{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 1455,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.36986301369863e-08,
      "logits/chosen": -2.8295512199401855,
      "logits/rejected": -2.9639337062835693,
      "logps/chosen": -242.64569091796875,
      "logps/rejected": -75.87144470214844,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.36986301369863e-07,
      "logits/chosen": -2.778250217437744,
      "logits/rejected": -2.813397169113159,
      "logps/chosen": -292.17218017578125,
      "logps/rejected": -78.88499450683594,
      "loss": 0.6942,
      "rewards/accuracies": 0.4097222089767456,
      "rewards/chosen": -0.001105638686567545,
      "rewards/margins": -0.0026314095593988895,
      "rewards/rejected": 0.0015257701743394136,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.73972602739726e-07,
      "logits/chosen": -2.76747465133667,
      "logits/rejected": -2.7585418224334717,
      "logps/chosen": -300.1101989746094,
      "logps/rejected": -81.14244079589844,
      "loss": 0.6932,
      "rewards/accuracies": 0.4437499940395355,
      "rewards/chosen": -0.005251293070614338,
      "rewards/margins": -0.003377618733793497,
      "rewards/rejected": -0.001873674220405519,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.10958904109589e-07,
      "logits/chosen": -2.776379108428955,
      "logits/rejected": -2.7856156826019287,
      "logps/chosen": -290.00897216796875,
      "logps/rejected": -78.96559143066406,
      "loss": 0.6885,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.005190015770494938,
      "rewards/margins": 0.011348642408847809,
      "rewards/rejected": -0.006158626172691584,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.47945205479452e-07,
      "logits/chosen": -2.7960293292999268,
      "logits/rejected": -2.814054250717163,
      "logps/chosen": -245.3889617919922,
      "logps/rejected": -84.38166046142578,
      "loss": 0.6823,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": 0.008710930123925209,
      "rewards/margins": 0.027255941182374954,
      "rewards/rejected": -0.018545005470514297,
      "step": 40
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.84931506849315e-07,
      "logits/chosen": -2.7724318504333496,
      "logits/rejected": -2.8249223232269287,
      "logps/chosen": -252.29598999023438,
      "logps/rejected": -80.92951965332031,
      "loss": 0.6751,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": 0.00529166916385293,
      "rewards/margins": 0.03346817195415497,
      "rewards/rejected": -0.028176506981253624,
      "step": 50
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.21917808219178e-07,
      "logits/chosen": -2.8436636924743652,
      "logits/rejected": -2.7858223915100098,
      "logps/chosen": -281.6982421875,
      "logps/rejected": -78.57672119140625,
      "loss": 0.6656,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": 0.0231462512165308,
      "rewards/margins": 0.06612209975719452,
      "rewards/rejected": -0.04297585040330887,
      "step": 60
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.58904109589041e-07,
      "logits/chosen": -2.8172953128814697,
      "logits/rejected": -2.813781261444092,
      "logps/chosen": -264.85235595703125,
      "logps/rejected": -74.86351013183594,
      "loss": 0.6541,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 0.026092741638422012,
      "rewards/margins": 0.0869915708899498,
      "rewards/rejected": -0.060898829251527786,
      "step": 70
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.095890410958904e-06,
      "logits/chosen": -2.8075528144836426,
      "logits/rejected": -2.8162968158721924,
      "logps/chosen": -255.69869995117188,
      "logps/rejected": -73.08162689208984,
      "loss": 0.6355,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 0.0321514829993248,
      "rewards/margins": 0.11635198444128036,
      "rewards/rejected": -0.08420050889253616,
      "step": 80
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.232876712328767e-06,
      "logits/chosen": -2.841773748397827,
      "logits/rejected": -2.825883388519287,
      "logps/chosen": -267.5022888183594,
      "logps/rejected": -85.21376037597656,
      "loss": 0.6119,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 0.030611341819167137,
      "rewards/margins": 0.16599974036216736,
      "rewards/rejected": -0.13538840413093567,
      "step": 90
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.36986301369863e-06,
      "logits/chosen": -2.852410316467285,
      "logits/rejected": -2.8519997596740723,
      "logps/chosen": -286.9059143066406,
      "logps/rejected": -82.15430450439453,
      "loss": 0.5808,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 0.057482797652482986,
      "rewards/margins": 0.26051220297813416,
      "rewards/rejected": -0.20302939414978027,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.5068493150684932e-06,
      "logits/chosen": -2.7674930095672607,
      "logits/rejected": -2.7689127922058105,
      "logps/chosen": -277.5893249511719,
      "logps/rejected": -84.25486755371094,
      "loss": 0.5528,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 0.05910978466272354,
      "rewards/margins": 0.3146067261695862,
      "rewards/rejected": -0.25549691915512085,
      "step": 110
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.643835616438356e-06,
      "logits/chosen": -2.7497572898864746,
      "logits/rejected": -2.808637857437134,
      "logps/chosen": -277.6904602050781,
      "logps/rejected": -82.65525817871094,
      "loss": 0.5245,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.08153042942285538,
      "rewards/margins": 0.4226166605949402,
      "rewards/rejected": -0.3410862386226654,
      "step": 120
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.780821917808219e-06,
      "logits/chosen": -2.765390634536743,
      "logits/rejected": -2.7765676975250244,
      "logps/chosen": -284.7118835449219,
      "logps/rejected": -90.37391662597656,
      "loss": 0.5057,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 0.1007312536239624,
      "rewards/margins": 0.49337419867515564,
      "rewards/rejected": -0.39264291524887085,
      "step": 130
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.917808219178082e-06,
      "logits/chosen": -2.8435187339782715,
      "logits/rejected": -2.848928213119507,
      "logps/chosen": -293.25323486328125,
      "logps/rejected": -88.73191833496094,
      "loss": 0.4691,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.09035232663154602,
      "rewards/margins": 0.6013891696929932,
      "rewards/rejected": -0.5110368132591248,
      "step": 140
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.9938884644766997e-06,
      "logits/chosen": -2.788867235183716,
      "logits/rejected": -2.797130584716797,
      "logps/chosen": -290.54638671875,
      "logps/rejected": -82.95954895019531,
      "loss": 0.4398,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": 0.09947594255208969,
      "rewards/margins": 0.6646188497543335,
      "rewards/rejected": -0.5651428699493408,
      "step": 150
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.978609625668449e-06,
      "logits/chosen": -2.809487819671631,
      "logits/rejected": -2.8850138187408447,
      "logps/chosen": -271.88629150390625,
      "logps/rejected": -83.17223358154297,
      "loss": 0.4204,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": 0.07071445882320404,
      "rewards/margins": 0.7117933034896851,
      "rewards/rejected": -0.6410789489746094,
      "step": 160
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.9633307868601984e-06,
      "logits/chosen": -2.7891199588775635,
      "logits/rejected": -2.789405584335327,
      "logps/chosen": -288.9054870605469,
      "logps/rejected": -86.50032043457031,
      "loss": 0.4053,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": 0.13311851024627686,
      "rewards/margins": 0.861064076423645,
      "rewards/rejected": -0.7279455065727234,
      "step": 170
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.948051948051948e-06,
      "logits/chosen": -2.760305643081665,
      "logits/rejected": -2.7756900787353516,
      "logps/chosen": -271.7988586425781,
      "logps/rejected": -88.26414489746094,
      "loss": 0.3764,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": 0.1391439437866211,
      "rewards/margins": 0.9942502975463867,
      "rewards/rejected": -0.8551063537597656,
      "step": 180
