{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9985553308292401,
  "eval_steps": 100,
  "global_step": 432,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023114706732158336,
      "grad_norm": 68.13874053955078,
      "learning_rate": 2.2727272727272726e-07,
      "logits/chosen": -0.33626726269721985,
      "logits/rejected": -0.31605297327041626,
      "logps/chosen": -269.3142395019531,
      "logps/rejected": -267.5635681152344,
      "loss": 2.9227,
      "nll_loss": 1.0585803985595703,
      "rewards/accuracies": 0.4906249940395355,
      "rewards/chosen": -26.93142318725586,
      "rewards/margins": -0.17506682872772217,
      "rewards/rejected": -26.756357192993164,
      "step": 10
    },
    {
      "epoch": 0.04622941346431667,
      "grad_norm": 57.248863220214844,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -0.3509574234485626,
      "logits/rejected": -0.3329581320285797,
      "logps/chosen": -260.00225830078125,
      "logps/rejected": -266.528076171875,
      "loss": 2.8595,
      "nll_loss": 0.9751935005187988,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -26.000228881835938,
      "rewards/margins": 0.6525786519050598,
      "rewards/rejected": -26.652807235717773,
      "step": 20
    },
    {
      "epoch": 0.06934412019647501,
      "grad_norm": 57.46038055419922,
      "learning_rate": 6.818181818181817e-07,
      "logits/chosen": -0.38174495100975037,
      "logits/rejected": -0.3690889775753021,
      "logps/chosen": -243.15576171875,
      "logps/rejected": -246.4366455078125,
      "loss": 2.6573,
      "nll_loss": 1.0130800008773804,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -24.315576553344727,
      "rewards/margins": 0.32808613777160645,
      "rewards/rejected": -24.64366340637207,
      "step": 30
    },
    {
      "epoch": 0.09245882692863334,
      "grad_norm": 51.81425476074219,
      "learning_rate": 9.09090909090909e-07,
      "logits/chosen": -0.6905701756477356,
      "logits/rejected": -0.673626184463501,
      "logps/chosen": -202.0510711669922,
      "logps/rejected": -203.85264587402344,
      "loss": 2.3834,
      "nll_loss": 0.8630102872848511,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -20.20510482788086,
      "rewards/margins": 0.18015719950199127,
      "rewards/rejected": -20.385265350341797,
      "step": 40
    },
    {
      "epoch": 0.11557353366079168,
      "grad_norm": 46.2398796081543,
      "learning_rate": 9.845360824742267e-07,
      "logits/chosen": -0.8133252263069153,
      "logits/rejected": -0.7886686325073242,
      "logps/chosen": -176.6295623779297,
      "logps/rejected": -175.64236450195312,
      "loss": 2.1663,
      "nll_loss": 0.46288958191871643,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -17.662960052490234,
      "rewards/margins": -0.09872126579284668,
      "rewards/rejected": -17.56423568725586,
      "step": 50
    },
    {
      "epoch": 0.13868824039295002,
      "grad_norm": 55.56767272949219,
      "learning_rate": 9.587628865979382e-07,
      "logits/chosen": -0.6391716003417969,
      "logits/rejected": -0.6422410607337952,
      "logps/chosen": -158.78402709960938,
      "logps/rejected": -159.15992736816406,
      "loss": 1.9369,
      "nll_loss": 0.4064036011695862,
      "rewards/accuracies": 0.5406249761581421,
      "rewards/chosen": -15.878405570983887,
      "rewards/margins": 0.037588153034448624,
      "rewards/rejected": -15.915992736816406,
      "step": 60
    },
    {
      "epoch": 0.16180294712510834,
      "grad_norm": 53.74827575683594,
      "learning_rate": 9.329896907216495e-07,
      "logits/chosen": -0.4799535274505615,
      "logits/rejected": -0.45562925934791565,
      "logps/chosen": -153.95602416992188,
      "logps/rejected": -156.0488739013672,
      "loss": 1.8829,
      "nll_loss": 0.33092719316482544,
