{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 201,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07462686567164178,
      "grad_norm": 53.96980806965151,
      "learning_rate": 5e-07,
      "logits/chosen": -2.7474961280822754,
      "logits/rejected": -2.740154504776001,
      "logps/chosen": -254.6457061767578,
      "logps/rejected": -205.4122314453125,
      "loss": 0.6904,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.015023870393633842,
      "rewards/margins": 0.000346918503055349,
      "rewards/rejected": 0.014676950871944427,
      "step": 5
    },
    {
      "epoch": 0.14925373134328357,
      "grad_norm": 51.127419187517475,
      "learning_rate": 1e-06,
      "logits/chosen": -2.70166277885437,
      "logits/rejected": -2.6838972568511963,
      "logps/chosen": -239.15487670898438,
      "logps/rejected": -208.0798797607422,
      "loss": 0.6437,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.468679815530777,
      "rewards/margins": 0.14893516898155212,
      "rewards/rejected": 0.31974464654922485,
      "step": 10
    },
    {
      "epoch": 0.22388059701492538,
      "grad_norm": 48.915622995434845,
      "learning_rate": 9.983100718730718e-07,
      "logits/chosen": -2.455461263656616,
      "logits/rejected": -2.436664342880249,
      "logps/chosen": -238.5790557861328,
      "logps/rejected": -190.64874267578125,
      "loss": 0.6643,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 1.17107093334198,
      "rewards/margins": 0.4588069021701813,
      "rewards/rejected": 0.7122639417648315,
      "step": 15
    },
    {
      "epoch": 0.29850746268656714,
      "grad_norm": 42.1310748636241,
      "learning_rate": 9.932517109205849e-07,
      "logits/chosen": -2.3171567916870117,
      "logits/rejected": -2.2968077659606934,
      "logps/chosen": -234.33566284179688,
      "logps/rejected": -199.1301727294922,
      "loss": 0.6155,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 1.4784119129180908,
      "rewards/margins": 0.7344081401824951,
      "rewards/rejected": 0.7440037131309509,
      "step": 20
    },
    {
      "epoch": 0.373134328358209,
      "grad_norm": 37.58643004926761,
      "learning_rate": 9.848591102083375e-07,
      "logits/chosen": -2.239626884460449,
      "logits/rejected": -2.2184576988220215,
      "logps/chosen": -270.282958984375,
      "logps/rejected": -196.48818969726562,
      "loss": 0.5953,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": 1.7982975244522095,
      "rewards/margins": 1.2364143133163452,
      "rewards/rejected": 0.5618831515312195,
      "step": 25
    },
    {
      "epoch": 0.44776119402985076,
      "grad_norm": 40.38923718028625,
      "learning_rate": 9.731890013043367e-07,
      "logits/chosen": -2.2369236946105957,
      "logits/rejected": -2.2441658973693848,
      "logps/chosen": -231.41775512695312,
      "logps/rejected": -221.48721313476562,
      "loss": 0.5606,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.3224443197250366,
      "rewards/margins": 1.2497867345809937,
      "rewards/rejected": 0.07265764474868774,
      "step": 30
    },
    {
      "epoch": 0.5223880597014925,
      "grad_norm": 47.56097946401291,
      "learning_rate": 9.583202707897073e-07,
      "logits/chosen": -2.3453598022460938,
      "logits/rejected": -2.339764356613159,
      "logps/chosen": -241.860107421875,
      "logps/rejected": -214.53897094726562,
      "loss": 0.65,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": 0.8914967775344849,
      "rewards/margins": 0.727935791015625,
      "rewards/rejected": 0.16356101632118225,
      "step": 35
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 42.0267103133091,
      "learning_rate": 9.403534270080829e-07,
      "logits/chosen": -2.3654093742370605,
      "logits/rejected": -2.353811264038086,
