{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 936,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -2.2888386249542236,
      "logits/rejected": -2.347537040710449,
      "logps/chosen": -240.19537353515625,
      "logps/rejected": -224.4659423828125,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -2.2613658905029297,
      "logits/rejected": -2.2750418186187744,
      "logps/chosen": -243.42933654785156,
      "logps/rejected": -246.80438232421875,
      "loss": 0.6983,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/chosen": -0.004907915368676186,
      "rewards/margins": -0.0037050736136734486,
      "rewards/rejected": -0.0012028426863253117,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -2.207406997680664,
      "logits/rejected": -2.1862308979034424,
      "logps/chosen": -252.49658203125,
      "logps/rejected": -264.63861083984375,
      "loss": 0.6883,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.01405362505465746,
      "rewards/margins": 0.04791956767439842,
      "rewards/rejected": -0.06197319179773331,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -2.1290104389190674,
      "logits/rejected": -2.124718189239502,
      "logps/chosen": -265.79962158203125,
      "logps/rejected": -293.8382263183594,
      "loss": 0.6817,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.23290427029132843,
      "rewards/margins": 0.05236431211233139,
      "rewards/rejected": -0.2852686047554016,
      "step": 30
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -2.1210882663726807,
      "logits/rejected": -2.1071760654449463,
      "logps/chosen": -239.9654083251953,
      "logps/rejected": -246.32162475585938,
      "loss": 0.6723,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.28943145275115967,
      "rewards/margins": 0.12118977308273315,
      "rewards/rejected": -0.4106212556362152,
      "step": 40
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -2.151090145111084,
      "logits/rejected": -2.1575913429260254,
      "logps/chosen": -243.2239990234375,
      "logps/rejected": -283.3126220703125,
      "loss": 0.6911,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.30816999077796936,
      "rewards/margins": 0.1214941143989563,
      "rewards/rejected": -0.42966407537460327,
      "step": 50
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -2.2355575561523438,
      "logits/rejected": -2.241624355316162,
      "logps/chosen": -225.8336181640625,
      "logps/rejected": -239.73583984375,
      "loss": 0.641,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.4416503310203552,
      "rewards/margins": 0.24247512221336365,
      "rewards/rejected": -0.684125542640686,
      "step": 60
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -2.284141778945923,
      "logits/rejected": -2.271517515182495,
      "logps/chosen": -247.878173828125,
      "logps/rejected": -271.75543212890625,
      "loss": 0.6634,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.4251589775085449,
      "rewards/margins": 0.20249144732952118,
      "rewards/rejected": -0.6276503801345825,
      "step": 70
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -2.1530141830444336,
      "logits/rejected": -2.1263160705566406,
      "logps/chosen": -250.94882202148438,
      "logps/rejected": -283.70452880859375,
      "loss": 0.7221,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.5067945122718811,
      "rewards/margins": 0.3924552798271179,
      "rewards/rejected": -0.8992497324943542,
      "step": 80
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -2.2971713542938232,
      "logits/rejected": -2.2914767265319824,
      "logps/chosen": -239.848388671875,
      "logps/rejected": -261.0388488769531,
      "loss": 0.6921,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.36695146560668945,
      "rewards/margins": 0.7086008787155151,
      "rewards/rejected": -1.0755524635314941,
      "step": 90
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.96437054631829e-07,
      "logits/chosen": -2.2555811405181885,
      "logits/rejected": -2.25207781791687,
      "logps/chosen": -267.49615478515625,
      "logps/rejected": -275.76678466796875,
      "loss": 0.6493,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.6560766100883484,
      "rewards/margins": 0.7053524851799011,
      "rewards/rejected": -1.361429214477539,
      "step": 100
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -2.28631591796875,
      "eval_logits/rejected": -2.316732883453369,
      "eval_logps/chosen": -247.2499237060547,
      "eval_logps/rejected": -264.13714599609375,
      "eval_loss": 0.7535801529884338,
      "eval_rewards/accuracies": 0.5277777910232544,
      "eval_rewards/chosen": -1.031434416770935,
