{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0044227465255923925,
  "eval_steps": 500,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.4021127119941484e-05,
      "grad_norm": 33.091712951660156,
      "learning_rate": 1.0000000000000002e-06,
      "logits/chosen": -3.133824586868286,
      "logits/rejected": -3.1356313228607178,
      "logps/chosen": -317.8847961425781,
      "logps/rejected": -306.8866271972656,
      "loss": 0.8544,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.10576057434082031,
      "rewards/margins": 0.2207910716533661,
      "rewards/rejected": -0.3265516459941864,
      "step": 1
    },
    {
      "epoch": 6.804225423988297e-05,
      "grad_norm": 38.49541473388672,
      "learning_rate": 2.0000000000000003e-06,
      "logits/chosen": -3.1433358192443848,
      "logits/rejected": -3.1124918460845947,
      "logps/chosen": -212.93820190429688,
      "logps/rejected": -198.22511291503906,
      "loss": 1.02,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.28638574481010437,
      "rewards/margins": -0.37813109159469604,
      "rewards/rejected": 0.0917452871799469,
      "step": 2
    },
    {
      "epoch": 0.00010206338135982445,
      "grad_norm": 27.203516006469727,
      "learning_rate": 3e-06,
      "logits/chosen": -3.2104451656341553,
      "logits/rejected": -3.2103281021118164,
      "logps/chosen": -235.89254760742188,
      "logps/rejected": -206.0758514404297,
      "loss": 0.605,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.40017586946487427,
      "rewards/margins": 0.24887371063232422,
      "rewards/rejected": 0.15130215883255005,
      "step": 3
    },
    {
      "epoch": 0.00013608450847976594,
      "grad_norm": 26.42736053466797,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -3.0741634368896484,
      "logits/rejected": -2.953117847442627,
      "logps/chosen": -428.7364196777344,
      "logps/rejected": -192.357666015625,
      "loss": 0.4914,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.4332427978515625,
      "rewards/margins": 0.5804362893104553,
      "rewards/rejected": -0.1471935510635376,
      "step": 4
    },
    {
      "epoch": 0.00017010563559970743,
      "grad_norm": 35.623634338378906,
      "learning_rate": 5e-06,
      "logits/chosen": -2.985285758972168,
      "logits/rejected": -2.8854246139526367,
      "logps/chosen": -342.38604736328125,
      "logps/rejected": -211.78237915039062,
      "loss": 0.8273,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -0.3475755751132965,
      "rewards/margins": -0.19371166825294495,
      "rewards/rejected": -0.15386392176151276,
      "step": 5
    },
    {
      "epoch": 0.0002041267627196489,
      "grad_norm": 31.70409393310547,
      "learning_rate": 6e-06,
      "logits/chosen": -3.074589967727661,
      "logits/rejected": -2.85581636428833,
      "logps/chosen": -593.6435546875,
      "logps/rejected": -259.5284729003906,
      "loss": 0.5473,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.2495414763689041,
      "rewards/margins": 0.35888367891311646,
      "rewards/rejected": -0.10934218764305115,
      "step": 6
    },
    {
      "epoch": 0.00023814788983959038,
      "grad_norm": 32.41214370727539,
      "learning_rate": 7.000000000000001e-06,
      "logits/chosen": -2.976675033569336,
      "logits/rejected": -2.978133201599121,
      "logps/chosen": -218.11688232421875,
      "logps/rejected": -231.7698974609375,
      "loss": 0.7124,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -0.08671779930591583,
      "rewards/margins": 0.21950224041938782,
      "rewards/rejected": -0.30622005462646484,
      "step": 7
    },
    {
      "epoch": 0.00027216901695953187,
      "grad_norm": 27.637351989746094,
      "learning_rate": 8.000000000000001e-06,
      "logits/chosen": -2.997600793838501,
      "logits/rejected": -3.1078853607177734,
      "logps/chosen": -165.71444702148438,
      "logps/rejected": -335.5938720703125,
      "loss": 0.6085,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.057393088936805725,
      "rewards/margins": 0.34109365940093994,
      "rewards/rejected": -0.3984867036342621,
      "step": 8
    },
    {
      "epoch": 0.00030619014407947336,
      "grad_norm": 32.246559143066406,
      "learning_rate": 9e-06,
      "logits/chosen": -3.0760390758514404,
      "logits/rejected": -2.938197135925293,
      "logps/chosen": -463.1333312988281,
      "logps/rejected": -173.99951171875,
      "loss": 0.6398,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.5581258535385132,
      "rewards/margins": 0.7653559446334839,
      "rewards/rejected": -0.20723000168800354,
      "step": 9
    },
    {
      "epoch": 0.00034021127119941485,
      "grad_norm": 39.84626388549805,
      "learning_rate": 1e-05,
      "logits/chosen": -3.192596673965454,
      "logits/rejected": -2.9751501083374023,
      "logps/chosen": -473.14532470703125,
      "logps/rejected": -110.90155029296875,
      "loss": 0.9794,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -0.3494076132774353,
      "rewards/margins": -0.31546899676322937,
      "rewards/rejected": -0.03393859416246414,
      "step": 10
    },
    {
      "epoch": 0.00037423239831935634,
      "grad_norm": 39.444889068603516,
      "learning_rate": 1.1000000000000001e-05,
      "logits/chosen": -3.096883535385132,
      "logits/rejected": -3.043616771697998,
      "logps/chosen": -391.7703552246094,
      "logps/rejected": -271.8331604003906,
      "loss": 0.8029,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.6240158081054688,
      "rewards/margins": 0.4651091992855072,
      "rewards/rejected": 0.15890654921531677,
      "step": 11
    },
    {
      "epoch": 0.0004082535254392978,
      "grad_norm": 40.78356170654297,
      "learning_rate": 1.2e-05,
      "logits/chosen": -3.000021457672119,
      "logits/rejected": -3.0893115997314453,
      "logps/chosen": -255.94607543945312,
      "logps/rejected": -405.3093566894531,
      "loss": 0.6297,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.33697396516799927,
      "rewards/margins": 0.3315456509590149,
      "rewards/rejected": 0.005428321659564972,
      "step": 12
    },
    {
      "epoch": 0.00044227465255923927,
      "grad_norm": 35.69674301147461,
      "learning_rate": 1.3000000000000001e-05,
      "logits/chosen": -3.011014699935913,
      "logits/rejected": -3.0214414596557617,
      "logps/chosen": -223.4480438232422,
      "logps/rejected": -261.62469482421875,
      "loss": 0.815,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.7416718006134033,
      "rewards/margins": 0.19626998901367188,
      "rewards/rejected": 0.5454018115997314,
      "step": 13
    },
    {
      "epoch": 0.00047629577967918076,
      "grad_norm": 32.88405990600586,
      "learning_rate": 1.4000000000000001e-05,
      "logits/chosen": -3.065549373626709,
      "logits/rejected": -3.1094508171081543,
      "logps/chosen": -169.08047485351562,
      "logps/rejected": -227.85162353515625,
      "loss": 0.8115,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 0.37659722566604614,
      "rewards/margins": -0.11390496790409088,
      "rewards/rejected": 0.4905022084712982,
      "step": 14
    },
    {
      "epoch": 0.0005103169067991223,
      "grad_norm": 20.918012619018555,
      "learning_rate": 1.5e-05,
      "logits/chosen": -3.045846462249756,
      "logits/rejected": -2.9114990234375,
      "logps/chosen": -573.1253662109375,
      "logps/rejected": -273.9326477050781,
      "loss": 0.3105,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7755428552627563,
      "rewards/margins": 1.082038164138794,
      "rewards/rejected": -0.3064952790737152,
      "step": 15
    },
    {
      "epoch": 0.0005443380339190637,
      "grad_norm": 38.434749603271484,
      "learning_rate": 1.6000000000000003e-05,
      "logits/chosen": -3.1250810623168945,
      "logits/rejected": -2.96527099609375,
      "logps/chosen": -396.9795837402344,
      "logps/rejected": -174.92918395996094,
      "loss": 0.9971,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.2244594693183899,
      "rewards/margins": -0.37520748376846313,
      "rewards/rejected": 0.5996670126914978,
      "step": 16
    },
    {
      "epoch": 0.0005783591610390052,
