{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.8691588785046729,
  "eval_steps": 50,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 55.02521133123827,
      "learning_rate": 5e-07,
      "logits/chosen": -2.7216238975524902,
      "logits/rejected": -2.7209055423736572,
      "logps/chosen": -268.4510192871094,
      "logps/rejected": -203.9590606689453,
      "loss": 0.6914,
      "rewards/accuracies": 0.33125001192092896,
      "rewards/chosen": 0.017813727259635925,
      "rewards/margins": 0.009322145953774452,
      "rewards/rejected": 0.008491581305861473,
      "step": 5
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 50.12554517439661,
      "learning_rate": 1e-06,
      "logits/chosen": -2.660832166671753,
      "logits/rejected": -2.6669700145721436,
      "logps/chosen": -256.93609619140625,
      "logps/rejected": -211.64938354492188,
      "loss": 0.6467,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 0.4671781659126282,
      "rewards/margins": 0.18480566143989563,
      "rewards/rejected": 0.28237253427505493,
      "step": 10
    },
    {
      "epoch": 0.14018691588785046,
      "grad_norm": 48.457281147169425,
      "learning_rate": 9.993623730611148e-07,
      "logits/chosen": -2.494657278060913,
      "logits/rejected": -2.5133018493652344,
      "logps/chosen": -232.6891632080078,
      "logps/rejected": -212.8677215576172,
      "loss": 0.6204,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": 1.2435152530670166,
      "rewards/margins": 0.6126660108566284,
      "rewards/rejected": 0.6308490037918091,
      "step": 15
    },
    {
      "epoch": 0.18691588785046728,
      "grad_norm": 42.23668651632898,
      "learning_rate": 9.97451118516912e-07,
      "logits/chosen": -2.3121209144592285,
      "logits/rejected": -2.302377462387085,
      "logps/chosen": -234.3399658203125,
      "logps/rejected": -191.3181610107422,
      "loss": 0.6223,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 1.3283790349960327,
      "rewards/margins": 0.9071598052978516,
      "rewards/rejected": 0.4212193489074707,
      "step": 20
    },
    {
      "epoch": 0.2336448598130841,
      "grad_norm": 49.68505743626908,
      "learning_rate": 9.94271111036929e-07,
      "logits/chosen": -2.2619333267211914,
      "logits/rejected": -2.2323482036590576,
      "logps/chosen": -230.17385864257812,
      "logps/rejected": -205.64108276367188,
      "loss": 0.616,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": 1.3353136777877808,
      "rewards/margins": 1.208017110824585,
      "rewards/rejected": 0.127296581864357,
      "step": 25
    },
    {
      "epoch": 0.2803738317757009,
      "grad_norm": 39.50573463077347,
      "learning_rate": 9.898304612549066e-07,
      "logits/chosen": -2.308243989944458,
      "logits/rejected": -2.2968266010284424,
      "logps/chosen": -241.26632690429688,
      "logps/rejected": -207.3319854736328,
      "loss": 0.5998,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": 0.9092999696731567,
      "rewards/margins": 0.9556086659431458,
      "rewards/rejected": -0.046308644115924835,
      "step": 30
    },
    {
      "epoch": 0.32710280373831774,
      "grad_norm": 42.78736900246308,
      "learning_rate": 9.841404950825536e-07,
      "logits/chosen": -2.3728604316711426,
      "logits/rejected": -2.3580102920532227,
      "logps/chosen": -256.8548583984375,
      "logps/rejected": -205.154052734375,
      "loss": 0.5996,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 1.0152027606964111,
      "rewards/margins": 1.2137099504470825,
      "rewards/rejected": -0.1985071450471878,
      "step": 35
    },
    {
      "epoch": 0.37383177570093457,
      "grad_norm": 41.73908597429494,
      "learning_rate": 9.77215724822721e-07,
      "logits/chosen": -2.4492850303649902,
      "logits/rejected": -2.4539356231689453,
