{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9485384932070811,
  "eval_steps": 500,
  "global_step": 72,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013174145738987238,
      "grad_norm": 0.39343583583831787,
      "learning_rate": 6.25e-08,
      "logits/chosen": 10.071717262268066,
      "logits/rejected": 10.610974311828613,
      "logps/chosen": -121.14067077636719,
      "logps/ref_chosen": -121.14067077636719,
      "logps/ref_rejected": -137.65684509277344,
      "logps/rejected": -137.65684509277344,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.026348291477974475,
      "grad_norm": 0.7521647214889526,
      "learning_rate": 1.25e-07,
      "logits/chosen": 10.222262382507324,
      "logits/rejected": 10.754176139831543,
      "logps/chosen": -116.48068237304688,
      "logps/ref_chosen": -116.48068237304688,
      "logps/ref_rejected": -130.27796936035156,
      "logps/rejected": -130.27796936035156,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 2
    },
    {
      "epoch": 0.03952243721696171,
      "grad_norm": 0.5880491137504578,
      "learning_rate": 1.875e-07,
      "logits/chosen": 10.01984977722168,
      "logits/rejected": 10.62405776977539,
      "logps/chosen": -123.36822509765625,
      "logps/ref_chosen": -122.6683349609375,
      "logps/ref_rejected": -132.69850158691406,
      "logps/rejected": -133.207275390625,
      "loss": 0.6941,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": -0.006998830940574408,
      "rewards/margins": -0.0019110905705019832,
      "rewards/rejected": -0.005087739787995815,
      "step": 3
    },
    {
      "epoch": 0.05269658295594895,
      "grad_norm": 0.4605408310890198,
      "learning_rate": 2.5e-07,
      "logits/chosen": 9.99990463256836,
      "logits/rejected": 10.736846923828125,
      "logps/chosen": -123.02133178710938,
      "logps/ref_chosen": -122.59739685058594,
      "logps/ref_rejected": -129.70767211914062,
      "logps/rejected": -129.98374938964844,
      "loss": 0.6939,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": -0.004239337984472513,
      "rewards/margins": -0.0014785109087824821,
      "rewards/rejected": -0.0027608266100287437,
      "step": 4
    },
    {
      "epoch": 0.06587072869493618,
      "grad_norm": 0.469856321811676,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": 10.075482368469238,
      "logits/rejected": 10.892666816711426,
      "logps/chosen": -117.554931640625,
      "logps/ref_chosen": -117.5941162109375,
      "logps/ref_rejected": -132.1708984375,
      "logps/rejected": -132.1227264404297,
      "loss": 0.6932,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": 0.0003917121794074774,
      "rewards/margins": -9.010493522509933e-05,
      "rewards/rejected": 0.0004818170564249158,
      "step": 5
    },
    {
      "epoch": 0.07904487443392343,
      "grad_norm": 0.5541566610336304,
      "learning_rate": 3.75e-07,
      "logits/chosen": 10.660999298095703,
      "logits/rejected": 11.290507316589355,
      "logps/chosen": -127.00320434570312,
      "logps/ref_chosen": -126.12411499023438,
      "logps/ref_rejected": -136.9976043701172,
      "logps/rejected": -137.75950622558594,
      "loss": 0.6938,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.00879070907831192,
      "rewards/margins": -0.0011717069428414106,
      "rewards/rejected": -0.007619001902639866,
      "step": 6
    },
    {
      "epoch": 0.09221902017291066,
      "grad_norm": 0.4422788619995117,
      "learning_rate": 4.375e-07,
      "logits/chosen": 9.958097457885742,
      "logits/rejected": 10.642163276672363,
      "logps/chosen": -115.61244201660156,
      "logps/ref_chosen": -115.08863830566406,
      "logps/ref_rejected": -125.91255187988281,
      "logps/rejected": -126.15577697753906,
      "loss": 0.6946,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.005238103214651346,
      "rewards/margins": -0.0028058765456080437,
      "rewards/rejected": -0.002432226436212659,
      "step": 7
    },
    {
      "epoch": 0.1053931659118979,
      "grad_norm": 0.5495327115058899,
      "learning_rate": 5e-07,
      "logits/chosen": 10.345503807067871,
      "logits/rejected": 10.97708797454834,
      "logps/chosen": -121.48406982421875,
      "logps/ref_chosen": -121.4114761352539,
      "logps/ref_rejected": -134.62770080566406,
      "logps/rejected": -134.7586212158203,
      "loss": 0.6929,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.0007259202538989484,
      "rewards/margins": 0.0005832896567881107,
      "rewards/rejected": -0.0013092098524793983,
      "step": 8
    },
    {
      "epoch": 0.11856731165088513,
      "grad_norm": 0.43922048807144165,
      "learning_rate": 4.997252228714278e-07,
      "logits/chosen": 10.159126281738281,
      "logits/rejected": 11.002123832702637,
      "logps/chosen": -122.51399230957031,
      "logps/ref_chosen": -121.59207153320312,
      "logps/ref_rejected": -134.70025634765625,
      "logps/rejected": -135.55740356445312,
      "loss": 0.6935,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": -0.00921926274895668,
      "rewards/margins": -0.0006479143630713224,
      "rewards/rejected": -0.008571348153054714,
      "step": 9
    },
    {
      "epoch": 0.13174145738987236,
      "grad_norm": 0.6643485426902771,
      "learning_rate": 4.989014955054745e-07,
      "logits/chosen": 9.9464111328125,
      "logits/rejected": 10.739057540893555,
      "logps/chosen": -117.91310119628906,
      "logps/ref_chosen": -117.16349029541016,
