{
  "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": -2.6824121475219727,
      "logits/rejected": -2.7049124240875244,
      "logps/chosen": -275.1597900390625,
      "logps/rejected": -271.6430969238281,
      "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": -2.6062471866607666,
      "logits/rejected": -2.633519172668457,
      "logps/chosen": -301.788818359375,
      "logps/rejected": -324.23992919921875,
      "loss": 0.6934,
      "rewards/accuracies": 0.4027777910232544,
      "rewards/chosen": -0.0003357415844220668,
      "rewards/margins": -0.00043744672439061105,
      "rewards/rejected": 0.00010170514724450186,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.000000000000001e-07,
      "logits/chosen": -2.5865533351898193,
      "logits/rejected": -2.589942693710327,
      "logps/chosen": -269.1646423339844,
      "logps/rejected": -289.1620178222656,
      "loss": 0.6934,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": -0.0017813893500715494,
      "rewards/margins": -0.00042842660332098603,
      "rewards/rejected": -0.0013529626885429025,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2000000000000002e-06,
      "logits/chosen": -2.5753893852233887,
      "logits/rejected": -2.58815860748291,
      "logps/chosen": -291.62603759765625,
      "logps/rejected": -311.4107971191406,
      "loss": 0.6933,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.0035930208396166563,
      "rewards/margins": -0.00022997017367742956,
      "rewards/rejected": -0.0033630509860813618,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6000000000000001e-06,
      "logits/chosen": -2.612914562225342,
      "logits/rejected": -2.619166135787964,
      "logps/chosen": -264.53851318359375,
      "logps/rejected": -273.5920104980469,
      "loss": 0.6929,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.005524822510778904,
      "rewards/margins": 0.0004894191515631974,
      "rewards/rejected": -0.006014241836965084,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0000000000000003e-06,
      "logits/chosen": -2.56376576423645,
      "logits/rejected": -2.5438730716705322,
      "logps/chosen": -264.94989013671875,
      "logps/rejected": -269.7061767578125,
      "loss": 0.6932,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.008829522877931595,
      "rewards/margins": -0.0001453599688829854,
      "rewards/rejected": -0.008684162981808186,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4000000000000003e-06,
      "logits/chosen": -2.630518674850464,
      "logits/rejected": -2.6339123249053955,
      "logps/chosen": -277.233642578125,
      "logps/rejected": -296.16107177734375,
      "loss": 0.6919,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.013874058611690998,
      "rewards/margins": 0.0025573372840881348,
      "rewards/rejected": -0.016431394964456558,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8000000000000003e-06,
      "logits/chosen": -2.623034954071045,
      "logits/rejected": -2.6133813858032227,
      "logps/chosen": -280.8143005371094,
      "logps/rejected": -286.6803283691406,
      "loss": 0.693,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.01927858218550682,
      "rewards/margins": 0.00032674067188054323,
      "rewards/rejected": -0.019605323672294617,
      "step": 70
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.2000000000000003e-06,
      "logits/chosen": -2.6520907878875732,
      "logits/rejected": -2.6506881713867188,
      "logps/chosen": -276.94451904296875,
      "logps/rejected": -297.6499328613281,
      "loss": 0.6924,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.017164330929517746,
      "rewards/margins": 0.0016351321246474981,
      "rewards/rejected": -0.018799465149641037,
      "step": 80
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6000000000000003e-06,
      "logits/chosen": -2.6500167846679688,
      "logits/rejected": -2.6388657093048096,
      "logps/chosen": -311.24853515625,
      "logps/rejected": -316.82427978515625,
      "loss": 0.6891,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.01826111041009426,
      "rewards/margins": 0.008519862778484821,
      "rewards/rejected": -0.026780972257256508,
      "step": 90
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -2.649533271789551,
      "logits/rejected": -2.647923231124878,
      "logps/chosen": -276.5948486328125,
      "logps/rejected": -290.6126708984375,
      "loss": 0.6909,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.025616105645895004,
      "rewards/margins": 0.00467633968219161,
      "rewards/rejected": -0.030292445793747902,
      "step": 100
    },
    {
      "epoch": 0.08,
      "eval_logits/chosen": -2.6257693767547607,
      "eval_logits/rejected": -2.622962474822998,
      "eval_logps/chosen": -222.508056640625,
      "eval_logps/rejected": -228.3364715576172,
      "eval_loss": 0.6922365427017212,
      "eval_rewards/accuracies": 0.5236666798591614,
      "eval_rewards/chosen": -0.01801561377942562,
      "eval_rewards/margins": 0.0019916673190891743,
      "eval_rewards/rejected": -0.020007280632853508,
      "eval_runtime": 1605.2167,
      "eval_samples_per_second": 1.865,
      "eval_steps_per_second": 0.234,
      "step": 100
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.4e-06,
      "logits/chosen": -2.6500422954559326,
      "logits/rejected": -2.6527457237243652,
      "logps/chosen": -299.06976318359375,
      "logps/rejected": -309.92266845703125,
      "loss": 0.6888,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.026254886761307716,
      "rewards/margins": 0.009161447174847126,
      "rewards/rejected": -0.03541633114218712,
      "step": 110
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.800000000000001e-06,
      "logits/chosen": -2.6758484840393066,
      "logits/rejected": -2.6881041526794434,
      "logps/chosen": -278.5473327636719,
      "logps/rejected": -296.552734375,
      "loss": 0.6901,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.03307301551103592,
      "rewards/margins": 0.006715013645589352,
      "rewards/rejected": -0.039788030087947845,
      "step": 120
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.999755876225375e-06,
      "logits/chosen": -2.6458847522735596,
      "logits/rejected": -2.6281027793884277,
      "logps/chosen": -294.466796875,
      "logps/rejected": -315.05035400390625,
      "loss": 0.6875,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.036008208990097046,
      "rewards/margins": 0.011871300637722015,
      "rewards/rejected": -0.04787950962781906,
      "step": 130
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.997803172081864e-06,
      "logits/chosen": -2.6800694465637207,
      "logits/rejected": -2.6806652545928955,
      "logps/chosen": -289.0479736328125,
      "logps/rejected": -302.6410217285156,
      "loss": 0.685,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.03994801267981529,
      "rewards/margins": 0.017537599429488182,
      "rewards/rejected": -0.057485610246658325,
      "step": 140
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9938992891651825e-06,
      "logits/chosen": -2.662532091140747,
      "logits/rejected": -2.652069568634033,
      "logps/chosen": -277.6954040527344,
      "logps/rejected": -300.7398986816406,
      "loss": 0.6833,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.05284956097602844,
      "rewards/margins": 0.02098439633846283,
      "rewards/rejected": -0.07383395731449127,
      "step": 150
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.988047277024456e-06,
      "logits/chosen": -2.7222561836242676,
      "logits/rejected": -2.732853412628174,
      "logps/chosen": -288.4869689941406,
      "logps/rejected": -303.99114990234375,
      "loss": 0.6812,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.07921075075864792,
      "rewards/margins": 0.024997711181640625,
      "rewards/rejected": -0.10420846939086914,