      "rewards/accuracies": 0.5218750238418579,
      "rewards/chosen": -15.395601272583008,
      "rewards/margins": 0.20928561687469482,
      "rewards/rejected": -15.604887008666992,
      "step": 70
    },
    {
      "epoch": 0.1849176538572667,
      "grad_norm": 45.28865432739258,
      "learning_rate": 9.072164948453608e-07,
      "logits/chosen": -0.39702308177948,
      "logits/rejected": -0.3713148832321167,
      "logps/chosen": -158.48983764648438,
      "logps/rejected": -161.58985900878906,
      "loss": 1.7248,
      "nll_loss": 0.2892971634864807,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -15.848983764648438,
      "rewards/margins": 0.31000271439552307,
      "rewards/rejected": -16.158987045288086,
      "step": 80
    },
    {
      "epoch": 0.208032360589425,
      "grad_norm": 47.74916076660156,
      "learning_rate": 8.814432989690721e-07,
      "logits/chosen": -0.39265576004981995,
      "logits/rejected": -0.3789977431297302,
      "logps/chosen": -151.72657775878906,
      "logps/rejected": -159.8419189453125,
      "loss": 1.625,
      "nll_loss": 0.24877457320690155,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -15.17265796661377,
      "rewards/margins": 0.8115337491035461,
      "rewards/rejected": -15.98419189453125,
      "step": 90
    },
    {
      "epoch": 0.23114706732158335,
      "grad_norm": 46.02494430541992,
      "learning_rate": 8.556701030927834e-07,
      "logits/chosen": -0.3637830317020416,
      "logits/rejected": -0.34602683782577515,
      "logps/chosen": -156.66448974609375,
      "logps/rejected": -159.7339630126953,
      "loss": 1.7848,
      "nll_loss": 0.3014821708202362,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -15.666448593139648,
      "rewards/margins": 0.306946337223053,
      "rewards/rejected": -15.973396301269531,
      "step": 100
    },
    {
      "epoch": 0.23114706732158335,
      "eval_logits/chosen": -0.3793767988681793,
      "eval_logits/rejected": -0.35158297419548035,
      "eval_logps/chosen": -153.7521209716797,
      "eval_logps/rejected": -157.6624755859375,
      "eval_loss": 1.6452385187149048,
      "eval_nll_loss": 0.27188077569007874,
      "eval_rewards/accuracies": 0.5804347991943359,
      "eval_rewards/chosen": -15.375213623046875,
      "eval_rewards/margins": 0.39103466272354126,
      "eval_rewards/rejected": -15.766247749328613,
      "eval_runtime": 77.4102,
      "eval_samples_per_second": 23.589,
      "eval_steps_per_second": 1.486,
      "step": 100
    },
    {
      "epoch": 0.2542617740537417,
      "grad_norm": 47.49018096923828,
      "learning_rate": 8.298969072164948e-07,
      "logits/chosen": -0.35000625252723694,
      "logits/rejected": -0.31752458214759827,
      "logps/chosen": -150.97109985351562,
      "logps/rejected": -155.34848022460938,
      "loss": 1.5701,
      "nll_loss": 0.26048144698143005,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -15.0971097946167,
      "rewards/margins": 0.4377376139163971,
      "rewards/rejected": -15.534846305847168,
      "step": 110
    },
    {
      "epoch": 0.27737648078590005,
      "grad_norm": 45.2691535949707,
      "learning_rate": 8.041237113402062e-07,
      "logits/chosen": -0.3519677221775055,
      "logits/rejected": -0.32791006565093994,
      "logps/chosen": -157.8858642578125,
      "logps/rejected": -159.35427856445312,
      "loss": 1.6168,
      "nll_loss": 0.3137766718864441,
      "rewards/accuracies": 0.534375011920929,
      "rewards/chosen": -15.788586616516113,
      "rewards/margins": 0.14684121310710907,
      "rewards/rejected": -15.93542766571045,
      "step": 120
    },
    {
      "epoch": 0.30049118751805837,
      "grad_norm": 57.24121856689453,
      "learning_rate": 7.783505154639175e-07,
      "logits/chosen": -0.42790165543556213,
      "logits/rejected": -0.40469226241111755,
      "logps/chosen": -157.43922424316406,
