{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9993998799759952,
  "eval_steps": 100,
  "global_step": 1249,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4e-08,
      "logits/chosen": -2.5970396995544434,
      "logits/rejected": -2.6002578735351562,
      "logps/chosen": -391.7848205566406,
      "logps/rejected": -379.88592529296875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.0000000000000003e-07,
      "logits/chosen": -2.6321280002593994,
      "logits/rejected": -2.654905319213867,
      "logps/chosen": -415.5525207519531,
      "logps/rejected": -451.3063049316406,
      "loss": 0.6934,
      "rewards/accuracies": 0.4166666567325592,
      "rewards/chosen": 3.593924702727236e-05,
      "rewards/margins": -0.0004077497578691691,
      "rewards/rejected": 0.00044368900125846267,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.000000000000001e-07,
      "logits/chosen": -2.5803229808807373,
      "logits/rejected": -2.6041855812072754,
      "logps/chosen": -378.57830810546875,
      "logps/rejected": -406.0473937988281,
      "loss": 0.6938,
      "rewards/accuracies": 0.41874998807907104,
      "rewards/chosen": -0.0005379965296015143,
      "rewards/margins": -0.0012370327021926641,
      "rewards/rejected": 0.0006990364054217935,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2000000000000002e-06,
      "logits/chosen": -2.5524168014526367,
      "logits/rejected": -2.610682249069214,
      "logps/chosen": -405.12225341796875,
      "logps/rejected": -437.0355529785156,
      "loss": 0.6931,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -2.3546338070445927e-06,
      "rewards/margins": 4.741999146062881e-05,
      "rewards/rejected": -4.97746623295825e-05,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6000000000000001e-06,
      "logits/chosen": -2.551060199737549,
      "logits/rejected": -2.5709078311920166,
      "logps/chosen": -374.8092346191406,
      "logps/rejected": -387.48931884765625,
      "loss": 0.6934,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0006756076472811401,
      "rewards/margins": -0.00045229861279949546,
      "rewards/rejected": -0.00022330903448164463,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0000000000000003e-06,
      "logits/chosen": -2.5436787605285645,
      "logits/rejected": -2.540389060974121,
      "logps/chosen": -366.8270568847656,
      "logps/rejected": -374.41497802734375,
      "loss": 0.6934,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.0006743489648215473,
      "rewards/margins": -0.00049586221575737,
      "rewards/rejected": -0.0001784868072718382,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4000000000000003e-06,
      "logits/chosen": -2.5643861293792725,
      "logits/rejected": -2.5933163166046143,
      "logps/chosen": -389.86444091796875,
      "logps/rejected": -415.7872619628906,
      "loss": 0.6932,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.0006498451111838222,
      "rewards/margins": -0.00017824865062721074,
      "rewards/rejected": -0.0004715964605566114,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8000000000000003e-06,
      "logits/chosen": -2.608799457550049,
      "logits/rejected": -2.6101136207580566,
      "logps/chosen": -390.1094970703125,
      "logps/rejected": -396.507568359375,
      "loss": 0.693,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.0011631803354248405,
      "rewards/margins": 0.00039177463622763753,
      "rewards/rejected": -0.0015549547970294952,
      "step": 70
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.2000000000000003e-06,
      "logits/chosen": -2.6286911964416504,
      "logits/rejected": -2.6467819213867188,
      "logps/chosen": -383.10284423828125,
      "logps/rejected": -406.47393798828125,
      "loss": 0.6933,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.0017519788816571236,
      "rewards/margins": -0.0002883758279494941,
      "rewards/rejected": -0.0014636029954999685,
      "step": 80
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6000000000000003e-06,
      "logits/chosen": -2.6428580284118652,
      "logits/rejected": -2.6388907432556152,
      "logps/chosen": -431.7904357910156,
      "logps/rejected": -446.31732177734375,
      "loss": 0.6927,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.0028546121902763844,
      "rewards/margins": 0.0008213006658479571,
      "rewards/rejected": -0.0036759127397090197,
      "step": 90
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -2.6136410236358643,
      "logits/rejected": -2.6054491996765137,
      "logps/chosen": -382.89886474609375,
      "logps/rejected": -399.26068115234375,
      "loss": 0.6932,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.003379733068868518,
      "rewards/margins": -0.00018416713282931596,
      "rewards/rejected": -0.0031955656595528126,
      "step": 100
    },
    {
      "epoch": 0.08,
      "eval_logits/chosen": -2.490821361541748,
      "eval_logits/rejected": -2.497763156890869,
      "eval_logps/chosen": -322.7948913574219,
      "eval_logps/rejected": -332.520751953125,
      "eval_loss": 0.6930087208747864,
      "eval_rewards/accuracies": 0.5220000147819519,
      "eval_rewards/chosen": -0.002995692892000079,
      "eval_rewards/margins": 0.0002869492454919964,
      "eval_rewards/rejected": -0.0032826424576342106,
      "eval_runtime": 1592.8954,
      "eval_samples_per_second": 1.88,
      "eval_steps_per_second": 0.235,
      "step": 100
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.4e-06,
      "logits/chosen": -2.6191463470458984,
      "logits/rejected": -2.626410722732544,
      "logps/chosen": -418.57061767578125,
      "logps/rejected": -431.57568359375,
      "loss": 0.6929,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.004728841129690409,
      "rewards/margins": 0.00041932202293537557,
      "rewards/rejected": -0.005148162599653006,
      "step": 110
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.800000000000001e-06,
      "logits/chosen": -2.616450309753418,
      "logits/rejected": -2.6316654682159424,
      "logps/chosen": -387.65936279296875,
      "logps/rejected": -406.0279235839844,
      "loss": 0.6924,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.006956800818443298,
      "rewards/margins": 0.00143801374360919,
      "rewards/rejected": -0.008394813165068626,
      "step": 120
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.999755876225375e-06,
      "logits/chosen": -2.5832996368408203,
      "logits/rejected": -2.5880000591278076,
      "logps/chosen": -411.5313415527344,
      "logps/rejected": -434.4432678222656,
      "loss": 0.6925,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.009584048762917519,
      "rewards/margins": 0.0013888883404433727,
      "rewards/rejected": -0.010972936637699604,
      "step": 130
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.997803172081864e-06,
      "logits/chosen": -2.5645198822021484,
      "logits/rejected": -2.5772173404693604,
      "logps/chosen": -400.5196533203125,
      "logps/rejected": -419.6172790527344,
      "loss": 0.6928,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.013501152396202087,
      "rewards/margins": 0.0006207471014931798,
      "rewards/rejected": -0.01412190031260252,
      "step": 140
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9938992891651825e-06,
      "logits/chosen": -2.581132411956787,
      "logits/rejected": -2.5872106552124023,
      "logps/chosen": -382.4533386230469,
      "logps/rejected": -414.8580627441406,
      "loss": 0.6922,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.016398200765252113,
      "rewards/margins": 0.0018568082014098763,
      "rewards/rejected": -0.01825500652194023,
      "step": 150
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.988047277024456e-06,
      "logits/chosen": -2.6221516132354736,
      "logits/rejected": -2.6419806480407715,
      "logps/chosen": -396.94757080078125,
      "logps/rejected": -416.6105041503906,