      "step": 160
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.980251707005417e-06,
      "logits/chosen": -2.719470500946045,
      "logits/rejected": -2.6922366619110107,
      "logps/chosen": -307.56341552734375,
      "logps/rejected": -315.6242370605469,
      "loss": 0.6891,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.10631716251373291,
      "rewards/margins": 0.0096724983304739,
      "rewards/rejected": -0.11598964780569077,
      "step": 170
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.970518668679459e-06,
      "logits/chosen": -2.7305121421813965,
      "logits/rejected": -2.715012311935425,
      "logps/chosen": -303.9269104003906,
      "logps/rejected": -311.3086853027344,
      "loss": 0.6821,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.08977223932743073,
      "rewards/margins": 0.02493356727063656,
      "rewards/rejected": -0.11470580101013184,
      "step": 180
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.958855765086722e-06,
      "logits/chosen": -2.7576236724853516,
      "logits/rejected": -2.754725933074951,
      "logps/chosen": -283.64117431640625,
      "logps/rejected": -293.0472717285156,
      "loss": 0.6843,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.058794498443603516,
      "rewards/margins": 0.02001199498772621,
      "rewards/rejected": -0.07880649715662003,
      "step": 190
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.945272106796919e-06,
      "logits/chosen": -2.7707252502441406,
      "logits/rejected": -2.7755463123321533,
      "logps/chosen": -284.79974365234375,
      "logps/rejected": -300.4746398925781,
      "loss": 0.6843,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.07238452136516571,
      "rewards/margins": 0.021619267761707306,
      "rewards/rejected": -0.09400378167629242,
      "step": 200
    },
    {
      "epoch": 0.16,
      "eval_logits/chosen": -2.7632017135620117,
      "eval_logits/rejected": -2.7602405548095703,
      "eval_logps/chosen": -228.6281280517578,
      "eval_logps/rejected": -235.652587890625,
      "eval_loss": 0.6873453259468079,
      "eval_rewards/accuracies": 0.5546666383743286,
      "eval_rewards/chosen": -0.07921627163887024,
      "eval_rewards/margins": 0.013952008448541164,
      "eval_rewards/rejected": -0.09316828101873398,
      "eval_runtime": 1606.4122,
      "eval_samples_per_second": 1.864,
      "eval_steps_per_second": 0.233,
      "step": 200
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.929778304792537e-06,
      "logits/chosen": -2.7544033527374268,
      "logits/rejected": -2.758392333984375,
      "logps/chosen": -311.075439453125,
      "logps/rejected": -315.38006591796875,
      "loss": 0.6759,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.09139387309551239,
      "rewards/margins": 0.04242325574159622,
      "rewards/rejected": -0.1338171362876892,
      "step": 210
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.912386462179987e-06,
      "logits/chosen": -2.783567190170288,
      "logits/rejected": -2.779317617416382,
      "logps/chosen": -298.64764404296875,
      "logps/rejected": -325.9937744140625,
      "loss": 0.6736,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.10082165896892548,
      "rewards/margins": 0.04564264789223671,
      "rewards/rejected": -0.14646431803703308,
      "step": 220
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.893110164735167e-06,
      "logits/chosen": -2.8824093341827393,
      "logits/rejected": -2.879509449005127,
      "logps/chosen": -305.2967224121094,
      "logps/rejected": -319.56695556640625,
      "loss": 0.6795,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.13284233212471008,
      "rewards/margins": 0.03221544623374939,
      "rewards/rejected": -0.16505777835845947,
      "step": 230
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.871964470290823e-06,
      "logits/chosen": -2.902723789215088,
      "logits/rejected": -2.9203548431396484,
      "logps/chosen": -309.1792907714844,
      "logps/rejected": -331.7867736816406,
      "loss": 0.6664,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.15928030014038086,
      "rewards/margins": 0.0632576271891594,
      "rewards/rejected": -0.22253794968128204,
      "step": 240
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.848965896974006e-06,
      "logits/chosen": -2.9424166679382324,
      "logits/rejected": -2.9350640773773193,
      "logps/chosen": -302.81573486328125,
      "logps/rejected": -325.1131896972656,
      "loss": 0.676,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.19259200990200043,
      "rewards/margins": 0.042614568024873734,
      "rewards/rejected": -0.23520657420158386,
      "step": 250
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8241324103028055e-06,
      "logits/chosen": -3.106982707977295,
      "logits/rejected": -3.080066204071045,
      "logps/chosen": -312.7992248535156,
      "logps/rejected": -328.4107360839844,
      "loss": 0.6624,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.24676008522510529,
      "rewards/margins": 0.07433095574378967,
      "rewards/rejected": -0.32109108567237854,
      "step": 260
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.797483409152438e-06,
      "logits/chosen": -3.210594892501831,
      "logits/rejected": -3.2006404399871826,
      "logps/chosen": -308.2965393066406,
      "logps/rejected": -333.50439453125,
      "loss": 0.6611,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.33079180121421814,
      "rewards/margins": 0.08313147723674774,
      "rewards/rejected": -0.4139232635498047,
      "step": 270
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.769039710601669e-06,
      "logits/chosen": -3.357815980911255,
      "logits/rejected": -3.363719940185547,
      "logps/chosen": -316.1602783203125,
      "logps/rejected": -338.216552734375,
      "loss": 0.6612,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.42240840196609497,
      "rewards/margins": 0.08246298134326935,
      "rewards/rejected": -0.5048713088035583,
      "step": 280
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.738823533671383e-06,
      "logits/chosen": -3.4880664348602295,
      "logits/rejected": -3.4769935607910156,
      "logps/chosen": -351.0909729003906,
      "logps/rejected": -368.14385986328125,
      "loss": 0.6803,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.5264540314674377,
      "rewards/margins": 0.04465585574507713,
      "rewards/rejected": -0.5711098909378052,
      "step": 290
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.706858481968017e-06,
      "logits/chosen": -3.45086407661438,
      "logits/rejected": -3.455495834350586,
      "logps/chosen": -340.1499938964844,
      "logps/rejected": -352.09429931640625,
      "loss": 0.68,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.5941203236579895,
      "rewards/margins": 0.046401310712099075,
      "rewards/rejected": -0.6405216455459595,
      "step": 300
    },
    {
      "epoch": 0.24,
      "eval_logits/chosen": -3.506669044494629,
      "eval_logits/rejected": -3.501077175140381,
      "eval_logps/chosen": -272.03741455078125,
      "eval_logps/rejected": -281.38311767578125,
      "eval_loss": 0.6839653253555298,
      "eval_rewards/accuracies": 0.5473333597183228,
      "eval_rewards/chosen": -0.5133091807365417,
      "eval_rewards/margins": 0.03716452047228813,
      "eval_rewards/rejected": -0.5504736304283142,
      "eval_runtime": 1603.1979,
      "eval_samples_per_second": 1.868,
      "eval_steps_per_second": 0.234,
      "step": 300
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.673169525245416e-06,
      "logits/chosen": -3.4355697631835938,
      "logits/rejected": -3.412278652191162,
      "logps/chosen": -336.8961181640625,
      "logps/rejected": -369.5231018066406,
      "loss": 0.6613,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.5616176724433899,