      "logps/rejected": -165.82485961914062,
      "loss": 1.6817,
      "nll_loss": 0.3016583323478699,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": -15.743922233581543,
      "rewards/margins": 0.8385635614395142,
      "rewards/rejected": -16.58248519897461,
      "step": 130
    },
    {
      "epoch": 0.3236058942502167,
      "grad_norm": 58.36777114868164,
      "learning_rate": 7.525773195876288e-07,
      "logits/chosen": -0.5298252105712891,
      "logits/rejected": -0.5220087170600891,
      "logps/chosen": -148.60704040527344,
      "logps/rejected": -154.1773681640625,
      "loss": 1.6595,
      "nll_loss": 0.2847565710544586,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -14.86070442199707,
      "rewards/margins": 0.5570319294929504,
      "rewards/rejected": -15.41773796081543,
      "step": 140
    },
    {
      "epoch": 0.34672060098237506,
      "grad_norm": 68.2182846069336,
      "learning_rate": 7.268041237113402e-07,
      "logits/chosen": -0.5616439580917358,
      "logits/rejected": -0.5490089654922485,
      "logps/chosen": -149.12344360351562,
      "logps/rejected": -154.98716735839844,
      "loss": 1.7257,
      "nll_loss": 0.2902756631374359,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -14.912343978881836,
      "rewards/margins": 0.5863727331161499,
      "rewards/rejected": -15.49871826171875,
      "step": 150
    },
    {
      "epoch": 0.3698353077145334,
      "grad_norm": 39.74797821044922,
      "learning_rate": 7.010309278350515e-07,
      "logits/chosen": -0.5092633962631226,
      "logits/rejected": -0.480968177318573,
      "logps/chosen": -163.02232360839844,
      "logps/rejected": -166.95147705078125,
      "loss": 1.5552,
      "nll_loss": 0.28835493326187134,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -16.302234649658203,
      "rewards/margins": 0.39291518926620483,
      "rewards/rejected": -16.695148468017578,
      "step": 160
    },
    {
      "epoch": 0.3929500144466917,
      "grad_norm": 47.49101257324219,
      "learning_rate": 6.752577319587629e-07,
      "logits/chosen": -0.3924413323402405,
      "logits/rejected": -0.3796409070491791,
      "logps/chosen": -158.3751220703125,
      "logps/rejected": -165.7510223388672,
      "loss": 1.4469,
      "nll_loss": 0.2873372733592987,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -15.837512016296387,
      "rewards/margins": 0.7375894784927368,
      "rewards/rejected": -16.575103759765625,
      "step": 170
    },
    {
      "epoch": 0.41606472117885,
      "grad_norm": 47.435874938964844,
      "learning_rate": 6.494845360824742e-07,
      "logits/chosen": -0.47805255651474,
      "logits/rejected": -0.4726547598838806,
      "logps/chosen": -152.86453247070312,
      "logps/rejected": -159.31887817382812,
      "loss": 1.4799,
      "nll_loss": 0.2764199376106262,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -15.286453247070312,
      "rewards/margins": 0.6454333066940308,
      "rewards/rejected": -15.93188762664795,
      "step": 180
    },
    {
      "epoch": 0.4391794279110084,
      "grad_norm": 53.53675842285156,
      "learning_rate": 6.237113402061855e-07,
      "logits/chosen": -0.5260998606681824,
      "logits/rejected": -0.5136414766311646,
      "logps/chosen": -162.34234619140625,
      "logps/rejected": -168.7532196044922,
      "loss": 1.5476,
      "nll_loss": 0.28763675689697266,
      "rewards/accuracies": 0.5843750238418579,
      "rewards/chosen": -16.234233856201172,
      "rewards/margins": 0.6410863995552063,
      "rewards/rejected": -16.875320434570312,
      "step": 190
    },
    {
      "epoch": 0.4622941346431667,
      "grad_norm": 45.457393646240234,
      "learning_rate": 5.979381443298969e-07,
      "logits/chosen": -0.4183273911476135,
      "logits/rejected": -0.4106271266937256,