      "grad_norm": 37.71885681152344,
      "learning_rate": 1.7000000000000003e-05,
      "logits/chosen": -2.970050096511841,
      "logits/rejected": -3.0346829891204834,
      "logps/chosen": -236.67071533203125,
      "logps/rejected": -403.0212097167969,
      "loss": 1.0307,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.9259212613105774,
      "rewards/margins": -0.23367927968502045,
      "rewards/rejected": 1.1596004962921143,
      "step": 17
    },
    {
      "epoch": 0.0006123802881589467,
      "grad_norm": 25.964290618896484,
      "learning_rate": 1.8e-05,
      "logits/chosen": -3.0690722465515137,
      "logits/rejected": -3.0348947048187256,
      "logps/chosen": -312.81317138671875,
      "logps/rejected": -259.230712890625,
      "loss": 0.6989,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.6348915100097656,
      "rewards/margins": 0.6418477892875671,
      "rewards/rejected": -0.006956290453672409,
      "step": 18
    },
    {
      "epoch": 0.0006464014152788882,
      "grad_norm": 27.076108932495117,
      "learning_rate": 1.9e-05,
      "logits/chosen": -3.135136842727661,
      "logits/rejected": -2.911625862121582,
      "logps/chosen": -473.7462158203125,
      "logps/rejected": -141.9491729736328,
      "loss": 0.6027,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.9424793720245361,
      "rewards/margins": 0.9380611181259155,
      "rewards/rejected": 0.00441819429397583,
      "step": 19
    },
    {
      "epoch": 0.0006804225423988297,
      "grad_norm": 51.820159912109375,
      "learning_rate": 2e-05,
      "logits/chosen": -2.8403027057647705,
      "logits/rejected": -3.0664803981781006,
      "logps/chosen": -219.79153442382812,
      "logps/rejected": -231.70004272460938,
      "loss": 1.1022,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.052263081073760986,
      "rewards/margins": -0.6957410573959351,
      "rewards/rejected": 0.7480041980743408,
      "step": 20
    },
    {
      "epoch": 0.0007144436695187712,
      "grad_norm": 45.9146614074707,
      "learning_rate": 2.1e-05,
      "logits/chosen": -2.9764764308929443,
      "logits/rejected": -2.9391565322875977,
      "logps/chosen": -458.0147705078125,
      "logps/rejected": -218.37648010253906,
      "loss": 1.0772,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.8049196004867554,
      "rewards/margins": -0.448437362909317,
      "rewards/rejected": 1.2533570528030396,
      "step": 21
    },
    {
      "epoch": 0.0007484647966387127,
      "grad_norm": 22.854082107543945,
      "learning_rate": 2.2000000000000003e-05,
      "logits/chosen": -3.1029181480407715,
      "logits/rejected": -3.044419527053833,
      "logps/chosen": -154.8139190673828,
      "logps/rejected": -155.980712890625,
      "loss": 0.4065,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.45841503143310547,
      "rewards/margins": 0.8753025531768799,
      "rewards/rejected": -0.41688746213912964,
      "step": 22
    },
    {
      "epoch": 0.0007824859237586541,
      "grad_norm": 43.94650650024414,
      "learning_rate": 2.3000000000000003e-05,
      "logits/chosen": -2.9984076023101807,
      "logits/rejected": -3.114562511444092,
      "logps/chosen": -263.8316650390625,
      "logps/rejected": -340.1427307128906,
      "loss": 1.0427,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 0.6933631896972656,
      "rewards/margins": -0.14637237787246704,
      "rewards/rejected": 0.8397356271743774,
      "step": 23
    },
    {
      "epoch": 0.0008165070508785956,
      "grad_norm": 49.11759948730469,
      "learning_rate": 2.4e-05,
      "logits/chosen": -3.025609016418457,
      "logits/rejected": -3.0236239433288574,
      "logps/chosen": -412.2383117675781,
      "logps/rejected": -643.138427734375,
      "loss": 1.0374,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.1175220012664795,
      "rewards/margins": -0.31503716111183167,
      "rewards/rejected": 1.4325592517852783,
      "step": 24
    },
    {
      "epoch": 0.000850528177998537,
      "grad_norm": 37.57383728027344,
      "learning_rate": 2.5e-05,
      "logits/chosen": -3.024618625640869,
      "logits/rejected": -3.123701572418213,
      "logps/chosen": -201.70530700683594,
      "logps/rejected": -224.99038696289062,
      "loss": 0.7477,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 0.9206753373146057,
      "rewards/margins": 0.5947309732437134,
      "rewards/rejected": 0.32594436407089233,
      "step": 25
    },
    {
      "epoch": 0.0008845493051184785,
      "grad_norm": 40.297794342041016,
      "learning_rate": 2.6000000000000002e-05,
      "logits/chosen": -3.1100215911865234,
      "logits/rejected": -3.0195024013519287,
      "logps/chosen": -226.2320098876953,
      "logps/rejected": -193.0676727294922,
      "loss": 0.8151,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.9122087359428406,
      "rewards/margins": 0.0203973650932312,
      "rewards/rejected": 0.8918113708496094,
      "step": 26
    },
    {
      "epoch": 0.00091857043223842,
      "grad_norm": 43.394229888916016,
      "learning_rate": 2.7000000000000002e-05,
      "logits/chosen": -3.0187630653381348,
      "logits/rejected": -3.053678035736084,
      "logps/chosen": -153.02883911132812,
      "logps/rejected": -260.0540771484375,
      "loss": 1.3457,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.7145153284072876,
      "rewards/margins": -0.806611180305481,
      "rewards/rejected": 1.5211265087127686,
      "step": 27
    },
    {
      "epoch": 0.0009525915593583615,
      "grad_norm": 28.59680938720703,
      "learning_rate": 2.8000000000000003e-05,
      "logits/chosen": -3.032409191131592,
      "logits/rejected": -3.0502161979675293,
      "logps/chosen": -240.1644287109375,
      "logps/rejected": -245.81436157226562,
      "loss": 0.5501,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.0826761722564697,
      "rewards/margins": 0.3433869779109955,
      "rewards/rejected": 1.7392891645431519,
      "step": 28
    },
    {
      "epoch": 0.000986612686478303,
      "grad_norm": 37.069583892822266,
      "learning_rate": 2.9e-05,
      "logits/chosen": -3.0516104698181152,
      "logits/rejected": -3.144774913787842,
      "logps/chosen": -296.03125,
      "logps/rejected": -292.9052429199219,
      "loss": 1.1087,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.8324771523475647,
      "rewards/margins": -0.34529268741607666,
      "rewards/rejected": 1.1777698993682861,
      "step": 29
    },
    {
      "epoch": 0.0010206338135982445,
      "grad_norm": 30.238658905029297,
      "learning_rate": 3e-05,
      "logits/chosen": -3.064671039581299,
      "logits/rejected": -2.8963613510131836,
      "logps/chosen": -415.6085510253906,
      "logps/rejected": -149.9669952392578,
      "loss": 0.4892,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.0249770879745483,
      "rewards/margins": 0.7549355030059814,
      "rewards/rejected": 0.27004164457321167,
      "step": 30
    },
    {
      "epoch": 0.001054654940718186,
      "grad_norm": 40.35402297973633,
      "learning_rate": 3.1e-05,
      "logits/chosen": -3.1762304306030273,
      "logits/rejected": -3.061676025390625,
      "logps/chosen": -585.9200439453125,
      "logps/rejected": -238.16307067871094,
      "loss": 0.587,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.726571559906006,
      "rewards/margins": 0.9808880090713501,
      "rewards/rejected": 1.7456837892532349,
      "step": 31
    },
    {
      "epoch": 0.0010886760678381275,
      "grad_norm": 33.78396987915039,
      "learning_rate": 3.2000000000000005e-05,
      "logits/chosen": -3.05906081199646,
      "logits/rejected": -2.9525132179260254,
      "logps/chosen": -580.87646484375,
      "logps/rejected": -223.65457153320312,
      "loss": 0.8173,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.3172895908355713,
      "rewards/margins": 0.07103176414966583,
      "rewards/rejected": 1.2462577819824219,
      "step": 32
    },
    {
      "epoch": 0.001122697194958069,
      "grad_norm": 40.97666549682617,
      "learning_rate": 3.3e-05,
      "logits/chosen": -3.084476947784424,
      "logits/rejected": -3.11270809173584,
      "logps/chosen": -285.040283203125,
      "logps/rejected": -251.92686462402344,