      "rewards/margins": 0.09071463346481323,
      "rewards/rejected": -0.6523322463035583,
      "step": 310
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.63778297989952e-06,
      "logits/chosen": -3.588507890701294,
      "logits/rejected": -3.570230484008789,
      "logps/chosen": -343.86956787109375,
      "logps/rejected": -364.1554260253906,
      "loss": 0.663,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5944578051567078,
      "rewards/margins": 0.08861590176820755,
      "rewards/rejected": -0.6830736994743347,
      "step": 320
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.60072648841109e-06,
      "logits/chosen": -3.744804859161377,
      "logits/rejected": -3.735940456390381,
      "logps/chosen": -364.733642578125,
      "logps/rejected": -394.76348876953125,
      "loss": 0.6249,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.7255194783210754,
      "rewards/margins": 0.18435899913311005,
      "rewards/rejected": -0.9098785519599915,
      "step": 330
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.562028997752574e-06,
      "logits/chosen": -3.9339213371276855,
      "logits/rejected": -3.9240550994873047,
      "logps/chosen": -382.0846252441406,
      "logps/rejected": -410.460205078125,
      "loss": 0.6783,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.9037133455276489,
      "rewards/margins": 0.0800480842590332,
      "rewards/rejected": -0.9837613105773926,
      "step": 340
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.521720736775947e-06,
      "logits/chosen": -3.9775962829589844,
      "logits/rejected": -4.001706123352051,
      "logps/chosen": -399.79351806640625,
      "logps/rejected": -413.55987548828125,
      "loss": 0.655,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.040808916091919,
      "rewards/margins": 0.1233123168349266,
      "rewards/rejected": -1.1641212701797485,
      "step": 350
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.479833192599198e-06,
      "logits/chosen": -3.9514153003692627,
      "logits/rejected": -3.929912567138672,
      "logps/chosen": -390.7148132324219,
      "logps/rejected": -413.5050354003906,
      "loss": 0.6615,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.901436984539032,
      "rewards/margins": 0.10929825156927109,
      "rewards/rejected": -1.010735273361206,
      "step": 360
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.436399086009928e-06,
      "logits/chosen": -3.7934257984161377,
      "logits/rejected": -3.7594401836395264,
      "logps/chosen": -366.05413818359375,
      "logps/rejected": -387.0696716308594,
      "loss": 0.6481,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.7858292460441589,
      "rewards/margins": 0.1209360808134079,
      "rewards/rejected": -0.9067652821540833,
      "step": 370
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.391452345905239e-06,
      "logits/chosen": -3.6905906200408936,
      "logits/rejected": -3.701895236968994,
      "logps/chosen": -377.90118408203125,
      "logps/rejected": -394.57147216796875,
      "loss": 0.6598,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.8731447458267212,
      "rewards/margins": 0.11739823967218399,
      "rewards/rejected": -0.990543007850647,
      "step": 380
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3450280827879125e-06,
      "logits/chosen": -3.7442736625671387,
      "logits/rejected": -3.769087553024292,
      "logps/chosen": -379.1444091796875,
      "logps/rejected": -399.7411804199219,
      "loss": 0.6601,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.9029915928840637,
      "rewards/margins": 0.11410637944936752,
      "rewards/rejected": -1.0170979499816895,
      "step": 390
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297162561339554e-06,
      "logits/chosen": -3.6351356506347656,
      "logits/rejected": -3.6010661125183105,
      "logps/chosen": -385.5706481933594,
      "logps/rejected": -412.40130615234375,
      "loss": 0.657,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.9261330366134644,
      "rewards/margins": 0.13883258402347565,
      "rewards/rejected": -1.0649656057357788,
      "step": 400
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -3.7522990703582764,
      "eval_logits/rejected": -3.7427072525024414,
      "eval_logps/chosen": -306.9489440917969,
      "eval_logps/rejected": -319.11859130859375,
      "eval_loss": 0.6807547211647034,
      "eval_rewards/accuracies": 0.5583333373069763,
      "eval_rewards/chosen": -0.862424373626709,
      "eval_rewards/margins": 0.06540393829345703,
      "eval_rewards/rejected": -0.9278281927108765,
      "eval_runtime": 1662.1993,
      "eval_samples_per_second": 1.801,
      "eval_steps_per_second": 0.226,
      "step": 400
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.247893172092157e-06,
      "logits/chosen": -3.6088128089904785,
      "logits/rejected": -3.6062328815460205,
      "logps/chosen": -375.0890197753906,
      "logps/rejected": -409.6904296875,
      "loss": 0.6588,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.948278546333313,
      "rewards/margins": 0.12885506451129913,
      "rewards/rejected": -1.0771336555480957,
      "step": 410
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.197258402220187e-06,
      "logits/chosen": -3.6457858085632324,
      "logits/rejected": -3.658412218093872,
      "logps/chosen": -383.5570068359375,
      "logps/rejected": -422.8628845214844,
      "loss": 0.6555,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.950586199760437,
      "rewards/margins": 0.34554851055145264,
      "rewards/rejected": -1.2961347103118896,
      "step": 420
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.145297805476023e-06,
      "logits/chosen": -3.6771697998046875,
      "logits/rejected": -3.6845245361328125,
      "logps/chosen": -379.49029541015625,
      "logps/rejected": -410.7177734375,
      "loss": 0.6399,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.9332340359687805,
      "rewards/margins": 0.1954960823059082,
      "rewards/rejected": -1.1287301778793335,
      "step": 430
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.092051971292228e-06,
      "logits/chosen": -3.715477705001831,
      "logits/rejected": -3.7073216438293457,
      "logps/chosen": -379.3064270019531,
      "logps/rejected": -404.5085144042969,
      "loss": 0.6564,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9848654866218567,
      "rewards/margins": 0.13351285457611084,
      "rewards/rejected": -1.1183784008026123,
      "step": 440
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.037562493074792e-06,
      "logits/chosen": -3.800858974456787,
      "logits/rejected": -3.8201656341552734,
      "logps/chosen": -404.48602294921875,
      "logps/rejected": -423.95330810546875,
      "loss": 0.6378,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.056680679321289,
      "rewards/margins": 0.18673589825630188,
      "rewards/rejected": -1.2434165477752686,
      "step": 450
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.981871935712112e-06,
      "logits/chosen": -3.9585278034210205,
      "logits/rejected": -3.9149413108825684,
      "logps/chosen": -382.9043884277344,
      "logps/rejected": -413.051025390625,
      "loss": 0.6556,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -1.0201416015625,
      "rewards/margins": 0.13921280205249786,
      "rewards/rejected": -1.1593544483184814,
      "step": 460
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.925023802325094e-06,
      "logits/chosen": -4.015332221984863,
      "logits/rejected": -3.9858086109161377,
      "logps/chosen": -397.7122802734375,
      "logps/rejected": -432.3436584472656,
      "loss": 0.6466,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.045425295829773,
      "rewards/margins": 0.2483961582183838,
      "rewards/rejected": -1.2938215732574463,
      "step": 470