      "logps/chosen": -158.37742614746094,
      "logps/rejected": -164.9311065673828,
      "loss": 1.5276,
      "nll_loss": 0.2878963351249695,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -15.837743759155273,
      "rewards/margins": 0.6553663015365601,
      "rewards/rejected": -16.49310874938965,
      "step": 200
    },
    {
      "epoch": 0.4622941346431667,
      "eval_logits/chosen": -0.4236670732498169,
      "eval_logits/rejected": -0.39834392070770264,
      "eval_logps/chosen": -158.0997314453125,
      "eval_logps/rejected": -164.43028259277344,
      "eval_loss": 1.5229449272155762,
      "eval_nll_loss": 0.27477577328681946,
      "eval_rewards/accuracies": 0.604347825050354,
      "eval_rewards/chosen": -15.809972763061523,
      "eval_rewards/margins": 0.6330567002296448,
      "eval_rewards/rejected": -16.443029403686523,
      "eval_runtime": 77.51,
      "eval_samples_per_second": 23.558,
      "eval_steps_per_second": 1.484,
      "step": 200
    },
    {
      "epoch": 0.48540884137532503,
      "grad_norm": 51.9904899597168,
      "learning_rate": 5.721649484536082e-07,
      "logits/chosen": -0.412663996219635,
      "logits/rejected": -0.385576069355011,
      "logps/chosen": -155.5344696044922,
      "logps/rejected": -157.98362731933594,
      "loss": 1.5112,
      "nll_loss": 0.2793150544166565,
      "rewards/accuracies": 0.5406249761581421,
      "rewards/chosen": -15.553448677062988,
      "rewards/margins": 0.24491462111473083,
      "rewards/rejected": -15.798362731933594,
      "step": 210
    },
    {
      "epoch": 0.5085235481074833,
      "grad_norm": 49.318214416503906,
      "learning_rate": 5.463917525773195e-07,
      "logits/chosen": -0.41389769315719604,
      "logits/rejected": -0.38615840673446655,
      "logps/chosen": -168.82632446289062,
      "logps/rejected": -174.5264129638672,
      "loss": 1.5696,
      "nll_loss": 0.2951294779777527,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -16.882633209228516,
      "rewards/margins": 0.5700088143348694,
      "rewards/rejected": -17.4526424407959,
      "step": 220
    },
    {
      "epoch": 0.5316382548396418,
      "grad_norm": 47.010562896728516,
      "learning_rate": 5.20618556701031e-07,
      "logits/chosen": -0.3984927237033844,
      "logits/rejected": -0.3791876435279846,
      "logps/chosen": -166.37368774414062,
      "logps/rejected": -173.02516174316406,
      "loss": 1.583,
      "nll_loss": 0.29688653349876404,
      "rewards/accuracies": 0.5843750238418579,
      "rewards/chosen": -16.637371063232422,
      "rewards/margins": 0.6651442050933838,
      "rewards/rejected": -17.302515029907227,
      "step": 230
    },
    {
      "epoch": 0.5547529615718001,
      "grad_norm": 50.885963439941406,
      "learning_rate": 4.948453608247422e-07,
      "logits/chosen": -0.37319958209991455,
      "logits/rejected": -0.35784170031547546,
      "logps/chosen": -166.8050537109375,
      "logps/rejected": -172.14004516601562,
      "loss": 1.4338,
      "nll_loss": 0.28680044412612915,
      "rewards/accuracies": 0.559374988079071,
      "rewards/chosen": -16.68050765991211,
      "rewards/margins": 0.5334986448287964,
      "rewards/rejected": -17.214006423950195,
      "step": 240
    },
    {
      "epoch": 0.5778676683039584,
      "grad_norm": 60.3321418762207,
      "learning_rate": 4.6907216494845357e-07,
      "logits/chosen": -0.41152358055114746,
      "logits/rejected": -0.3965645730495453,
      "logps/chosen": -161.13088989257812,
      "logps/rejected": -164.7410888671875,
      "loss": 1.5649,
      "nll_loss": 0.25410208106040955,
      "rewards/accuracies": 0.559374988079071,
      "rewards/chosen": -16.113088607788086,
      "rewards/margins": 0.361021488904953,
      "rewards/rejected": -16.474109649658203,
      "step": 250
    },
    {
      "epoch": 0.6009823750361167,
