{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 50,
  "global_step": 352,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.028409090909090908,
      "grad_norm": 63.70768739794084,
      "learning_rate": 1.3888888888888888e-07,
      "logits/chosen": -2.859254837036133,
      "logits/rejected": -2.642847776412964,
      "logps/chosen": -390.5564880371094,
      "logps/rejected": -607.8642578125,
      "loss": 1.3793,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.0020330864936113358,
      "rewards/margins": 0.013599475845694542,
      "rewards/rejected": -0.01156639028340578,
      "step": 10
    },
    {
      "epoch": 0.056818181818181816,
      "grad_norm": 37.53559671175899,
      "learning_rate": 2.7777777777777776e-07,
      "logits/chosen": -2.842411518096924,
      "logits/rejected": -2.694638967514038,
      "logps/chosen": -328.69854736328125,
      "logps/rejected": -775.2159423828125,
      "loss": 1.1119,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 0.08446629345417023,
      "rewards/margins": 0.6622289419174194,
      "rewards/rejected": -0.5777625441551208,
      "step": 20
    },
    {
      "epoch": 0.08522727272727272,
      "grad_norm": 11.233762764198163,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.8567159175872803,
      "logits/rejected": -2.6723504066467285,
      "logps/chosen": -296.57916259765625,
      "logps/rejected": -1129.5751953125,
      "loss": 0.4749,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.35702842473983765,
      "rewards/margins": 4.873236179351807,
      "rewards/rejected": -4.516207218170166,
      "step": 30
    },
    {
      "epoch": 0.11363636363636363,
      "grad_norm": 3.6495758175953084,
      "learning_rate": 4.998023493068254e-07,
      "logits/chosen": -2.814579725265503,
      "logits/rejected": -2.6064682006835938,
      "logps/chosen": -323.31414794921875,
      "logps/rejected": -2251.45361328125,
      "loss": 0.1291,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.23441331088542938,
      "rewards/margins": 15.669992446899414,
      "rewards/rejected": -15.435577392578125,
      "step": 40
    },
    {
      "epoch": 0.14204545454545456,
      "grad_norm": 0.3660230434321398,
      "learning_rate": 4.975823666181255e-07,
      "logits/chosen": -2.7671985626220703,
      "logits/rejected": -2.447579860687256,
      "logps/chosen": -483.5135803222656,
      "logps/rejected": -4943.55126953125,
      "loss": 0.0158,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.2087831497192383,
      "rewards/margins": 41.68732452392578,
      "rewards/rejected": -42.8961067199707,
      "step": 50
    },
    {
      "epoch": 0.14204545454545456,
      "eval_logits/chosen": -2.873523473739624,
      "eval_logits/rejected": -2.389453411102295,
      "eval_logps/chosen": -509.4383850097656,
      "eval_logps/rejected": -5863.54931640625,
      "eval_loss": 0.010370529256761074,
      "eval_rewards/accuracies": 0.9959677457809448,
      "eval_rewards/chosen": -1.4245665073394775,
      "eval_rewards/margins": 50.803043365478516,
      "eval_rewards/rejected": -52.22761154174805,
      "eval_runtime": 195.8611,
      "eval_samples_per_second": 19.938,
      "eval_steps_per_second": 0.317,
      "step": 50
    },
    {
      "epoch": 0.17045454545454544,
      "grad_norm": 0.9752123540317181,
      "learning_rate": 4.929173350101024e-07,
      "logits/chosen": -2.865919351577759,
      "logits/rejected": -2.1330180168151855,
      "logps/chosen": -531.2206420898438,
      "logps/rejected": -6913.0869140625,
      "loss": 0.0245,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -1.7132558822631836,
      "rewards/margins": 60.98906707763672,
      "rewards/rejected": -62.70232009887695,
      "step": 60
    },
    {
      "epoch": 0.19886363636363635,
      "grad_norm": 0.478800645918279,
      "learning_rate": 4.858533249305336e-07,
      "logits/chosen": -2.895341396331787,
      "logits/rejected": -1.7176485061645508,
      "logps/chosen": -551.4154052734375,
      "logps/rejected": -7227.5439453125,
      "loss": 0.0068,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.7225472927093506,
      "rewards/margins": 64.42552185058594,
      "rewards/rejected": -66.14808654785156,
      "step": 70
    },
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 10.076839489234553,
      "learning_rate": 4.764600984163808e-07,