      "eval_rewards/margins": 0.1443558931350708,
      "eval_rewards/rejected": -1.1757903099060059,
      "eval_runtime": 136.5011,
      "eval_samples_per_second": 7.326,
      "eval_steps_per_second": 0.462,
      "step": 100
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.904988123515439e-07,
      "logits/chosen": -2.3735289573669434,
      "logits/rejected": -2.3574609756469727,
      "logps/chosen": -258.567138671875,
      "logps/rejected": -265.116943359375,
      "loss": 0.7214,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.645125687122345,
      "rewards/margins": 0.558975338935852,
      "rewards/rejected": -1.2041009664535522,
      "step": 110
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.845605700712589e-07,
      "logits/chosen": -2.2169833183288574,
      "logits/rejected": -2.2219901084899902,
      "logps/chosen": -236.5016632080078,
      "logps/rejected": -263.465576171875,
      "loss": 0.7608,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.8542020916938782,
      "rewards/margins": 0.4452931880950928,
      "rewards/rejected": -1.2994953393936157,
      "step": 120
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.786223277909738e-07,
      "logits/chosen": -2.2680552005767822,
      "logits/rejected": -2.2671051025390625,
      "logps/chosen": -265.2757568359375,
      "logps/rejected": -281.46087646484375,
      "loss": 0.6982,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.42591819167137146,
      "rewards/margins": 1.4877907037734985,
      "rewards/rejected": -1.9137089252471924,
      "step": 130
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.7268408551068883e-07,
      "logits/chosen": -2.3235602378845215,
      "logits/rejected": -2.331632614135742,
      "logps/chosen": -277.58642578125,
      "logps/rejected": -309.63006591796875,
      "loss": 0.6538,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.0361998081207275,
      "rewards/margins": 1.2754342555999756,
      "rewards/rejected": -2.311634063720703,
      "step": 140
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.667458432304038e-07,
      "logits/chosen": -2.242255449295044,
      "logits/rejected": -2.247087001800537,
      "logps/chosen": -279.643310546875,
      "logps/rejected": -306.5362243652344,
      "loss": 0.7123,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -1.3043805360794067,
      "rewards/margins": 1.866286039352417,
      "rewards/rejected": -3.1706669330596924,
      "step": 150
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.6080760095011875e-07,
      "logits/chosen": -2.4138083457946777,
      "logits/rejected": -2.403261661529541,
      "logps/chosen": -267.54046630859375,
      "logps/rejected": -311.9237060546875,
      "loss": 0.7048,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.7491146326065063,
      "rewards/margins": 1.4739110469818115,
      "rewards/rejected": -2.2230257987976074,
      "step": 160
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.548693586698337e-07,
      "logits/chosen": -2.394935131072998,
      "logits/rejected": -2.387144088745117,
      "logps/chosen": -238.01992797851562,
      "logps/rejected": -249.9500274658203,
      "loss": 0.6621,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.7305070161819458,
      "rewards/margins": 1.7976157665252686,
      "rewards/rejected": -2.528122663497925,
      "step": 170
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.4893111638954866e-07,
      "logits/chosen": -2.358888626098633,
      "logits/rejected": -2.3674557209014893,
      "logps/chosen": -257.3042907714844,
      "logps/rejected": -276.6517028808594,
      "loss": 0.7177,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.2033367156982422,
      "rewards/margins": 1.2791545391082764,
      "rewards/rejected": -2.4824917316436768,
      "step": 180
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.429928741092636e-07,
      "logits/chosen": -2.336920738220215,
      "logits/rejected": -2.31685733795166,
      "logps/chosen": -249.4643096923828,
      "logps/rejected": -282.7246398925781,
      "loss": 0.7693,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.9033983945846558,
      "rewards/margins": 1.6113331317901611,
      "rewards/rejected": -2.5147316455841064,
      "step": 190
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.3705463182897863e-07,
      "logits/chosen": -2.2998836040496826,
      "logits/rejected": -2.2824947834014893,
      "logps/chosen": -255.98239135742188,
      "logps/rejected": -282.6729431152344,
      "loss": 0.6555,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.7247324585914612,
      "rewards/margins": 2.3264706134796143,