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.9327731092436974e-06,
      "logits/chosen": -2.7618346214294434,
      "logits/rejected": -2.7937939167022705,
      "logps/chosen": -274.2486877441406,
      "logps/rejected": -89.63475799560547,
      "loss": 0.3633,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.06488532572984695,
      "rewards/margins": 0.9964747428894043,
      "rewards/rejected": -0.9315894246101379,
      "step": 190
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.917494270435447e-06,
      "logits/chosen": -2.834862232208252,
      "logits/rejected": -2.8097951412200928,
      "logps/chosen": -267.5804748535156,
      "logps/rejected": -88.04959869384766,
      "loss": 0.3405,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 0.11280632019042969,
      "rewards/margins": 1.167088508605957,
      "rewards/rejected": -1.054282307624817,
      "step": 200
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.9022154316271963e-06,
      "logits/chosen": -2.789674997329712,
      "logits/rejected": -2.7489635944366455,
      "logps/chosen": -271.354736328125,
      "logps/rejected": -101.02803039550781,
      "loss": 0.3406,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": 0.08271731436252594,
      "rewards/margins": 1.1206178665161133,
      "rewards/rejected": -1.037900447845459,
      "step": 210
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.8869365928189456e-06,
      "logits/chosen": -2.8190276622772217,
      "logits/rejected": -2.7801876068115234,
      "logps/chosen": -252.3684844970703,
      "logps/rejected": -83.58897399902344,
      "loss": 0.3366,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": 0.047989681363105774,
      "rewards/margins": 1.1506556272506714,
      "rewards/rejected": -1.102665901184082,
      "step": 220
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.8716577540106951e-06,
      "logits/chosen": -2.7455031871795654,
      "logits/rejected": -2.7612051963806152,
      "logps/chosen": -296.0608825683594,
      "logps/rejected": -85.1841049194336,
      "loss": 0.3066,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.12809792160987854,
      "rewards/margins": 1.3899600505828857,
      "rewards/rejected": -1.2618623971939087,
      "step": 230
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.8563789152024445e-06,
      "logits/chosen": -2.775570869445801,
      "logits/rejected": -2.819958209991455,
      "logps/chosen": -282.4068908691406,
      "logps/rejected": -89.96455383300781,
      "loss": 0.3055,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.09846373647451401,
      "rewards/margins": 1.4890722036361694,
      "rewards/rejected": -1.390608549118042,
      "step": 240
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.841100076394194e-06,
      "logits/chosen": -2.799436092376709,
      "logits/rejected": -2.8345470428466797,
      "logps/chosen": -265.59735107421875,
      "logps/rejected": -96.5230941772461,
      "loss": 0.2986,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.011983467265963554,
      "rewards/margins": 1.4505140781402588,
      "rewards/rejected": -1.462497591972351,
      "step": 250
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.8258212375859433e-06,
      "logits/chosen": -2.767268419265747,
      "logits/rejected": -2.7291500568389893,
      "logps/chosen": -284.69512939453125,
      "logps/rejected": -84.67759704589844,
      "loss": 0.2975,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 0.13364484906196594,
      "rewards/margins": 1.5072095394134521,
      "rewards/rejected": -1.373564600944519,
      "step": 260
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.8105423987776928e-06,
      "logits/chosen": -2.740736961364746,
      "logits/rejected": -2.7466208934783936,
      "logps/chosen": -276.2334899902344,
      "logps/rejected": -102.246826171875,
      "loss": 0.271,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.018421288579702377,
      "rewards/margins": 1.600907325744629,
      "rewards/rejected": -1.5824859142303467,
      "step": 270
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.7952635599694424e-06,
      "logits/chosen": -2.763476610183716,
      "logits/rejected": -2.826669216156006,
      "logps/chosen": -256.71234130859375,
      "logps/rejected": -84.85465240478516,
      "loss": 0.2796,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.021867703646421432,
      "rewards/margins": 1.4527885913848877,
      "rewards/rejected": -1.474656343460083,
      "step": 280
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.7799847211611915e-06,
      "logits/chosen": -2.778783082962036,
      "logits/rejected": -2.7868287563323975,
      "logps/chosen": -278.37847900390625,
      "logps/rejected": -101.5394287109375,
      "loss": 0.2617,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.06040378659963608,
      "rewards/margins": 1.8995882272720337,
      "rewards/rejected": -1.8391841650009155,
      "step": 290
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.764705882352941e-06,
      "logits/chosen": -2.727238893508911,
      "logits/rejected": -2.723588466644287,
      "logps/chosen": -275.4689025878906,
      "logps/rejected": -93.48865509033203,
      "loss": 0.2612,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.02573985978960991,
      "rewards/margins": 1.606034517288208,
      "rewards/rejected": -1.5802944898605347,
      "step": 300
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7494270435446906e-06,
      "logits/chosen": -2.8035991191864014,
      "logits/rejected": -2.8192009925842285,
      "logps/chosen": -284.1933898925781,
      "logps/rejected": -100.13687896728516,
      "loss": 0.2521,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.07830613851547241,
      "rewards/margins": 1.9092552661895752,
      "rewards/rejected": -1.9875609874725342,
      "step": 310
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.7341482047364399e-06,
      "logits/chosen": -2.8154501914978027,
      "logits/rejected": -2.810351848602295,
      "logps/chosen": -274.22906494140625,
      "logps/rejected": -105.29779052734375,
      "loss": 0.2418,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.09999797493219376,
      "rewards/margins": 1.8004556894302368,
      "rewards/rejected": -1.9004535675048828,
      "step": 320
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.7188693659281894e-06,
      "logits/chosen": -2.7782671451568604,
      "logits/rejected": -2.7969181537628174,
      "logps/chosen": -253.5823516845703,
      "logps/rejected": -91.79356384277344,
      "loss": 0.238,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.16337540745735168,
      "rewards/margins": 1.8200305700302124,
      "rewards/rejected": -1.9834058284759521,
      "step": 330
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.703590527119939e-06,
      "logits/chosen": -2.7838528156280518,
      "logits/rejected": -2.7845559120178223,
      "logps/chosen": -276.6927795410156,
      "logps/rejected": -105.1791763305664,
      "loss": 0.2408,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.11859698593616486,
      "rewards/margins": 1.980875015258789,
      "rewards/rejected": -2.0994718074798584,
      "step": 340
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.688311688311688e-06,
      "logits/chosen": -2.8013651371002197,
      "logits/rejected": -2.7973175048828125,
      "logps/chosen": -280.02337646484375,
      "logps/rejected": -108.97808837890625,
      "loss": 0.2366,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.17565980553627014,
      "rewards/margins": 2.0567195415496826,
      "rewards/rejected": -2.23237943649292,
      "step": 350
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.6730328495034376e-06,
      "logits/chosen": -2.7511441707611084,
      "logits/rejected": -2.79335355758667,
      "logps/chosen": -295.89630126953125,
      "logps/rejected": -102.7479476928711,
      "loss": 0.2329,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.09856131672859192,
      "rewards/margins": 2.122011184692383,
      "rewards/rejected": -2.2205727100372314,
      "step": 360
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.6577540106951871e-06,
      "logits/chosen": -2.7538399696350098,
      "logits/rejected": -2.7933411598205566,
      "logps/chosen": -278.8785705566406,
      "logps/rejected": -95.23030090332031,
      "loss": 0.2309,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.1667167693376541,
      "rewards/margins": 2.049661159515381,
