_steps_per_second": 0.321,
      "step": 300
    },
    {
      "epoch": 0.8806818181818182,
      "grad_norm": 0.1560528226318461,
      "learning_rate": 2.147904716149135e-08,
      "logits/chosen": -0.37615886330604553,
      "logits/rejected": 2.69954252243042,
      "logps/chosen": -567.6695556640625,
      "logps/rejected": -8216.876953125,
      "loss": 0.0046,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0560302734375,
      "rewards/margins": 73.10771942138672,
      "rewards/rejected": -75.16374206542969,
      "step": 310
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.024456003641593862,
      "learning_rate": 1.254496706805433e-08,
      "logits/chosen": -0.5904131531715393,
      "logits/rejected": 2.5954599380493164,
      "logps/chosen": -595.1561279296875,
      "logps/rejected": -8192.4501953125,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.124830722808838,
      "rewards/margins": 73.27985382080078,
      "rewards/rejected": -75.40467834472656,
      "step": 320
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.24568296389199945,
      "learning_rate": 5.955921395237318e-09,
      "logits/chosen": -0.48061466217041016,
      "logits/rejected": 2.652150869369507,
      "logps/chosen": -554.1439208984375,
      "logps/rejected": -8231.5478515625,
      "loss": 0.0007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9887583255767822,
      "rewards/margins": 73.84918975830078,
      "rewards/rejected": -75.83795166015625,
      "step": 330
    },
    {
      "epoch": 0.9659090909090909,
      "grad_norm": 0.22303732669521928,
      "learning_rate": 1.7769815745066474e-09,
      "logits/chosen": -0.7278428673744202,
      "logits/rejected": 2.4799537658691406,
      "logps/chosen": -569.5009765625,
      "logps/rejected": -7504.6611328125,
      "loss": 0.0013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.005201816558838,
      "rewards/margins": 66.95375061035156,
      "rewards/rejected": -68.9589614868164,
      "step": 340
    },
    {
      "epoch": 0.9943181818181818,
      "grad_norm": 1.783067144450784,
      "learning_rate": 4.9417557483610875e-11,
      "logits/chosen": -0.47829318046569824,
      "logits/rejected": 2.6286473274230957,
      "logps/chosen": -601.5467529296875,
      "logps/rejected": -8092.9609375,
      "loss": 0.0058,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.272789716720581,
      "rewards/margins": 72.18123626708984,
      "rewards/rejected": -74.45402526855469,
      "step": 350
    },
    {
      "epoch": 0.9943181818181818,
      "eval_logits/chosen": -0.3615128695964813,
      "eval_logits/rejected": 2.7746219635009766,
      "eval_logps/chosen": -587.4920654296875,
      "eval_logps/rejected": -8119.24560546875,
      "eval_loss": 0.0010170801542699337,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.205103874206543,
      "eval_rewards/margins": 72.57946014404297,
      "eval_rewards/rejected": -74.78457641601562,
      "eval_runtime": 194.3785,
      "eval_samples_per_second": 20.09,
      "eval_steps_per_second": 0.319,
      "step": 350
    },
    {
      "epoch": 1.0,
      "step": 352,
      "total_flos": 0.0,
      "train_loss": 0.08915322791869006,
      "train_runtime": 10109.537,
      "train_samples_per_second": 4.451,
      "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
}