      "loss": 0.6922,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.019227243959903717,
      "rewards/margins": 0.0020019258372485638,
      "rewards/rejected": -0.021229172125458717,
      "step": 160
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.980251707005417e-06,
      "logits/chosen": -2.618068218231201,
      "logits/rejected": -2.6144068241119385,
      "logps/chosen": -425.07244873046875,
      "logps/rejected": -435.1886291503906,
      "loss": 0.6935,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": -0.02663121186196804,
      "rewards/margins": -0.0007048315601423383,
      "rewards/rejected": -0.0259263813495636,
      "step": 170
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.970518668679459e-06,
      "logits/chosen": -2.5808920860290527,
      "logits/rejected": -2.5735676288604736,
      "logps/chosen": -417.1087341308594,
      "logps/rejected": -421.8807678222656,
      "loss": 0.6932,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.026894057169556618,
      "rewards/margins": -6.247097917366773e-05,
      "rewards/rejected": -0.02683158591389656,
      "step": 180
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.958855765086722e-06,
      "logits/chosen": -2.606273651123047,
      "logits/rejected": -2.608921527862549,
      "logps/chosen": -391.02362060546875,
      "logps/rejected": -399.1781311035156,
      "loss": 0.6927,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.02517121471464634,
      "rewards/margins": 0.0010087827686220407,
      "rewards/rejected": -0.026179995387792587,
      "step": 190
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.945272106796919e-06,
      "logits/chosen": -2.578331708908081,
      "logits/rejected": -2.594592332839966,
      "logps/chosen": -388.1728515625,
      "logps/rejected": -412.2106018066406,
      "loss": 0.6921,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.022485502064228058,
      "rewards/margins": 0.0022102731745690107,
      "rewards/rejected": -0.024695774540305138,
      "step": 200
    },
    {
      "epoch": 0.16,
      "eval_logits/chosen": -2.4899628162384033,
      "eval_logits/rejected": -2.497047185897827,
      "eval_logps/chosen": -324.8166809082031,
      "eval_logps/rejected": -334.6196594238281,
      "eval_loss": 0.6926525831222534,
      "eval_rewards/accuracies": 0.5183333158493042,
      "eval_rewards/chosen": -0.023213516920804977,
      "eval_rewards/margins": 0.0010577173670753837,
      "eval_rewards/rejected": -0.02427123300731182,
      "eval_runtime": 1586.0043,
      "eval_samples_per_second": 1.888,
      "eval_steps_per_second": 0.236,
      "step": 200
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.929778304792537e-06,
      "logits/chosen": -2.5928235054016113,
      "logits/rejected": -2.601378917694092,
      "logps/chosen": -424.74066162109375,
      "logps/rejected": -424.7421875,
      "loss": 0.6914,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.029381360858678818,
      "rewards/margins": 0.003531560767441988,
      "rewards/rejected": -0.03291292116045952,
      "step": 210
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.912386462179987e-06,
      "logits/chosen": -2.630770206451416,
      "logits/rejected": -2.637812376022339,
      "logps/chosen": -404.63275146484375,
      "logps/rejected": -440.338134765625,
      "loss": 0.6902,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.035039953887462616,
      "rewards/margins": 0.005983405746519566,
      "rewards/rejected": -0.041023362427949905,
      "step": 220
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.893110164735167e-06,
      "logits/chosen": -2.6500868797302246,
      "logits/rejected": -2.6668245792388916,
      "logps/chosen": -419.52508544921875,
      "logps/rejected": -435.5750427246094,
      "loss": 0.6928,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.04090961441397667,
      "rewards/margins": 0.000937508069910109,
      "rewards/rejected": -0.04184712469577789,
      "step": 230
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.871964470290823e-06,
      "logits/chosen": -2.5765552520751953,
      "logits/rejected": -2.603816509246826,
      "logps/chosen": -419.22332763671875,
      "logps/rejected": -441.0245666503906,
      "loss": 0.692,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.03845367208123207,
      "rewards/margins": 0.0023850365541875362,
      "rewards/rejected": -0.040838707238435745,
      "step": 240
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.848965896974006e-06,
      "logits/chosen": -2.62872052192688,
      "logits/rejected": -2.652649402618408,
      "logps/chosen": -402.7193603515625,
      "logps/rejected": -426.7840270996094,
      "loss": 0.6924,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.038372453302145004,
      "rewards/margins": 0.0015891485381871462,
      "rewards/rejected": -0.03996160253882408,
      "step": 250
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8241324103028055e-06,
      "logits/chosen": -2.6180949211120605,
      "logits/rejected": -2.5828635692596436,
      "logps/chosen": -408.49200439453125,
      "logps/rejected": -419.191650390625,
      "loss": 0.6918,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.037816666066646576,
      "rewards/margins": 0.002941995393484831,
      "rewards/rejected": -0.040758661925792694,
      "step": 260
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.797483409152438e-06,
      "logits/chosen": -2.617037296295166,
      "logits/rejected": -2.636782169342041,
      "logps/chosen": -391.20391845703125,
      "logps/rejected": -412.9513244628906,
      "loss": 0.691,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.0356525257229805,
      "rewards/margins": 0.004444011487066746,
      "rewards/rejected": -0.04009654000401497,
      "step": 270
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.769039710601669e-06,
      "logits/chosen": -2.6231722831726074,
      "logits/rejected": -2.6144392490386963,
      "logps/chosen": -387.4334411621094,
      "logps/rejected": -403.40057373046875,
      "loss": 0.6898,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.03822268173098564,
      "rewards/margins": 0.00696157943457365,
      "rewards/rejected": -0.045184262096881866,
      "step": 280
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.738823533671383e-06,
      "logits/chosen": -2.614532709121704,
      "logits/rejected": -2.6345345973968506,
      "logps/chosen": -421.5420837402344,
      "logps/rejected": -442.87994384765625,
      "loss": 0.6916,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.04416601359844208,
      "rewards/margins": 0.003424243303015828,
      "rewards/rejected": -0.047590263187885284,
      "step": 290
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.706858481968017e-06,
      "logits/chosen": -2.5896575450897217,
      "logits/rejected": -2.5868849754333496,
      "logps/chosen": -402.4905700683594,
      "logps/rejected": -409.7896423339844,
      "loss": 0.6913,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.04842864349484444,
      "rewards/margins": 0.003919907845556736,
      "rewards/rejected": -0.052348554134368896,
      "step": 300
    },
    {
      "epoch": 0.24,
      "eval_logits/chosen": -2.4894840717315674,
      "eval_logits/rejected": -2.4966704845428467,
      "eval_logps/chosen": -326.6393127441406,
      "eval_logps/rejected": -336.60589599609375,
      "eval_loss": 0.6919258236885071,
      "eval_rewards/accuracies": 0.5339999794960022,
      "eval_rewards/chosen": -0.0414397232234478,
      "eval_rewards/margins": 0.0026943066623061895,
      "eval_rewards/rejected": -0.04413403198122978,
      "eval_runtime": 1585.7818,
      "eval_samples_per_second": 1.888,
      "eval_steps_per_second": 0.236,
      "step": 300
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.673169525245416e-06,
      "logits/chosen": -2.5968692302703857,
      "logits/rejected": -2.585181474685669,
      "logps/chosen": -404.8775939941406,