      "rewards/rejected": -2.2163777351379395,
      "step": 370
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.6424751718869364e-06,
      "logits/chosen": -2.7607922554016113,
      "logits/rejected": -2.763124942779541,
      "logps/chosen": -270.15484619140625,
      "logps/rejected": -103.03666687011719,
      "loss": 0.2252,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.10847791284322739,
      "rewards/margins": 2.2359797954559326,
      "rewards/rejected": -2.3444576263427734,
      "step": 380
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.627196333078686e-06,
      "logits/chosen": -2.7692110538482666,
      "logits/rejected": -2.7727513313293457,
      "logps/chosen": -284.1410217285156,
      "logps/rejected": -103.25160217285156,
      "loss": 0.2183,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.07571631669998169,
      "rewards/margins": 2.3671867847442627,
      "rewards/rejected": -2.4429030418395996,
      "step": 390
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.6119174942704355e-06,
      "logits/chosen": -2.713252544403076,
      "logits/rejected": -2.695067882537842,
      "logps/chosen": -246.3933563232422,
      "logps/rejected": -98.11837768554688,
      "loss": 0.2068,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.18476447463035583,
      "rewards/margins": 2.0507025718688965,
      "rewards/rejected": -2.2354671955108643,
      "step": 400
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5966386554621848e-06,
      "logits/chosen": -2.788705348968506,
      "logits/rejected": -2.7501461505889893,
      "logps/chosen": -268.0991516113281,
      "logps/rejected": -102.09769439697266,
      "loss": 0.1976,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.12583544850349426,
      "rewards/margins": 2.4059669971466064,
      "rewards/rejected": -2.5318026542663574,
      "step": 410
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5813598166539341e-06,
      "logits/chosen": -2.728642702102661,
      "logits/rejected": -2.7183501720428467,
      "logps/chosen": -277.0406188964844,
      "logps/rejected": -108.7503890991211,
      "loss": 0.2093,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.2682625353336334,
      "rewards/margins": 2.486672878265381,
      "rewards/rejected": -2.7549355030059814,
      "step": 420
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.5660809778456837e-06,
      "logits/chosen": -2.755733013153076,
      "logits/rejected": -2.794435977935791,
      "logps/chosen": -267.26983642578125,
      "logps/rejected": -103.68330383300781,
      "loss": 0.1918,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.2670840322971344,
      "rewards/margins": 2.529799461364746,
      "rewards/rejected": -2.7968833446502686,
      "step": 430
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.550802139037433e-06,
      "logits/chosen": -2.722480058670044,
      "logits/rejected": -2.7511894702911377,
      "logps/chosen": -271.193359375,
      "logps/rejected": -101.18013000488281,
      "loss": 0.1901,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.22587260603904724,
      "rewards/margins": 2.3586864471435547,
      "rewards/rejected": -2.584559202194214,
      "step": 440
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.5355233002291825e-06,
      "logits/chosen": -2.724370241165161,
      "logits/rejected": -2.754976511001587,
      "logps/chosen": -300.2513122558594,
      "logps/rejected": -107.4105224609375,
      "loss": 0.2001,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.2436836212873459,
      "rewards/margins": 2.5888543128967285,
      "rewards/rejected": -2.83253812789917,
      "step": 450
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.520244461420932e-06,
      "logits/chosen": -2.7703804969787598,
      "logits/rejected": -2.743251323699951,
      "logps/chosen": -304.1031799316406,
      "logps/rejected": -113.75921630859375,
      "loss": 0.2065,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.15126582980155945,
      "rewards/margins": 2.703249931335449,
      "rewards/rejected": -2.854515552520752,
      "step": 460
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.5049656226126814e-06,
      "logits/chosen": -2.7441253662109375,
      "logits/rejected": -2.7229297161102295,
      "logps/chosen": -240.0807342529297,
      "logps/rejected": -100.44654846191406,
      "loss": 0.2019,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.32086318731307983,
      "rewards/margins": 2.4921185970306396,
      "rewards/rejected": -2.8129820823669434,
      "step": 470
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.4896867838044307e-06,
      "logits/chosen": -2.704709529876709,
      "logits/rejected": -2.7237319946289062,
      "logps/chosen": -255.7570343017578,
      "logps/rejected": -113.80975341796875,
      "loss": 0.2024,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.40018701553344727,
      "rewards/margins": 2.4386959075927734,
      "rewards/rejected": -2.8388831615448,
      "step": 480
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -2.7680206298828125,
      "eval_logits/rejected": -2.7271738052368164,
      "eval_logps/chosen": -277.8106689453125,
      "eval_logps/rejected": -191.96604919433594,
      "eval_loss": 0.4197174608707428,
      "eval_rewards/accuracies": 0.80859375,
      "eval_rewards/chosen": -0.3973674178123474,
      "eval_rewards/margins": 1.4956284761428833,
      "eval_rewards/rejected": -1.892995834350586,
      "eval_runtime": 259.9183,
      "eval_samples_per_second": 7.695,
      "eval_steps_per_second": 0.062,
      "step": 485
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4744079449961802e-06,
      "logits/chosen": -2.7513904571533203,
      "logits/rejected": -2.8016788959503174,
      "logps/chosen": -279.7330017089844,
      "logps/rejected": -109.1887435913086,
      "loss": 0.1934,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.32768136262893677,
      "rewards/margins": 2.722477436065674,
      "rewards/rejected": -3.0501585006713867,
      "step": 490
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4591291061879296e-06,
      "logits/chosen": -2.705814838409424,
      "logits/rejected": -2.6782376766204834,
      "logps/chosen": -274.86480712890625,
      "logps/rejected": -113.91358947753906,
      "loss": 0.1797,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.2557370066642761,
      "rewards/margins": 2.544098377227783,
      "rewards/rejected": -2.799835205078125,
      "step": 500
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.443850267379679e-06,
      "logits/chosen": -2.7348856925964355,
      "logits/rejected": -2.7256247997283936,
      "logps/chosen": -260.6488342285156,
      "logps/rejected": -98.91087341308594,
      "loss": 0.1826,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -0.3437056541442871,
      "rewards/margins": 2.4448533058166504,
      "rewards/rejected": -2.7885591983795166,
      "step": 510
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.4285714285714286e-06,
      "logits/chosen": -2.799448013305664,
      "logits/rejected": -2.778778076171875,
      "logps/chosen": -279.9732360839844,
      "logps/rejected": -124.17852783203125,
      "loss": 0.1678,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.4281793236732483,
      "rewards/margins": 2.834944248199463,
      "rewards/rejected": -3.2631232738494873,
      "step": 520
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.413292589763178e-06,
      "logits/chosen": -2.7671663761138916,
      "logits/rejected": -2.7472527027130127,
      "logps/chosen": -306.42181396484375,
      "logps/rejected": -113.77569580078125,
      "loss": 0.1743,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.31418782472610474,
      "rewards/margins": 2.8712990283966064,
      "rewards/rejected": -3.1854867935180664,
      "step": 530
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.3980137509549275e-06,
      "logits/chosen": -2.7520015239715576,
      "logits/rejected": -2.772326946258545,
      "logps/chosen": -247.4855499267578,
      "logps/rejected": -101.20677185058594,
      "loss": 0.1779,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.4306362271308899,
      "rewards/margins": 2.6422040462493896,
      "rewards/rejected": -3.072840452194214,
      "step": 540
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3827349121466768e-06,
      "logits/chosen": -2.685715436935425,
      "logits/rejected": -2.705381155014038,