      "logits/chosen": -2.921323537826538,
      "logits/rejected": -1.122173547744751,
      "logps/chosen": -585.6353149414062,
      "logps/rejected": -8391.6171875,
      "loss": 0.0053,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.7931442260742188,
      "rewards/margins": 75.38569641113281,
      "rewards/rejected": -77.17884826660156,
      "step": 80
    },
    {
      "epoch": 0.2556818181818182,
      "grad_norm": 0.06316207964609313,
      "learning_rate": 4.6483042014491527e-07,
      "logits/chosen": -2.8797919750213623,
      "logits/rejected": -1.447209119796753,
      "logps/chosen": -596.3788452148438,
      "logps/rejected": -8167.92724609375,
      "loss": 0.0137,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.3426554203033447,
      "rewards/margins": 72.96277618408203,
      "rewards/rejected": -75.305419921875,
      "step": 90
    },
    {
      "epoch": 0.2840909090909091,
      "grad_norm": 1.4636731976721238,
      "learning_rate": 4.510791413176912e-07,
      "logits/chosen": -2.9729702472686768,
      "logits/rejected": -0.3022725284099579,
      "logps/chosen": -607.1732177734375,
      "logps/rejected": -10554.962890625,
      "loss": 0.0111,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.3770313262939453,
      "rewards/margins": 96.11477661132812,
      "rewards/rejected": -98.49180603027344,
      "step": 100
    },
    {
      "epoch": 0.2840909090909091,
      "eval_logits/chosen": -2.945009231567383,
      "eval_logits/rejected": 0.17861562967300415,
      "eval_logps/chosen": -592.9392700195312,
      "eval_logps/rejected": -9974.3623046875,
      "eval_loss": 0.0023908319417387247,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.259575366973877,
      "eval_rewards/margins": 91.07615661621094,
      "eval_rewards/rejected": -93.33573150634766,
      "eval_runtime": 193.9785,
      "eval_samples_per_second": 20.131,
      "eval_steps_per_second": 0.32,
      "step": 100
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.06640741403344876,
      "learning_rate": 4.353420654246546e-07,
      "logits/chosen": -2.9682483673095703,
      "logits/rejected": -0.10797711461782455,
      "logps/chosen": -584.2022094726562,
      "logps/rejected": -9037.5048828125,
      "loss": 0.0062,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.087810754776001,
      "rewards/margins": 82.29454803466797,
      "rewards/rejected": -84.3823471069336,
      "step": 110
    },
    {
      "epoch": 0.3409090909090909,
      "grad_norm": 0.11961136514165396,
      "learning_rate": 4.177746070897592e-07,
      "logits/chosen": -2.854703187942505,
      "logits/rejected": 0.2248195856809616,
      "logps/chosen": -577.8111572265625,
      "logps/rejected": -8910.1435546875,
      "loss": 0.0301,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9629430770874023,
      "rewards/margins": 80.76622009277344,
      "rewards/rejected": -82.72917175292969,
      "step": 120
    },
    {
      "epoch": 0.3693181818181818,
      "grad_norm": 0.44055468356671706,
      "learning_rate": 3.9855025724292763e-07,
      "logits/chosen": -2.659201145172119,
      "logits/rejected": 1.2345632314682007,
      "logps/chosen": -610.489990234375,
      "logps/rejected": -9278.4609375,
      "loss": 0.0023,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.4003846645355225,
      "rewards/margins": 83.97884368896484,
      "rewards/rejected": -86.37923431396484,
      "step": 130
    },
    {
      "epoch": 0.3977272727272727,
      "grad_norm": 2.2725494852752717,
      "learning_rate": 3.7785886977585555e-07,
      "logits/chosen": -2.1955535411834717,
      "logits/rejected": 1.7099689245224,
      "logps/chosen": -569.4786376953125,
      "logps/rejected": -11541.3544921875,
      "loss": 0.003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.034423589706421,
      "rewards/margins": 106.959228515625,
      "rewards/rejected": -108.99365234375,
      "step": 140
    },
    {
      "epoch": 0.42613636363636365,
      "grad_norm": 13.991084938762679,
      "learning_rate": 3.5590478660213206e-07,
      "logits/chosen": -1.8308794498443604,
      "logits/rejected": 2.1075541973114014,
      "logps/chosen": -663.3027954101562,
      "logps/rejected": -11290.779296875,
      "loss": 0.0039,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.852395534515381,
      "rewards/margins": 102.93562316894531,