      "logps/rejected": -433.09954833984375,
      "loss": 0.6902,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.04641267657279968,
      "rewards/margins": 0.00627540796995163,
      "rewards/rejected": -0.052688080817461014,
      "step": 310
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.63778297989952e-06,
      "logits/chosen": -2.604461193084717,
      "logits/rejected": -2.610708475112915,
      "logps/chosen": -403.58184814453125,
      "logps/rejected": -424.60772705078125,
      "loss": 0.691,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.05133752152323723,
      "rewards/margins": 0.004545481409877539,
      "rewards/rejected": -0.0558830089867115,
      "step": 320
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.60072648841109e-06,
      "logits/chosen": -2.589761734008789,
      "logits/rejected": -2.5855839252471924,
      "logps/chosen": -411.21026611328125,
      "logps/rejected": -430.5553283691406,
      "loss": 0.6872,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.0565360002219677,
      "rewards/margins": 0.012383319437503815,
      "rewards/rejected": -0.06891932338476181,
      "step": 330
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.562028997752574e-06,
      "logits/chosen": -2.5468997955322266,
      "logits/rejected": -2.5858261585235596,
      "logps/chosen": -409.3692321777344,
      "logps/rejected": -441.8675842285156,
      "loss": 0.6876,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.04863084852695465,
      "rewards/margins": 0.01160926092416048,
      "rewards/rejected": -0.06024010106921196,
      "step": 340
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.521720736775947e-06,
      "logits/chosen": -2.604952335357666,
      "logits/rejected": -2.596958637237549,
      "logps/chosen": -426.7057189941406,
      "logps/rejected": -427.89971923828125,
      "loss": 0.6907,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.06284764409065247,
      "rewards/margins": 0.005581230856478214,
      "rewards/rejected": -0.0684288740158081,
      "step": 350
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.479833192599198e-06,
      "logits/chosen": -2.6308584213256836,
      "logits/rejected": -2.627711057662964,
      "logps/chosen": -435.3694763183594,
      "logps/rejected": -454.00775146484375,
      "loss": 0.6867,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.06149282306432724,
      "rewards/margins": 0.013573038391768932,
      "rewards/rejected": -0.07506586611270905,
      "step": 360
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.436399086009928e-06,
      "logits/chosen": -2.6563782691955566,
      "logits/rejected": -2.6729133129119873,
      "logps/chosen": -409.5260925292969,
      "logps/rejected": -424.39208984375,
      "loss": 0.6874,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.07003921270370483,
      "rewards/margins": 0.012296813540160656,
      "rewards/rejected": -0.08233602344989777,
      "step": 370
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.391452345905239e-06,
      "logits/chosen": -2.5911147594451904,
      "logits/rejected": -2.6013450622558594,
      "logps/chosen": -420.21124267578125,
      "logps/rejected": -429.5535583496094,
      "loss": 0.6907,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.10565753281116486,
      "rewards/margins": 0.006660384126007557,
      "rewards/rejected": -0.11231790482997894,
      "step": 380
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3450280827879125e-06,
      "logits/chosen": -2.6037206649780273,
      "logits/rejected": -2.6106367111206055,
      "logps/chosen": -413.44122314453125,
      "logps/rejected": -427.99951171875,
      "loss": 0.6882,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.10301616042852402,
      "rewards/margins": 0.011568797752261162,
      "rewards/rejected": -0.11458496004343033,
      "step": 390
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297162561339554e-06,
      "logits/chosen": -2.522566556930542,
      "logits/rejected": -2.563248634338379,
      "logps/chosen": -421.3636169433594,
      "logps/rejected": -440.02276611328125,
      "loss": 0.6893,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.09894236922264099,
      "rewards/margins": 0.009308888576924801,
      "rewards/rejected": -0.10825125873088837,
      "step": 400
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -2.4952824115753174,
      "eval_logits/rejected": -2.5022895336151123,
      "eval_logps/chosen": -330.40167236328125,
      "eval_logps/rejected": -341.0243835449219,
      "eval_loss": 0.6891305446624756,
      "eval_rewards/accuracies": 0.5546666383743286,
      "eval_rewards/chosen": -0.0790637731552124,
      "eval_rewards/margins": 0.009254684671759605,
      "eval_rewards/rejected": -0.08831847459077835,
      "eval_runtime": 1591.081,
      "eval_samples_per_second": 1.882,
      "eval_steps_per_second": 0.236,
      "step": 400
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.247893172092157e-06,
      "logits/chosen": -2.5598597526550293,
      "logits/rejected": -2.5833513736724854,
      "logps/chosen": -404.2002868652344,
      "logps/rejected": -435.16357421875,
      "loss": 0.6857,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.10219142585992813,
      "rewards/margins": 0.016902286559343338,
      "rewards/rejected": -0.11909371614456177,
      "step": 410
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.197258402220187e-06,
      "logits/chosen": -2.5842721462249756,
      "logits/rejected": -2.5888819694519043,
      "logps/chosen": -414.70166015625,
      "logps/rejected": -425.90667724609375,
      "loss": 0.6896,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.12604860961437225,
      "rewards/margins": 0.009105135686695576,
      "rewards/rejected": -0.13515374064445496,
      "step": 420
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.145297805476023e-06,
      "logits/chosen": -2.639338493347168,
      "logits/rejected": -2.6249961853027344,
      "logps/chosen": -412.44232177734375,
      "logps/rejected": -427.6927185058594,
      "loss": 0.6859,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.11504076421260834,
      "rewards/margins": 0.016550499945878983,
      "rewards/rejected": -0.13159126043319702,
      "step": 430
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.092051971292228e-06,
      "logits/chosen": -2.598680019378662,
      "logits/rejected": -2.617053508758545,
      "logps/chosen": -404.762939453125,
      "logps/rejected": -424.2640686035156,
      "loss": 0.6839,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.1352802962064743,
      "rewards/margins": 0.02169608511030674,
      "rewards/rejected": -0.1569763869047165,
      "step": 440
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.037562493074792e-06,
      "logits/chosen": -2.5920491218566895,
      "logits/rejected": -2.59037446975708,
      "logps/chosen": -434.03118896484375,
      "logps/rejected": -439.4048767089844,
      "loss": 0.6832,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.1662554293870926,
      "rewards/margins": 0.024613002315163612,
      "rewards/rejected": -0.19086843729019165,
      "step": 450
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.981871935712112e-06,
      "logits/chosen": -2.6342484951019287,
      "logits/rejected": -2.6563720703125,
      "logps/chosen": -411.9359436035156,
      "logps/rejected": -435.57586669921875,
      "loss": 0.6833,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.16083207726478577,
      "rewards/margins": 0.023298023268580437,
      "rewards/rejected": -0.18413010239601135,
      "step": 460
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.925023802325094e-06,
      "logits/chosen": -2.664499044418335,
      "logits/rejected": -2.6798224449157715,
      "logps/chosen": -434.82354736328125,
      "logps/rejected": -453.67938232421875,
      "loss": 0.6776,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.18204860389232635,