      "logps/chosen": -269.1335144042969,
      "logps/rejected": -111.51566314697266,
      "loss": 0.1742,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.4332028329372406,
      "rewards/margins": 2.756269693374634,
      "rewards/rejected": -3.1894726753234863,
      "step": 550
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.3674560733384261e-06,
      "logits/chosen": -2.737140417098999,
      "logits/rejected": -2.7757163047790527,
      "logps/chosen": -294.99530029296875,
      "logps/rejected": -112.22420501708984,
      "loss": 0.163,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.3779616951942444,
      "rewards/margins": 3.002772331237793,
      "rewards/rejected": -3.3807339668273926,
      "step": 560
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.3521772345301757e-06,
      "logits/chosen": -2.765671730041504,
      "logits/rejected": -2.776446580886841,
      "logps/chosen": -291.44732666015625,
      "logps/rejected": -112.7077865600586,
      "loss": 0.1604,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.1892886459827423,
      "rewards/margins": 3.2649528980255127,
      "rewards/rejected": -3.4542412757873535,
      "step": 570
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3368983957219252e-06,
      "logits/chosen": -2.711365222930908,
      "logits/rejected": -2.761141300201416,
      "logps/chosen": -246.60989379882812,
      "logps/rejected": -106.89306640625,
      "loss": 0.1662,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -0.5674314498901367,
      "rewards/margins": 2.7119510173797607,
      "rewards/rejected": -3.2793827056884766,
      "step": 580
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3216195569136745e-06,
      "logits/chosen": -2.706845283508301,
      "logits/rejected": -2.7259907722473145,
      "logps/chosen": -266.33453369140625,
      "logps/rejected": -109.5294189453125,
      "loss": 0.1638,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.49850577116012573,
      "rewards/margins": 2.929001569747925,
      "rewards/rejected": -3.4275078773498535,
      "step": 590
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.306340718105424e-06,
      "logits/chosen": -2.7694671154022217,
      "logits/rejected": -2.7433857917785645,
      "logps/chosen": -279.09429931640625,
      "logps/rejected": -113.45039367675781,
      "loss": 0.1663,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.5474687218666077,
      "rewards/margins": 2.8283703327178955,
      "rewards/rejected": -3.3758392333984375,
      "step": 600
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.2910618792971732e-06,
      "logits/chosen": -2.724958896636963,
      "logits/rejected": -2.6607446670532227,
      "logps/chosen": -268.52862548828125,
      "logps/rejected": -99.50247955322266,
      "loss": 0.1663,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.39182907342910767,
      "rewards/margins": 2.8105075359344482,
      "rewards/rejected": -3.2023367881774902,
      "step": 610
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.2757830404889227e-06,
      "logits/chosen": -2.7382755279541016,
      "logits/rejected": -2.747498035430908,
      "logps/chosen": -295.7679748535156,
      "logps/rejected": -118.82535552978516,
      "loss": 0.1486,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.37939247488975525,
      "rewards/margins": 3.1222548484802246,
      "rewards/rejected": -3.5016472339630127,
      "step": 620
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.2605042016806722e-06,
      "logits/chosen": -2.7766032218933105,
      "logits/rejected": -2.7410993576049805,
      "logps/chosen": -282.5389099121094,
      "logps/rejected": -115.2748794555664,
      "loss": 0.153,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.4390658736228943,
      "rewards/margins": 3.1960551738739014,
      "rewards/rejected": -3.6351211071014404,
      "step": 630
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.2452253628724215e-06,
      "logits/chosen": -2.7750675678253174,
      "logits/rejected": -2.781956195831299,
      "logps/chosen": -299.9912109375,
      "logps/rejected": -124.12471008300781,
      "loss": 0.1546,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.4881526827812195,
      "rewards/margins": 3.3200020790100098,
      "rewards/rejected": -3.808154582977295,
      "step": 640
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.229946524064171e-06,
      "logits/chosen": -2.7297987937927246,
      "logits/rejected": -2.668787956237793,
      "logps/chosen": -285.3520202636719,
      "logps/rejected": -114.99342346191406,
      "loss": 0.1551,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.43964657187461853,
      "rewards/margins": 3.080749988555908,
      "rewards/rejected": -3.5203967094421387,
      "step": 650
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2146676852559206e-06,
      "logits/chosen": -2.70336651802063,
      "logits/rejected": -2.7211241722106934,
      "logps/chosen": -288.2897644042969,
      "logps/rejected": -117.46827697753906,
      "loss": 0.1598,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.5533924102783203,
      "rewards/margins": 3.033695697784424,
      "rewards/rejected": -3.587088108062744,
      "step": 660
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.19938884644767e-06,
      "logits/chosen": -2.7294211387634277,
      "logits/rejected": -2.776686191558838,
      "logps/chosen": -279.3514709472656,
      "logps/rejected": -110.04444885253906,
      "loss": 0.1681,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.5076431035995483,
      "rewards/margins": 3.0213823318481445,
      "rewards/rejected": -3.5290253162384033,
      "step": 670
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.1841100076394192e-06,
      "logits/chosen": -2.749238967895508,
      "logits/rejected": -2.7573282718658447,
      "logps/chosen": -270.27874755859375,
      "logps/rejected": -111.33082580566406,
      "loss": 0.1629,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -0.4736942648887634,
      "rewards/margins": 3.030879259109497,
      "rewards/rejected": -3.5045738220214844,
      "step": 680
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.1688311688311688e-06,
      "logits/chosen": -2.7193169593811035,
      "logits/rejected": -2.7363669872283936,
      "logps/chosen": -299.65753173828125,
      "logps/rejected": -120.58744812011719,
      "loss": 0.1436,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.4750828146934509,
      "rewards/margins": 3.319221019744873,
      "rewards/rejected": -3.794304370880127,
      "step": 690
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.153552330022918e-06,
      "logits/chosen": -2.7201449871063232,
      "logits/rejected": -2.7599716186523438,
      "logps/chosen": -285.97869873046875,
      "logps/rejected": -111.79976654052734,
      "loss": 0.1637,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -0.46921300888061523,
      "rewards/margins": 3.1174368858337402,
      "rewards/rejected": -3.5866501331329346,
      "step": 700
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.1382734912146676e-06,
      "logits/chosen": -2.683954954147339,
      "logits/rejected": -2.760031223297119,
      "logps/chosen": -285.8088073730469,
      "logps/rejected": -114.0859375,
      "loss": 0.1569,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.4851107597351074,
      "rewards/margins": 3.097794771194458,
      "rewards/rejected": -3.5829052925109863,
      "step": 710
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.1229946524064172e-06,
      "logits/chosen": -2.695244550704956,
      "logits/rejected": -2.7127084732055664,
      "logps/chosen": -258.01812744140625,
      "logps/rejected": -116.52490234375,
      "loss": 0.148,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.48574456572532654,
      "rewards/margins": 3.0761725902557373,
      "rewards/rejected": -3.5619170665740967,
      "step": 720
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.1077158135981665e-06,
      "logits/chosen": -2.6948394775390625,
      "logits/rejected": -2.7225587368011475,
      "logps/chosen": -247.1420440673828,
      "logps/rejected": -108.1529312133789,
      "loss": 0.1546,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.6709108352661133,
      "rewards/margins": 2.9054205417633057,
      "rewards/rejected": -3.5763309001922607,
      "step": 730
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.0924369747899158e-06,