      "logps/chosen": -238.9351043701172,
      "logps/rejected": -209.5872802734375,
      "loss": 0.6079,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 1.131967544555664,
      "rewards/margins": 1.2806609869003296,
      "rewards/rejected": -0.1486934870481491,
      "step": 40
    },
    {
      "epoch": 0.6716417910447762,
      "grad_norm": 39.949353314773276,
      "learning_rate": 9.19409920658098e-07,
      "logits/chosen": -2.212584972381592,
      "logits/rejected": -2.2196860313415527,
      "logps/chosen": -262.6395263671875,
      "logps/rejected": -202.49072265625,
      "loss": 0.6097,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": 1.2208731174468994,
      "rewards/margins": 1.1154453754425049,
      "rewards/rejected": 0.10542778670787811,
      "step": 45
    },
    {
      "epoch": 0.746268656716418,
      "grad_norm": 43.82292011113308,
      "learning_rate": 8.956313238215823e-07,
      "logits/chosen": -2.107001543045044,
      "logits/rejected": -2.0714831352233887,
      "logps/chosen": -242.4890899658203,
      "logps/rejected": -204.85910034179688,
      "loss": 0.5987,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": 1.1507846117019653,
      "rewards/margins": 1.0607472658157349,
      "rewards/rejected": 0.09003732353448868,
      "step": 50
    },
    {
      "epoch": 0.746268656716418,
      "eval_logits/chosen": -2.1097404956817627,
      "eval_logits/rejected": -2.074279546737671,
      "eval_logps/chosen": -264.0938720703125,
      "eval_logps/rejected": -229.32229614257812,
      "eval_loss": 0.6138201951980591,
      "eval_rewards/accuracies": 0.7583333253860474,
      "eval_rewards/chosen": 1.2434982061386108,
      "eval_rewards/margins": 1.4021955728530884,
      "eval_rewards/rejected": -0.15869754552841187,
      "eval_runtime": 129.58,
      "eval_samples_per_second": 14.663,
      "eval_steps_per_second": 0.232,
      "step": 50
    },
    {
      "epoch": 0.8208955223880597,
      "grad_norm": 40.5253665377123,
      "learning_rate": 8.691783729769873e-07,
      "logits/chosen": -2.115365505218506,
      "logits/rejected": -2.0831191539764404,
      "logps/chosen": -245.2086944580078,
      "logps/rejected": -208.19717407226562,
      "loss": 0.5896,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": 0.9533665776252747,
      "rewards/margins": 1.4577382802963257,
      "rewards/rejected": -0.5043715834617615,
      "step": 55
    },
    {
      "epoch": 0.8955223880597015,
      "grad_norm": 34.60112523621595,
      "learning_rate": 8.402298824670029e-07,
      "logits/chosen": -2.101773262023926,
      "logits/rejected": -2.0949249267578125,
      "logps/chosen": -249.70480346679688,
      "logps/rejected": -219.0111083984375,
      "loss": 0.5525,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 0.742239773273468,
      "rewards/margins": 1.2933545112609863,
      "rewards/rejected": -0.5511147379875183,
      "step": 60
    },
    {
      "epoch": 0.9701492537313433,
      "grad_norm": 33.604589067191306,
      "learning_rate": 8.089815357650089e-07,
      "logits/chosen": -2.090615749359131,
      "logits/rejected": -2.0645270347595215,
      "logps/chosen": -254.70431518554688,
      "logps/rejected": -216.178955078125,
      "loss": 0.5693,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": 0.8447690010070801,
      "rewards/margins": 1.3857734203338623,
      "rewards/rejected": -0.5410043001174927,
      "step": 65
    },
    {
      "epoch": 1.044776119402985,
      "grad_norm": 21.06317859535433,
      "learning_rate": 7.756445627110522e-07,
      "logits/chosen": -2.06718373298645,
      "logits/rejected": -2.051907539367676,
      "logps/chosen": -242.66635131835938,
      "logps/rejected": -222.59707641601562,
      "loss": 0.4108,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 1.4726606607437134,