      "loss": 1.0645,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.4826446771621704,
      "rewards/margins": -0.13718131184577942,
      "rewards/rejected": 1.6198259592056274,
      "step": 33
    },
    {
      "epoch": 0.0011567183220780105,
      "grad_norm": 33.32075881958008,
      "learning_rate": 3.4000000000000007e-05,
      "logits/chosen": -3.037482500076294,
      "logits/rejected": -3.0169787406921387,
      "logps/chosen": -467.7403869628906,
      "logps/rejected": -308.62274169921875,
      "loss": 0.5226,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.4038407802581787,
      "rewards/margins": 0.8624236583709717,
      "rewards/rejected": 1.541416883468628,
      "step": 34
    },
    {
      "epoch": 0.001190739449197952,
      "grad_norm": 26.35866928100586,
      "learning_rate": 3.5e-05,
      "logits/chosen": -2.969816207885742,
      "logits/rejected": -2.9069106578826904,
      "logps/chosen": -312.856689453125,
      "logps/rejected": -183.8594512939453,
      "loss": 0.4495,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.1757181882858276,
      "rewards/margins": 0.803225576877594,
      "rewards/rejected": 0.37249261140823364,
      "step": 35
    },
    {
      "epoch": 0.0012247605763178934,
      "grad_norm": 44.118099212646484,
      "learning_rate": 3.6e-05,
      "logits/chosen": -3.091865301132202,
      "logits/rejected": -3.0135951042175293,
      "logps/chosen": -724.1255493164062,
      "logps/rejected": -335.2945861816406,
      "loss": 0.6341,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.7523369789123535,
      "rewards/margins": 0.3912923336029053,
      "rewards/rejected": 1.3610445261001587,
      "step": 36
    },
    {
      "epoch": 0.001258781703437835,
      "grad_norm": 26.80854606628418,
      "learning_rate": 3.7e-05,
      "logits/chosen": -3.0334794521331787,
      "logits/rejected": -2.9878833293914795,
      "logps/chosen": -406.20745849609375,
      "logps/rejected": -377.2811584472656,
      "loss": 0.5577,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.6232261657714844,
      "rewards/margins": 0.9015346765518188,
      "rewards/rejected": 1.7216914892196655,
      "step": 37
    },
    {
      "epoch": 0.0012928028305577764,
      "grad_norm": 31.345367431640625,
      "learning_rate": 3.8e-05,
      "logits/chosen": -3.0986504554748535,
      "logits/rejected": -3.062534809112549,
      "logps/chosen": -412.54498291015625,
      "logps/rejected": -386.44158935546875,
      "loss": 0.5479,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.985120177268982,
      "rewards/margins": 0.8609979748725891,
      "rewards/rejected": 1.1241222620010376,
      "step": 38
    },
    {
      "epoch": 0.001326823957677718,
      "grad_norm": 15.26794719696045,
      "learning_rate": 3.9000000000000006e-05,
      "logits/chosen": -3.0730984210968018,
      "logits/rejected": -3.0951075553894043,
      "logps/chosen": -305.12982177734375,
      "logps/rejected": -285.45001220703125,
      "loss": 0.2547,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.55714750289917,
      "rewards/margins": 1.3561341762542725,
      "rewards/rejected": 1.2010130882263184,
      "step": 39
    },
    {
      "epoch": 0.0013608450847976594,
      "grad_norm": 22.881174087524414,
      "learning_rate": 4e-05,
      "logits/chosen": -3.162843942642212,
      "logits/rejected": -3.15628981590271,
      "logps/chosen": -277.48626708984375,
      "logps/rejected": -219.9507293701172,
      "loss": 0.5427,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.6990911960601807,
      "rewards/margins": 0.5041837692260742,
      "rewards/rejected": 2.1949074268341064,
      "step": 40
    },
    {
      "epoch": 0.001394866211917601,
      "grad_norm": 15.179062843322754,
      "learning_rate": 4.1e-05,
      "logits/chosen": -3.138144016265869,
      "logits/rejected": -3.1033153533935547,
      "logps/chosen": -234.6917724609375,
      "logps/rejected": -141.10888671875,
      "loss": 0.2626,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.799344539642334,
      "rewards/margins": 1.4650464057922363,
      "rewards/rejected": 1.3342978954315186,
      "step": 41
    },
    {
      "epoch": 0.0014288873390375424,
      "grad_norm": 21.09362030029297,
      "learning_rate": 4.2e-05,
      "logits/chosen": -3.0385375022888184,
      "logits/rejected": -3.083676338195801,
      "logps/chosen": -168.33697509765625,
      "logps/rejected": -179.30813598632812,
      "loss": 0.4002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.3925881385803223,
      "rewards/margins": 0.8477088809013367,
      "rewards/rejected": 1.5448791980743408,
      "step": 42
    },
    {
      "epoch": 0.0014629084661574839,
      "grad_norm": 38.255069732666016,
      "learning_rate": 4.3e-05,
      "logits/chosen": -3.080360174179077,
      "logits/rejected": -3.029470443725586,
      "logps/chosen": -267.5482177734375,
      "logps/rejected": -161.17111206054688,
      "loss": 0.8857,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.8575999736785889,
      "rewards/margins": -0.24116593599319458,
      "rewards/rejected": 2.0987658500671387,
      "step": 43
    },
    {
      "epoch": 0.0014969295932774254,
      "grad_norm": 27.97095489501953,
      "learning_rate": 4.4000000000000006e-05,
      "logits/chosen": -3.1322476863861084,
      "logits/rejected": -3.1334149837493896,
      "logps/chosen": -94.13690185546875,
      "logps/rejected": -78.98738098144531,
      "loss": 0.9987,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.6137428283691406,
      "rewards/margins": -0.4127828776836395,
      "rewards/rejected": 2.0265257358551025,
      "step": 44
    },
    {
      "epoch": 0.0015309507203973669,
      "grad_norm": 45.25028991699219,
      "learning_rate": 4.5e-05,
      "logits/chosen": -3.121274709701538,
      "logits/rejected": -3.1236798763275146,
      "logps/chosen": -516.3858642578125,
      "logps/rejected": -212.694580078125,
      "loss": 0.9457,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 2.419821262359619,
      "rewards/margins": -0.38284340500831604,
      "rewards/rejected": 2.8026647567749023,
      "step": 45
    },
    {
      "epoch": 0.0015649718475173081,
      "grad_norm": 23.58992576599121,
      "learning_rate": 4.600000000000001e-05,
      "logits/chosen": -3.113922357559204,
      "logits/rejected": -3.0506935119628906,
      "logps/chosen": -318.288330078125,
      "logps/rejected": -124.70735168457031,
      "loss": 0.5671,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.407118320465088,
      "rewards/margins": 0.30923187732696533,
      "rewards/rejected": 2.097886323928833,
      "step": 46
    },
    {
      "epoch": 0.0015989929746372496,
      "grad_norm": 36.93040084838867,
      "learning_rate": 4.7e-05,
      "logits/chosen": -3.079291582107544,
      "logits/rejected": -3.1158454418182373,
      "logps/chosen": -246.26065063476562,
      "logps/rejected": -336.46649169921875,
      "loss": 0.9352,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 2.214073657989502,
      "rewards/margins": -0.2623659074306488,
      "rewards/rejected": 2.4764394760131836,
      "step": 47
    },
    {
      "epoch": 0.0016330141017571911,
      "grad_norm": 32.5983772277832,
      "learning_rate": 4.8e-05,
      "logits/chosen": -3.155709981918335,
      "logits/rejected": -3.1829142570495605,
      "logps/chosen": -204.55328369140625,
      "logps/rejected": -219.52647399902344,
      "loss": 0.8805,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.155839681625366,
      "rewards/margins": -0.002133697271347046,
      "rewards/rejected": 2.157973289489746,
      "step": 48
    },
    {
      "epoch": 0.0016670352288771326,
      "grad_norm": 35.626869201660156,
      "learning_rate": 4.9e-05,
      "logits/chosen": -3.10532546043396,
      "logits/rejected": -3.113755464553833,
      "logps/chosen": -390.5479736328125,
      "logps/rejected": -351.7720031738281,
      "loss": 0.7657,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.5670056343078613,
      "rewards/margins": -0.04349440336227417,
      "rewards/rejected": 2.6105000972747803,
      "step": 49
    },
    {
      "epoch": 0.001701056355997074,