      "logps/chosen": -243.1707763671875,
      "logps/rejected": -213.95166015625,
      "loss": 0.6098,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": 1.0353302955627441,
      "rewards/margins": 1.2659428119659424,
      "rewards/rejected": -0.2306123673915863,
      "step": 40
    },
    {
      "epoch": 0.4205607476635514,
      "grad_norm": 41.530064757148224,
      "learning_rate": 9.69073812155662e-07,
      "logits/chosen": -2.5637125968933105,
      "logits/rejected": -2.5535428524017334,
      "logps/chosen": -244.7168731689453,
      "logps/rejected": -205.80923461914062,
      "loss": 0.5974,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": 0.8133939504623413,
      "rewards/margins": 0.9837163686752319,
      "rewards/rejected": -0.17032238841056824,
      "step": 45
    },
    {
      "epoch": 0.4672897196261682,
      "grad_norm": 38.26706141308248,
      "learning_rate": 9.597355230927788e-07,
      "logits/chosen": -2.5823917388916016,
      "logits/rejected": -2.562842607498169,
      "logps/chosen": -240.04067993164062,
      "logps/rejected": -209.23428344726562,
      "loss": 0.572,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": 0.9298027753829956,
      "rewards/margins": 1.2456680536270142,
      "rewards/rejected": -0.3158652186393738,
      "step": 50
    },
    {
      "epoch": 0.4672897196261682,
      "eval_logits/chosen": -2.530949115753174,
      "eval_logits/rejected": -2.529101610183716,
      "eval_logps/chosen": -245.5291748046875,
      "eval_logps/rejected": -217.46429443359375,
      "eval_loss": 0.5720326900482178,
      "eval_rewards/accuracies": 0.7578125,
      "eval_rewards/chosen": 1.0708366632461548,
      "eval_rewards/margins": 1.28933846950531,
      "eval_rewards/rejected": -0.2185017466545105,
      "eval_runtime": 202.2601,
      "eval_samples_per_second": 15.03,
      "eval_steps_per_second": 0.237,
      "step": 50
    },
    {
      "epoch": 0.514018691588785,
      "grad_norm": 40.54073508413725,
      "learning_rate": 9.4922467501275e-07,
      "logits/chosen": -2.495945930480957,
      "logits/rejected": -2.487422466278076,
      "logps/chosen": -250.51620483398438,
      "logps/rejected": -228.90200805664062,
      "loss": 0.5176,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": 1.0155770778656006,
      "rewards/margins": 1.9236654043197632,
      "rewards/rejected": -0.9080885648727417,
      "step": 55
    },
    {
      "epoch": 0.5607476635514018,
      "grad_norm": 38.23797310786567,
      "learning_rate": 9.375680759151206e-07,
      "logits/chosen": -2.474236249923706,
      "logits/rejected": -2.4737977981567383,
      "logps/chosen": -255.09298706054688,
      "logps/rejected": -200.73593139648438,
      "loss": 0.5654,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 1.0740002393722534,
      "rewards/margins": 1.5434155464172363,
      "rewards/rejected": -0.4694152772426605,
      "step": 60
    },
    {
      "epoch": 0.6074766355140186,
      "grad_norm": 42.648181943788025,
      "learning_rate": 9.247954560462927e-07,
      "logits/chosen": -2.505916118621826,
      "logits/rejected": -2.506608724594116,
      "logps/chosen": -255.432861328125,
      "logps/rejected": -205.4224090576172,
      "loss": 0.5628,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": 1.1411590576171875,
      "rewards/margins": 1.7762504816055298,
      "rewards/rejected": -0.6350914239883423,
      "step": 65
    },
    {
      "epoch": 0.6542056074766355,
      "grad_norm": 38.81572593341751,
      "learning_rate": 9.109393920723001e-07,
      "logits/chosen": -2.4328043460845947,
      "logits/rejected": -2.4342734813690186,
      "logps/chosen": -233.8389129638672,
      "logps/rejected": -212.91085815429688,
      "loss": 0.5378,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.6836588978767395,