      "logits/chosen": -2.776315450668335,
      "logits/rejected": -2.776728868484497,
      "logps/chosen": -270.6875915527344,
      "logps/rejected": -114.87939453125,
      "loss": 0.1503,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.6296391487121582,
      "rewards/margins": 3.1651525497436523,
      "rewards/rejected": -3.7947916984558105,
      "step": 740
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.0771581359816653e-06,
      "logits/chosen": -2.733975887298584,
      "logits/rejected": -2.6755785942077637,
      "logps/chosen": -253.8512420654297,
      "logps/rejected": -123.96826171875,
      "loss": 0.1483,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.5190945267677307,
      "rewards/margins": 3.2409470081329346,
      "rewards/rejected": -3.7600414752960205,
      "step": 750
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.0618792971734147e-06,
      "logits/chosen": -2.6686155796051025,
      "logits/rejected": -2.7576305866241455,
      "logps/chosen": -315.3510437011719,
      "logps/rejected": -116.48152160644531,
      "loss": 0.1401,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.4066368043422699,
      "rewards/margins": 3.404694080352783,
      "rewards/rejected": -3.811330795288086,
      "step": 760
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.0466004583651642e-06,
      "logits/chosen": -2.7223763465881348,
      "logits/rejected": -2.7043228149414062,
      "logps/chosen": -285.30767822265625,
      "logps/rejected": -123.26036071777344,
      "loss": 0.1394,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.6819589734077454,
      "rewards/margins": 3.2477848529815674,
      "rewards/rejected": -3.929744005203247,
      "step": 770
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.0313216195569137e-06,
      "logits/chosen": -2.710700511932373,
      "logits/rejected": -2.748891592025757,
      "logps/chosen": -267.1021423339844,
      "logps/rejected": -108.77484130859375,
      "loss": 0.15,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.7118343710899353,
      "rewards/margins": 3.120725154876709,
      "rewards/rejected": -3.832559585571289,
      "step": 780
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.016042780748663e-06,
      "logits/chosen": -2.6949081420898438,
      "logits/rejected": -2.6823809146881104,
      "logps/chosen": -278.98101806640625,
      "logps/rejected": -121.89730072021484,
      "loss": 0.1449,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.5939213037490845,
      "rewards/margins": 3.3531200885772705,
      "rewards/rejected": -3.9470412731170654,
      "step": 790
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0007639419404126e-06,
      "logits/chosen": -2.710453748703003,
      "logits/rejected": -2.720715045928955,
      "logps/chosen": -277.281982421875,
      "logps/rejected": -114.69615173339844,
      "loss": 0.1473,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.6140307188034058,
      "rewards/margins": 3.254765748977661,
      "rewards/rejected": -3.8687965869903564,
      "step": 800
    },
    {
      "epoch": 1.67,
      "learning_rate": 9.85485103132162e-07,
      "logits/chosen": -2.7251040935516357,
      "logits/rejected": -2.693974494934082,
      "logps/chosen": -286.97686767578125,
      "logps/rejected": -119.7470703125,
      "loss": 0.1369,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.5266898274421692,
      "rewards/margins": 3.49212384223938,
      "rewards/rejected": -4.018813133239746,
      "step": 810
    },
    {
      "epoch": 1.69,
      "learning_rate": 9.702062643239114e-07,
      "logits/chosen": -2.7554683685302734,
      "logits/rejected": -2.762106418609619,
      "logps/chosen": -289.8918762207031,
      "logps/rejected": -125.37986755371094,
      "loss": 0.1401,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.6475861668586731,
      "rewards/margins": 3.4364724159240723,
      "rewards/rejected": -4.08405876159668,
      "step": 820
    },
    {
      "epoch": 1.71,
      "learning_rate": 9.549274255156608e-07,
      "logits/chosen": -2.7048392295837402,
      "logits/rejected": -2.6878910064697266,
      "logps/chosen": -284.31219482421875,
      "logps/rejected": -122.46697998046875,
      "loss": 0.1402,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.6133157014846802,
      "rewards/margins": 3.361485242843628,
      "rewards/rejected": -3.9748013019561768,
      "step": 830
    },
    {
      "epoch": 1.73,
      "learning_rate": 9.396485867074102e-07,
      "logits/chosen": -2.6750662326812744,
      "logits/rejected": -2.696866750717163,
      "logps/chosen": -291.42938232421875,
      "logps/rejected": -113.9036865234375,
      "loss": 0.1509,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -0.6713167428970337,
      "rewards/margins": 3.037714719772339,
      "rewards/rejected": -3.709031581878662,
      "step": 840
    },
    {
      "epoch": 1.75,
      "learning_rate": 9.243697478991597e-07,
      "logits/chosen": -2.68489933013916,
      "logits/rejected": -2.6937882900238037,
      "logps/chosen": -254.6746063232422,
      "logps/rejected": -115.2020263671875,
      "loss": 0.1543,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -0.8679746389389038,
      "rewards/margins": 3.083875894546509,
      "rewards/rejected": -3.951850414276123,
      "step": 850
    },
    {
      "epoch": 1.77,
      "learning_rate": 9.09090909090909e-07,
      "logits/chosen": -2.6926121711730957,
      "logits/rejected": -2.698666572570801,
      "logps/chosen": -280.2935485839844,
      "logps/rejected": -116.54022216796875,
      "loss": 0.1349,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.43773943185806274,
      "rewards/margins": 3.511465072631836,
      "rewards/rejected": -3.949204683303833,
      "step": 860
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.938120702826585e-07,
      "logits/chosen": -2.740382671356201,
      "logits/rejected": -2.68933367729187,
      "logps/chosen": -276.387451171875,
      "logps/rejected": -115.27424621582031,
      "loss": 0.136,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.5380962491035461,
      "rewards/margins": 3.400437593460083,
      "rewards/rejected": -3.9385337829589844,
      "step": 870
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.785332314744079e-07,
      "logits/chosen": -2.7333688735961914,
      "logits/rejected": -2.771758794784546,
      "logps/chosen": -293.49578857421875,
      "logps/rejected": -126.72651672363281,
      "loss": 0.1379,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.6212050914764404,
      "rewards/margins": 3.3602237701416016,
      "rewards/rejected": -3.9814281463623047,
      "step": 880
    },
    {
      "epoch": 1.84,
      "learning_rate": 8.632543926661573e-07,
      "logits/chosen": -2.709484577178955,
      "logits/rejected": -2.7399744987487793,
      "logps/chosen": -309.5272521972656,
      "logps/rejected": -123.85481262207031,
      "loss": 0.1368,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -0.5241509675979614,
      "rewards/margins": 3.5228843688964844,
      "rewards/rejected": -4.0470356941223145,
      "step": 890
    },
    {
      "epoch": 1.86,
      "learning_rate": 8.479755538579067e-07,
      "logits/chosen": -2.7088303565979004,
      "logits/rejected": -2.7025811672210693,
      "logps/chosen": -271.97784423828125,
      "logps/rejected": -118.300537109375,
      "loss": 0.1319,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.5483931303024292,
      "rewards/margins": 3.45857310295105,
      "rewards/rejected": -4.006965637207031,
      "step": 900
    },
    {
      "epoch": 1.88,
      "learning_rate": 8.326967150496562e-07,
      "logits/chosen": -2.747811794281006,
      "logits/rejected": -2.6834397315979004,
      "logps/chosen": -301.4414367675781,
      "logps/rejected": -123.6478271484375,
      "loss": 0.1264,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.5513515472412109,
      "rewards/margins": 3.684753894805908,
      "rewards/rejected": -4.236104965209961,
      "step": 910
    },
    {
      "epoch": 1.9,
      "learning_rate": 8.174178762414056e-07,
      "logits/chosen": -2.7275288105010986,
      "logits/rejected": -2.7216854095458984,
      "logps/chosen": -300.93328857421875,
      "logps/rejected": -133.4979248046875,
      "loss": 0.1301,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.5985573530197144,
      "rewards/margins": 3.8093056678771973,
      "rewards/rejected": -4.407863140106201,