      "rewards/rejected": -3.0512032508850098,
      "step": 200
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.227078676223755,
      "eval_logits/rejected": -2.2540831565856934,
      "eval_logps/chosen": -251.6654815673828,
      "eval_logps/rejected": -268.0155029296875,
      "eval_loss": 0.8345099091529846,
      "eval_rewards/accuracies": 0.5595238208770752,
      "eval_rewards/chosen": -1.472989559173584,
      "eval_rewards/margins": 0.09063953906297684,
      "eval_rewards/rejected": -1.563629150390625,
      "eval_runtime": 135.0454,
      "eval_samples_per_second": 7.405,
      "eval_steps_per_second": 0.467,
      "step": 200
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.311163895486936e-07,
      "logits/chosen": -2.1886990070343018,
      "logits/rejected": -2.2061002254486084,
      "logps/chosen": -272.6381530761719,
      "logps/rejected": -307.3228759765625,
      "loss": 0.6607,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.7392950057983398,
      "rewards/margins": 2.356786012649536,
      "rewards/rejected": -3.096081018447876,
      "step": 210
    },
    {
      "epoch": 0.71,
      "learning_rate": 4.251781472684085e-07,
      "logits/chosen": -2.3205573558807373,
      "logits/rejected": -2.3125534057617188,
      "logps/chosen": -208.845458984375,
      "logps/rejected": -231.4398651123047,
      "loss": 0.6198,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.921154797077179,
      "rewards/margins": 1.9245474338531494,
      "rewards/rejected": -2.8457021713256836,
      "step": 220
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.192399049881235e-07,
      "logits/chosen": -2.2658634185791016,
      "logits/rejected": -2.2645907402038574,
      "logps/chosen": -258.8167419433594,
      "logps/rejected": -303.98541259765625,
      "loss": 0.6653,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.0780036449432373,
      "rewards/margins": 2.2675023078918457,
      "rewards/rejected": -3.345506191253662,
      "step": 230
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.1330166270783846e-07,
      "logits/chosen": -2.259265184402466,
      "logits/rejected": -2.259178638458252,
      "logps/chosen": -265.008544921875,
      "logps/rejected": -283.46844482421875,
      "loss": 0.6808,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.058840036392212,
      "rewards/margins": 1.759861946105957,
      "rewards/rejected": -2.818701982498169,
      "step": 240
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.0736342042755347e-07,
      "logits/chosen": -2.264968156814575,
      "logits/rejected": -2.2467730045318604,
      "logps/chosen": -245.5597686767578,
      "logps/rejected": -282.61566162109375,
      "loss": 0.6574,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.6839832067489624,
      "rewards/margins": 2.1135334968566895,
      "rewards/rejected": -2.7975165843963623,
      "step": 250
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.0142517814726837e-07,
      "logits/chosen": -2.2550175189971924,
      "logits/rejected": -2.2587451934814453,
      "logps/chosen": -264.4140319824219,
      "logps/rejected": -313.1862487792969,
      "loss": 0.6417,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.2853078842163086,
      "rewards/margins": 1.7770347595214844,
      "rewards/rejected": -3.062342882156372,
      "step": 260
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.9548693586698333e-07,
      "logits/chosen": -2.30830454826355,
      "logits/rejected": -2.3013198375701904,
      "logps/chosen": -253.91281127929688,
      "logps/rejected": -266.73382568359375,
      "loss": 0.6325,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.6817399859428406,
      "rewards/margins": 2.7105841636657715,
      "rewards/rejected": -3.3923239707946777,
      "step": 270
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.8954869358669834e-07,
      "logits/chosen": -2.331390142440796,
      "logits/rejected": -2.336571216583252,
      "logps/chosen": -247.58544921875,
      "logps/rejected": -284.62884521484375,
      "loss": 0.6308,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.9239405393600464,
      "rewards/margins": 2.58821439743042,
      "rewards/rejected": -3.512155532836914,
      "step": 280
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.836104513064133e-07,
      "logits/chosen": -2.3030402660369873,
      "logits/rejected": -2.290931224822998,
      "logps/chosen": -264.10296630859375,
      "logps/rejected": -310.730224609375,
      "loss": 0.6456,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.4110323190689087,
      "rewards/margins": 1.8175039291381836,
      "rewards/rejected": -3.2285361289978027,
      "step": 290