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.867062500284342e-06,
      "logits/chosen": -4.052220344543457,
      "logits/rejected": -4.021946907043457,
      "logps/chosen": -385.6368103027344,
      "logps/rejected": -421.88824462890625,
      "loss": 0.6432,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.0936793088912964,
      "rewards/margins": 0.17554286122322083,
      "rewards/rejected": -1.2692222595214844,
      "step": 480
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8080333065209885e-06,
      "logits/chosen": -4.067862510681152,
      "logits/rejected": -4.076433181762695,
      "logps/chosen": -389.60333251953125,
      "logps/rejected": -390.48321533203125,
      "loss": 0.7053,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -1.09650719165802,
      "rewards/margins": 0.06325383484363556,
      "rewards/rejected": -1.1597610712051392,
      "step": 490
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7479823321582624e-06,
      "logits/chosen": -3.942683458328247,
      "logits/rejected": -3.9045321941375732,
      "logps/chosen": -378.3946228027344,
      "logps/rejected": -416.160400390625,
      "loss": 0.6342,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.91758793592453,
      "rewards/margins": 0.18643911182880402,
      "rewards/rejected": -1.1040270328521729,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_logits/chosen": -4.11630392074585,
      "eval_logits/rejected": -4.106530666351318,
      "eval_logps/chosen": -313.0755920410156,
      "eval_logps/rejected": -326.7316589355469,
      "eval_loss": 0.6783922910690308,
      "eval_rewards/accuracies": 0.5580000281333923,
      "eval_rewards/chosen": -0.9236910939216614,
      "eval_rewards/margins": 0.08026820421218872,
      "eval_rewards/rejected": -1.00395929813385,
      "eval_runtime": 1601.521,
      "eval_samples_per_second": 1.869,
      "eval_steps_per_second": 0.234,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.686956486491419e-06,
      "logits/chosen": -3.9557979106903076,
      "logits/rejected": -3.950779676437378,
      "logps/chosen": -388.673583984375,
      "logps/rejected": -427.42962646484375,
      "loss": 0.6299,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.9443014860153198,
      "rewards/margins": 0.30186328291893005,
      "rewards/rejected": -1.2461649179458618,
      "step": 510
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.625003440344166e-06,
      "logits/chosen": -4.053747653961182,
      "logits/rejected": -4.082505226135254,
      "logps/chosen": -371.04339599609375,
      "logps/rejected": -383.62921142578125,
      "loss": 0.6644,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.9302754402160645,
      "rewards/margins": 0.08710617572069168,
      "rewards/rejected": -1.0173815488815308,
      "step": 520
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.562171588830231e-06,
      "logits/chosen": -4.011529445648193,
      "logits/rejected": -3.9905009269714355,
      "logps/chosen": -379.8844299316406,
      "logps/rejected": -406.698974609375,
      "loss": 0.699,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.9027377963066101,
      "rewards/margins": 0.07473494112491608,
      "rewards/rejected": -0.9774727821350098,
      "step": 530
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.4985100135491245e-06,
      "logits/chosen": -3.9951107501983643,
      "logits/rejected": -3.956082582473755,
      "logps/chosen": -383.92889404296875,
      "logps/rejected": -426.62347412109375,
      "loss": 0.6307,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.8952111005783081,
      "rewards/margins": 0.21770212054252625,
      "rewards/rejected": -1.1129133701324463,
      "step": 540
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4340684442456673e-06,
      "logits/chosen": -4.0311408042907715,
      "logits/rejected": -4.022164821624756,
      "logps/chosen": -383.31878662109375,
      "logps/rejected": -409.2886047363281,
      "loss": 0.6428,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.953955352306366,
      "rewards/margins": 0.1485908329486847,
      "rewards/rejected": -1.1025463342666626,
      "step": 550
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3688972199631974e-06,
      "logits/chosen": -4.0242533683776855,
      "logits/rejected": -4.047430992126465,
      "logps/chosen": -385.8880920410156,
      "logps/rejected": -412.330810546875,
      "loss": 0.6328,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.9757701754570007,
      "rewards/margins": 0.2515910267829895,
      "rewards/rejected": -1.2273612022399902,
      "step": 560
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3030472497208354e-06,
      "logits/chosen": -4.045630931854248,
      "logits/rejected": -4.000221252441406,
      "logps/chosen": -382.03607177734375,
      "logps/rejected": -447.540283203125,
      "loss": 0.6153,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.047086477279663,
      "rewards/margins": 0.2707362473011017,
      "rewards/rejected": -1.3178224563598633,
      "step": 570
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.236569972745492e-06,
      "logits/chosen": -4.149961471557617,
      "logits/rejected": -4.139186382293701,
      "logps/chosen": -379.27294921875,
      "logps/rejected": -405.08392333984375,
      "loss": 0.6644,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.05259108543396,
      "rewards/margins": 0.16109797358512878,
      "rewards/rejected": -1.213688850402832,
      "step": 580
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1695173182897126e-06,
      "logits/chosen": -4.157639026641846,
      "logits/rejected": -4.137297630310059,
      "logps/chosen": -400.1607666015625,
      "logps/rejected": -438.38275146484375,
      "loss": 0.6558,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.2017654180526733,
      "rewards/margins": 0.17644351720809937,
      "rewards/rejected": -1.378208875656128,
      "step": 590
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.10194166506673e-06,
      "logits/chosen": -4.194487571716309,
      "logits/rejected": -4.153265476226807,
      "logps/chosen": -393.9078063964844,
      "logps/rejected": -442.1280822753906,
      "loss": 0.6341,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.0909817218780518,
      "rewards/margins": 0.2533242404460907,
      "rewards/rejected": -1.3443058729171753,
      "step": 600
    },
    {
      "epoch": 0.48,
      "eval_logits/chosen": -4.4554667472839355,
      "eval_logits/rejected": -4.444517135620117,
      "eval_logps/chosen": -340.80841064453125,
      "eval_logps/rejected": -357.2098693847656,
      "eval_loss": 0.6766642928123474,
      "eval_rewards/accuracies": 0.5630000233650208,
      "eval_rewards/chosen": -1.2010191679000854,
      "eval_rewards/margins": 0.1077219545841217,
      "eval_rewards/rejected": -1.3087410926818848,
      "eval_runtime": 1605.4995,
      "eval_samples_per_second": 1.865,
      "eval_steps_per_second": 0.234,
      "step": 600
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0338958003344115e-06,
      "logits/chosen": -4.3319549560546875,
      "logits/rejected": -4.279036045074463,
      "logps/chosen": -399.0089416503906,
      "logps/rejected": -435.59075927734375,
      "loss": 0.692,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.2491410970687866,
      "rewards/margins": 0.2489248812198639,
      "rewards/rejected": -1.4980661869049072,
      "step": 610
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9654328786600823e-06,
      "logits/chosen": -4.308805465698242,
      "logits/rejected": -4.258028984069824,
      "logps/chosen": -398.9922180175781,
      "logps/rejected": -443.50775146484375,
      "loss": 0.6204,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.2274248600006104,
      "rewards/margins": 0.2125413715839386,
      "rewards/rejected": -1.4399662017822266,