      "rewards/rejected": -105.78802490234375,
      "step": 150
    },
    {
      "epoch": 0.42613636363636365,
      "eval_logits/chosen": -1.883123755455017,
      "eval_logits/rejected": 2.354902982711792,
      "eval_logps/chosen": -600.4949340820312,
      "eval_logps/rejected": -10689.58203125,
      "eval_loss": 0.0015989234670996666,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.3351328372955322,
      "eval_rewards/margins": 98.15279388427734,
      "eval_rewards/rejected": -100.48792266845703,
      "eval_runtime": 194.1289,
      "eval_samples_per_second": 20.116,
      "eval_steps_per_second": 0.319,
      "step": 150
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 15.634259463426035,
      "learning_rate": 3.3290481963801696e-07,
      "logits/chosen": -2.0796897411346436,
      "logits/rejected": 2.3097519874572754,
      "logps/chosen": -567.3259887695312,
      "logps/rejected": -9335.7529296875,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9718801975250244,
      "rewards/margins": 85.22074890136719,
      "rewards/rejected": -87.19263458251953,
      "step": 160
    },
    {
      "epoch": 0.48295454545454547,
      "grad_norm": 0.04468151503862641,
      "learning_rate": 3.0908610963322626e-07,
      "logits/chosen": -2.098179340362549,
      "logits/rejected": 1.9576606750488281,
      "logps/chosen": -608.5695190429688,
      "logps/rejected": -9132.7138671875,
      "loss": 0.0059,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.2095131874084473,
      "rewards/margins": 81.83451843261719,
      "rewards/rejected": -84.04402160644531,
      "step": 170
    },
    {
      "epoch": 0.5113636363636364,
      "grad_norm": 0.046664236017310436,
      "learning_rate": 2.846838829972671e-07,
      "logits/chosen": -2.053903102874756,
      "logits/rejected": 2.1386537551879883,
      "logps/chosen": -602.9306030273438,
      "logps/rejected": -8629.5322265625,
      "loss": 0.0205,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.4202630519866943,
      "rewards/margins": 77.7452163696289,
      "rewards/rejected": -80.16548156738281,
      "step": 180
    },
    {
      "epoch": 0.5397727272727273,
      "grad_norm": 0.2531340520912063,
      "learning_rate": 2.5993912877423147e-07,
      "logits/chosen": -1.8579347133636475,
      "logits/rejected": 1.7126737833023071,
      "logps/chosen": -551.639404296875,
      "logps/rejected": -9325.890625,
      "loss": 0.0012,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.060236692428589,
      "rewards/margins": 84.2825698852539,
      "rewards/rejected": -86.34281158447266,
      "step": 190
    },
    {
      "epoch": 0.5681818181818182,
      "grad_norm": 1.2128193169362667,
      "learning_rate": 2.3509621870754504e-07,
      "logits/chosen": -1.699134111404419,
      "logits/rejected": 2.7847816944122314,
      "logps/chosen": -572.9049072265625,
      "logps/rejected": -8628.181640625,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0777506828308105,
      "rewards/margins": 77.8956527709961,
      "rewards/rejected": -79.97340393066406,
      "step": 200
    },
    {
      "epoch": 0.5681818181818182,
      "eval_logits/chosen": -1.6725599765777588,
      "eval_logits/rejected": 2.5107614994049072,
      "eval_logps/chosen": -587.24755859375,
      "eval_logps/rejected": -9258.34375,
      "eval_loss": 0.0011689095990732312,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.202658176422119,
      "eval_rewards/margins": 83.97289276123047,
      "eval_rewards/rejected": -86.1755599975586,
      "eval_runtime": 193.9485,
      "eval_samples_per_second": 20.134,
      "eval_steps_per_second": 0.32,
      "step": 200
    },
    {
      "epoch": 0.5965909090909091,
      "grad_norm": 1.5143241412178876,
      "learning_rate": 2.1040049389819624e-07,
      "logits/chosen": -1.7269471883773804,
      "logits/rejected": 2.757033586502075,
      "logps/chosen": -597.5856323242188,
      "logps/rejected": -8867.4111328125,
      "loss": 0.0051,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0694241523742676,
      "rewards/margins": 80.28562927246094,
      "rewards/rejected": -82.35506439208984,
      "step": 210
    },
    {
      "epoch": 0.625,
      "grad_norm": 3.3477814516315405,
      "learning_rate": 1.8609584188988133e-07,
      "logits/chosen": -1.5605463981628418,
      "logits/rejected": 2.3806936740875244,