      "rewards/margins": 0.03722093254327774,
      "rewards/rejected": -0.2192695438861847,
      "step": 470
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.867062500284342e-06,
      "logits/chosen": -2.6116766929626465,
      "logits/rejected": -2.627462863922119,
      "logps/chosen": -422.6748962402344,
      "logps/rejected": -454.072509765625,
      "loss": 0.6764,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.23436252772808075,
      "rewards/margins": 0.041565194725990295,
      "rewards/rejected": -0.27592775225639343,
      "step": 480
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8080333065209885e-06,
      "logits/chosen": -2.6146063804626465,
      "logits/rejected": -2.6115143299102783,
      "logps/chosen": -428.1953125,
      "logps/rejected": -423.606201171875,
      "loss": 0.7007,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.31299567222595215,
      "rewards/margins": -0.0009805113077163696,
      "rewards/rejected": -0.3120151460170746,
      "step": 490
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7479823321582624e-06,
      "logits/chosen": -2.6379127502441406,
      "logits/rejected": -2.66550874710083,
      "logps/chosen": -430.17572021484375,
      "logps/rejected": -461.09942626953125,
      "loss": 0.6724,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.24496836960315704,
      "rewards/margins": 0.04783773794770241,
      "rewards/rejected": -0.29280608892440796,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_logits/chosen": -2.5028529167175293,
      "eval_logits/rejected": -2.5099613666534424,
      "eval_logps/chosen": -342.6784973144531,
      "eval_logps/rejected": -354.7255859375,
      "eval_loss": 0.684412956237793,
      "eval_rewards/accuracies": 0.5529999732971191,
      "eval_rewards/chosen": -0.20183168351650238,
      "eval_rewards/margins": 0.023498954251408577,
      "eval_rewards/rejected": -0.2253306657075882,
      "eval_runtime": 1585.7151,
      "eval_samples_per_second": 1.888,
      "eval_steps_per_second": 0.236,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.686956486491419e-06,
      "logits/chosen": -2.59777569770813,
      "logits/rejected": -2.600053071975708,
      "logps/chosen": -440.66217041015625,
      "logps/rejected": -454.6631774902344,
      "loss": 0.6921,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.2590482234954834,
      "rewards/margins": 0.0103686498478055,
      "rewards/rejected": -0.269416868686676,
      "step": 510
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.625003440344166e-06,
      "logits/chosen": -2.6117706298828125,
      "logits/rejected": -2.625525712966919,
      "logps/chosen": -416.2491760253906,
      "logps/rejected": -427.0350646972656,
      "loss": 0.6781,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.228049635887146,
      "rewards/margins": 0.037227652966976166,
      "rewards/rejected": -0.26527729630470276,
      "step": 520
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.562171588830231e-06,
      "logits/chosen": -2.670912265777588,
      "logits/rejected": -2.6928374767303467,
      "logps/chosen": -428.692626953125,
      "logps/rejected": -462.62274169921875,
      "loss": 0.677,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.22094754874706268,
      "rewards/margins": 0.041822899132966995,
      "rewards/rejected": -0.26277047395706177,
      "step": 530
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.4985100135491245e-06,
      "logits/chosen": -2.6906635761260986,
      "logits/rejected": -2.708909511566162,
      "logps/chosen": -444.14483642578125,
      "logps/rejected": -480.6028747558594,
      "loss": 0.6691,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.2567470967769623,
      "rewards/margins": 0.058775972574949265,
      "rewards/rejected": -0.31552308797836304,
      "step": 540
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4340684442456673e-06,
      "logits/chosen": -2.5944106578826904,
      "logits/rejected": -2.6143410205841064,
      "logps/chosen": -438.25738525390625,
      "logps/rejected": -459.01763916015625,
      "loss": 0.6759,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.3285837173461914,
      "rewards/margins": 0.04724721983075142,
      "rewards/rejected": -0.37583091855049133,
      "step": 550
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3688972199631974e-06,
      "logits/chosen": -2.6361145973205566,
      "logits/rejected": -2.613219976425171,
      "logps/chosen": -441.1446838378906,
      "logps/rejected": -451.8172912597656,
      "loss": 0.6769,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.3550431728363037,
      "rewards/margins": 0.05103546380996704,
      "rewards/rejected": -0.40607863664627075,
      "step": 560
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3030472497208354e-06,
      "logits/chosen": -2.6290247440338135,
      "logits/rejected": -2.652219295501709,
      "logps/chosen": -426.34234619140625,
      "logps/rejected": -484.0997009277344,
      "loss": 0.6598,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.3327481150627136,
      "rewards/margins": 0.08233290165662766,
      "rewards/rejected": -0.4150810241699219,
      "step": 570
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.236569972745492e-06,
      "logits/chosen": -2.640798807144165,
      "logits/rejected": -2.6551499366760254,
      "logps/chosen": -419.3091735839844,
      "logps/rejected": -434.71527099609375,
      "loss": 0.6828,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.31533199548721313,
      "rewards/margins": 0.03718234971165657,
      "rewards/rejected": -0.3525143265724182,
      "step": 580
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1695173182897126e-06,
      "logits/chosen": -2.5991015434265137,
      "logits/rejected": -2.6085472106933594,
      "logps/chosen": -435.3345642089844,
      "logps/rejected": -467.74078369140625,
      "loss": 0.6698,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.3784990906715393,
      "rewards/margins": 0.060858141630887985,
      "rewards/rejected": -0.4393572211265564,
      "step": 590
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.10194166506673e-06,
      "logits/chosen": -2.624173164367676,
      "logits/rejected": -2.649747848510742,
      "logps/chosen": -443.6612243652344,
      "logps/rejected": -470.52447509765625,
      "loss": 0.6849,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3967738747596741,
      "rewards/margins": 0.03684181347489357,
      "rewards/rejected": -0.43361568450927734,
      "step": 600
    },
    {
      "epoch": 0.48,
      "eval_logits/chosen": -2.534677505493164,
      "eval_logits/rejected": -2.5411627292633057,
      "eval_logps/chosen": -356.15911865234375,
      "eval_logps/rejected": -369.895751953125,
      "eval_loss": 0.6804664731025696,
      "eval_rewards/accuracies": 0.5596666932106018,
      "eval_rewards/chosen": -0.33663830161094666,
      "eval_rewards/margins": 0.04039384424686432,
      "eval_rewards/rejected": -0.3770321309566498,
      "eval_runtime": 1591.4153,
      "eval_samples_per_second": 1.881,
      "eval_steps_per_second": 0.236,
      "step": 600
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0338958003344115e-06,
      "logits/chosen": -2.628021717071533,
      "logits/rejected": -2.654170036315918,
      "logps/chosen": -423.56768798828125,
      "logps/rejected": -451.36041259765625,
      "loss": 0.6603,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.39979976415634155,
      "rewards/margins": 0.08663686364889145,
      "rewards/rejected": -0.486436665058136,
      "step": 610
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9654328786600823e-06,
      "logits/chosen": -2.6157093048095703,
      "logits/rejected": -2.6425511837005615,
      "logps/chosen": -437.9432678222656,
      "logps/rejected": -482.203125,
      "loss": 0.6558,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.4792413115501404,