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.7767220902612825e-07,
      "logits/chosen": -2.3290443420410156,
      "logits/rejected": -2.3188068866729736,
      "logps/chosen": -251.7384490966797,
      "logps/rejected": -287.4365234375,
      "loss": 0.6566,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.2217800617218018,
      "rewards/margins": 2.0904905796051025,
      "rewards/rejected": -3.312270402908325,
      "step": 300
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -2.352015733718872,
      "eval_logits/rejected": -2.3778603076934814,
      "eval_logps/chosen": -256.6260070800781,
      "eval_logps/rejected": -273.6545715332031,
      "eval_loss": 0.8076202869415283,
      "eval_rewards/accuracies": 0.5555555820465088,
      "eval_rewards/chosen": -1.9690420627593994,
      "eval_rewards/margins": 0.15849098563194275,
      "eval_rewards/rejected": -2.127532958984375,
      "eval_runtime": 135.1507,
      "eval_samples_per_second": 7.399,
      "eval_steps_per_second": 0.466,
      "step": 300
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.717339667458432e-07,
      "logits/chosen": -2.228579044342041,
      "logits/rejected": -2.2486958503723145,
      "logps/chosen": -269.064208984375,
      "logps/rejected": -314.79510498046875,
      "loss": 0.5843,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.7734891772270203,
      "rewards/margins": 3.1552462577819824,
      "rewards/rejected": -3.9287352561950684,
      "step": 310
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.6579572446555817e-07,
      "logits/chosen": -2.2717132568359375,
      "logits/rejected": -2.2769129276275635,
      "logps/chosen": -257.6423645019531,
      "logps/rejected": -314.797119140625,
      "loss": 0.3177,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -0.4064851403236389,
      "rewards/margins": 4.01033878326416,
      "rewards/rejected": -4.416823863983154,
      "step": 320
    },
    {
      "epoch": 1.06,
      "learning_rate": 3.598574821852731e-07,
      "logits/chosen": -2.247117280960083,
      "logits/rejected": -2.250030040740967,
      "logps/chosen": -235.4261016845703,
      "logps/rejected": -309.361572265625,
      "loss": 0.0693,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": 0.2905876338481903,
      "rewards/margins": 6.00634765625,
      "rewards/rejected": -5.715760231018066,
      "step": 330
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.5391923990498813e-07,
      "logits/chosen": -2.182969570159912,
      "logits/rejected": -2.16931414604187,
      "logps/chosen": -255.9242706298828,
      "logps/rejected": -364.0120544433594,
      "loss": 0.0276,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 0.9283174276351929,
      "rewards/margins": 8.556694030761719,
      "rewards/rejected": -7.628376007080078,
      "step": 340
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.479809976247031e-07,
      "logits/chosen": -2.167675495147705,
      "logits/rejected": -2.157602071762085,
      "logps/chosen": -235.40853881835938,
      "logps/rejected": -346.43328857421875,
      "loss": 0.0182,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6523984670639038,
      "rewards/margins": 9.692710876464844,
      "rewards/rejected": -8.040311813354492,
      "step": 350
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.42042755344418e-07,
      "logits/chosen": -2.1937975883483887,
      "logits/rejected": -2.2003121376037598,
      "logps/chosen": -211.8225555419922,
      "logps/rejected": -345.91094970703125,
      "loss": 0.0124,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 2.343766927719116,
      "rewards/margins": 12.03510570526123,
      "rewards/rejected": -9.691339492797852,
      "step": 360
    },
    {
      "epoch": 1.19,
      "learning_rate": 3.36104513064133e-07,
      "logits/chosen": -2.2343811988830566,
      "logits/rejected": -2.242821216583252,
      "logps/chosen": -194.0260772705078,
      "logps/rejected": -351.4122009277344,
      "loss": 0.0116,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.336714267730713,
      "rewards/margins": 12.730189323425293,
      "rewards/rejected": -10.393475532531738,
      "step": 370
    },
    {
      "epoch": 1.22,
      "learning_rate": 3.3016627078384796e-07,
      "logits/chosen": -2.2088000774383545,
      "logits/rejected": -2.2172210216522217,
      "logps/chosen": -214.9263153076172,
      "logps/rejected": -374.9606018066406,
      "loss": 0.0099,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.942856788635254,
      "rewards/margins": 14.011993408203125,
      "rewards/rejected": -11.069137573242188,
      "step": 380
    },
    {
      "epoch": 1.25,