      "rewards/margins": 1.282029390335083,
      "rewards/rejected": -0.5983705520629883,
      "step": 70
    },
    {
      "epoch": 0.7009345794392523,
      "grad_norm": 32.317612654080975,
      "learning_rate": 8.960352239917699e-07,
      "logits/chosen": -2.450084924697876,
      "logits/rejected": -2.401425361633301,
      "logps/chosen": -240.6315460205078,
      "logps/rejected": -227.21084594726562,
      "loss": 0.5154,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": 0.6731350421905518,
      "rewards/margins": 1.5562646389007568,
      "rewards/rejected": -0.8831297755241394,
      "step": 75
    },
    {
      "epoch": 0.7476635514018691,
      "grad_norm": 30.17721204804764,
      "learning_rate": 8.801209650009814e-07,
      "logits/chosen": -2.4172046184539795,
      "logits/rejected": -2.400567054748535,
      "logps/chosen": -245.4665985107422,
      "logps/rejected": -214.18515014648438,
      "loss": 0.5206,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": 0.9583255052566528,
      "rewards/margins": 1.7562158107757568,
      "rewards/rejected": -0.7978904843330383,
      "step": 80
    },
    {
      "epoch": 0.794392523364486,
      "grad_norm": 32.24535114623233,
      "learning_rate": 8.632372045409141e-07,
      "logits/chosen": -2.320589065551758,
      "logits/rejected": -2.3311946392059326,
      "logps/chosen": -245.598388671875,
      "logps/rejected": -234.7646026611328,
      "loss": 0.5472,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": 1.1685658693313599,
      "rewards/margins": 1.754003882408142,
      "rewards/rejected": -0.5854381322860718,
      "step": 85
    },
    {
      "epoch": 0.8411214953271028,
      "grad_norm": 34.061291659967246,
      "learning_rate": 8.454270047735642e-07,
      "logits/chosen": -2.329784870147705,
      "logits/rejected": -2.304997682571411,
      "logps/chosen": -238.0483856201172,
      "logps/rejected": -195.24313354492188,
      "loss": 0.5291,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.6335947513580322,
      "rewards/margins": 1.5654070377349854,
      "rewards/rejected": -0.9318124055862427,
      "step": 90
    },
    {
      "epoch": 0.8878504672897196,
      "grad_norm": 31.735542564732725,
      "learning_rate": 8.267357907515661e-07,
      "logits/chosen": -2.298316478729248,
      "logits/rejected": -2.2975010871887207,
      "logps/chosen": -246.3526153564453,
      "logps/rejected": -206.03524780273438,
      "loss": 0.5418,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": 0.5418449640274048,
      "rewards/margins": 1.539952039718628,
      "rewards/rejected": -0.9981070756912231,
      "step": 95
    },
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 32.65915062987667,
      "learning_rate": 8.072112345612433e-07,
      "logits/chosen": -2.2663910388946533,
      "logits/rejected": -2.218681812286377,
      "logps/chosen": -246.5704803466797,
      "logps/rejected": -218.6560821533203,
      "loss": 0.4997,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.8927062153816223,
      "rewards/margins": 2.236736297607422,
      "rewards/rejected": -1.3440301418304443,
      "step": 100
    },
    {
      "epoch": 0.9345794392523364,
      "eval_logits/chosen": -2.2007782459259033,
      "eval_logits/rejected": -2.177567720413208,
      "eval_logps/chosen": -247.5850067138672,
      "eval_logps/rejected": -224.4142608642578,
      "eval_loss": 0.5101521015167236,
      "eval_rewards/accuracies": 0.7864583134651184,
      "eval_rewards/chosen": 0.8652558326721191,
      "eval_rewards/margins": 1.7787574529647827,
      "eval_rewards/rejected": -0.9135015606880188,
      "eval_runtime": 201.8599,
      "eval_samples_per_second": 15.06,
      "eval_steps_per_second": 0.238,
      "step": 100
    },
    {
      "epoch": 0.9813084112149533,