      "rewards/margins": 2.182919979095459,
      "rewards/rejected": -0.710259199142456,
      "step": 70
    },
    {
      "epoch": 1.1194029850746268,
      "grad_norm": 22.259738178667774,
      "learning_rate": 7.404443116588547e-07,
      "logits/chosen": -2.0406711101531982,
      "logits/rejected": -2.018770217895508,
      "logps/chosen": -241.16098022460938,
      "logps/rejected": -227.0569610595703,
      "loss": 0.2542,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 1.9895944595336914,
      "rewards/margins": 2.734738349914551,
      "rewards/rejected": -0.7451435327529907,
      "step": 75
    },
    {
      "epoch": 1.1940298507462686,
      "grad_norm": 16.79071490181188,
      "learning_rate": 7.036187261857288e-07,
      "logits/chosen": -2.0667386054992676,
      "logits/rejected": -2.020956516265869,
      "logps/chosen": -238.8300018310547,
      "logps/rejected": -230.9183349609375,
      "loss": 0.2169,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 2.387232542037964,
      "rewards/margins": 3.326624631881714,
      "rewards/rejected": -0.9393925666809082,
      "step": 80
    },
    {
      "epoch": 1.2686567164179103,
      "grad_norm": 26.173042165666224,
      "learning_rate": 6.654167366624008e-07,
      "logits/chosen": -2.1021151542663574,
      "logits/rejected": -2.0837912559509277,
      "logps/chosen": -225.0214080810547,
      "logps/rejected": -230.99365234375,
      "loss": 0.2518,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 2.4148898124694824,
      "rewards/margins": 3.303870439529419,
      "rewards/rejected": -0.8889807462692261,
      "step": 85
    },
    {
      "epoch": 1.3432835820895521,
      "grad_norm": 22.231137216437176,
      "learning_rate": 6.260965775552713e-07,
      "logits/chosen": -2.164511203765869,
      "logits/rejected": -2.1222901344299316,
      "logps/chosen": -232.6430206298828,
      "logps/rejected": -203.33914184570312,
      "loss": 0.2452,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": 2.4132513999938965,
      "rewards/margins": 3.234947919845581,
      "rewards/rejected": -0.8216965794563293,
      "step": 90
    },
    {
      "epoch": 1.417910447761194,
      "grad_norm": 25.44451135006177,
      "learning_rate": 5.859240418356614e-07,
      "logits/chosen": -2.181170701980591,
      "logits/rejected": -2.1437950134277344,
      "logps/chosen": -227.32174682617188,
      "logps/rejected": -220.1398162841797,
      "loss": 0.3253,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 2.5023834705352783,
      "rewards/margins": 3.4593710899353027,
      "rewards/rejected": -0.9569879770278931,
      "step": 95
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 21.85215171836367,
      "learning_rate": 5.451706842957421e-07,
      "logits/chosen": -2.190605640411377,
      "logits/rejected": -2.1743760108947754,
      "logps/chosen": -222.10366821289062,
      "logps/rejected": -216.30953979492188,
      "loss": 0.2776,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 2.2073616981506348,
      "rewards/margins": 3.289036989212036,
      "rewards/rejected": -1.08167564868927,
      "step": 100
    },
    {
      "epoch": 1.4925373134328357,
      "eval_logits/chosen": -2.2346103191375732,
      "eval_logits/rejected": -2.197340488433838,
      "eval_logps/chosen": -259.05035400390625,
      "eval_logps/rejected": -231.1164093017578,
      "eval_loss": 0.5806639194488525,
      "eval_rewards/accuracies": 0.75,
      "eval_rewards/chosen": 1.7478455305099487,
      "eval_rewards/margins": 2.0859532356262207,
      "eval_rewards/rejected": -0.33810776472091675,
      "eval_runtime": 127.669,
      "eval_samples_per_second": 14.882,
      "eval_steps_per_second": 0.235,
      "step": 100
    },
    {
      "epoch": 1.5671641791044775,
      "grad_norm": 27.554218423838524,