      "rewards/margins": 0.10463404655456543,
      "rewards/rejected": -0.5838753581047058,
      "step": 620
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.896606380398402e-06,
      "logits/chosen": -2.6413679122924805,
      "logits/rejected": -2.6552627086639404,
      "logps/chosen": -455.11260986328125,
      "logps/rejected": -468.00885009765625,
      "loss": 0.6878,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.5618987083435059,
      "rewards/margins": 0.03895387798547745,
      "rewards/rejected": -0.6008526086807251,
      "step": 630
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.827470069914772e-06,
      "logits/chosen": -2.681443214416504,
      "logits/rejected": -2.6886324882507324,
      "logps/chosen": -459.51953125,
      "logps/rejected": -491.306396484375,
      "loss": 0.6753,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.5284784436225891,
      "rewards/margins": 0.06967002153396606,
      "rewards/rejected": -0.5981484651565552,
      "step": 640
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7580779535868675e-06,
      "logits/chosen": -2.630887508392334,
      "logits/rejected": -2.645677328109741,
      "logps/chosen": -457.000732421875,
      "logps/rejected": -472.52545166015625,
      "loss": 0.6748,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.47293543815612793,
      "rewards/margins": 0.06035799905657768,
      "rewards/rejected": -0.5332934260368347,
      "step": 650
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.688484237617129e-06,
      "logits/chosen": -2.6493170261383057,
      "logits/rejected": -2.68170166015625,
      "logps/chosen": -453.5931091308594,
      "logps/rejected": -479.94818115234375,
      "loss": 0.6716,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.4414174556732178,
      "rewards/margins": 0.06612064689397812,
      "rewards/rejected": -0.5075381398200989,
      "step": 660
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6187432856891585e-06,
      "logits/chosen": -2.6594104766845703,
      "logits/rejected": -2.6713552474975586,
      "logps/chosen": -456.5660095214844,
      "logps/rejected": -488.9071350097656,
      "loss": 0.6665,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.44480568170547485,
      "rewards/margins": 0.07257355749607086,
      "rewards/rejected": -0.5173791646957397,
      "step": 670
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.548909576501096e-06,
      "logits/chosen": -2.7302539348602295,
      "logits/rejected": -2.7218852043151855,
      "logps/chosen": -458.1416931152344,
      "logps/rejected": -482.7391662597656,
      "loss": 0.6629,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.4737418591976166,
      "rewards/margins": 0.08715833723545074,
      "rewards/rejected": -0.5609001517295837,
      "step": 680
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4790376612091503e-06,
      "logits/chosen": -2.7039058208465576,
      "logits/rejected": -2.7273173332214355,
      "logps/chosen": -473.97064208984375,
      "logps/rejected": -498.3961486816406,
      "loss": 0.6715,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.5430811643600464,
      "rewards/margins": 0.08151289075613022,
      "rewards/rejected": -0.6245940327644348,
      "step": 690
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.40918212081453e-06,
      "logits/chosen": -2.6450188159942627,
      "logits/rejected": -2.667860507965088,
      "logps/chosen": -424.064208984375,
      "logps/rejected": -463.056640625,
      "loss": 0.6503,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.4541637897491455,
      "rewards/margins": 0.10934968292713165,
      "rewards/rejected": -0.5635135173797607,
      "step": 700
    },
    {
      "epoch": 0.56,
      "eval_logits/chosen": -2.543064832687378,
      "eval_logits/rejected": -2.5492172241210938,
      "eval_logps/chosen": -366.25225830078125,
      "eval_logps/rejected": -381.38427734375,
      "eval_loss": 0.677395224571228,
      "eval_rewards/accuracies": 0.5630000233650208,
      "eval_rewards/chosen": -0.4375694692134857,
      "eval_rewards/margins": 0.054348159581422806,
      "eval_rewards/rejected": -0.4919176399707794,
      "eval_runtime": 1584.8471,
      "eval_samples_per_second": 1.889,
      "eval_steps_per_second": 0.237,
      "step": 700
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3393975235270654e-06,
      "logits/chosen": -2.6429007053375244,
      "logits/rejected": -2.635617256164551,
      "logps/chosen": -460.6632385253906,
      "logps/rejected": -495.57098388671875,
      "loss": 0.6491,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.5247341990470886,
      "rewards/margins": 0.12096454948186874,
      "rewards/rejected": -0.6456986665725708,
      "step": 710
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2697383821388153e-06,
      "logits/chosen": -2.665684700012207,
      "logits/rejected": -2.63784122467041,
      "logps/chosen": -454.69110107421875,
      "logps/rejected": -466.15020751953125,
      "loss": 0.6856,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.49990567564964294,
      "rewards/margins": 0.04791254550218582,
      "rewards/rejected": -0.547818124294281,
      "step": 720
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2002591114409657e-06,
      "logits/chosen": -2.672154426574707,
      "logits/rejected": -2.6631722450256348,
      "logps/chosen": -453.4434509277344,
      "logps/rejected": -476.5977478027344,
      "loss": 0.6716,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.4569157660007477,
      "rewards/margins": 0.07301469147205353,
      "rewards/rejected": -0.5299304723739624,
      "step": 730
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.131013985717285e-06,
      "logits/chosen": -2.6686697006225586,
      "logits/rejected": -2.662213087081909,
      "logps/chosen": -456.8734436035156,
      "logps/rejected": -498.776123046875,
      "loss": 0.6499,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.38933560252189636,
      "rewards/margins": 0.1104939803481102,
      "rewards/rejected": -0.49982959032058716,
      "step": 740
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.062057096347338e-06,
      "logits/chosen": -2.6757967472076416,
      "logits/rejected": -2.683659553527832,
      "logps/chosen": -449.1112365722656,
      "logps/rejected": -460.0345764160156,
      "loss": 0.6687,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.40121787786483765,
      "rewards/margins": 0.07739736139774323,
      "rewards/rejected": -0.47861528396606445,
      "step": 750
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9934423095525733e-06,
      "logits/chosen": -2.660757541656494,
      "logits/rejected": -2.641571521759033,
      "logps/chosen": -453.0924377441406,
      "logps/rejected": -462.0196228027344,
      "loss": 0.6656,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.3819692134857178,
      "rewards/margins": 0.08219519257545471,
      "rewards/rejected": -0.4641643464565277,
      "step": 760
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9252232243182986e-06,
      "logits/chosen": -2.6494193077087402,
      "logits/rejected": -2.6644535064697266,
      "logps/chosen": -420.4537658691406,
      "logps/rejected": -452.49700927734375,
      "loss": 0.6659,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.4124923348426819,
      "rewards/margins": 0.08118386566638947,
      "rewards/rejected": -0.49367618560791016,
      "step": 770
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8574531305244043e-06,
      "logits/chosen": -2.674877882003784,
      "logits/rejected": -2.698631763458252,
      "logps/chosen": -459.5841369628906,
      "logps/rejected": -493.99969482421875,
      "loss": 0.6453,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.4147106111049652,
      "rewards/margins": 0.12960824370384216,
      "rewards/rejected": -0.5443187952041626,
      "step": 780
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.7901849673175559e-06,