      "logps/ref_rejected": -129.98167419433594,
      "logps/rejected": -130.65924072265625,
      "loss": 0.6935,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.007496046833693981,
      "rewards/margins": -0.0007203805143944919,
      "rewards/rejected": -0.00677566509693861,
      "step": 10
    },
    {
      "epoch": 0.14491560312885962,
      "grad_norm": 0.4432956576347351,
      "learning_rate": 4.975306286336627e-07,
      "logits/chosen": 10.095781326293945,
      "logits/rejected": 11.001167297363281,
      "logps/chosen": -123.26506042480469,
      "logps/ref_chosen": -121.95927429199219,
      "logps/ref_rejected": -136.18655395507812,
      "logps/rejected": -137.39865112304688,
      "loss": 0.6936,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.01305788941681385,
      "rewards/margins": -0.0009369202307425439,
      "rewards/rejected": -0.012120969593524933,
      "step": 11
    },
    {
      "epoch": 0.15808974886784685,
      "grad_norm": 0.3955974280834198,
      "learning_rate": 4.956156357188939e-07,
      "logits/chosen": 10.129995346069336,
      "logits/rejected": 10.673677444458008,
      "logps/chosen": -126.05357360839844,
      "logps/ref_chosen": -124.12315368652344,
      "logps/ref_rejected": -134.275390625,
      "logps/rejected": -136.24940490722656,
      "loss": 0.693,
      "rewards/accuracies": 0.5234375,
      "rewards/chosen": -0.01930420845746994,
      "rewards/margins": 0.00043603626545518637,
      "rewards/rejected": -0.01974024437367916,
      "step": 12
    },
    {
      "epoch": 0.17126389460683408,
      "grad_norm": 0.4873650074005127,
      "learning_rate": 4.931607263312032e-07,
      "logits/chosen": 9.8243989944458,
      "logits/rejected": 10.843228340148926,
      "logps/chosen": -119.30509185791016,
      "logps/ref_chosen": -116.83765411376953,
      "logps/ref_rejected": -130.78997802734375,
      "logps/rejected": -133.15672302246094,
      "loss": 0.6937,
      "rewards/accuracies": 0.4453125,
      "rewards/chosen": -0.024674497544765472,
      "rewards/margins": -0.0010069820564240217,
      "rewards/rejected": -0.02366751804947853,
      "step": 13
    },
    {
      "epoch": 0.1844380403458213,
      "grad_norm": 0.8769639134407043,
      "learning_rate": 4.9017129689421e-07,
      "logits/chosen": 10.427848815917969,
      "logits/rejected": 11.37716293334961,
      "logps/chosen": -120.9067611694336,
      "logps/ref_chosen": -118.43791961669922,
      "logps/ref_rejected": -132.5309600830078,
      "logps/rejected": -134.91236877441406,
      "loss": 0.6936,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.02468838170170784,
      "rewards/margins": -0.0008744060760363936,
      "rewards/rejected": -0.023813974112272263,
      "step": 14
    },
    {
      "epoch": 0.19761218608480857,
      "grad_norm": 0.8736073970794678,
      "learning_rate": 4.866539188226085e-07,
      "logits/chosen": 9.956391334533691,
      "logits/rejected": 10.684510231018066,
      "logps/chosen": -128.198486328125,
      "logps/ref_chosen": -124.82101440429688,
      "logps/ref_rejected": -135.32565307617188,
      "logps/rejected": -138.8860626220703,
      "loss": 0.6923,
      "rewards/accuracies": 0.5234375,
      "rewards/chosen": -0.033774565905332565,
      "rewards/margins": 0.0018295131158083677,
      "rewards/rejected": -0.0356040820479393,
      "step": 15
    },
    {
      "epoch": 0.2107863318237958,
      "grad_norm": 0.7293412089347839,
      "learning_rate": 4.826163240767716e-07,
      "logits/chosen": 10.677412033081055,
      "logits/rejected": 11.254134178161621,
      "logps/chosen": -122.79344940185547,
      "logps/ref_chosen": -119.69990539550781,
      "logps/ref_rejected": -130.34449768066406,
      "logps/rejected": -133.39932250976562,
      "loss": 0.6934,
      "rewards/accuracies": 0.484375,
      "rewards/chosen": -0.030935294926166534,
      "rewards/margins": -0.0003869622596539557,
      "rewards/rejected": -0.030548332259058952,
      "step": 16
    },
    {
      "epoch": 0.22396047756278303,
      "grad_norm": 0.6406402587890625,
      "learning_rate": 4.780673881662242e-07,
      "logits/chosen": 10.172043800354004,
      "logits/rejected": 10.859190940856934,
      "logps/chosen": -118.11813354492188,
      "logps/ref_chosen": -114.52975463867188,
      "logps/ref_rejected": -129.5846405029297,
      "logps/rejected": -133.20779418945312,
      "loss": 0.6931,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.035883828997612,
      "rewards/margins": 0.00034771906211972237,
      "rewards/rejected": -0.036231543868780136,
      "step": 17
    },
    {
      "epoch": 0.23713462330177026,
      "grad_norm": 0.7559142708778381,
      "learning_rate": 4.730171106393466e-07,
      "logits/chosen": 10.747884750366211,
      "logits/rejected": 11.173224449157715,
      "logps/chosen": -121.89351654052734,
      "logps/ref_chosen": -117.67997741699219,
      "logps/ref_rejected": -128.00650024414062,
      "logps/rejected": -132.31655883789062,
      "loss": 0.6928,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.042135339230298996,
      "rewards/margins": 0.0009651560103520751,
      "rewards/rejected": -0.043100498616695404,
      "step": 18
    },
    {
      "epoch": 0.2503087690407575,
      "grad_norm": 0.49808746576309204,
      "learning_rate": 4.6747659310219757e-07,
      "logits/chosen": 10.380050659179688,
      "logits/rejected": 10.91146469116211,
      "logps/chosen": -125.04395294189453,