      "learning_rate": 5.041119859162068e-07,
      "logits/chosen": -2.2345571517944336,
      "logits/rejected": -2.21101450920105,
      "logps/chosen": -231.67910766601562,
      "logps/rejected": -249.38589477539062,
      "loss": 0.358,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 1.9362834692001343,
      "rewards/margins": 3.1361162662506104,
      "rewards/rejected": -1.199832797050476,
      "step": 105
    },
    {
      "epoch": 1.6417910447761193,
      "grad_norm": 20.80791618185839,
      "learning_rate": 4.630254916940423e-07,
      "logits/chosen": -2.2466845512390137,
      "logits/rejected": -2.251269817352295,
      "logps/chosen": -249.7599639892578,
      "logps/rejected": -220.67337036132812,
      "loss": 0.266,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 2.3526227474212646,
      "rewards/margins": 3.386568069458008,
      "rewards/rejected": -1.0339449644088745,
      "step": 110
    },
    {
      "epoch": 1.716417910447761,
      "grad_norm": 22.347799579180975,
      "learning_rate": 4.2218893451814e-07,
      "logits/chosen": -2.283097505569458,
      "logits/rejected": -2.249800443649292,
      "logps/chosen": -233.9537811279297,
      "logps/rejected": -241.9879150390625,
      "loss": 0.2804,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 2.4147448539733887,
      "rewards/margins": 3.5848922729492188,
      "rewards/rejected": -1.1701476573944092,
      "step": 115
    },
    {
      "epoch": 1.7910447761194028,
      "grad_norm": 24.070594223050126,
      "learning_rate": 3.8187835777481375e-07,
      "logits/chosen": -2.2594428062438965,
      "logits/rejected": -2.242706537246704,
      "logps/chosen": -248.45553588867188,
      "logps/rejected": -246.9784698486328,
      "loss": 0.2957,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 2.410540819168091,
      "rewards/margins": 3.638995409011841,
      "rewards/rejected": -1.228454351425171,
      "step": 120
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 25.340900232171183,
      "learning_rate": 3.423662493738687e-07,
      "logits/chosen": -2.2234294414520264,
      "logits/rejected": -2.2221410274505615,
      "logps/chosen": -257.01434326171875,
      "logps/rejected": -228.7210235595703,
      "loss": 0.3255,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 2.2142786979675293,
      "rewards/margins": 3.4862682819366455,
      "rewards/rejected": -1.2719895839691162,
      "step": 125
    },
    {
      "epoch": 1.9402985074626866,
      "grad_norm": 21.51851322735578,
      "learning_rate": 3.039196998086687e-07,
      "logits/chosen": -2.2037501335144043,
      "logits/rejected": -2.169529676437378,
      "logps/chosen": -229.593017578125,
      "logps/rejected": -212.1902618408203,
      "loss": 0.2822,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 1.9124352931976318,
      "rewards/margins": 3.246814727783203,
      "rewards/rejected": -1.33437979221344,
      "step": 130
    },
    {
      "epoch": 2.014925373134328,
      "grad_norm": 18.57620212781393,
      "learning_rate": 2.667985967011878e-07,
      "logits/chosen": -2.2134411334991455,
      "logits/rejected": -2.183706760406494,
      "logps/chosen": -232.1904754638672,
      "logps/rejected": -220.9305877685547,
      "loss": 0.2702,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": 2.022249698638916,
      "rewards/margins": 3.3063716888427734,
      "rewards/rejected": -1.2841222286224365,
      "step": 135
    },
    {
      "epoch": 2.08955223880597,
      "grad_norm": 15.42241480180504,
      "learning_rate": 2.3125386803640183e-07,
      "logits/chosen": -2.2169346809387207,
      "logits/rejected": -2.1884984970092773,
      "logps/chosen": -233.9433135986328,
      "logps/rejected": -226.44033813476562,
      "loss": 0.165,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": 2.141209602355957,