      "grad_norm": 37.51042938232422,
      "learning_rate": 5e-05,
      "logits/chosen": -3.122190237045288,
      "logits/rejected": -3.0837321281433105,
      "logps/chosen": -251.68600463867188,
      "logps/rejected": -242.4346923828125,
      "loss": 0.9539,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.0532169342041016,
      "rewards/margins": -0.32518482208251953,
      "rewards/rejected": 2.378401756286621,
      "step": 50
    },
    {
      "epoch": 0.0017350774831170156,
      "grad_norm": 43.16423416137695,
      "learning_rate": 4.999939076763487e-05,
      "logits/chosen": -3.0424001216888428,
      "logits/rejected": -3.0246453285217285,
      "logps/chosen": -228.8073272705078,
      "logps/rejected": -399.57781982421875,
      "loss": 1.204,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.1423535346984863,
      "rewards/margins": -0.628890335559845,
      "rewards/rejected": 2.7712438106536865,
      "step": 51
    },
    {
      "epoch": 0.001769098610236957,
      "grad_norm": 32.16966247558594,
      "learning_rate": 4.999756310023261e-05,
      "logits/chosen": -3.049051284790039,
      "logits/rejected": -3.061192750930786,
      "logps/chosen": -219.27503967285156,
      "logps/rejected": -238.98226928710938,
      "loss": 0.9616,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.1868481636047363,
      "rewards/margins": -0.2618050277233124,
      "rewards/rejected": 2.448652744293213,
      "step": 52
    },
    {
      "epoch": 0.0018031197373568986,
      "grad_norm": 38.5218505859375,
      "learning_rate": 4.999451708687114e-05,
      "logits/chosen": -3.088409185409546,
      "logits/rejected": -3.0845394134521484,
      "logps/chosen": -131.70074462890625,
      "logps/rejected": -205.93624877929688,
      "loss": 1.0419,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.822286605834961,
      "rewards/margins": -0.18211710453033447,
      "rewards/rejected": 2.004403829574585,
      "step": 53
    },
    {
      "epoch": 0.00183714086447684,
      "grad_norm": 31.46576690673828,
      "learning_rate": 4.999025287600886e-05,
      "logits/chosen": -3.143968105316162,
      "logits/rejected": -3.032649278640747,
      "logps/chosen": -403.4815673828125,
      "logps/rejected": -195.69281005859375,
      "loss": 0.567,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.2466557025909424,
      "rewards/margins": 0.6339659094810486,
      "rewards/rejected": 1.612689733505249,
      "step": 54
    },
    {
      "epoch": 0.0018711619915967816,
      "grad_norm": 22.339370727539062,
      "learning_rate": 4.99847706754774e-05,
      "logits/chosen": -3.1364409923553467,
      "logits/rejected": -3.0678646564483643,
      "logps/chosen": -272.07330322265625,
      "logps/rejected": -94.81082153320312,
      "loss": 0.4856,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9992039203643799,
      "rewards/margins": 0.5075141191482544,
      "rewards/rejected": 1.491689682006836,
      "step": 55
    },
    {
      "epoch": 0.001905183118716723,
      "grad_norm": 38.88309860229492,
      "learning_rate": 4.997807075247146e-05,
      "logits/chosen": -3.0754241943359375,
      "logits/rejected": -3.1445722579956055,
      "logps/chosen": -161.1916961669922,
      "logps/rejected": -230.43589782714844,
      "loss": 0.8925,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.070554256439209,
      "rewards/margins": -0.027551889419555664,
      "rewards/rejected": 2.0981061458587646,
      "step": 56
    },
    {
      "epoch": 0.0019392042458366645,
      "grad_norm": 29.006982803344727,
      "learning_rate": 4.997015343353585e-05,
      "logits/chosen": -3.079509973526001,
      "logits/rejected": -3.119426727294922,
      "logps/chosen": -236.40664672851562,
      "logps/rejected": -259.257568359375,
      "loss": 0.5812,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.4261646270751953,
      "rewards/margins": 0.7371253371238708,
      "rewards/rejected": 1.6890392303466797,
      "step": 57
    },
    {
      "epoch": 0.001973225372956606,
      "grad_norm": 18.193532943725586,
      "learning_rate": 4.996101910454953e-05,
      "logits/chosen": -3.070995569229126,
      "logits/rejected": -2.8317549228668213,
      "logps/chosen": -730.03759765625,
      "logps/rejected": -133.17929077148438,
      "loss": 0.3401,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 3.711276054382324,
      "rewards/margins": 1.7915010452270508,
      "rewards/rejected": 1.9197750091552734,
      "step": 58
    },
    {
      "epoch": 0.0020072465000765477,
      "grad_norm": 35.50777053833008,
      "learning_rate": 4.995066821070679e-05,
      "logits/chosen": -3.093824863433838,
      "logits/rejected": -3.083796739578247,
      "logps/chosen": -238.11883544921875,
      "logps/rejected": -151.69790649414062,
      "loss": 0.6251,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.0124969482421875,
      "rewards/margins": 0.640554666519165,
      "rewards/rejected": 1.371942400932312,
      "step": 59
    },
    {
      "epoch": 0.002041267627196489,
      "grad_norm": 40.34638977050781,
      "learning_rate": 4.993910125649561e-05,
      "logits/chosen": -3.1637063026428223,
      "logits/rejected": -3.055285930633545,
      "logps/chosen": -365.89434814453125,
      "logps/rejected": -187.11392211914062,
      "loss": 1.1514,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.089510679244995,
      "rewards/margins": -0.3508464992046356,
      "rewards/rejected": 2.440357208251953,
      "step": 60
    },
    {
      "epoch": 0.0020752887543164307,
      "grad_norm": 31.27332878112793,
      "learning_rate": 4.992631880567301e-05,
      "logits/chosen": -2.9931535720825195,
      "logits/rejected": -2.994088649749756,
      "logps/chosen": -127.87042236328125,
      "logps/rejected": -236.82327270507812,
      "loss": 0.8745,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.9324662685394287,
      "rewards/margins": -0.040864601731300354,
      "rewards/rejected": 1.9733309745788574,
      "step": 61
    },
    {
      "epoch": 0.002109309881436372,
      "grad_norm": 24.1131591796875,
      "learning_rate": 4.991232148123761e-05,
      "logits/chosen": -3.094635009765625,
      "logits/rejected": -3.060533046722412,
      "logps/chosen": -369.5483093261719,
      "logps/rejected": -167.15155029296875,
      "loss": 0.46,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.38918399810791,
      "rewards/margins": 1.1014902591705322,
      "rewards/rejected": 1.2876936197280884,
      "step": 62
    },
    {
      "epoch": 0.0021433310085563133,
      "grad_norm": 49.346961975097656,
      "learning_rate": 4.989710996539926e-05,
      "logits/chosen": -3.1134653091430664,
      "logits/rejected": -3.0323257446289062,
      "logps/chosen": -273.39361572265625,
      "logps/rejected": -157.84652709960938,
      "loss": 1.023,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.1458964347839355,
      "rewards/margins": -0.5081260204315186,
      "rewards/rejected": 1.654022455215454,
      "step": 63
    },
    {
      "epoch": 0.002177352135676255,
      "grad_norm": 41.145633697509766,
      "learning_rate": 4.988068499954578e-05,
      "logits/chosen": -2.9834160804748535,
      "logits/rejected": -2.9958994388580322,
      "logps/chosen": -334.5246887207031,
      "logps/rejected": -386.3387145996094,
      "loss": 1.5388,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.305959939956665,
      "rewards/margins": 0.07403802871704102,
      "rewards/rejected": 2.231921911239624,
      "step": 64
    },
    {
      "epoch": 0.0022113732627961962,
      "grad_norm": 33.07326126098633,
      "learning_rate": 4.9863047384206835e-05,
      "logits/chosen": -3.0006370544433594,
      "logits/rejected": -3.020918846130371,
      "logps/chosen": -344.7704162597656,
      "logps/rejected": -200.45114135742188,
      "loss": 0.5845,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.643819808959961,
      "rewards/margins": 0.6079027652740479,
      "rewards/rejected": 1.035917043685913,
      "step": 65
    },
    {
      "epoch": 0.002245394389916138,
      "grad_norm": 51.295345306396484,
      "learning_rate": 4.984419797901491e-05,
      "logits/chosen": -3.039350986480713,