      "logps/ref_chosen": -120.92308044433594,
      "logps/ref_rejected": -133.8301544189453,
      "logps/rejected": -138.79052734375,
      "loss": 0.689,
      "rewards/accuracies": 0.6328125,
      "rewards/chosen": -0.0412086620926857,
      "rewards/margins": 0.008395083248615265,
      "rewards/rejected": -0.04960374906659126,
      "step": 19
    },
    {
      "epoch": 0.2634829147797447,
      "grad_norm": 0.5601547956466675,
      "learning_rate": 4.6145801481477433e-07,
      "logits/chosen": 10.563058853149414,
      "logits/rejected": 11.37236499786377,
      "logps/chosen": -126.93672180175781,
      "logps/ref_chosen": -121.84554290771484,
      "logps/ref_rejected": -133.343017578125,
      "logps/rejected": -138.12808227539062,
      "loss": 0.6948,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.050911907106637955,
      "rewards/margins": -0.0030612414702773094,
      "rewards/rejected": -0.04785066470503807,
      "step": 20
    },
    {
      "epoch": 0.276657060518732,
      "grad_norm": 0.41351014375686646,
      "learning_rate": 4.549746059183561e-07,
      "logits/chosen": 9.687071800231934,
      "logits/rejected": 10.572582244873047,
      "logps/chosen": -124.77998352050781,
      "logps/ref_chosen": -119.20828247070312,
      "logps/ref_rejected": -134.38436889648438,
      "logps/rejected": -140.02293395996094,
      "loss": 0.6929,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.055717017501592636,
      "rewards/margins": 0.0006685962434858084,
      "rewards/rejected": -0.056385621428489685,
      "step": 21
    },
    {
      "epoch": 0.28983120625771924,
      "grad_norm": 0.6027271747589111,
      "learning_rate": 4.480406183527823e-07,
      "logits/chosen": 10.03482437133789,
      "logits/rejected": 10.820859909057617,
      "logps/chosen": -121.17003631591797,
      "logps/ref_chosen": -114.62059020996094,
      "logps/ref_rejected": -128.0896759033203,
      "logps/rejected": -134.1069793701172,
      "loss": 0.6959,
      "rewards/accuracies": 0.484375,
      "rewards/chosen": -0.06549445539712906,
      "rewards/margins": -0.005321440752595663,
      "rewards/rejected": -0.06017300859093666,
      "step": 22
    },
    {
      "epoch": 0.3030053519967065,
      "grad_norm": 0.5877695083618164,
      "learning_rate": 4.4067129452759546e-07,
      "logits/chosen": 10.00536060333252,
      "logits/rejected": 10.877461433410645,
      "logps/chosen": -124.2905044555664,
      "logps/ref_chosen": -117.84042358398438,
      "logps/ref_rejected": -131.79171752929688,
      "logps/rejected": -137.91160583496094,
      "loss": 0.6949,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.06450086086988449,
      "rewards/margins": -0.0033018956892192364,
      "rewards/rejected": -0.061198972165584564,
      "step": 23
    },
    {
      "epoch": 0.3161794977356937,
      "grad_norm": 0.5177574157714844,
      "learning_rate": 4.3288283381591725e-07,
      "logits/chosen": 10.104101181030273,
      "logits/rejected": 10.70304012298584,
      "logps/chosen": -129.96224975585938,
      "logps/ref_chosen": -123.75523376464844,
      "logps/ref_rejected": -138.6237030029297,
      "logps/rejected": -144.98983764648438,
      "loss": 0.6925,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.062070075422525406,
      "rewards/margins": 0.0015914504183456302,
      "rewards/rejected": -0.06366152316331863,
      "step": 24
    },
    {
      "epoch": 0.32935364347468093,
      "grad_norm": 0.57741379737854,
      "learning_rate": 4.246923569447104e-07,
      "logits/chosen": 10.097904205322266,
      "logits/rejected": 10.884883880615234,
      "logps/chosen": -130.04393005371094,
      "logps/ref_chosen": -122.53610229492188,
      "logps/ref_rejected": -141.00828552246094,
      "logps/rejected": -148.18405151367188,
      "loss": 0.695,
      "rewards/accuracies": 0.4609375,
      "rewards/chosen": -0.07507827132940292,
      "rewards/margins": -0.003320657880976796,
      "rewards/rejected": -0.07175761461257935,
      "step": 25
    },
    {
      "epoch": 0.34252778921366817,
      "grad_norm": 0.5819231867790222,
      "learning_rate": 4.161178683597054e-07,
      "logits/chosen": 10.32345199584961,
      "logits/rejected": 11.267877578735352,
      "logps/chosen": -132.95077514648438,
      "logps/ref_chosen": -124.1744384765625,
      "logps/ref_rejected": -139.68605041503906,
      "logps/rejected": -148.0929718017578,
      "loss": 0.6952,
      "rewards/accuracies": 0.484375,
      "rewards/chosen": -0.08776339888572693,
      "rewards/margins": -0.0036940835416316986,
      "rewards/rejected": -0.08406931161880493,
      "step": 26
    },
    {
      "epoch": 0.3557019349526554,
      "grad_norm": 0.6049931645393372,
      "learning_rate": 4.0717821664772124e-07,
      "logits/chosen": 10.15705394744873,
      "logits/rejected": 11.114046096801758,
      "logps/chosen": -125.23123931884766,
      "logps/ref_chosen": -117.1941146850586,
      "logps/ref_rejected": -129.4031982421875,
      "logps/rejected": -137.76333618164062,
      "loss": 0.6918,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.08037131279706955,
      "rewards/margins": 0.0032299975864589214,
      "rewards/rejected": -0.08360131084918976,
      "step": 27
    },
    {
      "epoch": 0.3688760806916426,
      "grad_norm": 0.5475514531135559,
      "learning_rate": 3.978930531033806e-07,
      "logits/chosen": 9.436954498291016,
      "logits/rejected": 10.49735164642334,