      "grad_norm": 49.894596099853516,
      "learning_rate": 4.432989690721649e-07,
      "logits/chosen": -0.518215537071228,
      "logits/rejected": -0.504486083984375,
      "logps/chosen": -157.352294921875,
      "logps/rejected": -166.1837921142578,
      "loss": 1.4283,
      "nll_loss": 0.2800624370574951,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -15.7352294921875,
      "rewards/margins": 0.8831487894058228,
      "rewards/rejected": -16.618377685546875,
      "step": 260
    },
    {
      "epoch": 0.624097081768275,
      "grad_norm": 44.237037658691406,
      "learning_rate": 4.175257731958763e-07,
      "logits/chosen": -0.4858783185482025,
      "logits/rejected": -0.47672492265701294,
      "logps/chosen": -158.05084228515625,
      "logps/rejected": -165.44720458984375,
      "loss": 1.4637,
      "nll_loss": 0.2714413106441498,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -15.805084228515625,
      "rewards/margins": 0.739635169506073,
      "rewards/rejected": -16.544719696044922,
      "step": 270
    },
    {
      "epoch": 0.6472117885004334,
      "grad_norm": 52.68519973754883,
      "learning_rate": 3.917525773195876e-07,
      "logits/chosen": -0.4353243410587311,
      "logits/rejected": -0.4047181010246277,
      "logps/chosen": -162.55743408203125,
      "logps/rejected": -171.99114990234375,
      "loss": 1.4694,
      "nll_loss": 0.2985231876373291,
      "rewards/accuracies": 0.6468750238418579,
      "rewards/chosen": -16.2557430267334,
      "rewards/margins": 0.9433721303939819,
      "rewards/rejected": -17.199115753173828,
      "step": 280
    },
    {
      "epoch": 0.6703264952325917,
      "grad_norm": 54.75222396850586,
      "learning_rate": 3.659793814432989e-07,
      "logits/chosen": -0.4927333891391754,
      "logits/rejected": -0.46107035875320435,
      "logps/chosen": -165.6143035888672,
      "logps/rejected": -171.44073486328125,
      "loss": 1.4147,
      "nll_loss": 0.2945927381515503,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -16.561431884765625,
      "rewards/margins": 0.5826419591903687,
      "rewards/rejected": -17.144071578979492,
      "step": 290
    },
    {
      "epoch": 0.6934412019647501,
      "grad_norm": 44.112884521484375,
      "learning_rate": 3.402061855670103e-07,
      "logits/chosen": -0.5113806128501892,
      "logits/rejected": -0.5009027719497681,
      "logps/chosen": -168.73641967773438,
      "logps/rejected": -178.53773498535156,
      "loss": 1.4811,
      "nll_loss": 0.3009123206138611,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -16.873641967773438,
      "rewards/margins": 0.9801331758499146,
      "rewards/rejected": -17.853775024414062,
      "step": 300
    },
    {
      "epoch": 0.6934412019647501,
      "eval_logits/chosen": -0.4338991940021515,
      "eval_logits/rejected": -0.40692025423049927,
      "eval_logps/chosen": -160.7057342529297,
      "eval_logps/rejected": -168.00125122070312,
      "eval_loss": 1.463964819908142,
      "eval_nll_loss": 0.28037506341934204,
      "eval_rewards/accuracies": 0.613043487071991,
      "eval_rewards/chosen": -16.070573806762695,
      "eval_rewards/margins": 0.7295539975166321,
      "eval_rewards/rejected": -16.800127029418945,
      "eval_runtime": 77.4432,
      "eval_samples_per_second": 23.579,
      "eval_steps_per_second": 1.485,
      "step": 300
    },
    {
      "epoch": 0.7165559086969084,
      "grad_norm": 47.5919303894043,
      "learning_rate": 3.1443298969072163e-07,
      "logits/chosen": -0.47527360916137695,
      "logits/rejected": -0.4661695957183838,
      "logps/chosen": -170.21279907226562,
      "logps/rejected": -174.0437774658203,
      "loss": 1.5403,
      "nll_loss": 0.3031921982765198,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -17.021282196044922,
      "rewards/margins": 0.38309773802757263,