      "logits/chosen": -2.69256329536438,
      "logits/rejected": -2.7218129634857178,
      "logps/chosen": -455.54736328125,
      "logps/rejected": -486.2936096191406,
      "loss": 0.66,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.38235577940940857,
      "rewards/margins": 0.09029395878314972,
      "rewards/rejected": -0.4726497530937195,
      "step": 790
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7234712817573555e-06,
      "logits/chosen": -2.670511484146118,
      "logits/rejected": -2.6233620643615723,
      "logps/chosen": -479.3063049316406,
      "logps/rejected": -484.30718994140625,
      "loss": 0.6841,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.45734840631484985,
      "rewards/margins": 0.05288088321685791,
      "rewards/rejected": -0.510229229927063,
      "step": 800
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.559168577194214,
      "eval_logits/rejected": -2.56618070602417,
      "eval_logps/chosen": -354.3205871582031,
      "eval_logps/rejected": -370.067626953125,
      "eval_loss": 0.6735012531280518,
      "eval_rewards/accuracies": 0.5913333296775818,
      "eval_rewards/chosen": -0.31825292110443115,
      "eval_rewards/margins": 0.06049799174070358,
      "eval_rewards/rejected": -0.37875089049339294,
      "eval_runtime": 1585.6649,
      "eval_samples_per_second": 1.888,
      "eval_steps_per_second": 0.236,
      "step": 800
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6573641877687936e-06,
      "logits/chosen": -2.688758373260498,
      "logits/rejected": -2.7048513889312744,
      "logps/chosen": -433.21484375,
      "logps/rejected": -472.264892578125,
      "loss": 0.6559,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.3647409975528717,
      "rewards/margins": 0.10189852863550186,
      "rewards/rejected": -0.46663951873779297,
      "step": 810
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.591915325433034e-06,
      "logits/chosen": -2.705111265182495,
      "logits/rejected": -2.690777540206909,
      "logps/chosen": -436.6507263183594,
      "logps/rejected": -449.94891357421875,
      "loss": 0.6732,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.39074164628982544,
      "rewards/margins": 0.06702588498592377,
      "rewards/rejected": -0.4577675461769104,
      "step": 820
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5271758206483664e-06,
      "logits/chosen": -2.7012417316436768,
      "logits/rejected": -2.692389726638794,
      "logps/chosen": -457.00238037109375,
      "logps/rejected": -475.9202575683594,
      "loss": 0.6669,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.4308633804321289,
      "rewards/margins": 0.09242836385965347,
      "rewards/rejected": -0.5232917666435242,
      "step": 830
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4631962451927966e-06,
      "logits/chosen": -2.675565719604492,
      "logits/rejected": -2.708937168121338,
      "logps/chosen": -453.7381286621094,
      "logps/rejected": -485.38360595703125,
      "loss": 0.6646,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.42438727617263794,
      "rewards/margins": 0.09076879173517227,
      "rewards/rejected": -0.515156090259552,
      "step": 840
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.4000265772195032e-06,
      "logits/chosen": -2.706509590148926,
      "logits/rejected": -2.7089927196502686,
      "logps/chosen": -460.16326904296875,
      "logps/rejected": -492.61163330078125,
      "loss": 0.6616,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.4706672132015228,
      "rewards/margins": 0.09530263394117355,
      "rewards/rejected": -0.5659698247909546,
      "step": 850
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3377161622160077e-06,
      "logits/chosen": -2.688765048980713,
      "logits/rejected": -2.686649799346924,
      "logps/chosen": -452.43035888671875,
      "logps/rejected": -478.95855712890625,
      "loss": 0.6575,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.4564797878265381,
      "rewards/margins": 0.09994712471961975,
      "rewards/rejected": -0.5564268827438354,
      "step": 860
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.276313674457553e-06,
      "logits/chosen": -2.687105417251587,
      "logits/rejected": -2.6988847255706787,
      "logps/chosen": -431.041259765625,
      "logps/rejected": -468.3299865722656,
      "loss": 0.644,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.4324752390384674,
      "rewards/margins": 0.13809339702129364,
      "rewards/rejected": -0.5705686211585999,
      "step": 870
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2158670789848095e-06,
      "logits/chosen": -2.6659626960754395,
      "logits/rejected": -2.6399731636047363,
      "logps/chosen": -455.6881408691406,
      "logps/rejected": -471.60479736328125,
      "loss": 0.6906,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.4899720549583435,
      "rewards/margins": 0.03855130821466446,
      "rewards/rejected": -0.528523325920105,
      "step": 880
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1564235941356016e-06,
      "logits/chosen": -2.636784791946411,
      "logits/rejected": -2.665184497833252,
      "logps/chosen": -443.228759765625,
      "logps/rejected": -486.3818359375,
      "loss": 0.6375,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.49858155846595764,
      "rewards/margins": 0.15248027443885803,
      "rewards/rejected": -0.6510618329048157,
      "step": 890
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.0980296546599254e-06,
      "logits/chosen": -2.6810688972473145,
      "logits/rejected": -2.66475248336792,
      "logps/chosen": -478.31512451171875,
      "logps/rejected": -483.4214782714844,
      "loss": 0.6773,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.5605891942977905,
      "rewards/margins": 0.07103095203638077,
      "rewards/rejected": -0.6316200494766235,
      "step": 900
    },
    {
      "epoch": 0.72,
      "eval_logits/chosen": -2.5706326961517334,
      "eval_logits/rejected": -2.577380418777466,
      "eval_logps/chosen": -362.3546142578125,
      "eval_logps/rejected": -378.96929931640625,
      "eval_loss": 0.6724240183830261,
      "eval_rewards/accuracies": 0.5886666774749756,
      "eval_rewards/chosen": -0.3985931873321533,
      "eval_rewards/margins": 0.06917466968297958,
      "eval_rewards/rejected": -0.4677678346633911,
      "eval_runtime": 1591.0139,
      "eval_samples_per_second": 1.882,
      "eval_steps_per_second": 0.236,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.040730875447083e-06,
      "logits/chosen": -2.6869008541107178,
      "logits/rejected": -2.6931540966033936,
      "logps/chosen": -458.8529357910156,
      "logps/rejected": -475.8218688964844,
      "loss": 0.6623,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.47533708810806274,
      "rewards/margins": 0.09188753366470337,
      "rewards/rejected": -0.5672245621681213,
      "step": 910
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.845720158932414e-07,
      "logits/chosen": -2.655848741531372,
      "logits/rejected": -2.650817632675171,
      "logps/chosen": -427.18896484375,
      "logps/rejected": -438.2098693847656,
      "loss": 0.6659,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.4379213750362396,
      "rewards/margins": 0.079404316842556,
      "rewards/rejected": -0.5173256993293762,
      "step": 920
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.295969449372796e-07,
      "logits/chosen": -2.682236433029175,
      "logits/rejected": -2.706988573074341,
      "logps/chosen": -434.9564514160156,
      "logps/rejected": -458.70831298828125,
      "loss": 0.6677,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.38662606477737427,
      "rewards/margins": 0.08299224078655243,
      "rewards/rejected": -0.4696182608604431,
      "step": 930
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.758486067922176e-07,