      "logits/rejected": -3.030729293823242,
      "logps/chosen": -307.3952331542969,
      "logps/rejected": -238.56224060058594,
      "loss": 1.2331,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 0.6686372756958008,
      "rewards/margins": -0.6910279989242554,
      "rewards/rejected": 1.3596653938293457,
      "step": 66
    },
    {
      "epoch": 0.0022794155170360792,
      "grad_norm": 27.732166290283203,
      "learning_rate": 4.982413770266342e-05,
      "logits/chosen": -3.092078685760498,
      "logits/rejected": -3.1104204654693604,
      "logps/chosen": -206.7743682861328,
      "logps/rejected": -112.66790771484375,
      "loss": 0.5552,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.273049235343933,
      "rewards/margins": 0.48671725392341614,
      "rewards/rejected": 0.7863319516181946,
      "step": 67
    },
    {
      "epoch": 0.002313436644156021,
      "grad_norm": 46.03068161010742,
      "learning_rate": 4.980286753286195e-05,
      "logits/chosen": -2.9219119548797607,
      "logits/rejected": -2.973864793777466,
      "logps/chosen": -455.3492431640625,
      "logps/rejected": -422.7337646484375,
      "loss": 0.8967,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.788267731666565,
      "rewards/margins": -0.24189908802509308,
      "rewards/rejected": 2.0301668643951416,
      "step": 68
    },
    {
      "epoch": 0.002347457771275962,
      "grad_norm": 24.625539779663086,
      "learning_rate": 4.978038850628854e-05,
      "logits/chosen": -2.9824914932250977,
      "logits/rejected": -2.923306703567505,
      "logps/chosen": -320.2785339355469,
      "logps/rejected": -199.95407104492188,
      "loss": 0.5542,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.4030070304870605,
      "rewards/margins": 1.2227776050567627,
      "rewards/rejected": 1.180229663848877,
      "step": 69
    },
    {
      "epoch": 0.002381478898395904,
      "grad_norm": 52.87644958496094,
      "learning_rate": 4.975670171853926e-05,
      "logits/chosen": -2.9576680660247803,
      "logits/rejected": -3.113696336746216,
      "logps/chosen": -226.1188201904297,
      "logps/rejected": -472.0656433105469,
      "loss": 1.5976,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.8890628814697266,
      "rewards/margins": -1.2377982139587402,
      "rewards/rejected": 3.126861095428467,
      "step": 70
    },
    {
      "epoch": 0.002415500025515845,
      "grad_norm": 41.89684295654297,
      "learning_rate": 4.9731808324074717e-05,
      "logits/chosen": -2.963118553161621,
      "logits/rejected": -2.95782732963562,
      "logps/chosen": -376.9123229980469,
      "logps/rejected": -408.6007080078125,
      "loss": 0.8305,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.2748968601226807,
      "rewards/margins": 1.4728538990020752,
      "rewards/rejected": 0.8020429611206055,
      "step": 71
    },
    {
      "epoch": 0.002449521152635787,
      "grad_norm": 33.66620635986328,
      "learning_rate": 4.9705709536163824e-05,
      "logits/chosen": -2.951019287109375,
      "logits/rejected": -2.8525712490081787,
      "logps/chosen": -334.75689697265625,
      "logps/rejected": -256.43963623046875,
      "loss": 0.5852,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.2168543338775635,
      "rewards/margins": 1.3764129877090454,
      "rewards/rejected": 0.8404413461685181,
      "step": 72
    },
    {
      "epoch": 0.002483542279755728,
      "grad_norm": 48.537376403808594,
      "learning_rate": 4.96784066268247e-05,
      "logits/chosen": -2.96649169921875,
      "logits/rejected": -2.877800703048706,
      "logps/chosen": -342.9018859863281,
      "logps/rejected": -166.73826599121094,
      "loss": 1.3147,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 1.4513309001922607,
      "rewards/margins": -0.9881281852722168,
      "rewards/rejected": 2.4394590854644775,
      "step": 73
    },
    {
      "epoch": 0.00251756340687567,
      "grad_norm": 35.8758544921875,
      "learning_rate": 4.964990092676263e-05,
      "logits/chosen": -3.007376194000244,
      "logits/rejected": -3.0710248947143555,
      "logps/chosen": -322.75604248046875,
      "logps/rejected": -288.9136047363281,
      "loss": 1.0483,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.328976035118103,
      "rewards/margins": -0.057621777057647705,
      "rewards/rejected": 1.3865978717803955,
      "step": 74
    },
    {
      "epoch": 0.002551584533995611,
      "grad_norm": 52.42377471923828,
      "learning_rate": 4.962019382530521e-05,
      "logits/chosen": -2.952801465988159,
      "logits/rejected": -2.9580321311950684,
      "logps/chosen": -272.15576171875,
      "logps/rejected": -230.2147674560547,
      "loss": 1.4341,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.2654911279678345,
      "rewards/margins": -0.9303737282752991,
      "rewards/rejected": 2.1958649158477783,
      "step": 75
    },
    {
      "epoch": 0.002585605661115553,
      "grad_norm": 54.157386779785156,
      "learning_rate": 4.9589286770334654e-05,
      "logits/chosen": -2.997375965118408,
      "logits/rejected": -3.091611385345459,
      "logps/chosen": -219.73159790039062,
      "logps/rejected": -503.797119140625,
      "loss": 1.3568,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.520772933959961,
      "rewards/margins": -0.4154503643512726,
      "rewards/rejected": 1.9362232685089111,
      "step": 76
    },
    {
      "epoch": 0.002619626788235494,
      "grad_norm": 40.647274017333984,
      "learning_rate": 4.9557181268217227e-05,
      "logits/chosen": -3.0128493309020996,
      "logits/rejected": -2.9981260299682617,
      "logps/chosen": -319.1977233886719,
      "logps/rejected": -250.11123657226562,
      "loss": 0.7922,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.9633705615997314,
      "rewards/margins": -0.027412444353103638,
      "rewards/rejected": 1.9907830953598022,
      "step": 77
    },
    {
      "epoch": 0.002653647915355436,
      "grad_norm": 46.99341583251953,
      "learning_rate": 4.952387888372979e-05,
      "logits/chosen": -3.0307843685150146,
      "logits/rejected": -3.070122480392456,
      "logps/chosen": -134.52757263183594,
      "logps/rejected": -225.51242065429688,
      "loss": 1.182,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 1.2856706380844116,
      "rewards/margins": -0.7930065393447876,
      "rewards/rejected": 2.078677177429199,
      "step": 78
    },
    {
      "epoch": 0.002687669042475377,
      "grad_norm": 26.92460823059082,
      "learning_rate": 4.94893812399836e-05,
      "logits/chosen": -2.995112895965576,
      "logits/rejected": -2.950528144836426,
      "logps/chosen": -560.2005615234375,
      "logps/rejected": -383.92523193359375,
      "loss": 0.5477,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.8620712757110596,
      "rewards/margins": 0.3632301390171051,
      "rewards/rejected": 1.4988411664962769,
      "step": 79
    },
    {
      "epoch": 0.002721690169595319,
      "grad_norm": 18.80535125732422,
      "learning_rate": 4.9453690018345144e-05,
      "logits/chosen": -3.0887956619262695,
      "logits/rejected": -3.013925790786743,
      "logps/chosen": -201.7890167236328,
      "logps/rejected": -120.17092895507812,
      "loss": 0.6049,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.207493305206299,
      "rewards/margins": 0.5624717473983765,
      "rewards/rejected": 1.645021677017212,
      "step": 80
    },
    {
      "epoch": 0.00275571129671526,
      "grad_norm": 35.741817474365234,
      "learning_rate": 4.94168069583542e-05,
      "logits/chosen": -3.087347984313965,
      "logits/rejected": -2.9956464767456055,
      "logps/chosen": -485.24749755859375,
      "logps/rejected": -269.6015930175781,
      "loss": 1.0357,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.650765299797058,
      "rewards/margins": -0.448674738407135,
      "rewards/rejected": 2.099440097808838,
      "step": 81
    },
    {
      "epoch": 0.002789732423835202,
      "grad_norm": 25.36280059814453,
      "learning_rate": 4.937873385763908e-05,
      "logits/chosen": -2.9839279651641846,
      "logits/rejected": -3.0026051998138428,
      "logps/chosen": -221.44424438476562,