      "logps/chosen": -627.6144409179688,
      "logps/rejected": -7452.0048828125,
      "loss": 0.0066,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.6301381587982178,
      "rewards/margins": 65.39354705810547,
      "rewards/rejected": -68.02368927001953,
      "step": 220
    },
    {
      "epoch": 0.6534090909090909,
      "grad_norm": 86.10043361080564,
      "learning_rate": 1.624222881090439e-07,
      "logits/chosen": -1.8570022583007812,
      "logits/rejected": 2.0753836631774902,
      "logps/chosen": -650.1534423828125,
      "logps/rejected": -8688.783203125,
      "loss": 0.0216,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.5548887252807617,
      "rewards/margins": 77.47712707519531,
      "rewards/rejected": -80.03202056884766,
      "step": 230
    },
    {
      "epoch": 0.6818181818181818,
      "grad_norm": 0.6247509460136759,
      "learning_rate": 1.3961362544602212e-07,
      "logits/chosen": -1.4957786798477173,
      "logits/rejected": 2.78035306930542,
      "logps/chosen": -612.0357666015625,
      "logps/rejected": -7844.0234375,
      "loss": 0.0027,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.360233783721924,
      "rewards/margins": 70.1392822265625,
      "rewards/rejected": -72.49951171875,
      "step": 240
    },
    {
      "epoch": 0.7102272727272727,
      "grad_norm": 0.8355271945684738,
      "learning_rate": 1.1789510538684522e-07,
      "logits/chosen": -1.3467962741851807,
      "logits/rejected": 3.0541701316833496,
      "logps/chosen": -582.3834838867188,
      "logps/rejected": -10293.1650390625,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.3034989833831787,
      "rewards/margins": 93.33232116699219,
      "rewards/rejected": -95.63582611083984,
      "step": 250
    },
    {
      "epoch": 0.7102272727272727,
      "eval_logits/chosen": -1.0149785280227661,
      "eval_logits/rejected": 3.758545160293579,
      "eval_logps/chosen": -596.009521484375,
      "eval_logps/rejected": -8959.7470703125,
      "eval_loss": 0.0008426356362178922,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.290278196334839,
      "eval_rewards/margins": 80.89930725097656,
      "eval_rewards/rejected": -83.18959045410156,
      "eval_runtime": 193.2341,
      "eval_samples_per_second": 20.209,
      "eval_steps_per_second": 0.321,
      "step": 250
    },
    {
      "epoch": 0.7386363636363636,
      "grad_norm": 0.3552207920354382,
      "learning_rate": 9.748121349736891e-08,
      "logits/chosen": -0.6737051010131836,
      "logits/rejected": 3.9108054637908936,
      "logps/chosen": -637.5172119140625,
      "logps/rejected": -9320.220703125,
      "loss": 0.0011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.5368902683258057,
      "rewards/margins": 83.88404083251953,
      "rewards/rejected": -86.4209213256836,
      "step": 260
    },
    {
      "epoch": 0.7670454545454546,
      "grad_norm": 0.09771710883654211,
      "learning_rate": 7.857355122839673e-08,
      "logits/chosen": -0.7171241044998169,
      "logits/rejected": 3.9581894874572754,
      "logps/chosen": -603.0701293945312,
      "logps/rejected": -8722.890625,
      "loss": 0.0013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.222205638885498,
      "rewards/margins": 78.31710052490234,
      "rewards/rejected": -80.539306640625,
      "step": 270
    },
    {
      "epoch": 0.7954545454545454,
      "grad_norm": 0.20312109356571723,
      "learning_rate": 6.135884496044244e-08,
      "logits/chosen": -0.48570650815963745,
      "logits/rejected": 4.524535655975342,
      "logps/chosen": -585.2769775390625,
      "logps/rejected": -8691.55859375,
      "loss": 0.0015,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0610411167144775,
      "rewards/margins": 78.40565490722656,
      "rewards/rejected": -80.46669006347656,
      "step": 280
    },
    {
      "epoch": 0.8238636363636364,
      "grad_norm": 0.0562319233114269,
      "learning_rate": 4.600710195020982e-08,
      "logits/chosen": -0.22825559973716736,
      "logits/rejected": 4.695797443389893,
      "logps/chosen": -607.1776123046875,
      "logps/rejected": -8615.1923828125,
      "loss": 0.0052,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.1803441047668457,
      "rewards/margins": 77.55899810791016,
      "rewards/rejected": -79.73933410644531,
      "step": 290
    },
    {