      "grad_norm": 28.6005139133492,
      "learning_rate": 7.869031337345827e-07,
      "logits/chosen": -2.1810142993927,
      "logits/rejected": -2.1466403007507324,
      "logps/chosen": -273.2081298828125,
      "logps/rejected": -224.4601593017578,
      "loss": 0.474,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.8940876126289368,
      "rewards/margins": 2.1252331733703613,
      "rewards/rejected": -1.2311456203460693,
      "step": 105
    },
    {
      "epoch": 1.02803738317757,
      "grad_norm": 17.790000678929353,
      "learning_rate": 7.658632842402432e-07,
      "logits/chosen": -2.1617965698242188,
      "logits/rejected": -2.118025064468384,
      "logps/chosen": -248.0548858642578,
      "logps/rejected": -209.18603515625,
      "loss": 0.3432,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": 0.46495524048805237,
      "rewards/margins": 2.547100067138672,
      "rewards/rejected": -2.0821449756622314,
      "step": 110
    },
    {
      "epoch": 1.074766355140187,
      "grad_norm": 18.641161706325903,
      "learning_rate": 7.441453483775353e-07,
      "logits/chosen": -2.2370879650115967,
      "logits/rejected": -2.1639482975006104,
      "logps/chosen": -252.3055419921875,
      "logps/rejected": -231.727294921875,
      "loss": 0.2131,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": 1.1295344829559326,
      "rewards/margins": 2.7920079231262207,
      "rewards/rejected": -1.6624739170074463,
      "step": 115
    },
    {
      "epoch": 1.1214953271028036,
      "grad_norm": 22.27642795647513,
      "learning_rate": 7.218047179103112e-07,
      "logits/chosen": -2.2182936668395996,
      "logits/rejected": -2.2140285968780518,
      "logps/chosen": -248.0435028076172,
      "logps/rejected": -233.4251251220703,
      "loss": 0.2493,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 1.1873703002929688,
      "rewards/margins": 3.075556755065918,
      "rewards/rejected": -1.8881866931915283,
      "step": 120
    },
    {
      "epoch": 1.1682242990654206,
      "grad_norm": 24.129663206875104,
      "learning_rate": 6.988983727898413e-07,
      "logits/chosen": -2.2516915798187256,
      "logits/rejected": -2.2213852405548096,
      "logps/chosen": -239.94125366210938,
      "logps/rejected": -213.9071807861328,
      "loss": 0.2434,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 1.149621844291687,
      "rewards/margins": 2.956752300262451,
      "rewards/rejected": -1.8071304559707642,
      "step": 125
    },
    {
      "epoch": 1.2149532710280373,
      "grad_norm": 19.098258762584823,
      "learning_rate": 6.754847358270066e-07,
      "logits/chosen": -2.268832206726074,
      "logits/rejected": -2.2357370853424072,
      "logps/chosen": -252.8026885986328,
      "logps/rejected": -218.2414093017578,
      "loss": 0.2656,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 1.428450345993042,
      "rewards/margins": 3.458483934402466,
      "rewards/rejected": -2.030033826828003,
      "step": 130
    },
    {
      "epoch": 1.2616822429906542,
      "grad_norm": 16.69096458544893,
      "learning_rate": 6.516235236844661e-07,
      "logits/chosen": -2.2503199577331543,
      "logits/rejected": -2.223175048828125,
      "logps/chosen": -251.1974639892578,
      "logps/rejected": -224.8771209716797,
      "loss": 0.229,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 1.4385788440704346,
      "rewards/margins": 3.5852439403533936,
      "rewards/rejected": -2.146665573120117,
      "step": 135
    },
    {
      "epoch": 1.308411214953271,
      "grad_norm": 24.370271103680654,
      "learning_rate": 6.273755945688457e-07,
      "logits/chosen": -2.30786395072937,
      "logits/rejected": -2.2820160388946533,
      "logps/chosen": -247.8080596923828,
      "logps/rejected": -242.51596069335938,