      "step": 920
    },
    {
      "epoch": 1.92,
      "learning_rate": 8.02139037433155e-07,
      "logits/chosen": -2.7658352851867676,
      "logits/rejected": -2.738119602203369,
      "logps/chosen": -272.04449462890625,
      "logps/rejected": -120.6009521484375,
      "loss": 0.1252,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.6955040693283081,
      "rewards/margins": 3.5386176109313965,
      "rewards/rejected": -4.234121799468994,
      "step": 930
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.868601986249045e-07,
      "logits/chosen": -2.7340145111083984,
      "logits/rejected": -2.7205958366394043,
      "logps/chosen": -286.5509948730469,
      "logps/rejected": -126.727294921875,
      "loss": 0.1468,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.8658866882324219,
      "rewards/margins": 3.4253039360046387,
      "rewards/rejected": -4.2911906242370605,
      "step": 940
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.71581359816654e-07,
      "logits/chosen": -2.7004799842834473,
      "logits/rejected": -2.6438944339752197,
      "logps/chosen": -226.2438507080078,
      "logps/rejected": -113.9751205444336,
      "loss": 0.1347,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.913210391998291,
      "rewards/margins": 3.264578342437744,
      "rewards/rejected": -4.177788734436035,
      "step": 950
    },
    {
      "epoch": 1.98,
      "learning_rate": 7.563025210084033e-07,
      "logits/chosen": -2.7258245944976807,
      "logits/rejected": -2.792160987854004,
      "logps/chosen": -270.41314697265625,
      "logps/rejected": -115.57466125488281,
      "loss": 0.1153,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.7112643122673035,
      "rewards/margins": 3.4535746574401855,
      "rewards/rejected": -4.164839267730713,
      "step": 960
    },
    {
      "epoch": 2.0,
      "learning_rate": 7.410236822001527e-07,
      "logits/chosen": -2.682445526123047,
      "logits/rejected": -2.680621862411499,
      "logps/chosen": -300.4261474609375,
      "logps/rejected": -123.0173110961914,
      "loss": 0.1305,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.544654130935669,
      "rewards/margins": 3.4888763427734375,
      "rewards/rejected": -4.033530235290527,
      "step": 970
    },
    {
      "epoch": 2.0,
      "eval_logits/chosen": -2.730585813522339,
      "eval_logits/rejected": -2.6866092681884766,
      "eval_logps/chosen": -281.4207763671875,
      "eval_logps/rejected": -201.63304138183594,
      "eval_loss": 0.3693665862083435,
      "eval_rewards/accuracies": 0.82421875,
      "eval_rewards/chosen": -0.7583777904510498,
      "eval_rewards/margins": 2.1013174057006836,
      "eval_rewards/rejected": -2.8596951961517334,
      "eval_runtime": 259.2152,
      "eval_samples_per_second": 7.716,
      "eval_steps_per_second": 0.062,
      "step": 970
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.257448433919023e-07,
      "logits/chosen": -2.7307159900665283,
      "logits/rejected": -2.760425090789795,
      "logps/chosen": -263.69964599609375,
      "logps/rejected": -121.00809478759766,
      "loss": 0.1237,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.7474689483642578,
      "rewards/margins": 3.526792526245117,
      "rewards/rejected": -4.274261474609375,
      "step": 980
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.104660045836516e-07,
      "logits/chosen": -2.688023090362549,
      "logits/rejected": -2.6981234550476074,
      "logps/chosen": -283.63311767578125,
      "logps/rejected": -119.1172866821289,
      "loss": 0.1167,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.5560388565063477,
      "rewards/margins": 3.7879276275634766,
      "rewards/rejected": -4.343966484069824,
      "step": 990
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.95187165775401e-07,
      "logits/chosen": -2.6954102516174316,
      "logits/rejected": -2.696408748626709,
      "logps/chosen": -277.02764892578125,
      "logps/rejected": -115.6692886352539,
      "loss": 0.1243,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.6953684091567993,
      "rewards/margins": 3.4351089000701904,
      "rewards/rejected": -4.130476951599121,
      "step": 1000
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.799083269671506e-07,
      "logits/chosen": -2.6869771480560303,
      "logits/rejected": -2.7037670612335205,
      "logps/chosen": -251.7620086669922,
      "logps/rejected": -117.48094177246094,
      "loss": 0.1317,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.8689507246017456,
      "rewards/margins": 3.475069046020508,
      "rewards/rejected": -4.344019889831543,
      "step": 1010
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.646294881588999e-07,
      "logits/chosen": -2.6988840103149414,
      "logits/rejected": -2.6869451999664307,
      "logps/chosen": -287.7177734375,
      "logps/rejected": -123.6144790649414,
      "loss": 0.119,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.7602699995040894,
      "rewards/margins": 3.628121852874756,
      "rewards/rejected": -4.388391971588135,
      "step": 1020
    },
    {
      "epoch": 2.12,
      "learning_rate": 6.493506493506493e-07,
      "logits/chosen": -2.7356371879577637,
      "logits/rejected": -2.8164889812469482,
      "logps/chosen": -305.6545715332031,
      "logps/rejected": -119.76715087890625,
      "loss": 0.1255,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.7907172441482544,
      "rewards/margins": 3.6126227378845215,
      "rewards/rejected": -4.403339862823486,
      "step": 1030
    },
    {
      "epoch": 2.14,
      "learning_rate": 6.340718105423987e-07,
      "logits/chosen": -2.647362232208252,
      "logits/rejected": -2.6018130779266357,
      "logps/chosen": -281.76092529296875,
      "logps/rejected": -118.23252868652344,
      "loss": 0.121,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.64565110206604,
      "rewards/margins": 3.727663516998291,
      "rewards/rejected": -4.37331485748291,
      "step": 1040
    },
    {
      "epoch": 2.16,
      "learning_rate": 6.187929717341482e-07,
      "logits/chosen": -2.7285995483398438,
      "logits/rejected": -2.67795467376709,
      "logps/chosen": -273.88555908203125,
      "logps/rejected": -112.9361801147461,
      "loss": 0.1223,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.5947860479354858,
      "rewards/margins": 3.604292392730713,
      "rewards/rejected": -4.199078559875488,
      "step": 1050
    },
    {
      "epoch": 2.19,
      "learning_rate": 6.035141329258976e-07,
      "logits/chosen": -2.718034505844116,
      "logits/rejected": -2.6936326026916504,
      "logps/chosen": -287.3531799316406,
      "logps/rejected": -123.18647766113281,
      "loss": 0.1196,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.7472411394119263,
      "rewards/margins": 3.734687328338623,
      "rewards/rejected": -4.481928825378418,
      "step": 1060
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.88235294117647e-07,
      "logits/chosen": -2.699462413787842,
      "logits/rejected": -2.667182683944702,
      "logps/chosen": -255.7397003173828,
      "logps/rejected": -114.04801177978516,
      "loss": 0.1287,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -0.706480860710144,
      "rewards/margins": 3.5075461864471436,
      "rewards/rejected": -4.214027404785156,
      "step": 1070
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.729564553093965e-07,
      "logits/chosen": -2.719064712524414,
      "logits/rejected": -2.724663734436035,
      "logps/chosen": -283.41607666015625,
      "logps/rejected": -120.31791687011719,
      "loss": 0.1205,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.8808507919311523,
      "rewards/margins": 3.6230416297912598,
      "rewards/rejected": -4.503891944885254,
      "step": 1080
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.576776165011459e-07,
      "logits/chosen": -2.6613950729370117,
      "logits/rejected": -2.669309377670288,
      "logps/chosen": -274.6718444824219,
      "logps/rejected": -123.7230453491211,
      "loss": 0.1295,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.8109513521194458,
      "rewards/margins": 3.535714626312256,
      "rewards/rejected": -4.346665859222412,
      "step": 1090
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.423987776928953e-07,
      "logits/chosen": -2.752354860305786,
      "logits/rejected": -2.689492702484131,
      "logps/chosen": -297.46575927734375,