      "learning_rate": 3.2422802850356297e-07,
      "logits/chosen": -2.129809856414795,
      "logits/rejected": -2.1051840782165527,
      "logps/chosen": -223.37490844726562,
      "logps/rejected": -409.17919921875,
      "loss": 0.0104,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.214216709136963,
      "rewards/margins": 16.285297393798828,
      "rewards/rejected": -13.071080207824707,
      "step": 390
    },
    {
      "epoch": 1.28,
      "learning_rate": 3.182897862232779e-07,
      "logits/chosen": -2.1976237297058105,
      "logits/rejected": -2.194974422454834,
      "logps/chosen": -197.03897094726562,
      "logps/rejected": -380.13031005859375,
      "loss": 0.0057,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.860673427581787,
      "rewards/margins": 16.22535514831543,
      "rewards/rejected": -13.364683151245117,
      "step": 400
    },
    {
      "epoch": 1.28,
      "eval_logits/chosen": -2.170393705368042,
      "eval_logits/rejected": -2.197235345840454,
      "eval_logps/chosen": -270.4623718261719,
      "eval_logps/rejected": -289.9275207519531,
      "eval_loss": 0.8791230320930481,
      "eval_rewards/accuracies": 0.5634920597076416,
      "eval_rewards/chosen": -3.3526790142059326,
      "eval_rewards/margins": 0.4021467864513397,
      "eval_rewards/rejected": -3.7548255920410156,
      "eval_runtime": 135.1943,
      "eval_samples_per_second": 7.397,
      "eval_steps_per_second": 0.466,
      "step": 400
    },
    {
      "epoch": 1.31,
      "learning_rate": 3.1235154394299283e-07,
      "logits/chosen": -2.1489205360412598,
      "logits/rejected": -2.1540191173553467,
      "logps/chosen": -219.0214385986328,
      "logps/rejected": -397.30364990234375,
      "loss": 0.0043,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.2062506675720215,
      "rewards/margins": 16.805278778076172,
      "rewards/rejected": -13.599026679992676,
      "step": 410
    },
    {
      "epoch": 1.35,
      "learning_rate": 3.0641330166270784e-07,
      "logits/chosen": -2.115307092666626,
      "logits/rejected": -2.0995349884033203,
      "logps/chosen": -234.65438842773438,
      "logps/rejected": -407.8638000488281,
      "loss": 0.0098,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 2.6958987712860107,
      "rewards/margins": 17.416133880615234,
      "rewards/rejected": -14.720235824584961,
      "step": 420
    },
    {
      "epoch": 1.38,
      "learning_rate": 3.004750593824228e-07,
      "logits/chosen": -2.134051561355591,
      "logits/rejected": -2.148725986480713,
      "logps/chosen": -195.26100158691406,
      "logps/rejected": -378.8183898925781,
      "loss": 0.0072,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.5557029247283936,
      "rewards/margins": 17.337064743041992,
      "rewards/rejected": -14.781362533569336,
      "step": 430
    },
    {
      "epoch": 1.41,
      "learning_rate": 2.9453681710213776e-07,
      "logits/chosen": -2.1074020862579346,
      "logits/rejected": -2.1130149364471436,
      "logps/chosen": -247.6088104248047,
      "logps/rejected": -419.13232421875,
      "loss": 0.0058,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.3471179008483887,
      "rewards/margins": 16.78860855102539,
      "rewards/rejected": -14.441492080688477,
      "step": 440
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.885985748218527e-07,
      "logits/chosen": -2.1506831645965576,
      "logits/rejected": -2.1653859615325928,
      "logps/chosen": -245.6567840576172,
      "logps/rejected": -420.876220703125,
      "loss": 0.0079,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.4259517192840576,
      "rewards/margins": 16.992084503173828,
      "rewards/rejected": -14.566131591796875,
      "step": 450
    },
    {
      "epoch": 1.47,
      "learning_rate": 2.8266033254156767e-07,
      "logits/chosen": -2.109947443008423,
      "logits/rejected": -2.1249232292175293,
      "logps/chosen": -241.10916137695312,
      "logps/rejected": -439.56781005859375,
      "loss": 0.0051,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.2505855560302734,
      "rewards/margins": 16.7042179107666,
      "rewards/rejected": -14.453630447387695,
      "step": 460
    },
    {
      "epoch": 1.51,
      "learning_rate": 2.7672209026128263e-07,
      "logits/chosen": -2.1982929706573486,
      "logits/rejected": -2.1973724365234375,
      "logps/chosen": -238.69497680664062,
      "logps/rejected": -403.54791259765625,
      "loss": 0.0144,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 1.3100935220718384,
      "rewards/margins": 14.033093452453613,
      "rewards/rejected": -12.723000526428223,
      "step": 470
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.7078384798099764e-07,