      "step": 620
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.896606380398402e-06,
      "logits/chosen": -4.377072334289551,
      "logits/rejected": -4.414391994476318,
      "logps/chosen": -420.06884765625,
      "logps/rejected": -445.095458984375,
      "loss": 0.6656,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.3716115951538086,
      "rewards/margins": 0.1795826405286789,
      "rewards/rejected": -1.551194190979004,
      "step": 630
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.827470069914772e-06,
      "logits/chosen": -4.2714762687683105,
      "logits/rejected": -4.23270320892334,
      "logps/chosen": -426.16943359375,
      "logps/rejected": -454.91278076171875,
      "loss": 0.6689,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -1.4019109010696411,
      "rewards/margins": 0.15336069464683533,
      "rewards/rejected": -1.5552715063095093,
      "step": 640
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7580779535868675e-06,
      "logits/chosen": -4.235100746154785,
      "logits/rejected": -4.236804008483887,
      "logps/chosen": -408.59002685546875,
      "logps/rejected": -438.03204345703125,
      "loss": 0.6618,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.2248241901397705,
      "rewards/margins": 0.1827341616153717,
      "rewards/rejected": -1.4075584411621094,
      "step": 650
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.688484237617129e-06,
      "logits/chosen": -4.117595672607422,
      "logits/rejected": -4.08989143371582,
      "logps/chosen": -397.66925048828125,
      "logps/rejected": -433.625732421875,
      "loss": 0.6206,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.0825097560882568,
      "rewards/margins": 0.2466181069612503,
      "rewards/rejected": -1.3291277885437012,
      "step": 660
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6187432856891585e-06,
      "logits/chosen": -4.0606913566589355,
      "logits/rejected": -4.052728176116943,
      "logps/chosen": -407.27392578125,
      "logps/rejected": -445.926025390625,
      "loss": 0.6409,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.151308298110962,
      "rewards/margins": 0.19334861636161804,
      "rewards/rejected": -1.344657063484192,
      "step": 670
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.548909576501096e-06,
      "logits/chosen": -4.1063737869262695,
      "logits/rejected": -4.097175121307373,
      "logps/chosen": -412.988037109375,
      "logps/rejected": -442.6796875,
      "loss": 0.6362,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -1.2146575450897217,
      "rewards/margins": 0.1937810331583023,
      "rewards/rejected": -1.4084386825561523,
      "step": 680
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4790376612091503e-06,
      "logits/chosen": -4.220733642578125,
      "logits/rejected": -4.182897090911865,
      "logps/chosen": -444.1158142089844,
      "logps/rejected": -476.25286865234375,
      "loss": 0.6217,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.4591256380081177,
      "rewards/margins": 0.2400621622800827,
      "rewards/rejected": -1.6991876363754272,
      "step": 690
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.40918212081453e-06,
      "logits/chosen": -4.308882713317871,
      "logits/rejected": -4.258685111999512,
      "logps/chosen": -410.50714111328125,
      "logps/rejected": -470.98712158203125,
      "loss": 0.573,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.4014512300491333,
      "rewards/margins": 0.4143608510494232,
      "rewards/rejected": -1.8158118724822998,
      "step": 700
    },
    {
      "epoch": 0.56,
      "eval_logits/chosen": -4.460686683654785,
      "eval_logits/rejected": -4.447921276092529,
      "eval_logps/chosen": -376.8887939453125,
      "eval_logps/rejected": -395.554931640625,
      "eval_loss": 0.680784285068512,
      "eval_rewards/accuracies": 0.5633333325386047,
      "eval_rewards/chosen": -1.561822772026062,
      "eval_rewards/margins": 0.1303686946630478,
      "eval_rewards/rejected": -1.6921913623809814,
      "eval_runtime": 1600.8511,
      "eval_samples_per_second": 1.87,
      "eval_steps_per_second": 0.234,
      "step": 700
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3393975235270654e-06,
      "logits/chosen": -4.24044132232666,
      "logits/rejected": -4.221581935882568,
      "logps/chosen": -451.12353515625,
      "logps/rejected": -492.0777893066406,
      "loss": 0.6601,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -1.5504473447799683,
      "rewards/margins": 0.23769822716712952,
      "rewards/rejected": -1.788145661354065,
      "step": 710
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2697383821388153e-06,
      "logits/chosen": -4.228194236755371,
      "logits/rejected": -4.242656230926514,
      "logps/chosen": -432.177001953125,
      "logps/rejected": -457.5282287597656,
      "loss": 0.6498,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -1.4517402648925781,
      "rewards/margins": 0.22140760719776154,
      "rewards/rejected": -1.6731477975845337,
      "step": 720
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2002591114409657e-06,
      "logits/chosen": -4.1490349769592285,
      "logits/rejected": -4.144831657409668,
      "logps/chosen": -428.9891662597656,
      "logps/rejected": -466.90228271484375,
      "loss": 0.6282,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.3725817203521729,
      "rewards/margins": 0.26138487458229065,
      "rewards/rejected": -1.6339666843414307,
      "step": 730
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.131013985717285e-06,
      "logits/chosen": -4.210469722747803,
      "logits/rejected": -4.1584882736206055,
      "logps/chosen": -442.4186096191406,
      "logps/rejected": -490.4705505371094,
      "loss": 0.6473,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.4414502382278442,
      "rewards/margins": 0.2493906468153,
      "rewards/rejected": -1.690840721130371,
      "step": 740
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.062057096347338e-06,
      "logits/chosen": -4.1790266036987305,
      "logits/rejected": -4.148745059967041,
      "logps/chosen": -418.1016540527344,
      "logps/rejected": -432.9185485839844,
      "loss": 0.6612,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.2657266855239868,
      "rewards/margins": 0.14516393840312958,
      "rewards/rejected": -1.4108905792236328,
      "step": 750
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9934423095525733e-06,
      "logits/chosen": -4.040548324584961,
      "logits/rejected": -4.0535197257995605,
      "logps/chosen": -413.2955017089844,
      "logps/rejected": -438.1127014160156,
      "loss": 0.6539,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.1577858924865723,
      "rewards/margins": 0.2499733865261078,
      "rewards/rejected": -1.4077593088150024,
      "step": 760
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9252232243182986e-06,
      "logits/chosen": -4.132304668426514,
      "logits/rejected": -4.086350440979004,
      "logps/chosen": -374.2073974609375,
      "logps/rejected": -421.70947265625,
      "loss": 0.6035,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.0375020503997803,
      "rewards/margins": 0.3012349009513855,
      "rewards/rejected": -1.3387370109558105,
      "step": 770
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8574531305244043e-06,
      "logits/chosen": -3.9333484172821045,
      "logits/rejected": -3.921320676803589,
      "logps/chosen": -413.23236083984375,
      "logps/rejected": -459.419677734375,
      "loss": 0.5945,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.1705034971237183,
      "rewards/margins": 0.31472960114479065,
      "rewards/rejected": -1.485233187675476,
      "step": 780
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.7901849673175559e-06,
      "logits/chosen": -4.004987716674805,