      "rewards/rejected": -17.404376983642578,
      "step": 310
    },
    {
      "epoch": 0.7396706154290668,
      "grad_norm": 45.896297454833984,
      "learning_rate": 2.8865979381443296e-07,
      "logits/chosen": -0.4675866961479187,
      "logits/rejected": -0.4545253813266754,
      "logps/chosen": -159.13848876953125,
      "logps/rejected": -168.03915405273438,
      "loss": 1.5256,
      "nll_loss": 0.28240206837654114,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -15.913850784301758,
      "rewards/margins": 0.8900658488273621,
      "rewards/rejected": -16.80391502380371,
      "step": 320
    },
    {
      "epoch": 0.7627853221612251,
      "grad_norm": 46.80994415283203,
      "learning_rate": 2.6288659793814435e-07,
      "logits/chosen": -0.4704606533050537,
      "logits/rejected": -0.4714192748069763,
      "logps/chosen": -158.95199584960938,
      "logps/rejected": -165.59017944335938,
      "loss": 1.5045,
      "nll_loss": 0.2972811162471771,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -15.8951997756958,
      "rewards/margins": 0.6638190150260925,
      "rewards/rejected": -16.559019088745117,
      "step": 330
    },
    {
      "epoch": 0.7859000288933834,
      "grad_norm": 44.699893951416016,
      "learning_rate": 2.3711340206185566e-07,
      "logits/chosen": -0.47059255838394165,
      "logits/rejected": -0.45767131447792053,
      "logps/chosen": -162.30491638183594,
      "logps/rejected": -167.87808227539062,
      "loss": 1.3925,
      "nll_loss": 0.2751705050468445,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -16.230493545532227,
      "rewards/margins": 0.5573164820671082,
      "rewards/rejected": -16.787809371948242,
      "step": 340
    },
    {
      "epoch": 0.8090147356255417,
      "grad_norm": 46.135292053222656,
      "learning_rate": 2.11340206185567e-07,
      "logits/chosen": -0.4483928680419922,
      "logits/rejected": -0.4473996162414551,
      "logps/chosen": -163.79013061523438,
      "logps/rejected": -172.65423583984375,
      "loss": 1.4903,
      "nll_loss": 0.2803964912891388,
      "rewards/accuracies": 0.609375,
      "rewards/chosen": -16.379013061523438,
      "rewards/margins": 0.8864116668701172,
      "rewards/rejected": -17.265422821044922,
      "step": 350
    },
    {
      "epoch": 0.8321294423577,
      "grad_norm": 56.81486511230469,
      "learning_rate": 1.8556701030927835e-07,
      "logits/chosen": -0.4406563639640808,
      "logits/rejected": -0.4262049198150635,
      "logps/chosen": -168.64210510253906,
      "logps/rejected": -174.23231506347656,
      "loss": 1.535,
      "nll_loss": 0.27415817975997925,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": -16.86421012878418,
      "rewards/margins": 0.559019923210144,
      "rewards/rejected": -17.423233032226562,
      "step": 360
    },
    {
      "epoch": 0.8552441490898585,
      "grad_norm": 50.85494613647461,
      "learning_rate": 1.5979381443298966e-07,
      "logits/chosen": -0.4312410354614258,
      "logits/rejected": -0.4031241834163666,
      "logps/chosen": -162.82699584960938,
      "logps/rejected": -172.81155395507812,
      "loss": 1.4015,
      "nll_loss": 0.2656095623970032,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -16.282699584960938,
      "rewards/margins": 0.9984554052352905,
      "rewards/rejected": -17.28115463256836,
      "step": 370
    },
    {
      "epoch": 0.8783588558220168,
      "grad_norm": 54.44953918457031,
      "learning_rate": 1.3402061855670102e-07,
      "logits/chosen": -0.38503849506378174,
      "logits/rejected": -0.36510857939720154,
      "logps/chosen": -159.20938110351562,
      "logps/rejected": -166.0490264892578,
      "loss": 1.4196,
      "nll_loss": 0.2949269711971283,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -15.920938491821289,
      "rewards/margins": 0.6839638352394104,