      "logps/chosen": -125.24280548095703,
      "logps/ref_chosen": -117.17620086669922,
      "logps/ref_rejected": -130.76107788085938,
      "logps/rejected": -138.6796875,
      "loss": 0.6941,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.08066616207361221,
      "rewards/margins": -0.001479951897636056,
      "rewards/rejected": -0.07918620854616165,
      "step": 28
    },
    {
      "epoch": 0.3820502264306299,
      "grad_norm": 0.599656343460083,
      "learning_rate": 3.882827885312998e-07,
      "logits/chosen": 10.068204879760742,
      "logits/rejected": 10.938701629638672,
      "logps/chosen": -130.98472595214844,
      "logps/ref_chosen": -123.2500228881836,
      "logps/ref_rejected": -133.91970825195312,
      "logps/rejected": -141.19216918945312,
      "loss": 0.6957,
      "rewards/accuracies": 0.453125,
      "rewards/chosen": -0.07734709233045578,
      "rewards/margins": -0.004622358828783035,
      "rewards/rejected": -0.07272473722696304,
      "step": 29
    },
    {
      "epoch": 0.39522437216961714,
      "grad_norm": 0.4804610311985016,
      "learning_rate": 3.7836854837871044e-07,
      "logits/chosen": 10.131808280944824,
      "logits/rejected": 11.42764663696289,
      "logps/chosen": -127.49568176269531,
      "logps/ref_chosen": -118.52604675292969,
      "logps/ref_rejected": -139.1146240234375,
      "logps/rejected": -147.99575805664062,
      "loss": 0.6938,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.08969634026288986,
      "rewards/margins": -0.0008849686128087342,
      "rewards/rejected": -0.08881138265132904,
      "step": 30
    },
    {
      "epoch": 0.4083985179086044,
      "grad_norm": 0.4208291471004486,
      "learning_rate": 3.681721262971413e-07,
      "logits/chosen": 9.75320816040039,
      "logits/rejected": 10.846428871154785,
      "logps/chosen": -119.75810241699219,
      "logps/ref_chosen": -111.525146484375,
      "logps/ref_rejected": -128.1785888671875,
      "logps/rejected": -136.38653564453125,
      "loss": 0.6935,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.08232954144477844,
      "rewards/margins": -0.00024991348618641496,
      "rewards/rejected": -0.08207963407039642,
      "step": 31
    },
    {
      "epoch": 0.4215726636475916,
      "grad_norm": 0.7168395519256592,
      "learning_rate": 3.577159362352426e-07,
      "logits/chosen": 9.947325706481934,
      "logits/rejected": 11.155643463134766,
      "logps/chosen": -124.97373962402344,
      "logps/ref_chosen": -116.80255126953125,
      "logps/ref_rejected": -134.80767822265625,
      "logps/rejected": -142.88174438476562,
      "loss": 0.6939,
      "rewards/accuracies": 0.4296875,
      "rewards/chosen": -0.08171181380748749,
      "rewards/margins": -0.0009711601305752993,
      "rewards/rejected": -0.08074064552783966,
      "step": 32
    },
    {
      "epoch": 0.43474680938657884,
      "grad_norm": 1.1801953315734863,
      "learning_rate": 3.470229631680624e-07,
      "logits/chosen": 10.00582504272461,
      "logits/rejected": 10.817474365234375,
      "logps/chosen": -127.24813842773438,
      "logps/ref_chosen": -118.635009765625,
      "logps/ref_rejected": -133.8279571533203,
      "logps/rejected": -142.4803924560547,
      "loss": 0.6932,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": -0.08613133430480957,
      "rewards/margins": 0.0003930249949917197,
      "rewards/rejected": -0.0865243598818779,
      "step": 33
    },
    {
      "epoch": 0.44792095512556607,
      "grad_norm": 0.5862898230552673,
      "learning_rate": 3.361167125710832e-07,
      "logits/chosen": 10.005105972290039,
      "logits/rejected": 10.864886283874512,
      "logps/chosen": -135.84164428710938,
      "logps/ref_chosen": -126.42659759521484,
      "logps/ref_rejected": -143.6361846923828,
      "logps/rejected": -153.5630340576172,
      "loss": 0.6909,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.09415031969547272,
      "rewards/margins": 0.005118116270750761,
      "rewards/rejected": -0.09926842898130417,
      "step": 34
    },
    {
      "epoch": 0.4610951008645533,
      "grad_norm": 0.7007870674133301,
      "learning_rate": 3.2502115875008516e-07,
      "logits/chosen": 10.54973030090332,
      "logits/rejected": 11.243837356567383,
      "logps/chosen": -131.86981201171875,
      "logps/ref_chosen": -123.14965057373047,
      "logps/ref_rejected": -134.22947692871094,
      "logps/rejected": -143.7413787841797,
      "loss": 0.6895,
      "rewards/accuracies": 0.5703125,
      "rewards/chosen": -0.0872015431523323,
      "rewards/margins": 0.007917709648609161,
      "rewards/rejected": -0.09511925280094147,
      "step": 35
    },
    {
      "epoch": 0.47426924660354053,
      "grad_norm": 0.7300606966018677,
      "learning_rate": 3.137606921404191e-07,
      "logits/chosen": 10.125234603881836,
      "logits/rejected": 10.481938362121582,
      "logps/chosen": -127.8335189819336,
      "logps/ref_chosen": -118.79129791259766,
      "logps/ref_rejected": -126.64965057373047,
      "logps/rejected": -135.0525665283203,
      "loss": 0.6966,
      "rewards/accuracies": 0.4296875,
      "rewards/chosen": -0.09042223542928696,
      "rewards/margins": -0.006393034942448139,
      "rewards/rejected": -0.0840291976928711,
      "step": 36
    },
    {
      "epoch": 0.4874433923425278,
      "grad_norm": 0.456684947013855,
      "learning_rate": 3.0236006569153616e-07,
      "logits/chosen": 10.322757720947266,
      "logits/rejected": 10.932893753051758,