      "loss": 0.2745,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": 1.6472270488739014,
      "rewards/margins": 3.4453282356262207,
      "rewards/rejected": -1.7981010675430298,
      "step": 140
    },
    {
      "epoch": 1.355140186915888,
      "grad_norm": 21.195332919733744,
      "learning_rate": 6.02802793011411e-07,
      "logits/chosen": -2.304081916809082,
      "logits/rejected": -2.2989423274993896,
      "logps/chosen": -221.5400848388672,
      "logps/rejected": -236.2010955810547,
      "loss": 0.2682,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 1.4084885120391846,
      "rewards/margins": 3.7067673206329346,
      "rewards/rejected": -2.29827880859375,
      "step": 145
    },
    {
      "epoch": 1.4018691588785046,
      "grad_norm": 22.224652369004666,
      "learning_rate": 5.779677921331093e-07,
      "logits/chosen": -2.2803831100463867,
      "logits/rejected": -2.2711875438690186,
      "logps/chosen": -247.6720733642578,
      "logps/rejected": -217.12509155273438,
      "loss": 0.2873,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": 1.7340141534805298,
      "rewards/margins": 3.378777265548706,
      "rewards/rejected": -1.6447633504867554,
      "step": 150
    },
    {
      "epoch": 1.4018691588785046,
      "eval_logits/chosen": -2.275045394897461,
      "eval_logits/rejected": -2.263206720352173,
      "eval_logps/chosen": -245.67860412597656,
      "eval_logps/rejected": -227.55787658691406,
      "eval_loss": 0.5675327181816101,
      "eval_rewards/accuracies": 0.7890625,
      "eval_rewards/chosen": 1.055895209312439,
      "eval_rewards/margins": 2.2837564945220947,
      "eval_rewards/rejected": -1.2278612852096558,
      "eval_runtime": 202.0364,
      "eval_samples_per_second": 15.047,
      "eval_steps_per_second": 0.238,
      "step": 150
    },
    {
      "epoch": 1.4485981308411215,
      "grad_norm": 24.024418676682174,
      "learning_rate": 5.529339337962897e-07,
      "logits/chosen": -2.26741361618042,
      "logits/rejected": -2.2619667053222656,
      "logps/chosen": -228.74258422851562,
      "logps/rejected": -199.869873046875,
      "loss": 0.3185,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 1.8829383850097656,
      "rewards/margins": 3.547306776046753,
      "rewards/rejected": -1.6643686294555664,
      "step": 155
    },
    {
      "epoch": 1.4953271028037383,
      "grad_norm": 22.765211926437665,
      "learning_rate": 5.277650670507915e-07,
      "logits/chosen": -2.2662367820739746,
      "logits/rejected": -2.241522789001465,
      "logps/chosen": -241.50253295898438,
      "logps/rejected": -211.10791015625,
      "loss": 0.2582,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 1.677080512046814,
      "rewards/margins": 3.3089568614959717,
      "rewards/rejected": -1.6318763494491577,
      "step": 160
    },
    {
      "epoch": 1.542056074766355,
      "grad_norm": 20.952241807232628,
      "learning_rate": 5.025253852864471e-07,
      "logits/chosen": -2.2016148567199707,
      "logits/rejected": -2.2075283527374268,
      "logps/chosen": -247.7741241455078,
      "logps/rejected": -224.11892700195312,
      "loss": 0.2816,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": 1.5324174165725708,
      "rewards/margins": 3.230978488922119,
      "rewards/rejected": -1.6985607147216797,
      "step": 165
    },
    {
      "epoch": 1.588785046728972,
      "grad_norm": 23.145444455236966,
      "learning_rate": 4.77279262507344e-07,
      "logits/chosen": -2.2021024227142334,
      "logits/rejected": -2.1827890872955322,
      "logps/chosen": -243.9816131591797,
      "logps/rejected": -249.20703125,
      "loss": 0.2787,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 1.669757604598999,
      "rewards/margins": 3.6465446949005127,
      "rewards/rejected": -1.9767868518829346,
      "step": 170