      "logps/rejected": -123.75872802734375,
      "loss": 0.1112,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.8303642272949219,
      "rewards/margins": 3.775070905685425,
      "rewards/rejected": -4.605435371398926,
      "step": 1100
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.271199388846448e-07,
      "logits/chosen": -2.7163376808166504,
      "logits/rejected": -2.7011306285858154,
      "logps/chosen": -268.9468078613281,
      "logps/rejected": -129.1416473388672,
      "loss": 0.1147,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.7546231150627136,
      "rewards/margins": 3.6865787506103516,
      "rewards/rejected": -4.441201686859131,
      "step": 1110
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.118411000763941e-07,
      "logits/chosen": -2.7277450561523438,
      "logits/rejected": -2.6973800659179688,
      "logps/chosen": -282.5606689453125,
      "logps/rejected": -124.7701416015625,
      "loss": 0.1192,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.5854411125183105,
      "rewards/margins": 3.8541629314422607,
      "rewards/rejected": -4.439603805541992,
      "step": 1120
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.965622612681436e-07,
      "logits/chosen": -2.6750576496124268,
      "logits/rejected": -2.7182953357696533,
      "logps/chosen": -274.384765625,
      "logps/rejected": -122.15213775634766,
      "loss": 0.1222,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.7071852684020996,
      "rewards/margins": 3.846698045730591,
      "rewards/rejected": -4.5538835525512695,
      "step": 1130
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.81283422459893e-07,
      "logits/chosen": -2.676004648208618,
      "logits/rejected": -2.6993675231933594,
      "logps/chosen": -263.89752197265625,
      "logps/rejected": -116.71295166015625,
      "loss": 0.1107,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -0.7977498769760132,
      "rewards/margins": 3.523756742477417,
      "rewards/rejected": -4.321506500244141,
      "step": 1140
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.660045836516425e-07,
      "logits/chosen": -2.7055160999298096,
      "logits/rejected": -2.7121243476867676,
      "logps/chosen": -289.3218994140625,
      "logps/rejected": -121.4267807006836,
      "loss": 0.1189,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.6630896329879761,
      "rewards/margins": 3.811305522918701,
      "rewards/rejected": -4.474394798278809,
      "step": 1150
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.5072574484339185e-07,
      "logits/chosen": -2.6670591831207275,
      "logits/rejected": -2.690974712371826,
      "logps/chosen": -278.7057189941406,
      "logps/rejected": -117.63764953613281,
      "loss": 0.1214,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.798798143863678,
      "rewards/margins": 3.643995761871338,
      "rewards/rejected": -4.442793846130371,
      "step": 1160
    },
    {
      "epoch": 2.41,
      "learning_rate": 4.3544690603514133e-07,
      "logits/chosen": -2.686856269836426,
      "logits/rejected": -2.6309545040130615,
      "logps/chosen": -271.1042785644531,
      "logps/rejected": -131.63259887695312,
      "loss": 0.1233,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.7242040038108826,
      "rewards/margins": 3.6964995861053467,
      "rewards/rejected": -4.420703411102295,
      "step": 1170
    },
    {
      "epoch": 2.43,
      "learning_rate": 4.2016806722689076e-07,
      "logits/chosen": -2.677506923675537,
      "logits/rejected": -2.7353405952453613,
      "logps/chosen": -304.0357360839844,
      "logps/rejected": -128.03981018066406,
      "loss": 0.1109,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.6244128942489624,
      "rewards/margins": 3.888216733932495,
      "rewards/rejected": -4.512629985809326,
      "step": 1180
    },
    {
      "epoch": 2.45,
      "learning_rate": 4.0488922841864013e-07,
      "logits/chosen": -2.738661527633667,
      "logits/rejected": -2.696394443511963,
      "logps/chosen": -263.1926574707031,
      "logps/rejected": -117.48931884765625,
      "loss": 0.1266,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.903983473777771,
      "rewards/margins": 3.5497829914093018,
      "rewards/rejected": -4.453766345977783,
      "step": 1190
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.896103896103896e-07,
      "logits/chosen": -2.721292018890381,
      "logits/rejected": -2.6752657890319824,
      "logps/chosen": -270.2406005859375,
      "logps/rejected": -121.4746322631836,
      "loss": 0.1118,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.7563194036483765,
      "rewards/margins": 3.8605034351348877,
      "rewards/rejected": -4.616823673248291,
      "step": 1200
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.7433155080213904e-07,
      "logits/chosen": -2.6903328895568848,
      "logits/rejected": -2.7279880046844482,
      "logps/chosen": -263.3211975097656,
      "logps/rejected": -120.23514556884766,
      "loss": 0.1132,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.7410348057746887,
      "rewards/margins": 3.5494637489318848,
      "rewards/rejected": -4.290497779846191,
      "step": 1210
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.590527119938884e-07,
      "logits/chosen": -2.697645425796509,
      "logits/rejected": -2.676959753036499,
      "logps/chosen": -275.2843322753906,
      "logps/rejected": -126.24308776855469,
      "loss": 0.1157,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -0.9119027256965637,
      "rewards/margins": 3.6408069133758545,
      "rewards/rejected": -4.552709102630615,
      "step": 1220
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.437738731856379e-07,
      "logits/chosen": -2.704871654510498,
      "logits/rejected": -2.6901638507843018,
      "logps/chosen": -292.1809387207031,
      "logps/rejected": -129.85531616210938,
      "loss": 0.1134,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.8088245391845703,
      "rewards/margins": 3.841012477874756,
      "rewards/rejected": -4.649837493896484,
      "step": 1230
    },
    {
      "epoch": 2.56,
      "learning_rate": 3.2849503437738727e-07,
      "logits/chosen": -2.6897072792053223,
      "logits/rejected": -2.6975724697113037,
      "logps/chosen": -289.3966369628906,
      "logps/rejected": -130.11814880371094,
      "loss": 0.1215,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.8563445210456848,
      "rewards/margins": 3.8016419410705566,
      "rewards/rejected": -4.657986640930176,
      "step": 1240
    },
    {
      "epoch": 2.58,
      "learning_rate": 3.1321619556913675e-07,
      "logits/chosen": -2.6511547565460205,
      "logits/rejected": -2.6918509006500244,
      "logps/chosen": -304.6783447265625,
      "logps/rejected": -128.91067504882812,
      "loss": 0.1256,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -0.6992138028144836,
      "rewards/margins": 3.894549608230591,
      "rewards/rejected": -4.593764305114746,
      "step": 1250
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.9793735676088617e-07,
      "logits/chosen": -2.7180018424987793,
      "logits/rejected": -2.73219633102417,
      "logps/chosen": -287.21697998046875,
      "logps/rejected": -128.64976501464844,
      "loss": 0.1276,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.9189088940620422,
      "rewards/margins": 3.821709394454956,
      "rewards/rejected": -4.7406182289123535,
      "step": 1260
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.8265851795263555e-07,
      "logits/chosen": -2.6880316734313965,
      "logits/rejected": -2.6963746547698975,
      "logps/chosen": -270.5711364746094,
      "logps/rejected": -109.63471984863281,
      "loss": 0.1227,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.8319600820541382,
      "rewards/margins": 3.4531593322753906,
      "rewards/rejected": -4.285120010375977,
      "step": 1270
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.6737967914438503e-07,
      "logits/chosen": -2.718834400177002,
      "logits/rejected": -2.730088472366333,
      "logps/chosen": -285.2646179199219,
      "logps/rejected": -129.10108947753906,
      "loss": 0.1177,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.7013461589813232,
      "rewards/margins": 4.131698131561279,
      "rewards/rejected": -4.833044052124023,
      "step": 1280
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.5210084033613445e-07,
      "logits/chosen": -2.6872308254241943,