      "logits/rejected": -3.96270751953125,
      "logps/chosen": -415.433349609375,
      "logps/rejected": -450.65447998046875,
      "loss": 0.6413,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.1887439489364624,
      "rewards/margins": 0.20863094925880432,
      "rewards/rejected": -1.3973749876022339,
      "step": 790
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7234712817573555e-06,
      "logits/chosen": -3.995453357696533,
      "logits/rejected": -3.992877960205078,
      "logps/chosen": -452.4390563964844,
      "logps/rejected": -477.8495178222656,
      "loss": 0.6384,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.3979339599609375,
      "rewards/margins": 0.2388983964920044,
      "rewards/rejected": -1.6368324756622314,
      "step": 800
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -4.182389259338379,
      "eval_logits/rejected": -4.171318531036377,
      "eval_logps/chosen": -357.1873474121094,
      "eval_logps/rejected": -375.4593200683594,
      "eval_loss": 0.6752617955207825,
      "eval_rewards/accuracies": 0.5659999847412109,
      "eval_rewards/chosen": -1.3648083209991455,
      "eval_rewards/margins": 0.12642760574817657,
      "eval_rewards/rejected": -1.4912358522415161,
      "eval_runtime": 1604.6041,
      "eval_samples_per_second": 1.866,
      "eval_steps_per_second": 0.234,
      "step": 800
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6573641877687936e-06,
      "logits/chosen": -4.01367712020874,
      "logits/rejected": -3.9943137168884277,
      "logps/chosen": -420.02734375,
      "logps/rejected": -469.4754333496094,
      "loss": 0.612,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.3465534448623657,
      "rewards/margins": 0.27429622411727905,
      "rewards/rejected": -1.620849370956421,
      "step": 810
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.591915325433034e-06,
      "logits/chosen": -4.069581031799316,
      "logits/rejected": -4.076575756072998,
      "logps/chosen": -411.4861755371094,
      "logps/rejected": -448.5120544433594,
      "loss": 0.6138,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.289924144744873,
      "rewards/margins": 0.3182913661003113,
      "rewards/rejected": -1.608215570449829,
      "step": 820
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5271758206483664e-06,
      "logits/chosen": -4.08115291595459,
      "logits/rejected": -4.065010070800781,
      "logps/chosen": -438.2860412597656,
      "logps/rejected": -472.0508728027344,
      "loss": 0.6431,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -1.458389163017273,
      "rewards/margins": 0.23057003319263458,
      "rewards/rejected": -1.6889593601226807,
      "step": 830
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4631962451927966e-06,
      "logits/chosen": -3.989629030227661,
      "logits/rejected": -3.970881938934326,
      "logps/chosen": -431.90972900390625,
      "logps/rejected": -473.61505126953125,
      "loss": 0.6104,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.3902056217193604,
      "rewards/margins": 0.27469927072525024,
      "rewards/rejected": -1.6649048328399658,
      "step": 840
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.4000265772195032e-06,
      "logits/chosen": -4.167831897735596,
      "logits/rejected": -4.113857746124268,
      "logps/chosen": -430.6526794433594,
      "logps/rejected": -475.681884765625,
      "loss": 0.6188,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.3779888153076172,
      "rewards/margins": 0.2821890711784363,
      "rewards/rejected": -1.6601779460906982,
      "step": 850
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3377161622160077e-06,
      "logits/chosen": -4.1135406494140625,
      "logits/rejected": -4.103851795196533,
      "logps/chosen": -431.333740234375,
      "logps/rejected": -472.18505859375,
      "loss": 0.6059,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -1.4065897464752197,
      "rewards/margins": 0.28285306692123413,
      "rewards/rejected": -1.6894428730010986,
      "step": 860
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.276313674457553e-06,
      "logits/chosen": -4.243846416473389,
      "logits/rejected": -4.234537601470947,
      "logps/chosen": -417.1517639160156,
      "logps/rejected": -473.24517822265625,
      "loss": 0.574,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.4315975904464722,
      "rewards/margins": 0.36709946393966675,
      "rewards/rejected": -1.7986972332000732,
      "step": 870
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2158670789848095e-06,
      "logits/chosen": -4.32526159286499,
      "logits/rejected": -4.3212785720825195,
      "logps/chosen": -463.43548583984375,
      "logps/rejected": -504.515380859375,
      "loss": 0.609,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.7021582126617432,
      "rewards/margins": 0.33437836170196533,
      "rewards/rejected": -2.036536693572998,
      "step": 880
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1564235941356016e-06,
      "logits/chosen": -4.47339391708374,
      "logits/rejected": -4.411566257476807,
      "logps/chosen": -455.85986328125,
      "logps/rejected": -513.0088500976562,
      "loss": 0.6417,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -1.7953081130981445,
      "rewards/margins": 0.3589649796485901,
      "rewards/rejected": -2.154273271560669,
      "step": 890
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.0980296546599254e-06,
      "logits/chosen": -4.28812313079834,
      "logits/rejected": -4.28458309173584,
      "logps/chosen": -480.78045654296875,
      "logps/rejected": -516.3783569335938,
      "loss": 0.6728,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.8175837993621826,
      "rewards/margins": 0.40267056226730347,
      "rewards/rejected": -2.220254421234131,
      "step": 900
    },
    {
      "epoch": 0.72,
      "eval_logits/chosen": -4.539231300354004,
      "eval_logits/rejected": -4.5250043869018555,
      "eval_logps/chosen": -389.2689208984375,
      "eval_logps/rejected": -409.6128234863281,
      "eval_loss": 0.6791353821754456,
      "eval_rewards/accuracies": 0.5713333487510681,
      "eval_rewards/chosen": -1.6856240034103394,
      "eval_rewards/margins": 0.14714699983596802,
      "eval_rewards/rejected": -1.8327711820602417,
      "eval_runtime": 1601.9204,
      "eval_samples_per_second": 1.869,
      "eval_steps_per_second": 0.234,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.040730875447083e-06,
      "logits/chosen": -4.249950885772705,
      "logits/rejected": -4.256103515625,
      "logps/chosen": -454.0740661621094,
      "logps/rejected": -481.5208435058594,
      "loss": 0.6463,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.5884063243865967,
      "rewards/margins": 0.2100685089826584,
      "rewards/rejected": -1.798474669456482,
      "step": 910
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.845720158932414e-07,
      "logits/chosen": -4.217880725860596,
      "logits/rejected": -4.230615615844727,
      "logps/chosen": -421.2108459472656,
      "logps/rejected": -443.8993225097656,
      "loss": 0.6443,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.46346914768219,
      "rewards/margins": 0.2133568972349167,
      "rewards/rejected": -1.6768258810043335,
      "step": 920
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.295969449372796e-07,
      "logits/chosen": -4.250065803527832,
      "logits/rejected": -4.226481914520264,
      "logps/chosen": -418.61883544921875,
      "logps/rejected": -455.81170654296875,
      "loss": 0.6211,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.3508899211883545,
      "rewards/margins": 0.25358742475509644,
      "rewards/rejected": -1.6044775247573853,
      "step": 930
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.758486067922176e-07,
      "logits/chosen": -4.2414469718933105,