    },
    {
      "epoch": 1.6355140186915889,
      "grad_norm": 26.563830659774606,
      "learning_rate": 4.5209108914542714e-07,
      "logits/chosen": -2.1774230003356934,
      "logits/rejected": -2.1725821495056152,
      "logps/chosen": -224.1102294921875,
      "logps/rejected": -234.7071075439453,
      "loss": 0.3123,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": 1.2634233236312866,
      "rewards/margins": 3.5007872581481934,
      "rewards/rejected": -2.2373640537261963,
      "step": 175
    },
    {
      "epoch": 1.6822429906542056,
      "grad_norm": 25.254447136991615,
      "learning_rate": 4.2702510783220475e-07,
      "logits/chosen": -2.168032169342041,
      "logits/rejected": -2.1451544761657715,
      "logps/chosen": -228.2743682861328,
      "logps/rejected": -211.07705688476562,
      "loss": 0.2948,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 1.1032750606536865,
      "rewards/margins": 3.2025279998779297,
      "rewards/rejected": -2.0992531776428223,
      "step": 180
    },
    {
      "epoch": 1.7289719626168223,
      "grad_norm": 20.89817797522474,
      "learning_rate": 4.0214524954741586e-07,
      "logits/chosen": -2.1714885234832764,
      "logits/rejected": -2.148820400238037,
      "logps/chosen": -250.5221405029297,
      "logps/rejected": -223.18399047851562,
      "loss": 0.2802,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 1.3431367874145508,
      "rewards/margins": 3.7224392890930176,
      "rewards/rejected": -2.3793022632598877,
      "step": 185
    },
    {
      "epoch": 1.7757009345794392,
      "grad_norm": 17.107545008827852,
      "learning_rate": 3.7751497056257305e-07,
      "logits/chosen": -2.1603405475616455,
      "logits/rejected": -2.145948648452759,
      "logps/chosen": -234.90872192382812,
      "logps/rejected": -240.06298828125,
      "loss": 0.269,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 1.084341049194336,
      "rewards/margins": 3.5826897621154785,
      "rewards/rejected": -2.4983482360839844,
      "step": 190
    },
    {
      "epoch": 1.8224299065420562,
      "grad_norm": 27.137427341683352,
      "learning_rate": 3.531970905952478e-07,
      "logits/chosen": -2.1491293907165527,
      "logits/rejected": -2.1209685802459717,
      "logps/chosen": -221.9265899658203,
      "logps/rejected": -215.990478515625,
      "loss": 0.2937,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 0.9956735372543335,
      "rewards/margins": 3.271005630493164,
      "rewards/rejected": -2.275331974029541,
      "step": 195
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 23.508295318902285,
      "learning_rate": 3.2925363258689553e-07,
      "logits/chosen": -2.161498785018921,
      "logits/rejected": -2.1209306716918945,
      "logps/chosen": -248.2667999267578,
      "logps/rejected": -227.0295867919922,
      "loss": 0.2853,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": 1.2955918312072754,
      "rewards/margins": 3.347618818283081,
      "rewards/rejected": -2.0520269870758057,
      "step": 200
    },
    {
      "epoch": 1.8691588785046729,
      "eval_logits/chosen": -2.147773504257202,
      "eval_logits/rejected": -2.1250855922698975,
      "eval_logps/chosen": -249.0491485595703,
      "eval_logps/rejected": -232.39312744140625,
      "eval_loss": 0.5163093209266663,
      "eval_rewards/accuracies": 0.8203125,
      "eval_rewards/chosen": 0.718841552734375,
      "eval_rewards/margins": 2.4302282333374023,
      "eval_rewards/rejected": -1.711386799812317,
      "eval_runtime": 201.455,
      "eval_samples_per_second": 15.09,
      "eval_steps_per_second": 0.238,
      "step": 200
    }
  ],
  "logging_steps": 5,
  "max_steps": 321,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2358113407598592.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}