      "logits/chosen": -2.1538589000701904,
      "logits/rejected": -2.1524319648742676,
      "logps/chosen": -219.76461791992188,
      "logps/rejected": -362.3617248535156,
      "loss": 0.0089,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.142836093902588,
      "rewards/margins": 14.598767280578613,
      "rewards/rejected": -12.45592975616455,
      "step": 480
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.648456057007126e-07,
      "logits/chosen": -2.270354747772217,
      "logits/rejected": -2.285727024078369,
      "logps/chosen": -213.4247283935547,
      "logps/rejected": -350.7535400390625,
      "loss": 0.009,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 2.24265456199646,
      "rewards/margins": 13.511209487915039,
      "rewards/rejected": -11.268553733825684,
      "step": 490
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.589073634204275e-07,
      "logits/chosen": -2.189033031463623,
      "logits/rejected": -2.1983439922332764,
      "logps/chosen": -226.820556640625,
      "logps/rejected": -391.2894592285156,
      "loss": 0.0063,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.644188642501831,
      "rewards/margins": 15.23480224609375,
      "rewards/rejected": -12.59061336517334,
      "step": 500
    },
    {
      "epoch": 1.6,
      "eval_logits/chosen": -2.2120492458343506,
      "eval_logits/rejected": -2.239100217819214,
      "eval_logps/chosen": -266.6351318359375,
      "eval_logps/rejected": -285.4739074707031,
      "eval_loss": 0.8692338466644287,
      "eval_rewards/accuracies": 0.567460298538208,
      "eval_rewards/chosen": -2.969956398010254,
      "eval_rewards/margins": 0.3395082652568817,
      "eval_rewards/rejected": -3.309464693069458,
      "eval_runtime": 135.0534,
      "eval_samples_per_second": 7.404,
      "eval_steps_per_second": 0.466,
      "step": 500
    },
    {
      "epoch": 1.63,
      "learning_rate": 2.529691211401425e-07,
      "logits/chosen": -2.2285079956054688,
      "logits/rejected": -2.2011256217956543,
      "logps/chosen": -215.3155059814453,
      "logps/rejected": -366.4723205566406,
      "loss": 0.0093,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.619319438934326,
      "rewards/margins": 14.285768508911133,
      "rewards/rejected": -11.666448593139648,
      "step": 510
    },
    {
      "epoch": 1.67,
      "learning_rate": 2.4703087885985747e-07,
      "logits/chosen": -2.180443525314331,
      "logits/rejected": -2.196728467941284,
      "logps/chosen": -226.239013671875,
      "logps/rejected": -395.51397705078125,
      "loss": 0.0083,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.7000768184661865,
      "rewards/margins": 15.548223495483398,
      "rewards/rejected": -12.84814739227295,
      "step": 520
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.410926365795724e-07,
      "logits/chosen": -2.1908416748046875,
      "logits/rejected": -2.1960384845733643,
      "logps/chosen": -209.208984375,
      "logps/rejected": -333.3621520996094,
      "loss": 0.0081,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.2140251398086548,
      "rewards/margins": 11.866113662719727,
      "rewards/rejected": -10.652088165283203,
      "step": 530
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.351543942992874e-07,
      "logits/chosen": -2.1653943061828613,
      "logits/rejected": -2.171555995941162,
      "logps/chosen": -209.2273712158203,
      "logps/rejected": -370.8827819824219,
      "loss": 0.0056,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8547580242156982,
      "rewards/margins": 14.104257583618164,
      "rewards/rejected": -12.24949836730957,
      "step": 540
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.2921615201900234e-07,
      "logits/chosen": -2.179865837097168,
      "logits/rejected": -2.187969446182251,
      "logps/chosen": -248.9722137451172,
      "logps/rejected": -397.516357421875,
      "loss": 0.011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.0791749954223633,
      "rewards/margins": 15.069662094116211,
      "rewards/rejected": -12.99048900604248,
      "step": 550
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.2327790973871732e-07,
      "logits/chosen": -2.2084298133850098,
      "logits/rejected": -2.196798801422119,
      "logps/chosen": -216.34432983398438,
      "logps/rejected": -364.7033386230469,
      "loss": 0.0114,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6750681400299072,
      "rewards/margins": 12.608332633972168,
      "rewards/rejected": -10.933263778686523,
      "step": 560
    },
    {
      "epoch": 1.83,
      "learning_rate": 2.173396674584323e-07,
      "logits/chosen": -2.1891276836395264,