      "logps/rejected": -193.8582305908203,
      "loss": 0.9423,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.4655239582061768,
      "rewards/margins": -0.0549183189868927,
      "rewards/rejected": 1.520442247390747,
      "step": 82
    },
    {
      "epoch": 0.002823753550955143,
      "grad_norm": 12.532795906066895,
      "learning_rate": 4.933947257182901e-05,
      "logits/chosen": -3.0249555110931396,
      "logits/rejected": -3.0148353576660156,
      "logps/chosen": -386.8382263183594,
      "logps/rejected": -106.46695709228516,
      "loss": 0.3462,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 3.745790481567383,
      "rewards/margins": 1.7120976448059082,
      "rewards/rejected": 2.0336928367614746,
      "step": 83
    },
    {
      "epoch": 0.002857774678075085,
      "grad_norm": 39.60961151123047,
      "learning_rate": 4.929902501446366e-05,
      "logits/chosen": -3.058694362640381,
      "logits/rejected": -3.0021092891693115,
      "logps/chosen": -269.97113037109375,
      "logps/rejected": -247.43728637695312,
      "loss": 1.5676,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 1.230128526687622,
      "rewards/margins": -1.297512412071228,
      "rewards/rejected": 2.5276408195495605,
      "step": 84
    },
    {
      "epoch": 0.002891795805195026,
      "grad_norm": 19.17630958557129,
      "learning_rate": 4.925739315689991e-05,
      "logits/chosen": -3.1123204231262207,
      "logits/rejected": -3.076711416244507,
      "logps/chosen": -323.8536376953125,
      "logps/rejected": -180.48394775390625,
      "loss": 0.5561,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.0740671157836914,
      "rewards/margins": 0.7356327772140503,
      "rewards/rejected": 1.3384342193603516,
      "step": 85
    },
    {
      "epoch": 0.0029258169323149678,
      "grad_norm": 15.29592227935791,
      "learning_rate": 4.9214579028215776e-05,
      "logits/chosen": -2.995969772338867,
      "logits/rejected": -2.975437879562378,
      "logps/chosen": -198.14756774902344,
      "logps/rejected": -97.39586639404297,
      "loss": 0.4645,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.481008529663086,
      "rewards/margins": 0.6242896914482117,
      "rewards/rejected": 1.8567187786102295,
      "step": 86
    },
    {
      "epoch": 0.002959838059434909,
      "grad_norm": 37.80276870727539,
      "learning_rate": 4.917058471511149e-05,
      "logits/chosen": -3.0480551719665527,
      "logits/rejected": -3.0806996822357178,
      "logps/chosen": -224.3466796875,
      "logps/rejected": -420.4508361816406,
      "loss": 1.1967,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.6259909868240356,
      "rewards/margins": -0.6426498889923096,
      "rewards/rejected": 2.2686409950256348,
      "step": 87
    },
    {
      "epoch": 0.0029938591865548508,
      "grad_norm": 20.813907623291016,
      "learning_rate": 4.912541236180779e-05,
      "logits/chosen": -3.045354127883911,
      "logits/rejected": -3.0551252365112305,
      "logps/chosen": -190.39498901367188,
      "logps/rejected": -417.9426574707031,
      "loss": 0.4819,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.733407497406006,
      "rewards/margins": 0.5519695281982422,
      "rewards/rejected": 2.1814379692077637,
      "step": 88
    },
    {
      "epoch": 0.003027880313674792,
      "grad_norm": 18.250917434692383,
      "learning_rate": 4.907906416994146e-05,
      "logits/chosen": -3.1118156909942627,
      "logits/rejected": -3.124022960662842,
      "logps/chosen": -423.84698486328125,
      "logps/rejected": -217.9359130859375,
      "loss": 0.3913,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.9902727603912354,
      "rewards/margins": 1.3608616590499878,
      "rewards/rejected": 1.629410982131958,
      "step": 89
    },
    {
      "epoch": 0.0030619014407947337,
      "grad_norm": 18.630115509033203,
      "learning_rate": 4.9031542398457974e-05,
      "logits/chosen": -3.0090408325195312,
      "logits/rejected": -3.0369369983673096,
      "logps/chosen": -435.40478515625,
      "logps/rejected": -191.86627197265625,
      "loss": 0.7079,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.947678327560425,
      "rewards/margins": 0.8492887020111084,
      "rewards/rejected": 2.0983893871307373,
      "step": 90
    },
    {
      "epoch": 0.003095922567914675,
      "grad_norm": 19.552978515625,
      "learning_rate": 4.898284936350144e-05,
      "logits/chosen": -3.0316624641418457,
      "logits/rejected": -3.011552572250366,
      "logps/chosen": -432.43402099609375,
      "logps/rejected": -214.85935974121094,
      "loss": 0.7816,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.7960076332092285,
      "rewards/margins": 0.5724576115608215,
      "rewards/rejected": 2.2235498428344727,
      "step": 91
    },
    {
      "epoch": 0.0031299436950346163,
      "grad_norm": 7.452019691467285,
      "learning_rate": 4.893298743830168e-05,
      "logits/chosen": -3.038140296936035,
      "logits/rejected": -3.00756573677063,
      "logps/chosen": -168.3916473388672,
      "logps/rejected": -119.79571533203125,
      "loss": 0.2097,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.8402457237243652,
      "rewards/margins": 1.5049405097961426,
      "rewards/rejected": 1.3353054523468018,
      "step": 92
    },
    {
      "epoch": 0.003163964822154558,
      "grad_norm": 16.27883529663086,
      "learning_rate": 4.888195905305859e-05,
      "logits/chosen": -3.030277729034424,
      "logits/rejected": -3.0196878910064697,
      "logps/chosen": -295.2430114746094,
      "logps/rejected": -229.96961975097656,
      "loss": 0.5378,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.0215210914611816,
      "rewards/margins": 0.6857559084892273,
      "rewards/rejected": 1.3357652425765991,
      "step": 93
    },
    {
      "epoch": 0.0031979859492744993,
      "grad_norm": 27.870994567871094,
      "learning_rate": 4.882976669482367e-05,
      "logits/chosen": -3.0157508850097656,
      "logits/rejected": -3.062469005584717,
      "logps/chosen": -594.1408081054688,
      "logps/rejected": -284.8695068359375,
      "loss": 0.829,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 2.395214796066284,
      "rewards/margins": -0.06876510381698608,
      "rewards/rejected": 2.463979959487915,
      "step": 94
    },
    {
      "epoch": 0.003232007076394441,
      "grad_norm": 34.57992172241211,
      "learning_rate": 4.877641290737884e-05,
      "logits/chosen": -3.0241267681121826,
      "logits/rejected": -3.072436809539795,
      "logps/chosen": -206.46072387695312,
      "logps/rejected": -490.99163818359375,
      "loss": 0.9507,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 3.0148510932922363,
      "rewards/margins": -0.41296035051345825,
      "rewards/rejected": 3.427811622619629,
      "step": 95
    },
    {
      "epoch": 0.0032660282035143822,
      "grad_norm": 31.449804306030273,
      "learning_rate": 4.8721900291112415e-05,
      "logits/chosen": -2.978365659713745,
      "logits/rejected": -2.9975037574768066,
      "logps/chosen": -353.8739929199219,
      "logps/rejected": -265.1501159667969,
      "loss": 0.7224,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 3.355400562286377,
      "rewards/margins": 0.41021838784217834,
      "rewards/rejected": 2.9451825618743896,
      "step": 96
    },
    {
      "epoch": 0.003300049330634324,
      "grad_norm": 20.54676628112793,
      "learning_rate": 4.8666231502892415e-05,
      "logits/chosen": -3.0715813636779785,
      "logits/rejected": -3.0223703384399414,
      "logps/chosen": -417.56280517578125,
      "logps/rejected": -126.94705963134766,
      "loss": 0.5149,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.667041540145874,
      "rewards/margins": 0.9312877655029297,
      "rewards/rejected": 1.7357536554336548,
      "step": 97
    },
    {
      "epoch": 0.0033340704577542652,
      "grad_norm": 34.95238494873047,
      "learning_rate": 4.860940925593703e-05,
      "logits/chosen": -3.038242816925049,
      "logits/rejected": -3.026169776916504,
      "logps/chosen": -320.06005859375,
      "logps/rejected": -534.7603759765625,
      "loss": 0.6213,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 3.1372852325439453,