      "logps/chosen": -130.44381713867188,
      "logps/ref_chosen": -121.899169921875,
      "logps/ref_rejected": -136.85321044921875,
      "logps/rejected": -145.57542419433594,
      "loss": 0.6925,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.08544634282588959,
      "rewards/margins": 0.00177572516258806,
      "rewards/rejected": -0.08722206950187683,
      "step": 37
    },
    {
      "epoch": 0.500617538081515,
      "grad_norm": 0.6402201056480408,
      "learning_rate": 2.9084434045463254e-07,
      "logits/chosen": 10.014504432678223,
      "logits/rejected": 10.873030662536621,
      "logps/chosen": -127.30659484863281,
      "logps/ref_chosen": -118.75363159179688,
      "logps/ref_rejected": -132.82818603515625,
      "logps/rejected": -141.9580535888672,
      "loss": 0.6905,
      "rewards/accuracies": 0.5546875,
      "rewards/chosen": -0.08552955836057663,
      "rewards/margins": 0.005769058130681515,
      "rewards/rejected": -0.09129861742258072,
      "step": 38
    },
    {
      "epoch": 0.5137916838205022,
      "grad_norm": 0.5188658833503723,
      "learning_rate": 2.7923883049302066e-07,
      "logits/chosen": 10.178832054138184,
      "logits/rejected": 10.932839393615723,
      "logps/chosen": -137.26002502441406,
      "logps/ref_chosen": -128.23240661621094,
      "logps/ref_rejected": -140.05517578125,
      "logps/rejected": -149.08457946777344,
      "loss": 0.6934,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": -0.09027623385190964,
      "rewards/margins": 1.788110239431262e-05,
      "rewards/rejected": -0.09029410034418106,
      "step": 39
    },
    {
      "epoch": 0.5269658295594895,
      "grad_norm": 0.7729969024658203,
      "learning_rate": 2.6756904723632324e-07,
      "logits/chosen": 10.076566696166992,
      "logits/rejected": 11.339007377624512,
      "logps/chosen": -127.69380187988281,
      "logps/ref_chosen": -118.38938903808594,
      "logps/ref_rejected": -138.8988037109375,
      "logps/rejected": -147.9398956298828,
      "loss": 0.6947,
      "rewards/accuracies": 0.484375,
      "rewards/chosen": -0.09304402768611908,
      "rewards/margins": -0.0026330682449042797,
      "rewards/rejected": -0.09041095525026321,
      "step": 40
    },
    {
      "epoch": 0.5401399752984768,
      "grad_norm": 0.6295668482780457,
      "learning_rate": 2.5586064340081516e-07,
      "logits/chosen": 10.364312171936035,
      "logits/rejected": 10.864094734191895,
      "logps/chosen": -126.18923950195312,
      "logps/ref_chosen": -117.82182312011719,
      "logps/ref_rejected": -132.89553833007812,
      "logps/rejected": -142.4484405517578,
      "loss": 0.6875,
      "rewards/accuracies": 0.5703125,
      "rewards/chosen": -0.08367416262626648,
      "rewards/margins": 0.011854931712150574,
      "rewards/rejected": -0.09552909433841705,
      "step": 41
    },
    {
      "epoch": 0.553314121037464,
      "grad_norm": 0.6686902046203613,
      "learning_rate": 2.4413935659918487e-07,
      "logits/chosen": 9.487873077392578,
      "logits/rejected": 10.435559272766113,
      "logps/chosen": -125.37492370605469,
      "logps/ref_chosen": -117.0536117553711,
      "logps/ref_rejected": -130.59812927246094,
      "logps/rejected": -139.56155395507812,
      "loss": 0.6903,
      "rewards/accuracies": 0.5234375,
      "rewards/chosen": -0.08321310579776764,
      "rewards/margins": 0.006421199534088373,
      "rewards/rejected": -0.08963430672883987,
      "step": 42
    },
    {
      "epoch": 0.5664882667764513,
      "grad_norm": 0.5477384924888611,
      "learning_rate": 2.3243095276367684e-07,
      "logits/chosen": 9.582094192504883,
      "logits/rejected": 10.506773948669434,
      "logps/chosen": -124.29164123535156,
      "logps/ref_chosen": -115.851806640625,
      "logps/ref_rejected": -130.8096160888672,
      "logps/rejected": -139.1727294921875,
      "loss": 0.6938,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.08439842611551285,
      "rewards/margins": -0.0007674552034586668,
      "rewards/rejected": -0.08363097161054611,
      "step": 43
    },
    {
      "epoch": 0.5796624125154385,
      "grad_norm": 0.4657438397407532,
      "learning_rate": 2.2076116950697937e-07,
      "logits/chosen": 9.838776588439941,
      "logits/rejected": 10.566040992736816,
      "logps/chosen": -126.56464385986328,
      "logps/ref_chosen": -117.90168762207031,
      "logps/ref_rejected": -132.960693359375,
      "logps/rejected": -142.1176300048828,
      "loss": 0.6909,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.08662942796945572,
      "rewards/margins": 0.004940015729516745,
      "rewards/rejected": -0.0915694385766983,
      "step": 44
    },
    {
      "epoch": 0.5928365582544257,
      "grad_norm": 0.4596782922744751,
      "learning_rate": 2.091556595453674e-07,
      "logits/chosen": 9.941704750061035,
      "logits/rejected": 10.477331161499023,
      "logps/chosen": -127.3547592163086,
      "logps/ref_chosen": -118.84819030761719,
      "logps/ref_rejected": -134.4351806640625,
      "logps/rejected": -143.46334838867188,
      "loss": 0.6908,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.08506564050912857,
      "rewards/margins": 0.005216139368712902,
      "rewards/rejected": -0.09028176963329315,
      "step": 45
    },
    {
      "epoch": 0.606010703993413,
      "grad_norm": 0.7369089722633362,
      "learning_rate": 1.9763993430846392e-07,
      "logits/chosen": 9.956421852111816,