      "logits/rejected": -2.71856689453125,
      "logps/chosen": -264.56591796875,
      "logps/rejected": -117.89599609375,
      "loss": 0.115,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.6815407872200012,
      "rewards/margins": 3.6640231609344482,
      "rewards/rejected": -4.3455634117126465,
      "step": 1290
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.3682200152788388e-07,
      "logits/chosen": -2.723666191101074,
      "logits/rejected": -2.7219574451446533,
      "logps/chosen": -317.8732604980469,
      "logps/rejected": -120.55049896240234,
      "loss": 0.1072,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.7161759734153748,
      "rewards/margins": 3.9293036460876465,
      "rewards/rejected": -4.645480155944824,
      "step": 1300
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.215431627196333e-07,
      "logits/chosen": -2.6789979934692383,
      "logits/rejected": -2.6807262897491455,
      "logps/chosen": -260.33062744140625,
      "logps/rejected": -120.8504409790039,
      "loss": 0.1201,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -0.854164719581604,
      "rewards/margins": 3.6763319969177246,
      "rewards/rejected": -4.530496597290039,
      "step": 1310
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.0626432391138274e-07,
      "logits/chosen": -2.703617572784424,
      "logits/rejected": -2.696929931640625,
      "logps/chosen": -250.5372772216797,
      "logps/rejected": -126.94450378417969,
      "loss": 0.1262,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -1.0846295356750488,
      "rewards/margins": 3.540682554244995,
      "rewards/rejected": -4.625311374664307,
      "step": 1320
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.9098548510313214e-07,
      "logits/chosen": -2.7046430110931396,
      "logits/rejected": -2.691488027572632,
      "logps/chosen": -267.9579162597656,
      "logps/rejected": -123.9790267944336,
      "loss": 0.1141,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.6386908292770386,
      "rewards/margins": 4.031415939331055,
      "rewards/rejected": -4.670106887817383,
      "step": 1330
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.757066462948816e-07,
      "logits/chosen": -2.689009189605713,
      "logits/rejected": -2.7033090591430664,
      "logps/chosen": -266.8208312988281,
      "logps/rejected": -126.87858581542969,
      "loss": 0.1157,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.6870437860488892,
      "rewards/margins": 3.979443073272705,
      "rewards/rejected": -4.666487216949463,
      "step": 1340
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.6042780748663102e-07,
      "logits/chosen": -2.70473051071167,
      "logits/rejected": -2.6794614791870117,
      "logps/chosen": -278.357421875,
      "logps/rejected": -125.93829345703125,
      "loss": 0.1152,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -0.9934309124946594,
      "rewards/margins": 3.67195200920105,
      "rewards/rejected": -4.665382385253906,
      "step": 1350
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.4514896867838044e-07,
      "logits/chosen": -2.6879210472106934,
      "logits/rejected": -2.7055411338806152,
      "logps/chosen": -301.1689147949219,
      "logps/rejected": -137.4368896484375,
      "loss": 0.1134,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.8130217790603638,
      "rewards/margins": 3.979782819747925,
      "rewards/rejected": -4.792804718017578,
      "step": 1360
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2987012987012984e-07,
      "logits/chosen": -2.6889195442199707,
      "logits/rejected": -2.677527666091919,
      "logps/chosen": -300.85272216796875,
      "logps/rejected": -131.1885986328125,
      "loss": 0.1181,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.8304030299186707,
      "rewards/margins": 4.027594089508057,
      "rewards/rejected": -4.857996940612793,
      "step": 1370
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.1459129106187928e-07,
      "logits/chosen": -2.675020694732666,
      "logits/rejected": -2.6782567501068115,
      "logps/chosen": -329.69696044921875,
      "logps/rejected": -135.40345764160156,
      "loss": 0.1182,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.631290078163147,
      "rewards/margins": 3.959303379058838,
      "rewards/rejected": -4.5905938148498535,
      "step": 1380
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.931245225362872e-08,
      "logits/chosen": -2.6736457347869873,
      "logits/rejected": -2.6810030937194824,
      "logps/chosen": -274.8864440917969,
      "logps/rejected": -124.03629302978516,
      "loss": 0.1258,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.7999323606491089,
      "rewards/margins": 3.9392266273498535,
      "rewards/rejected": -4.73915958404541,
      "step": 1390
    },
    {
      "epoch": 2.89,
      "learning_rate": 8.403361344537815e-08,
      "logits/chosen": -2.699571132659912,
      "logits/rejected": -2.701141834259033,
      "logps/chosen": -283.51751708984375,
      "logps/rejected": -130.84814453125,
      "loss": 0.1212,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.7443857192993164,
      "rewards/margins": 3.96293568611145,
      "rewards/rejected": -4.707321643829346,
      "step": 1400
    },
    {
      "epoch": 2.91,
      "learning_rate": 6.875477463712758e-08,
      "logits/chosen": -2.690229892730713,
      "logits/rejected": -2.7361502647399902,
      "logps/chosen": -310.0445556640625,
      "logps/rejected": -141.15748596191406,
      "loss": 0.1115,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.7573784589767456,
      "rewards/margins": 4.050500869750977,
      "rewards/rejected": -4.807879447937012,
      "step": 1410
    },
    {
      "epoch": 2.93,
      "learning_rate": 5.3475935828877e-08,
      "logits/chosen": -2.7038116455078125,
      "logits/rejected": -2.715642213821411,
      "logps/chosen": -294.73486328125,
      "logps/rejected": -127.1126480102539,
      "loss": 0.1176,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.720859944820404,
      "rewards/margins": 3.844661235809326,
      "rewards/rejected": -4.565520763397217,
      "step": 1420
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.8197097020626426e-08,
      "logits/chosen": -2.695003032684326,
      "logits/rejected": -2.7012951374053955,
      "logps/chosen": -257.13287353515625,
      "logps/rejected": -120.28269958496094,
      "loss": 0.1259,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -0.8394849896430969,
      "rewards/margins": 3.808011531829834,
      "rewards/rejected": -4.647497177124023,
      "step": 1430
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.291825821237586e-08,
      "logits/chosen": -2.7045774459838867,
      "logits/rejected": -2.679112672805786,
      "logps/chosen": -280.9797058105469,
      "logps/rejected": -127.39398193359375,
      "loss": 0.1096,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.8950273394584656,
      "rewards/margins": 3.8749961853027344,
      "rewards/rejected": -4.770023345947266,
      "step": 1440
    },
    {
      "epoch": 2.99,
      "learning_rate": 7.639419404125286e-09,
      "logits/chosen": -2.7298693656921387,
      "logits/rejected": -2.673051357269287,
      "logps/chosen": -329.364990234375,
      "logps/rejected": -131.12486267089844,
      "loss": 0.109,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -0.5651420950889587,
      "rewards/margins": 4.195387840270996,
      "rewards/rejected": -4.760529518127441,
      "step": 1450
    },
    {
      "epoch": 3.0,
      "eval_logits/chosen": -2.715555429458618,
      "eval_logits/rejected": -2.6699323654174805,
      "eval_logps/chosen": -282.458740234375,
      "eval_logps/rejected": -204.27066040039062,
      "eval_loss": 0.3553008437156677,
      "eval_rewards/accuracies": 0.828125,
      "eval_rewards/chosen": -0.8621728420257568,
      "eval_rewards/margins": 2.261284112930298,
      "eval_rewards/rejected": -3.123457193374634,
      "eval_runtime": 259.2547,
      "eval_samples_per_second": 7.714,
      "eval_steps_per_second": 0.062,
      "step": 1455
    },
    {
      "epoch": 3.0,
      "step": 1455,
      "total_flos": 0.0,
      "train_loss": 0.21351368668972423,
      "train_runtime": 46913.4477,
      "train_samples_per_second": 3.969,
      "train_steps_per_second": 0.031
    }
  ],
  "logging_steps": 10,
  "max_steps": 1455,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}