      "logits/rejected": -4.200159549713135,
      "logps/chosen": -413.689697265625,
      "logps/rejected": -464.592529296875,
      "loss": 0.5956,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.3547637462615967,
      "rewards/margins": 0.32075968384742737,
      "rewards/rejected": -1.6755234003067017,
      "step": 940
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.233689873990006e-07,
      "logits/chosen": -4.249865531921387,
      "logits/rejected": -4.214231491088867,
      "logps/chosen": -419.63397216796875,
      "logps/rejected": -491.7757873535156,
      "loss": 0.5892,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.4119402170181274,
      "rewards/margins": 0.5434682369232178,
      "rewards/rejected": -1.9554083347320557,
      "step": 950
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.721990816288555e-07,
      "logits/chosen": -4.227639198303223,
      "logits/rejected": -4.187970161437988,
      "logps/chosen": -395.7589416503906,
      "logps/rejected": -436.3963928222656,
      "loss": 0.612,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.3264906406402588,
      "rewards/margins": 0.2781200408935547,
      "rewards/rejected": -1.604610800743103,
      "step": 960
    },
    {
      "epoch": 0.78,
      "learning_rate": 7.223788612598148e-07,
      "logits/chosen": -4.25420618057251,
      "logits/rejected": -4.232513427734375,
      "logps/chosen": -437.45721435546875,
      "logps/rejected": -472.1407165527344,
      "loss": 0.6314,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.4481703042984009,
      "rewards/margins": 0.1885295808315277,
      "rewards/rejected": -1.636699914932251,
      "step": 970
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.73947243752448e-07,
      "logits/chosen": -4.329432964324951,
      "logits/rejected": -4.338817596435547,
      "logps/chosen": -462.8741760253906,
      "logps/rejected": -490.8958435058594,
      "loss": 0.6649,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.6740070581436157,
      "rewards/margins": 0.3544973134994507,
      "rewards/rejected": -2.0285041332244873,
      "step": 980
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.269420618491759e-07,
      "logits/chosen": -4.2786760330200195,
      "logits/rejected": -4.245335578918457,
      "logps/chosen": -423.904296875,
      "logps/rejected": -463.173095703125,
      "loss": 0.6426,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.5292352437973022,
      "rewards/margins": 0.23503565788269043,
      "rewards/rejected": -1.7642710208892822,
      "step": 990
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.814000340209267e-07,
      "logits/chosen": -4.25986385345459,
      "logits/rejected": -4.194713592529297,
      "logps/chosen": -439.45086669921875,
      "logps/rejected": -495.15771484375,
      "loss": 0.603,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.533198356628418,
      "rewards/margins": 0.34696659445762634,
      "rewards/rejected": -1.8801651000976562,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -4.503380298614502,
      "eval_logits/rejected": -4.4899396896362305,
      "eval_logps/chosen": -382.5414733886719,
      "eval_logps/rejected": -403.2298889160156,
      "eval_loss": 0.676650881767273,
      "eval_rewards/accuracies": 0.5770000219345093,
      "eval_rewards/chosen": -1.618349313735962,
      "eval_rewards/margins": 0.15059219300746918,
      "eval_rewards/rejected": -1.7689412832260132,
      "eval_runtime": 1634.1825,
      "eval_samples_per_second": 1.832,
      "eval_steps_per_second": 0.229,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.373567357842111e-07,
      "logits/chosen": -4.2339959144592285,
      "logits/rejected": -4.205721378326416,
      "logps/chosen": -433.2574768066406,
      "logps/rejected": -476.2079162597656,
      "loss": 0.5976,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.4623594284057617,
      "rewards/margins": 0.33363303542137146,
      "rewards/rejected": -1.7959926128387451,
      "step": 1010
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.948465719110226e-07,
      "logits/chosen": -4.32258415222168,
      "logits/rejected": -4.291365623474121,
      "logps/chosen": -420.07330322265625,
      "logps/rejected": -451.5908203125,
      "loss": 0.6421,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.45734441280365,
      "rewards/margins": 0.24639002978801727,
      "rewards/rejected": -1.7037346363067627,
      "step": 1020
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.539027495532766e-07,
      "logits/chosen": -4.28394889831543,
      "logits/rejected": -4.300284385681152,
      "logps/chosen": -417.7969665527344,
      "logps/rejected": -456.79180908203125,
      "loss": 0.6168,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.4292632341384888,
      "rewards/margins": 0.28898704051971436,
      "rewards/rejected": -1.718250036239624,
      "step": 1030
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.14557252302783e-07,
      "logits/chosen": -4.28948450088501,
      "logits/rejected": -4.251836776733398,
      "logps/chosen": -435.3079528808594,
      "logps/rejected": -474.75006103515625,
      "loss": 0.6739,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.5503196716308594,
      "rewards/margins": 0.23696406185626984,
      "rewards/rejected": -1.7872836589813232,
      "step": 1040
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7684081520700884e-07,
      "logits/chosen": -4.184053897857666,
      "logits/rejected": -4.1847357749938965,
      "logps/chosen": -452.0301818847656,
      "logps/rejected": -472.8980407714844,
      "loss": 0.6398,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.5176578760147095,
      "rewards/margins": 0.2665901780128479,
      "rewards/rejected": -1.7842479944229126,
      "step": 1050
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.407829007601507e-07,
      "logits/chosen": -4.219012260437012,
      "logits/rejected": -4.17362117767334,
      "logps/chosen": -430.40997314453125,
      "logps/rejected": -481.66363525390625,
      "loss": 0.6108,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.4400875568389893,
      "rewards/margins": 0.3422803282737732,
      "rewards/rejected": -1.7823677062988281,
      "step": 1060
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.064116758882724e-07,
      "logits/chosen": -4.185604572296143,
      "logits/rejected": -4.131557941436768,
      "logps/chosen": -446.9971618652344,
      "logps/rejected": -507.7557067871094,
      "loss": 0.607,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.5457991361618042,
      "rewards/margins": 0.41560330986976624,
      "rewards/rejected": -1.9614025354385376,
      "step": 1070
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.737539899464908e-07,
      "logits/chosen": -4.246437072753906,
      "logits/rejected": -4.235133647918701,
      "logps/chosen": -405.3433532714844,
      "logps/rejected": -456.494140625,
      "loss": 0.6183,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.4034773111343384,
      "rewards/margins": 0.37458181381225586,
      "rewards/rejected": -1.7780590057373047,
      "step": 1080
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4283535374538645e-07,
      "logits/chosen": -4.18662691116333,
      "logits/rejected": -4.184709072113037,
      "logps/chosen": -436.64849853515625,
      "logps/rejected": -474.5104064941406,
      "loss": 0.6166,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.4191557168960571,
      "rewards/margins": 0.30806541442871094,
      "rewards/rejected": -1.727220892906189,
      "step": 1090
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1367991962303298e-07,
      "logits/chosen": -4.172500133514404,
      "logits/rejected": -4.158700942993164,
      "logps/chosen": -419.20977783203125,
      "logps/rejected": -447.9820251464844,
      "loss": 0.6732,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.5145409107208252,