      "rewards/rejected": -16.604902267456055,
      "step": 380
    },
    {
      "epoch": 0.9014735625541751,
      "grad_norm": 53.84355926513672,
      "learning_rate": 1.0824742268041237e-07,
      "logits/chosen": -0.4367571473121643,
      "logits/rejected": -0.41515684127807617,
      "logps/chosen": -162.70645141601562,
      "logps/rejected": -173.05177307128906,
      "loss": 1.4525,
      "nll_loss": 0.27491894364356995,
      "rewards/accuracies": 0.6156250238418579,
      "rewards/chosen": -16.27064323425293,
      "rewards/margins": 1.0345335006713867,
      "rewards/rejected": -17.30517578125,
      "step": 390
    },
    {
      "epoch": 0.9245882692863334,
      "grad_norm": 51.0332145690918,
      "learning_rate": 8.24742268041237e-08,
      "logits/chosen": -0.4105447232723236,
      "logits/rejected": -0.38223332166671753,
      "logps/chosen": -167.57431030273438,
      "logps/rejected": -172.93923950195312,
      "loss": 1.4642,
      "nll_loss": 0.3052617907524109,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -16.757429122924805,
      "rewards/margins": 0.5364928841590881,
      "rewards/rejected": -17.293922424316406,
      "step": 400
    },
    {
      "epoch": 0.9245882692863334,
      "eval_logits/chosen": -0.38119494915008545,
      "eval_logits/rejected": -0.35089170932769775,
      "eval_logps/chosen": -161.57652282714844,
      "eval_logps/rejected": -169.120361328125,
      "eval_loss": 1.4428884983062744,
      "eval_nll_loss": 0.2844657897949219,
      "eval_rewards/accuracies": 0.6304348111152649,
      "eval_rewards/chosen": -16.157651901245117,
      "eval_rewards/margins": 0.7543851137161255,
      "eval_rewards/rejected": -16.912038803100586,
      "eval_runtime": 77.5027,
      "eval_samples_per_second": 23.56,
      "eval_steps_per_second": 1.484,
      "step": 400
    },
    {
      "epoch": 0.9477029760184917,
      "grad_norm": 46.96026611328125,
      "learning_rate": 5.670103092783505e-08,
      "logits/chosen": -0.3439410626888275,
      "logits/rejected": -0.3307141661643982,
      "logps/chosen": -154.14857482910156,
      "logps/rejected": -163.5908966064453,
      "loss": 1.3967,
      "nll_loss": 0.2838439345359802,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -15.4148588180542,
      "rewards/margins": 0.9442328214645386,
      "rewards/rejected": -16.35909080505371,
      "step": 410
    },
    {
      "epoch": 0.9708176827506501,
      "grad_norm": 44.309818267822266,
      "learning_rate": 3.092783505154639e-08,
      "logits/chosen": -0.3803669214248657,
      "logits/rejected": -0.3566874861717224,
      "logps/chosen": -156.36138916015625,
      "logps/rejected": -162.5567626953125,
      "loss": 1.4043,
      "nll_loss": 0.28173336386680603,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -15.636138916015625,
      "rewards/margins": 0.6195372939109802,
      "rewards/rejected": -16.25567626953125,
      "step": 420
    },
    {
      "epoch": 0.9939323894828085,
      "grad_norm": 48.28641128540039,
      "learning_rate": 5.154639175257731e-09,
      "logits/chosen": -0.3961271345615387,
      "logits/rejected": -0.40863722562789917,
      "logps/chosen": -164.62960815429688,
      "logps/rejected": -171.9886474609375,
      "loss": 1.4854,
      "nll_loss": 0.2709375023841858,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -16.46295928955078,
      "rewards/margins": 0.7359048128128052,
      "rewards/rejected": -17.198863983154297,
      "step": 430
    },
    {
      "epoch": 0.9985553308292401,
      "step": 432,
      "total_flos": 0.0,
      "train_loss": 1.6653077276768509,
      "train_runtime": 9934.0195,
      "train_samples_per_second": 5.574,
      "train_steps_per_second": 0.043
    }
  ],
  "logging_steps": 10,
  "max_steps": 432,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}