      "rewards/margins": 0.3597479462623596,
      "rewards/rejected": 2.7775373458862305,
      "step": 98
    },
    {
      "epoch": 0.003368091584874207,
      "grad_norm": 22.04984474182129,
      "learning_rate": 4.855143631968242e-05,
      "logits/chosen": -3.0394222736358643,
      "logits/rejected": -3.040712356567383,
      "logps/chosen": -405.9694519042969,
      "logps/rejected": -224.75143432617188,
      "loss": 0.472,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 3.6535189151763916,
      "rewards/margins": 1.3992540836334229,
      "rewards/rejected": 2.2542648315429688,
      "step": 99
    },
    {
      "epoch": 0.003402112711994148,
      "grad_norm": 10.456045150756836,
      "learning_rate": 4.849231551964771e-05,
      "logits/chosen": -3.0558083057403564,
      "logits/rejected": -3.0164895057678223,
      "logps/chosen": -215.77474975585938,
      "logps/rejected": -114.40177917480469,
      "loss": 0.3431,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 3.2260184288024902,
      "rewards/margins": 1.2816791534423828,
      "rewards/rejected": 1.9443392753601074,
      "step": 100
    },
    {
      "epoch": 0.00343613383911409,
      "grad_norm": 26.428396224975586,
      "learning_rate": 4.843204973729729e-05,
      "logits/chosen": -2.9279229640960693,
      "logits/rejected": -2.941478729248047,
      "logps/chosen": -344.0908203125,
      "logps/rejected": -369.3677978515625,
      "loss": 0.851,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 4.361503601074219,
      "rewards/margins": 2.4121603965759277,
      "rewards/rejected": 1.949343204498291,
      "step": 101
    },
    {
      "epoch": 0.003470154966234031,
      "grad_norm": 20.163902282714844,
      "learning_rate": 4.837064190990036e-05,
      "logits/chosen": -2.994917869567871,
      "logits/rejected": -2.9446635246276855,
      "logps/chosen": -348.66937255859375,
      "logps/rejected": -345.49578857421875,
      "loss": 0.4779,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 3.1643588542938232,
      "rewards/margins": 1.304919958114624,
      "rewards/rejected": 1.8594388961791992,
      "step": 102
    },
    {
      "epoch": 0.003504176093353973,
      "grad_norm": 10.98054027557373,
      "learning_rate": 4.830809503038781e-05,
      "logits/chosen": -2.9521918296813965,
      "logits/rejected": -2.9645028114318848,
      "logps/chosen": -226.67506408691406,
      "logps/rejected": -326.85455322265625,
      "loss": 0.1768,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.5958943367004395,
      "rewards/margins": 1.691298007965088,
      "rewards/rejected": 1.904596209526062,
      "step": 103
    },
    {
      "epoch": 0.003538197220473914,
      "grad_norm": 10.843141555786133,
      "learning_rate": 4.8244412147206284e-05,
      "logits/chosen": -2.9416680335998535,
      "logits/rejected": -2.935615062713623,
      "logps/chosen": -242.29098510742188,
      "logps/rejected": -196.54017639160156,
      "loss": 0.4134,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.624265670776367,
      "rewards/margins": 1.845750331878662,
      "rewards/rejected": 0.778515100479126,
      "step": 104
    },
    {
      "epoch": 0.003572218347593856,
      "grad_norm": 51.45275115966797,
      "learning_rate": 4.817959636416969e-05,
      "logits/chosen": -2.987889289855957,
      "logits/rejected": -2.9992475509643555,
      "logps/chosen": -265.65509033203125,
      "logps/rejected": -504.4512023925781,
      "loss": 2.0434,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.9749891757965088,
      "rewards/margins": -1.6412996053695679,
      "rewards/rejected": 3.6162891387939453,
      "step": 105
    },
    {
      "epoch": 0.003606239474713797,
      "grad_norm": 34.740169525146484,
      "learning_rate": 4.8113650840307834e-05,
      "logits/chosen": -2.9495849609375,
      "logits/rejected": -2.9650793075561523,
      "logps/chosen": -267.8492126464844,
      "logps/rejected": -229.05166625976562,
      "loss": 1.4427,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.6815515756607056,
      "rewards/margins": -1.025644063949585,
      "rewards/rejected": 2.70719575881958,
      "step": 106
    },
    {
      "epoch": 0.003640260601833739,
      "grad_norm": 33.90592956542969,
      "learning_rate": 4.8046578789712515e-05,
      "logits/chosen": -2.9517102241516113,
      "logits/rejected": -2.892892360687256,
      "logps/chosen": -271.4212646484375,
      "logps/rejected": -432.76318359375,
      "loss": 0.9206,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.004805564880371,
      "rewards/margins": -0.28335341811180115,
      "rewards/rejected": 2.288158893585205,
      "step": 107
    },
    {
      "epoch": 0.00367428172895368,
      "grad_norm": 26.77821922302246,
      "learning_rate": 4.797838348138086e-05,
      "logits/chosen": -2.961052656173706,
      "logits/rejected": -2.963571548461914,
      "logps/chosen": -235.24127197265625,
      "logps/rejected": -195.713623046875,
      "loss": 0.9046,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 0.960392951965332,
      "rewards/margins": -0.3249235153198242,
      "rewards/rejected": 1.2853164672851562,
      "step": 108
    },
    {
      "epoch": 0.003708302856073622,
      "grad_norm": 20.291223526000977,
      "learning_rate": 4.790906823905599e-05,
      "logits/chosen": -2.922367811203003,
      "logits/rejected": -2.9089303016662598,
      "logps/chosen": -432.6478576660156,
      "logps/rejected": -147.66995239257812,
      "loss": 0.5442,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.312082052230835,
      "rewards/margins": 0.7899919748306274,
      "rewards/rejected": 1.522089958190918,
      "step": 109
    },
    {
      "epoch": 0.003742323983193563,
      "grad_norm": 21.555723190307617,
      "learning_rate": 4.783863644106502e-05,
      "logits/chosen": -2.893828868865967,
      "logits/rejected": -2.9130947589874268,
      "logps/chosen": -188.0413360595703,
      "logps/rejected": -163.48797607421875,
      "loss": 0.7379,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.8718959093093872,
      "rewards/margins": 0.09508019685745239,
      "rewards/rejected": 1.77681565284729,
      "step": 110
    },
    {
      "epoch": 0.003776345110313505,
      "grad_norm": 32.17390060424805,
      "learning_rate": 4.776709152015443e-05,
      "logits/chosen": -2.9426565170288086,
      "logits/rejected": -2.961993455886841,
      "logps/chosen": -292.1125183105469,
      "logps/rejected": -345.0081481933594,
      "loss": 1.1772,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.0569343566894531,
      "rewards/margins": -0.44999200105667114,
      "rewards/rejected": 1.5069262981414795,
      "step": 111
    },
    {
      "epoch": 0.003810366237433446,
      "grad_norm": 33.43811798095703,
      "learning_rate": 4.769443696332272e-05,
      "logits/chosen": -2.935262680053711,
      "logits/rejected": -2.9003987312316895,
      "logps/chosen": -299.31158447265625,
      "logps/rejected": -369.65911865234375,
      "loss": 0.9117,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.7615939378738403,
      "rewards/margins": 0.04655948281288147,
      "rewards/rejected": 1.7150346040725708,
      "step": 112
    },
    {
      "epoch": 0.003844387364553388,
      "grad_norm": 14.72221851348877,
      "learning_rate": 4.762067631165049e-05,
      "logits/chosen": -2.9283885955810547,
      "logits/rejected": -2.9265716075897217,
      "logps/chosen": -172.1368865966797,
      "logps/rejected": -198.41604614257812,
      "loss": 0.4199,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.7335453033447266,
      "rewards/margins": 0.9735270738601685,
      "rewards/rejected": 1.7600181102752686,
      "step": 113
    },
    {
      "epoch": 0.003878408491673329,
      "grad_norm": 26.454801559448242,
      "learning_rate": 4.754581316012785e-05,
      "logits/chosen": -3.0016942024230957,
      "logits/rejected": -2.9508540630340576,
      "logps/chosen": -301.06683349609375,
      "logps/rejected": -207.55111694335938,
      "loss": 0.6857,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 2.0953211784362793,
      "rewards/margins": 0.12809863686561584,
      "rewards/rejected": 1.9672226905822754,