      "logits/chosen": -2.68632435798645,
      "logits/rejected": -2.700981616973877,
      "logps/chosen": -425.8504943847656,
      "logps/rejected": -465.1412048339844,
      "loss": 0.651,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.372600257396698,
      "rewards/margins": 0.1192459687590599,
      "rewards/rejected": -0.4918461740016937,
      "step": 940
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.233689873990006e-07,
      "logits/chosen": -2.661367654800415,
      "logits/rejected": -2.663963794708252,
      "logps/chosen": -430.2080993652344,
      "logps/rejected": -466.76092529296875,
      "loss": 0.6451,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.3554816246032715,
      "rewards/margins": 0.12667140364646912,
      "rewards/rejected": -0.4821530282497406,
      "step": 950
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.721990816288555e-07,
      "logits/chosen": -2.7383949756622314,
      "logits/rejected": -2.7589166164398193,
      "logps/chosen": -413.57135009765625,
      "logps/rejected": -441.5545349121094,
      "loss": 0.6724,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.371616005897522,
      "rewards/margins": 0.06721927225589752,
      "rewards/rejected": -0.4388352930545807,
      "step": 960
    },
    {
      "epoch": 0.78,
      "learning_rate": 7.223788612598148e-07,
      "logits/chosen": -2.7295823097229004,
      "logits/rejected": -2.740691661834717,
      "logps/chosen": -443.89776611328125,
      "logps/rejected": -474.42254638671875,
      "loss": 0.6495,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.332157701253891,
      "rewards/margins": 0.1200638860464096,
      "rewards/rejected": -0.4522215723991394,
      "step": 970
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.73947243752448e-07,
      "logits/chosen": -2.70650053024292,
      "logits/rejected": -2.6953940391540527,
      "logps/chosen": -460.09918212890625,
      "logps/rejected": -455.58489990234375,
      "loss": 0.6806,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.4144526422023773,
      "rewards/margins": 0.05592719465494156,
      "rewards/rejected": -0.4703798294067383,
      "step": 980
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.269420618491759e-07,
      "logits/chosen": -2.7356605529785156,
      "logits/rejected": -2.737133026123047,
      "logps/chosen": -423.57537841796875,
      "logps/rejected": -452.10260009765625,
      "loss": 0.6665,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.3517349660396576,
      "rewards/margins": 0.08071886003017426,
      "rewards/rejected": -0.43245387077331543,
      "step": 990
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.814000340209267e-07,
      "logits/chosen": -2.7007267475128174,
      "logits/rejected": -2.700061798095703,
      "logps/chosen": -442.41705322265625,
      "logps/rejected": -477.9020080566406,
      "loss": 0.657,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.34119993448257446,
      "rewards/margins": 0.10114644467830658,
      "rewards/rejected": -0.44234639406204224,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -2.570801258087158,
      "eval_logits/rejected": -2.578376293182373,
      "eval_logps/chosen": -350.23724365234375,
      "eval_logps/rejected": -366.5909118652344,
      "eval_loss": 0.6711118221282959,
      "eval_rewards/accuracies": 0.5996666550636292,
      "eval_rewards/chosen": -0.277418851852417,
      "eval_rewards/margins": 0.06656476855278015,
      "eval_rewards/rejected": -0.34398359060287476,
      "eval_runtime": 1581.2525,
      "eval_samples_per_second": 1.893,
      "eval_steps_per_second": 0.237,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.373567357842111e-07,
      "logits/chosen": -2.6614608764648438,
      "logits/rejected": -2.694235324859619,
      "logps/chosen": -432.1556091308594,
      "logps/rejected": -458.6222229003906,
      "loss": 0.6625,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.2784090042114258,
      "rewards/margins": 0.0895313173532486,
      "rewards/rejected": -0.36794033646583557,
      "step": 1010
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.948465719110226e-07,
      "logits/chosen": -2.6836206912994385,
      "logits/rejected": -2.697563409805298,
      "logps/chosen": -415.11041259765625,
      "logps/rejected": -438.9033203125,
      "loss": 0.6496,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.31501251459121704,
      "rewards/margins": 0.11396019160747528,
      "rewards/rejected": -0.4289727210998535,
      "step": 1020
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.539027495532766e-07,
      "logits/chosen": -2.6660382747650146,
      "logits/rejected": -2.6685073375701904,
      "logps/chosen": -425.77490234375,
      "logps/rejected": -446.0552673339844,
      "loss": 0.6661,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.35333576798439026,
      "rewards/margins": 0.08899492025375366,
      "rewards/rejected": -0.4423306882381439,
      "step": 1030
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.14557252302783e-07,
      "logits/chosen": -2.6789193153381348,
      "logits/rejected": -2.6913952827453613,
      "logps/chosen": -429.2127990722656,
      "logps/rejected": -461.7294006347656,
      "loss": 0.6592,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.3878576159477234,
      "rewards/margins": 0.0975080356001854,
      "rewards/rejected": -0.48536568880081177,
      "step": 1040
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7684081520700884e-07,
      "logits/chosen": -2.6964402198791504,
      "logits/rejected": -2.70263934135437,
      "logps/chosen": -462.135986328125,
      "logps/rejected": -464.80035400390625,
      "loss": 0.6855,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.437302827835083,
      "rewards/margins": 0.041490085422992706,
      "rewards/rejected": -0.4787929654121399,
      "step": 1050
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.407829007601507e-07,
      "logits/chosen": -2.6650023460388184,
      "logits/rejected": -2.6826632022857666,
      "logps/chosen": -443.31982421875,
      "logps/rejected": -482.0733947753906,
      "loss": 0.6466,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.37219929695129395,
      "rewards/margins": 0.13468633592128754,
      "rewards/rejected": -0.5068856477737427,
      "step": 1060
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.064116758882724e-07,
      "logits/chosen": -2.704041004180908,
      "logits/rejected": -2.712378978729248,
      "logps/chosen": -449.221435546875,
      "logps/rejected": -488.82733154296875,
      "loss": 0.6569,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.38919639587402344,
      "rewards/margins": 0.10973678529262543,
      "rewards/rejected": -0.49893322587013245,
      "step": 1070
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.737539899464908e-07,
      "logits/chosen": -2.66728138923645,
      "logits/rejected": -2.6824071407318115,
      "logps/chosen": -408.9554138183594,
      "logps/rejected": -435.3258361816406,
      "loss": 0.6621,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.34276992082595825,
      "rewards/margins": 0.08587387949228287,
      "rewards/rejected": -0.4286437928676605,
      "step": 1080
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4283535374538645e-07,
      "logits/chosen": -2.731504440307617,
      "logits/rejected": -2.733990430831909,
      "logps/chosen": -450.702392578125,
      "logps/rejected": -473.677490234375,
      "loss": 0.6486,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.3203676640987396,
      "rewards/margins": 0.12553098797798157,
      "rewards/rejected": -0.4458986818790436,
      "step": 1090
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1367991962303298e-07,
      "logits/chosen": -2.6853950023651123,
      "logits/rejected": -2.702267646789551,
      "logps/chosen": -415.25048828125,
      "logps/rejected": -443.6421813964844,