      "rewards/margins": 0.15635111927986145,
      "rewards/rejected": -1.6708920001983643,
      "step": 1100
    },
    {
      "epoch": 0.88,
      "eval_logits/chosen": -4.4235520362854,
      "eval_logits/rejected": -4.410409927368164,
      "eval_logps/chosen": -377.40155029296875,
      "eval_logps/rejected": -397.9567565917969,
      "eval_loss": 0.6752503514289856,
      "eval_rewards/accuracies": 0.5776666402816772,
      "eval_rewards/chosen": -1.5669503211975098,
      "eval_rewards/margins": 0.1492597907781601,
      "eval_rewards/rejected": -1.7162100076675415,
      "eval_runtime": 1708.8293,
      "eval_samples_per_second": 1.752,
      "eval_steps_per_second": 0.219,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.8631046257820278e-07,
      "logits/chosen": -4.177311897277832,
      "logits/rejected": -4.192288398742676,
      "logps/chosen": -433.6348571777344,
      "logps/rejected": -464.0807189941406,
      "loss": 0.6238,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.4329723119735718,
      "rewards/margins": 0.2695394456386566,
      "rewards/rejected": -1.7025117874145508,
      "step": 1110
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6074836247950143e-07,
      "logits/chosen": -4.152982234954834,
      "logits/rejected": -4.155800819396973,
      "logps/chosen": -444.71722412109375,
      "logps/rejected": -471.8829650878906,
      "loss": 0.6421,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.4441962242126465,
      "rewards/margins": 0.21269936859607697,
      "rewards/rejected": -1.656895399093628,
      "step": 1120
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.370135873643097e-07,
      "logits/chosen": -4.171530723571777,
      "logits/rejected": -4.211108684539795,
      "logps/chosen": -438.52178955078125,
      "logps/rejected": -458.4681701660156,
      "loss": 0.6246,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.4780899286270142,
      "rewards/margins": 0.28698402643203735,
      "rewards/rejected": -1.7650740146636963,
      "step": 1130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1512467784059372e-07,
      "logits/chosen": -4.241854667663574,
      "logits/rejected": -4.201731204986572,
      "logps/chosen": -400.08306884765625,
      "logps/rejected": -440.25250244140625,
      "loss": 0.6219,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.385571837425232,
      "rewards/margins": 0.30431246757507324,
      "rewards/rejected": -1.6898844242095947,
      "step": 1140
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.509873260376251e-08,
      "logits/chosen": -4.173992156982422,
      "logits/rejected": -4.1186909675598145,
      "logps/chosen": -429.03570556640625,
      "logps/rejected": -502.2874450683594,
      "loss": 0.5799,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.4350130558013916,
      "rewards/margins": 0.4506250321865082,
      "rewards/rejected": -1.8856379985809326,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.695139507988559e-08,
      "logits/chosen": -4.202746391296387,
      "logits/rejected": -4.225730895996094,
      "logps/chosen": -442.8606872558594,
      "logps/rejected": -478.89776611328125,
      "loss": 0.6358,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.4460103511810303,
      "rewards/margins": 0.257608562707901,
      "rewards/rejected": -1.7036190032958984,
      "step": 1160
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.069684120570684e-08,
      "logits/chosen": -4.227932929992676,
      "logits/rejected": -4.133112907409668,
      "logps/chosen": -433.41192626953125,
      "logps/rejected": -483.49383544921875,
      "loss": 0.6375,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.5354173183441162,
      "rewards/margins": 0.30013877153396606,
      "rewards/rejected": -1.8355562686920166,
      "step": 1170
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.634776835499871e-08,
      "logits/chosen": -4.156412124633789,
      "logits/rejected": -4.1273393630981445,
      "logps/chosen": -413.9664611816406,
      "logps/rejected": -457.55877685546875,
      "loss": 0.6273,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.4416838884353638,
      "rewards/margins": 0.2954239845275879,
      "rewards/rejected": -1.7371078729629517,
      "step": 1180
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.3915385419908964e-08,
      "logits/chosen": -4.120321750640869,
      "logits/rejected": -4.165139675140381,
      "logps/chosen": -432.268310546875,
      "logps/rejected": -465.0489807128906,
      "loss": 0.6372,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.4545971155166626,
      "rewards/margins": 0.2788635790348053,
      "rewards/rejected": -1.7334604263305664,
      "step": 1190
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.3409404055043938e-08,
      "logits/chosen": -4.251112937927246,
      "logits/rejected": -4.216982841491699,
      "logps/chosen": -437.51922607421875,
      "logps/rejected": -471.106201171875,
      "loss": 0.6431,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -1.5188627243041992,
      "rewards/margins": 0.23063072562217712,
      "rewards/rejected": -1.7494935989379883,
      "step": 1200
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -4.397921085357666,
      "eval_logits/rejected": -4.384759902954102,
      "eval_logps/chosen": -375.8670959472656,
      "eval_logps/rejected": -396.2835693359375,
      "eval_loss": 0.6749910712242126,
      "eval_rewards/accuracies": 0.5773333311080933,
      "eval_rewards/chosen": -1.5516059398651123,
      "eval_rewards/margins": 0.14787256717681885,
      "eval_rewards/rejected": -1.6994785070419312,
      "eval_runtime": 1629.1995,
      "eval_samples_per_second": 1.838,
      "eval_steps_per_second": 0.23,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4838031091134186e-08,
      "logits/chosen": -4.187882423400879,
      "logits/rejected": -4.119401454925537,
      "logps/chosen": -407.01824951171875,
      "logps/rejected": -472.28631591796875,
      "loss": 0.6076,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.4217784404754639,
      "rewards/margins": 0.3864634931087494,
      "rewards/rejected": -1.808241844177246,
      "step": 1210
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.207962124201774e-09,
      "logits/chosen": -4.208897590637207,
      "logits/rejected": -4.17116641998291,
      "logps/chosen": -435.45013427734375,
      "logps/rejected": -472.75421142578125,
      "loss": 0.6003,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.3684009313583374,
      "rewards/margins": 0.32645002007484436,
      "rewards/rejected": -1.6948511600494385,
      "step": 1220
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.5243762852441023e-09,
      "logits/chosen": -4.157495021820068,
      "logits/rejected": -4.110323429107666,
      "logps/chosen": -427.61651611328125,
      "logps/rejected": -471.6131286621094,
      "loss": 0.6286,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.4408422708511353,
      "rewards/margins": 0.2736855149269104,
      "rewards/rejected": -1.7145278453826904,
      "step": 1230
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.909321945129278e-10,
      "logits/chosen": -4.126371383666992,
      "logits/rejected": -4.085521221160889,
      "logps/chosen": -441.79095458984375,
      "logps/rejected": -493.57049560546875,
      "loss": 0.5948,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.3874802589416504,
      "rewards/margins": 0.3715583086013794,
      "rewards/rejected": -1.7590383291244507,
      "step": 1240
    },
    {
      "epoch": 1.0,
      "step": 1249,
      "total_flos": 0.0,
      "train_loss": 0.647387065536218,
      "train_runtime": 42673.6748,
      "train_samples_per_second": 0.469,
      "train_steps_per_second": 0.029
    }
  ],
  "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
}