      "rewards/margins": 3.7715439796447754,
      "rewards/rejected": -1.6303342580795288,
      "step": 140
    },
    {
      "epoch": 2.1641791044776117,
      "grad_norm": 11.320977426687946,
      "learning_rate": 1.9752578596124952e-07,
      "logits/chosen": -2.1993801593780518,
      "logits/rejected": -2.1754441261291504,
      "logps/chosen": -246.0238494873047,
      "logps/rejected": -220.15823364257812,
      "loss": 0.1649,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 2.1452529430389404,
      "rewards/margins": 3.4021663665771484,
      "rewards/rejected": -1.2569139003753662,
      "step": 145
    },
    {
      "epoch": 2.2388059701492535,
      "grad_norm": 15.28748494666602,
      "learning_rate": 1.6584234261399532e-07,
      "logits/chosen": -2.1991629600524902,
      "logits/rejected": -2.154266834259033,
      "logps/chosen": -231.54934692382812,
      "logps/rejected": -234.1011505126953,
      "loss": 0.1711,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 2.044983386993408,
      "rewards/margins": 3.682882785797119,
      "rewards/rejected": -1.637899398803711,
      "step": 150
    },
    {
      "epoch": 2.2388059701492535,
      "eval_logits/chosen": -2.209343910217285,
      "eval_logits/rejected": -2.168731212615967,
      "eval_logps/chosen": -262.0248718261719,
      "eval_logps/rejected": -235.22265625,
      "eval_loss": 0.5594516396522522,
      "eval_rewards/accuracies": 0.7833333611488342,
      "eval_rewards/chosen": 1.4503962993621826,
      "eval_rewards/margins": 2.1991307735443115,
      "eval_rewards/rejected": -0.7487342953681946,
      "eval_runtime": 127.6538,
      "eval_samples_per_second": 14.884,
      "eval_steps_per_second": 0.235,
      "step": 150
    },
    {
      "epoch": 2.3134328358208958,
      "grad_norm": 16.156825257003483,
      "learning_rate": 1.3641770896292082e-07,
      "logits/chosen": -2.1902785301208496,
      "logits/rejected": -2.166799545288086,
      "logps/chosen": -232.46304321289062,
      "logps/rejected": -227.48110961914062,
      "loss": 0.1739,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 2.0732829570770264,
      "rewards/margins": 3.8429579734802246,
      "rewards/rejected": -1.7696750164031982,
      "step": 155
    },
    {
      "epoch": 2.388059701492537,
      "grad_norm": 16.89696456987175,
      "learning_rate": 1.0945078707215221e-07,
      "logits/chosen": -2.2025253772735596,
      "logits/rejected": -2.205261707305908,
      "logps/chosen": -239.1471710205078,
      "logps/rejected": -239.80917358398438,
      "loss": 0.1445,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": 2.2130126953125,
      "rewards/margins": 3.790947437286377,
      "rewards/rejected": -1.5779345035552979,
      "step": 160
    },
    {
      "epoch": 2.4626865671641793,
      "grad_norm": 18.913466393129188,
      "learning_rate": 8.512386558088919e-08,
      "logits/chosen": -2.226701021194458,
      "logits/rejected": -2.2018561363220215,
      "logps/chosen": -238.5213623046875,
      "logps/rejected": -236.8737335205078,
      "loss": 0.1616,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 2.0778307914733887,
      "rewards/margins": 3.6287314891815186,
      "rewards/rejected": -1.5509008169174194,
      "step": 165
    },
    {
      "epoch": 2.5373134328358207,
      "grad_norm": 22.4417883150168,
      "learning_rate": 6.360138748461013e-08,
      "logits/chosen": -2.236025333404541,
      "logits/rejected": -2.201603412628174,
      "logps/chosen": -212.53872680664062,
      "logps/rejected": -229.65371704101562,
      "loss": 0.1627,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 1.9044567346572876,
      "rewards/margins": 3.8053181171417236,
      "rewards/rejected": -1.900861144065857,
      "step": 170
    },