      "logits/rejected": 10.419254302978516,
      "logps/chosen": -124.58769989013672,
      "logps/ref_chosen": -116.23255920410156,
      "logps/ref_rejected": -127.91217041015625,
      "logps/rejected": -137.61070251464844,
      "loss": 0.6867,
      "rewards/accuracies": 0.609375,
      "rewards/chosen": -0.08355137705802917,
      "rewards/margins": 0.013433952815830708,
      "rewards/rejected": -0.0969853326678276,
      "step": 46
    },
    {
      "epoch": 0.6191848497324002,
      "grad_norm": 0.6935871243476868,
      "learning_rate": 1.862393078595809e-07,
      "logits/chosen": 9.769856452941895,
      "logits/rejected": 10.882231712341309,
      "logps/chosen": -133.2343292236328,
      "logps/ref_chosen": -123.88105773925781,
      "logps/ref_rejected": -140.62420654296875,
      "logps/rejected": -150.24195861816406,
      "loss": 0.692,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.09353267401456833,
      "rewards/margins": 0.0026449086144566536,
      "rewards/rejected": -0.0961775854229927,
      "step": 47
    },
    {
      "epoch": 0.6323589954713874,
      "grad_norm": 0.9227128624916077,
      "learning_rate": 1.7497884124991485e-07,
      "logits/chosen": 10.461740493774414,
      "logits/rejected": 11.23796272277832,
      "logps/chosen": -129.02268981933594,
      "logps/ref_chosen": -119.78947448730469,
      "logps/ref_rejected": -134.12991333007812,
      "logps/rejected": -143.42581176757812,
      "loss": 0.693,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": -0.09233222156763077,
      "rewards/margins": 0.0006269109435379505,
      "rewards/rejected": -0.09295912832021713,
      "step": 48
    },
    {
      "epoch": 0.6455331412103746,
      "grad_norm": 0.6264681816101074,
      "learning_rate": 1.6388328742891678e-07,
      "logits/chosen": 10.495881080627441,
      "logits/rejected": 11.157975196838379,
      "logps/chosen": -129.8017578125,
      "logps/ref_chosen": -121.04522705078125,
      "logps/ref_rejected": -132.03546142578125,
      "logps/rejected": -141.43582153320312,
      "loss": 0.6902,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.08756528049707413,
      "rewards/margins": 0.00643829395994544,
      "rewards/rejected": -0.09400356560945511,
      "step": 49
    },
    {
      "epoch": 0.6587072869493619,
      "grad_norm": 0.4232679009437561,
      "learning_rate": 1.5297703683193753e-07,
      "logits/chosen": 10.161746978759766,
      "logits/rejected": 10.93832015991211,
      "logps/chosen": -117.66082000732422,
      "logps/ref_chosen": -109.65755462646484,
      "logps/ref_rejected": -123.57516479492188,
      "logps/rejected": -132.5537872314453,
      "loss": 0.6886,
      "rewards/accuracies": 0.5703125,
      "rewards/chosen": -0.08003270626068115,
      "rewards/margins": 0.009753433056175709,
      "rewards/rejected": -0.08978613466024399,
      "step": 50
    },
    {
      "epoch": 0.6718814326883491,
      "grad_norm": 0.5076274275779724,
      "learning_rate": 1.422840637647574e-07,
      "logits/chosen": 10.146079063415527,
      "logits/rejected": 10.626852035522461,
      "logps/chosen": -126.62266540527344,
      "logps/ref_chosen": -117.81820678710938,
      "logps/ref_rejected": -132.0400390625,
      "logps/rejected": -141.37049865722656,
      "loss": 0.6908,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.08804453164339066,
      "rewards/margins": 0.00526001350954175,
      "rewards/rejected": -0.09330454468727112,
      "step": 51
    },
    {
      "epoch": 0.6850555784273363,
      "grad_norm": 0.41372328996658325,
      "learning_rate": 1.3182787370285865e-07,
      "logits/chosen": 9.487447738647461,
      "logits/rejected": 10.429905891418457,
      "logps/chosen": -121.18466186523438,
      "logps/ref_chosen": -112.1561508178711,
      "logps/ref_rejected": -131.75282287597656,
      "logps/rejected": -141.48947143554688,
      "loss": 0.69,
      "rewards/accuracies": 0.5859375,
      "rewards/chosen": -0.09028512984514236,
      "rewards/margins": 0.007081407587975264,
      "rewards/rejected": -0.09736653417348862,
      "step": 52
    },
    {
      "epoch": 0.6982297241663236,
      "grad_norm": 0.7098668217658997,
      "learning_rate": 1.2163145162128946e-07,
      "logits/chosen": 9.994986534118652,
      "logits/rejected": 10.834747314453125,
      "logps/chosen": -129.7722625732422,
      "logps/ref_chosen": -120.89462280273438,
      "logps/ref_rejected": -131.4730682373047,
      "logps/rejected": -140.58242797851562,
      "loss": 0.6923,
      "rewards/accuracies": 0.484375,
      "rewards/chosen": -0.08877623826265335,
      "rewards/margins": 0.0023174649104475975,
      "rewards/rejected": -0.09109370410442352,
      "step": 53
    },
    {
      "epoch": 0.7114038699053108,
      "grad_norm": 0.6442684531211853,
      "learning_rate": 1.1171721146870014e-07,
      "logits/chosen": 10.26471996307373,
      "logits/rejected": 10.965478897094727,
      "logps/chosen": -132.84579467773438,
      "logps/ref_chosen": -122.57451629638672,
      "logps/ref_rejected": -140.17933654785156,
      "logps/rejected": -150.3111572265625,
      "loss": 0.6942,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.10271281003952026,
      "rewards/margins": -0.001394760562106967,
      "rewards/rejected": -0.10131805390119553,
      "step": 54
    },
    {
      "epoch": 0.724578015644298,
      "grad_norm": 0.43760424852371216,
      "learning_rate": 1.0210694689661939e-07,
      "logits/chosen": 10.27353572845459,