      "epoch": 0.8522727272727273,
      "grad_norm": 1.3409016161535439,
      "learning_rate": 3.2669931390104374e-08,
      "logits/chosen": -0.32437822222709656,
      "logits/rejected": 4.543366432189941,
      "logps/chosen": -559.5152587890625,
      "logps/rejected": -9425.8134765625,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.032606601715088,
      "rewards/margins": 86.01726531982422,
      "rewards/rejected": -88.04988098144531,
      "step": 300
    },
    {
      "epoch": 0.8522727272727273,
      "eval_logits/chosen": -0.12213863432407379,
      "eval_logits/rejected": 4.708926677703857,
      "eval_logps/chosen": -586.3375854492188,
      "eval_logps/rejected": -9036.201171875,
      "eval_loss": 0.0007051606080494821,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.1935579776763916,
      "eval_rewards/margins": 81.76056671142578,
      "eval_rewards/rejected": -83.95413208007812,
      "eval_runtime": 194.0614,
      "eval_samples_per_second": 20.122,
      "eval_steps_per_second": 0.319,
      "step": 300
    },
    {
      "epoch": 0.8806818181818182,
      "grad_norm": 0.1502360317259883,
      "learning_rate": 2.147904716149135e-08,
      "logits/chosen": -0.03174494951963425,
      "logits/rejected": 4.578262805938721,
      "logps/chosen": -568.2017822265625,
      "logps/rejected": -9189.7119140625,
      "loss": 0.0037,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.061352014541626,
      "rewards/margins": 82.83074188232422,
      "rewards/rejected": -84.89209747314453,
      "step": 310
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.12997178207019558,
      "learning_rate": 1.254496706805433e-08,
      "logits/chosen": -0.2779918313026428,
      "logits/rejected": 4.609086036682129,
      "logps/chosen": -596.7515869140625,
      "logps/rejected": -9248.6181640625,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.140784978866577,
      "rewards/margins": 83.82556915283203,
      "rewards/rejected": -85.96635437011719,
      "step": 320
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.20243054667676932,
      "learning_rate": 5.955921395237318e-09,
      "logits/chosen": -0.13022509217262268,
      "logits/rejected": 4.544893741607666,
      "logps/chosen": -555.28271484375,
      "logps/rejected": -9316.8115234375,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.0001461505889893,
      "rewards/margins": 84.69043731689453,
      "rewards/rejected": -86.69058990478516,
      "step": 330
    },
    {
      "epoch": 0.9659090909090909,
      "grad_norm": 0.25686140623112713,
      "learning_rate": 1.7769815745066474e-09,
      "logits/chosen": -0.500297486782074,
      "logits/rejected": 4.298487186431885,
      "logps/chosen": -568.9089965820312,
      "logps/rejected": -8478.435546875,
      "loss": 0.0026,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.9992825984954834,
      "rewards/margins": 76.69742584228516,
      "rewards/rejected": -78.69671630859375,
      "step": 340
    },
    {
      "epoch": 0.9943181818181818,
      "grad_norm": 2.1488990981394114,
      "learning_rate": 4.9417557483610875e-11,
      "logits/chosen": -0.16308510303497314,
      "logits/rejected": 4.496396064758301,
      "logps/chosen": -605.3216552734375,
      "logps/rejected": -9169.884765625,
      "loss": 0.008,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.310539484024048,
      "rewards/margins": 82.91272735595703,
      "rewards/rejected": -85.2232666015625,
      "step": 350
    },
    {
      "epoch": 0.9943181818181818,
      "eval_logits/chosen": -0.04419805109500885,
      "eval_logits/rejected": 4.755471229553223,
      "eval_logps/chosen": -588.0638427734375,
      "eval_logps/rejected": -9137.26171875,
      "eval_loss": 0.0007233908982016146,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -2.2108209133148193,
      "eval_rewards/margins": 82.75391387939453,
      "eval_rewards/rejected": -84.96473693847656,
      "eval_runtime": 194.5802,
      "eval_samples_per_second": 20.069,
      "eval_steps_per_second": 0.319,
      "step": 350
    },
    {
      "epoch": 1.0,
      "step": 352,
      "total_flos": 0.0,
      "train_loss": 0.0941242430602539,
      "train_runtime": 10036.5915,
      "train_samples_per_second": 4.484,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 10,
  "max_steps": 352,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}