      "loss": 0.6577,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.361422061920166,
      "rewards/margins": 0.10248668491840363,
      "rewards/rejected": -0.46390876173973083,
      "step": 1100
    },
    {
      "epoch": 0.88,
      "eval_logits/chosen": -2.5729336738586426,
      "eval_logits/rejected": -2.580476999282837,
      "eval_logps/chosen": -351.83758544921875,
      "eval_logps/rejected": -368.46795654296875,
      "eval_loss": 0.6706418395042419,
      "eval_rewards/accuracies": 0.5993333458900452,
      "eval_rewards/chosen": -0.29342252016067505,
      "eval_rewards/margins": 0.06933192163705826,
      "eval_rewards/rejected": -0.3627544641494751,
      "eval_runtime": 1589.982,
      "eval_samples_per_second": 1.883,
      "eval_steps_per_second": 0.236,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.8631046257820278e-07,
      "logits/chosen": -2.713442802429199,
      "logits/rejected": -2.7153563499450684,
      "logps/chosen": -448.9380798339844,
      "logps/rejected": -459.6236267089844,
      "loss": 0.6822,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.40070003271102905,
      "rewards/margins": 0.04853721335530281,
      "rewards/rejected": -0.44923728704452515,
      "step": 1110
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6074836247950143e-07,
      "logits/chosen": -2.7305519580841064,
      "logits/rejected": -2.727928638458252,
      "logps/chosen": -458.64208984375,
      "logps/rejected": -479.4064025878906,
      "loss": 0.6573,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.35037961602211,
      "rewards/margins": 0.09948080033063889,
      "rewards/rejected": -0.4498603940010071,
      "step": 1120
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.370135873643097e-07,
      "logits/chosen": -2.6915557384490967,
      "logits/rejected": -2.675395965576172,
      "logps/chosen": -444.5713806152344,
      "logps/rejected": -442.8504943847656,
      "loss": 0.6738,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3383481502532959,
      "rewards/margins": 0.06244407966732979,
      "rewards/rejected": -0.4007922112941742,
      "step": 1130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1512467784059372e-07,
      "logits/chosen": -2.697908639907837,
      "logits/rejected": -2.7043871879577637,
      "logps/chosen": -409.3360595703125,
      "logps/rejected": -430.56939697265625,
      "loss": 0.6764,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.353982150554657,
      "rewards/margins": 0.07203380763530731,
      "rewards/rejected": -0.4260159432888031,
      "step": 1140
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.509873260376251e-08,
      "logits/chosen": -2.6704964637756348,
      "logits/rejected": -2.680117130279541,
      "logps/chosen": -426.01568603515625,
      "logps/rejected": -481.14056396484375,
      "loss": 0.6301,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.292033851146698,
      "rewards/margins": 0.16145673394203186,
      "rewards/rejected": -0.45349058508872986,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.695139507988559e-08,
      "logits/chosen": -2.690162181854248,
      "logits/rejected": -2.6840763092041016,
      "logps/chosen": -455.21429443359375,
      "logps/rejected": -471.9693908691406,
      "loss": 0.6911,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.39844316244125366,
      "rewards/margins": 0.04045991227030754,
      "rewards/rejected": -0.4389030337333679,
      "step": 1160
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.069684120570684e-08,
      "logits/chosen": -2.6355466842651367,
      "logits/rejected": -2.6722214221954346,
      "logps/chosen": -428.15887451171875,
      "logps/rejected": -469.9241638183594,
      "loss": 0.6556,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.31940728425979614,
      "rewards/margins": 0.10891290009021759,
      "rewards/rejected": -0.42832016944885254,
      "step": 1170
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.634776835499871e-08,
      "logits/chosen": -2.648555278778076,
      "logits/rejected": -2.648024082183838,
      "logps/chosen": -414.77166748046875,
      "logps/rejected": -438.7622985839844,
      "loss": 0.6786,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.33630624413490295,
      "rewards/margins": 0.059269875288009644,
      "rewards/rejected": -0.3955760896205902,
      "step": 1180
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.3915385419908964e-08,
      "logits/chosen": -2.724853992462158,
      "logits/rejected": -2.7020461559295654,
      "logps/chosen": -443.90399169921875,
      "logps/rejected": -462.37384033203125,
      "loss": 0.667,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.35228732228279114,
      "rewards/margins": 0.08694271743297577,
      "rewards/rejected": -0.4392300546169281,
      "step": 1190
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.3409404055043938e-08,
      "logits/chosen": -2.6520884037017822,
      "logits/rejected": -2.677919387817383,
      "logps/chosen": -431.74053955078125,
      "logps/rejected": -462.65179443359375,
      "loss": 0.6444,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.3235710561275482,
      "rewards/margins": 0.12747570872306824,
      "rewards/rejected": -0.45104676485061646,
      "step": 1200
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -2.5725319385528564,
      "eval_logits/rejected": -2.580101490020752,
      "eval_logps/chosen": -351.09490966796875,
      "eval_logps/rejected": -367.6591796875,
      "eval_loss": 0.670768678188324,
      "eval_rewards/accuracies": 0.5993333458900452,
      "eval_rewards/chosen": -0.2859960198402405,
      "eval_rewards/margins": 0.06867069751024246,
      "eval_rewards/rejected": -0.35466670989990234,
      "eval_runtime": 1585.5273,
      "eval_samples_per_second": 1.888,
      "eval_steps_per_second": 0.237,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4838031091134186e-08,
      "logits/chosen": -2.6522107124328613,
      "logits/rejected": -2.6851000785827637,
      "logps/chosen": -404.271484375,
      "logps/rejected": -455.322998046875,
      "loss": 0.6397,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.3024134933948517,
      "rewards/margins": 0.13375119864940643,
      "rewards/rejected": -0.4361646771430969,
      "step": 1210
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.207962124201774e-09,
      "logits/chosen": -2.687406063079834,
      "logits/rejected": -2.6915080547332764,
      "logps/chosen": -446.8517150878906,
      "logps/rejected": -463.1326599121094,
      "loss": 0.6692,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.33363211154937744,
      "rewards/margins": 0.08175511658191681,
      "rewards/rejected": -0.41538724303245544,
      "step": 1220
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.5243762852441023e-09,
      "logits/chosen": -2.68792462348938,
      "logits/rejected": -2.705341339111328,
      "logps/chosen": -433.271484375,
      "logps/rejected": -466.77362060546875,
      "loss": 0.6609,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.34372276067733765,
      "rewards/margins": 0.1005091518163681,
      "rewards/rejected": -0.44423192739486694,
      "step": 1230
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.909321945129278e-10,
      "logits/chosen": -2.704862117767334,
      "logits/rejected": -2.7122840881347656,
      "logps/chosen": -461.7477111816406,
      "logps/rejected": -496.6314392089844,
      "loss": 0.6514,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.3258623480796814,
      "rewards/margins": 0.12274013459682465,
      "rewards/rejected": -0.4486025273799896,
      "step": 1240
    },
    {
      "epoch": 1.0,
      "step": 1249,
      "total_flos": 0.0,
      "train_loss": 0.67484232365751,
      "train_runtime": 40604.4701,
      "train_samples_per_second": 0.492,
      "train_steps_per_second": 0.031
    }
  ],
  "logging_steps": 10,
  "max_steps": 1249,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}