      "logits/rejected": 10.934257507324219,
      "logps/chosen": -128.82815551757812,
      "logps/ref_chosen": -119.25200653076172,
      "logps/ref_rejected": -132.94203186035156,
      "logps/rejected": -143.47572326660156,
      "loss": 0.6887,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -0.09576155245304108,
      "rewards/margins": 0.00957532785832882,
      "rewards/rejected": -0.10533688962459564,
      "step": 55
    },
    {
      "epoch": 0.7377521613832853,
      "grad_norm": 0.5378979444503784,
      "learning_rate": 9.282178335227883e-08,
      "logits/chosen": 9.814579010009766,
      "logits/rejected": 10.796792030334473,
      "logps/chosen": -129.24404907226562,
      "logps/ref_chosen": -119.66452026367188,
      "logps/ref_rejected": -134.48817443847656,
      "logps/rejected": -144.27352905273438,
      "loss": 0.6923,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.09579524397850037,
      "rewards/margins": 0.002058264799416065,
      "rewards/rejected": -0.09785348922014236,
      "step": 56
    },
    {
      "epoch": 0.7509263071222725,
      "grad_norm": 0.486501008272171,
      "learning_rate": 8.388213164029459e-08,
      "logits/chosen": 10.484691619873047,
      "logits/rejected": 11.245904922485352,
      "logps/chosen": -129.30291748046875,
      "logps/ref_chosen": -119.82748413085938,
      "logps/ref_rejected": -132.91250610351562,
      "logps/rejected": -142.54058837890625,
      "loss": 0.6927,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.09475436806678772,
      "rewards/margins": 0.0015264188405126333,
      "rewards/rejected": -0.09628079831600189,
      "step": 57
    },
    {
      "epoch": 0.7641004528612598,
      "grad_norm": 0.5657970309257507,
      "learning_rate": 7.530764305528958e-08,
      "logits/chosen": 9.822986602783203,
      "logits/rejected": 10.402128219604492,
      "logps/chosen": -126.34912872314453,
      "logps/ref_chosen": -116.2895278930664,
      "logps/ref_rejected": -130.56680297851562,
      "logps/rejected": -140.8634033203125,
      "loss": 0.6922,
      "rewards/accuracies": 0.5546875,
      "rewards/chosen": -0.1005958765745163,
      "rewards/margins": 0.002369978930801153,
      "rewards/rejected": -0.10296584665775299,
      "step": 58
    },
    {
      "epoch": 0.7772745986002471,
      "grad_norm": 0.4375581741333008,
      "learning_rate": 6.711716618408281e-08,
      "logits/chosen": 9.962854385375977,
      "logits/rejected": 10.844944953918457,
      "logps/chosen": -128.9604034423828,
      "logps/ref_chosen": -118.71016693115234,
      "logps/ref_rejected": -136.29736328125,
      "logps/rejected": -146.73483276367188,
      "loss": 0.6925,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": -0.10250238329172134,
      "rewards/margins": 0.0018722245004028082,
      "rewards/rejected": -0.10437458753585815,
      "step": 59
    },
    {
      "epoch": 0.7904487443392343,
      "grad_norm": 0.4616609513759613,
      "learning_rate": 5.932870547240454e-08,
      "logits/chosen": 10.142905235290527,
      "logits/rejected": 11.08536148071289,
      "logps/chosen": -129.35299682617188,
      "logps/ref_chosen": -119.55867767333984,
      "logps/ref_rejected": -136.89230346679688,
      "logps/rejected": -147.8154754638672,
      "loss": 0.6879,
      "rewards/accuracies": 0.5703125,
      "rewards/chosen": -0.09794314205646515,
      "rewards/margins": 0.011288601905107498,
      "rewards/rejected": -0.10923174023628235,
      "step": 60
    },
    {
      "epoch": 0.8036228900782215,
      "grad_norm": 0.7305423617362976,
      "learning_rate": 5.1959381647217665e-08,
      "logits/chosen": 9.79527473449707,
      "logits/rejected": 10.603086471557617,
      "logps/chosen": -129.09518432617188,
      "logps/ref_chosen": -118.58332061767578,
      "logps/ref_rejected": -132.3976287841797,
      "logps/rejected": -143.35614013671875,
      "loss": 0.6912,
      "rewards/accuracies": 0.5078125,
      "rewards/chosen": -0.10511861741542816,
      "rewards/margins": 0.004466457758098841,
      "rewards/rejected": -0.10958506911993027,
      "step": 61
    },
    {
      "epoch": 0.8167970358172087,
      "grad_norm": 0.5044032335281372,
      "learning_rate": 4.502539408164385e-08,
      "logits/chosen": 9.983121871948242,
      "logits/rejected": 10.669804573059082,
      "logps/chosen": -132.02255249023438,
      "logps/ref_chosen": -122.35013580322266,
      "logps/ref_rejected": -131.44406127929688,
      "logps/rejected": -140.59652709960938,
      "loss": 0.696,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.09672413021326065,
      "rewards/margins": -0.0051995753310620785,
      "rewards/rejected": -0.09152455627918243,
      "step": 62
    },
    {
      "epoch": 0.829971181556196,
      "grad_norm": 0.4333811402320862,
      "learning_rate": 3.854198518522564e-08,
      "logits/chosen": 9.989500045776367,
      "logits/rejected": 10.936676979064941,
      "logps/chosen": -129.6297149658203,
      "logps/ref_chosen": -119.4000015258789,
      "logps/ref_rejected": -133.5243682861328,
      "logps/rejected": -144.4652557373047,
      "loss": 0.69,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.10229712724685669,
      "rewards/margins": 0.007111664395779371,
      "rewards/rejected": -0.10940880328416824,
      "step": 63
    },
    {
      "epoch": 0.8431453272951832,
      "grad_norm": 0.49472713470458984,
      "learning_rate": 3.2523406897802444e-08,
      "logits/chosen": 10.366823196411133,