      "step": 114
    },
    {
      "epoch": 0.00391242961879327,
      "grad_norm": 25.513513565063477,
      "learning_rate": 4.7469851157479177e-05,
      "logits/chosen": -2.9934637546539307,
      "logits/rejected": -3.0119104385375977,
      "logps/chosen": -660.9867553710938,
      "logps/rejected": -195.93348693847656,
      "loss": 0.5308,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.8832085132598877,
      "rewards/margins": 0.5043414831161499,
      "rewards/rejected": 2.3788669109344482,
      "step": 115
    },
    {
      "epoch": 0.003946450745913212,
      "grad_norm": 16.804649353027344,
      "learning_rate": 4.7392794005985326e-05,
      "logits/chosen": -2.970472574234009,
      "logits/rejected": -2.901583194732666,
      "logps/chosen": -214.0200958251953,
      "logps/rejected": -311.00836181640625,
      "loss": 0.5421,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.2354931831359863,
      "rewards/margins": 0.49688705801963806,
      "rewards/rejected": 1.7386059761047363,
      "step": 116
    },
    {
      "epoch": 0.003980471873033154,
      "grad_norm": 23.340364456176758,
      "learning_rate": 4.731464546130314e-05,
      "logits/chosen": -3.0023207664489746,
      "logits/rejected": -2.985395908355713,
      "logps/chosen": -91.83181762695312,
      "logps/rejected": -107.75381469726562,
      "loss": 1.0601,
      "rewards/accuracies": 0.25,
      "rewards/chosen": 1.4270564317703247,
      "rewards/margins": -0.45852407813072205,
      "rewards/rejected": 1.88558030128479,
      "step": 117
    },
    {
      "epoch": 0.0040144930001530955,
      "grad_norm": 18.47270965576172,
      "learning_rate": 4.723540933228244e-05,
      "logits/chosen": -2.899160861968994,
      "logits/rejected": -2.9231584072113037,
      "logps/chosen": -432.610595703125,
      "logps/rejected": -297.2770690917969,
      "loss": 0.5938,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 3.0332915782928467,
      "rewards/margins": 1.6466760635375977,
      "rewards/rejected": 1.3866153955459595,
      "step": 118
    },
    {
      "epoch": 0.004048514127273036,
      "grad_norm": 27.180835723876953,
      "learning_rate": 4.715508948078037e-05,
      "logits/chosen": -3.047947645187378,
      "logits/rejected": -3.05257248878479,
      "logps/chosen": -174.9864959716797,
      "logps/rejected": -159.964111328125,
      "loss": 1.3,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.1789213418960571,
      "rewards/margins": -0.4154563546180725,
      "rewards/rejected": 1.5943776369094849,
      "step": 119
    },
    {
      "epoch": 0.004082535254392978,
      "grad_norm": 36.64414978027344,
      "learning_rate": 4.707368982147318e-05,
      "logits/chosen": -3.0685150623321533,
      "logits/rejected": -3.0044500827789307,
      "logps/chosen": -522.5695190429688,
      "logps/rejected": -213.46484375,
      "loss": 0.7759,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.9243935346603394,
      "rewards/margins": 0.016531378030776978,
      "rewards/rejected": 1.9078620672225952,
      "step": 120
    },
    {
      "epoch": 0.00411655638151292,
      "grad_norm": 7.666904449462891,
      "learning_rate": 4.6991214321665414e-05,
      "logits/chosen": -2.9543278217315674,
      "logits/rejected": -2.9607834815979004,
      "logps/chosen": -258.1210021972656,
      "logps/rejected": -198.5812225341797,
      "loss": 0.1974,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.4223241806030273,
      "rewards/margins": 2.2252554893493652,
      "rewards/rejected": 0.1970686912536621,
      "step": 121
    },
    {
      "epoch": 0.0041505775086328614,
      "grad_norm": 19.53645896911621,
      "learning_rate": 4.690766700109659e-05,
      "logits/chosen": -2.9542903900146484,
      "logits/rejected": -2.9877171516418457,
      "logps/chosen": -532.9715576171875,
      "logps/rejected": -224.84103393554688,
      "loss": 0.4557,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.20086407661438,
      "rewards/margins": 0.7117323875427246,
      "rewards/rejected": 1.4891316890716553,
      "step": 122
    },
    {
      "epoch": 0.004184598635752802,
      "grad_norm": 16.438650131225586,
      "learning_rate": 4.682305193174524e-05,
      "logits/chosen": -3.004117727279663,
      "logits/rejected": -2.9864871501922607,
      "logps/chosen": -207.78421020507812,
      "logps/rejected": -124.6191635131836,
      "loss": 0.5509,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.6520434617996216,
      "rewards/margins": 0.7864970564842224,
      "rewards/rejected": 0.865546464920044,
      "step": 123
    },
    {
      "epoch": 0.004218619762872744,
      "grad_norm": 14.840346336364746,
      "learning_rate": 4.6737373237630476e-05,
      "logits/chosen": -3.018615245819092,
      "logits/rejected": -3.0123519897460938,
      "logps/chosen": -115.13555908203125,
      "logps/rejected": -88.72488403320312,
      "loss": 0.5478,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 1.4716414213180542,
      "rewards/margins": 0.9200279712677002,
      "rewards/rejected": 0.5516133308410645,
      "step": 124
    },
    {
      "epoch": 0.004252640889992686,
      "grad_norm": 25.749723434448242,
      "learning_rate": 4.665063509461097e-05,
      "logits/chosen": -3.0335187911987305,
      "logits/rejected": -2.9879348278045654,
      "logps/chosen": -428.715576171875,
      "logps/rejected": -315.72943115234375,
      "loss": 0.9619,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.7211081981658936,
      "rewards/margins": 0.08669501543045044,
      "rewards/rejected": 2.634413242340088,
      "step": 125
    },
    {
      "epoch": 0.0042866620171126265,
      "grad_norm": 11.342658042907715,
      "learning_rate": 4.656284173018144e-05,
      "logits/chosen": -3.0191352367401123,
      "logits/rejected": -2.9867730140686035,
      "logps/chosen": -282.3454895019531,
      "logps/rejected": -163.94921875,
      "loss": 0.3235,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.988835573196411,
      "rewards/margins": 1.4542242288589478,
      "rewards/rejected": 1.5346113443374634,
      "step": 126
    },
    {
      "epoch": 0.004320683144232568,
      "grad_norm": 22.939863204956055,
      "learning_rate": 4.6473997423266614e-05,
      "logits/chosen": -2.9171319007873535,
      "logits/rejected": -2.9400453567504883,
      "logps/chosen": -294.5677490234375,
      "logps/rejected": -219.3333282470703,
      "loss": 0.5542,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 2.0927131175994873,
      "rewards/margins": 0.75679612159729,
      "rewards/rejected": 1.3359171152114868,
      "step": 127
    },
    {
      "epoch": 0.00435470427135251,
      "grad_norm": 10.7678804397583,
      "learning_rate": 4.638410650401267e-05,
      "logits/chosen": -3.050967216491699,
      "logits/rejected": -2.973550796508789,
      "logps/chosen": -427.3642883300781,
      "logps/rejected": -214.50796508789062,
      "loss": 0.2155,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.5660037994384766,
      "rewards/margins": 2.3330581188201904,
      "rewards/rejected": 1.2329455614089966,
      "step": 128
    },
    {
      "epoch": 0.004388725398472452,
      "grad_norm": 25.86681365966797,
      "learning_rate": 4.629317335357619e-05,
      "logits/chosen": -3.0208535194396973,
      "logits/rejected": -3.0417640209198,
      "logps/chosen": -103.72763061523438,
      "logps/rejected": -151.43617248535156,
      "loss": 1.3143,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.1442210078239441,
      "rewards/margins": -0.46983587741851807,
      "rewards/rejected": 0.32561492919921875,
      "step": 129
    },
    {
      "epoch": 0.0044227465255923925,
      "grad_norm": 5.466204643249512,
      "learning_rate": 4.620120240391065e-05,
      "logits/chosen": -3.016930341720581,
      "logits/rejected": -2.971389055252075,
      "logps/chosen": -496.30938720703125,
      "logps/rejected": -235.5283660888672,
      "loss": 0.133,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.568811416625977,
      "rewards/margins": 2.570054054260254,
      "rewards/rejected": 1.9987574815750122,
      "step": 130
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}