    {
      "epoch": 2.611940298507463,
      "grad_norm": 14.68248054835397,
      "learning_rate": 4.5028838547699346e-08,
      "logits/chosen": -2.233783721923828,
      "logits/rejected": -2.2161941528320312,
      "logps/chosen": -245.54971313476562,
      "logps/rejected": -241.5737762451172,
      "loss": 0.1576,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.3260607719421387,
      "rewards/margins": 4.2508111000061035,
      "rewards/rejected": -1.9247499704360962,
      "step": 175
    },
    {
      "epoch": 2.6865671641791042,
      "grad_norm": 19.533167496739686,
      "learning_rate": 2.9531763861505964e-08,
      "logits/chosen": -2.21659517288208,
      "logits/rejected": -2.2015933990478516,
      "logps/chosen": -231.3876953125,
      "logps/rejected": -225.3466033935547,
      "loss": 0.147,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.0927014350891113,
      "rewards/margins": 3.892148494720459,
      "rewards/rejected": -1.7994472980499268,
      "step": 180
    },
    {
      "epoch": 2.7611940298507465,
      "grad_norm": 18.890485527883804,
      "learning_rate": 1.7214919195619125e-08,
      "logits/chosen": -2.2142136096954346,
      "logits/rejected": -2.1983494758605957,
      "logps/chosen": -241.61471557617188,
      "logps/rejected": -247.6813507080078,
      "loss": 0.1541,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 2.099193572998047,
      "rewards/margins": 3.850362777709961,
      "rewards/rejected": -1.751169204711914,
      "step": 185
    },
    {
      "epoch": 2.835820895522388,
      "grad_norm": 14.760180739764596,
      "learning_rate": 8.161562878982398e-09,
      "logits/chosen": -2.202644109725952,
      "logits/rejected": -2.200831651687622,
      "logps/chosen": -241.77230834960938,
      "logps/rejected": -239.11038208007812,
      "loss": 0.1525,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.383692502975464,
      "rewards/margins": 4.061740398406982,
      "rewards/rejected": -1.6780484914779663,
      "step": 190
    },
    {
      "epoch": 2.91044776119403,
      "grad_norm": 14.370730470038882,
      "learning_rate": 2.432892997526026e-09,
      "logits/chosen": -2.225444793701172,
      "logits/rejected": -2.201166868209839,
      "logps/chosen": -254.4960479736328,
      "logps/rejected": -243.36727905273438,
      "loss": 0.1523,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 2.8620870113372803,
      "rewards/margins": 4.515549659729004,
      "rewards/rejected": -1.6534621715545654,
      "step": 195
    },
    {
      "epoch": 2.9850746268656714,
      "grad_norm": 22.652598111500406,
      "learning_rate": 6.763371270035457e-11,
      "logits/chosen": -2.227086305618286,
      "logits/rejected": -2.2074146270751953,
      "logps/chosen": -225.73452758789062,
      "logps/rejected": -239.37252807617188,
      "loss": 0.1884,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 2.3307242393493652,
      "rewards/margins": 3.682234287261963,
      "rewards/rejected": -1.3515098094940186,
      "step": 200
    },
    {
      "epoch": 2.9850746268656714,
      "eval_logits/chosen": -2.2318949699401855,
      "eval_logits/rejected": -2.192718505859375,
      "eval_logps/chosen": -259.9425354003906,
      "eval_logps/rejected": -234.73751831054688,
      "eval_loss": 0.568375825881958,
      "eval_rewards/accuracies": 0.7916666865348816,
      "eval_rewards/chosen": 1.6586296558380127,
      "eval_rewards/margins": 2.3588478565216064,
      "eval_rewards/rejected": -0.7002180218696594,
      "eval_runtime": 127.6931,
      "eval_samples_per_second": 14.879,
      "eval_steps_per_second": 0.235,
      "step": 200
    }
  ],
  "logging_steps": 5,
  "max_steps": 201,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2369906314051584.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}