      "logits/rejected": 10.912262916564941,
      "logps/chosen": -131.99964904785156,
      "logps/ref_chosen": -122.98699188232422,
      "logps/ref_rejected": -134.78515625,
      "logps/rejected": -145.1294708251953,
      "loss": 0.6868,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.09012652933597565,
      "rewards/margins": 0.013316763564944267,
      "rewards/rejected": -0.10344328731298447,
      "step": 64
    },
    {
      "epoch": 0.8563194730341704,
      "grad_norm": 0.41725417971611023,
      "learning_rate": 2.6982889360653376e-08,
      "logits/chosen": 9.988856315612793,
      "logits/rejected": 10.997881889343262,
      "logps/chosen": -135.51519775390625,
      "logps/ref_chosen": -125.97425079345703,
      "logps/ref_rejected": -139.60340881347656,
      "logps/rejected": -149.2346954345703,
      "loss": 0.6929,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.09540940076112747,
      "rewards/margins": 0.0009036160772666335,
      "rewards/rejected": -0.09631301462650299,
      "step": 65
    },
    {
      "epoch": 0.8694936187731577,
      "grad_norm": 0.4466845393180847,
      "learning_rate": 2.1932611833775843e-08,
      "logits/chosen": 9.85665512084961,
      "logits/rejected": 10.717552185058594,
      "logps/chosen": -129.7871551513672,
      "logps/ref_chosen": -119.83372497558594,
      "logps/ref_rejected": -138.48184204101562,
      "logps/rejected": -149.67349243164062,
      "loss": 0.6873,
      "rewards/accuracies": 0.5859375,
      "rewards/chosen": -0.09953439235687256,
      "rewards/margins": 0.012382248416543007,
      "rewards/rejected": -0.11191663891077042,
      "step": 66
    },
    {
      "epoch": 0.8826677645121449,
      "grad_norm": 1.0083649158477783,
      "learning_rate": 1.738367592322837e-08,
      "logits/chosen": 9.97527027130127,
      "logits/rejected": 10.858278274536133,
      "logps/chosen": -129.06263732910156,
      "logps/ref_chosen": -119.95507049560547,
      "logps/ref_rejected": -135.12179565429688,
      "logps/rejected": -144.7335662841797,
      "loss": 0.6909,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.09107556939125061,
      "rewards/margins": 0.005042179953306913,
      "rewards/rejected": -0.09611774981021881,
      "step": 67
    },
    {
      "epoch": 0.8958419102511321,
      "grad_norm": 0.9535924792289734,
      "learning_rate": 1.3346081177391472e-08,
      "logits/chosen": 10.458059310913086,
      "logits/rejected": 10.82265567779541,
      "logps/chosen": -136.9637451171875,
      "logps/ref_chosen": -127.0732421875,
      "logps/ref_rejected": -134.5887908935547,
      "logps/rejected": -144.5718994140625,
      "loss": 0.6931,
      "rewards/accuracies": 0.453125,
      "rewards/chosen": -0.09890485554933548,
      "rewards/margins": 0.0009263246320188046,
      "rewards/rejected": -0.09983118623495102,
      "step": 68
    },
    {
      "epoch": 0.9090160559901194,
      "grad_norm": 0.6569604873657227,
      "learning_rate": 9.828703105789981e-09,
      "logits/chosen": 10.355306625366211,
      "logits/rejected": 11.095146179199219,
      "logps/chosen": -127.32408142089844,
      "logps/ref_chosen": -118.01029968261719,
      "logps/ref_rejected": -131.78138732910156,
      "logps/rejected": -141.7933807373047,
      "loss": 0.69,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.09313800930976868,
      "rewards/margins": 0.006981834769248962,
      "rewards/rejected": -0.10011984407901764,
      "step": 69
    },
    {
      "epoch": 0.9221902017291066,
      "grad_norm": 0.832196056842804,
      "learning_rate": 6.839273668796747e-09,
      "logits/chosen": 9.993429183959961,
      "logits/rejected": 10.590087890625,
      "logps/chosen": -127.79402160644531,
      "logps/ref_chosen": -117.97954559326172,
      "logps/ref_rejected": -134.86376953125,
      "logps/rejected": -145.40037536621094,
      "loss": 0.6899,
      "rewards/accuracies": 0.5546875,
      "rewards/chosen": -0.09814472496509552,
      "rewards/margins": 0.00722128339111805,
      "rewards/rejected": -0.10536602139472961,
      "step": 70
    },
    {
      "epoch": 0.9353643474680938,
      "grad_norm": 0.5360449552536011,
      "learning_rate": 4.384364281105973e-09,
      "logits/chosen": 9.948233604431152,
      "logits/rejected": 11.157342910766602,
      "logps/chosen": -121.68891906738281,
      "logps/ref_chosen": -112.08084869384766,
      "logps/ref_rejected": -127.31222534179688,
      "logps/rejected": -137.7093963623047,
      "loss": 0.6896,
      "rewards/accuracies": 0.5859375,
      "rewards/chosen": -0.09608055651187897,
      "rewards/margins": 0.007891225628554821,
      "rewards/rejected": -0.10397178679704666,
      "step": 71
    },
    {
      "epoch": 0.9485384932070811,
      "grad_norm": 0.6390266418457031,
      "learning_rate": 2.469371366337264e-09,
      "logits/chosen": 10.44871711730957,
      "logits/rejected": 11.414918899536133,
      "logps/chosen": -129.57125854492188,
      "logps/ref_chosen": -120.54607391357422,
      "logps/ref_rejected": -133.43760681152344,
      "logps/rejected": -143.47120666503906,
      "loss": 0.6884,
      "rewards/accuracies": 0.5390625,
      "rewards/chosen": -0.09025204181671143,
      "rewards/margins": 0.010083984583616257,
      "rewards/rejected": -0.10033603012561798,
      "step": 72
    }
  ],
  "logging_steps": 1,
  "max_steps": 75,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 12,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}