      "logits/rejected": -2.2081406116485596,
      "logps/chosen": -233.3393096923828,
      "logps/rejected": -390.85699462890625,
      "loss": 0.0068,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.547383189201355,
      "rewards/margins": 13.556729316711426,
      "rewards/rejected": -12.009345054626465,
      "step": 570
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.1140142517814726e-07,
      "logits/chosen": -2.184476375579834,
      "logits/rejected": -2.1820874214172363,
      "logps/chosen": -235.4527130126953,
      "logps/rejected": -346.69952392578125,
      "loss": 0.0142,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5227887630462646,
      "rewards/margins": 11.664176940917969,
      "rewards/rejected": -10.141386032104492,
      "step": 580
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.0546318289786222e-07,
      "logits/chosen": -2.1869351863861084,
      "logits/rejected": -2.1961607933044434,
      "logps/chosen": -219.57962036132812,
      "logps/rejected": -345.61663818359375,
      "loss": 0.0171,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.060570478439331,
      "rewards/margins": 12.773801803588867,
      "rewards/rejected": -10.71323013305664,
      "step": 590
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.9952494061757718e-07,
      "logits/chosen": -2.1666817665100098,
      "logits/rejected": -2.162493944168091,
      "logps/chosen": -220.46829223632812,
      "logps/rejected": -381.7313537597656,
      "loss": 0.0164,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8761663436889648,
      "rewards/margins": 12.851763725280762,
      "rewards/rejected": -10.97559642791748,
      "step": 600
    },
    {
      "epoch": 1.92,
      "eval_logits/chosen": -2.16733717918396,
      "eval_logits/rejected": -2.193927526473999,
      "eval_logps/chosen": -270.457275390625,
      "eval_logps/rejected": -290.05157470703125,
      "eval_loss": 0.8890212774276733,
      "eval_rewards/accuracies": 0.5873016119003296,
      "eval_rewards/chosen": -3.352169990539551,
      "eval_rewards/margins": 0.4150641858577728,
      "eval_rewards/rejected": -3.7672338485717773,
      "eval_runtime": 135.1266,
      "eval_samples_per_second": 7.4,
      "eval_steps_per_second": 0.466,
      "step": 600
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.9358669833729216e-07,
      "logits/chosen": -2.165018320083618,
      "logits/rejected": -2.1551880836486816,
      "logps/chosen": -236.99209594726562,
      "logps/rejected": -379.29034423828125,
      "loss": 0.0063,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8915096521377563,
      "rewards/margins": 12.898310661315918,
      "rewards/rejected": -11.006799697875977,
      "step": 610
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.876484560570071e-07,
      "logits/chosen": -2.112156391143799,
      "logits/rejected": -2.1361286640167236,
      "logps/chosen": -238.260986328125,
      "logps/rejected": -367.57647705078125,
      "loss": 0.0089,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.1359070539474487,
      "rewards/margins": 12.050466537475586,
      "rewards/rejected": -10.914558410644531,
      "step": 620
    },
    {
      "epoch": 2.02,
      "learning_rate": 1.8171021377672207e-07,
      "logits/chosen": -2.1914753913879395,
      "logits/rejected": -2.2092552185058594,
      "logps/chosen": -239.82968139648438,
      "logps/rejected": -379.8808898925781,
      "loss": 0.0065,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7776199579238892,
      "rewards/margins": 11.117961883544922,
      "rewards/rejected": -10.340343475341797,
      "step": 630
    },
    {
      "epoch": 2.05,
      "learning_rate": 1.7577197149643706e-07,
      "logits/chosen": -2.124424695968628,
      "logits/rejected": -2.1311514377593994,
      "logps/chosen": -251.8069610595703,
      "logps/rejected": -337.7670593261719,
      "loss": 0.0089,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.6031948924064636,
      "rewards/margins": 7.557145595550537,
      "rewards/rejected": -8.160341262817383,
      "step": 640
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.6983372921615202e-07,
      "logits/chosen": -2.106081008911133,
      "logits/rejected": -2.0959670543670654,
      "logps/chosen": -249.3319854736328,
      "logps/rejected": -362.3397521972656,
      "loss": 0.0062,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.243983656167984,
      "rewards/margins": 9.672110557556152,
      "rewards/rejected": -9.916093826293945,
      "step": 650
    },
    {
      "epoch": 2.12,
      "learning_rate": 1.6389548693586697e-07,
      "logits/chosen": -2.0426225662231445,
      "logits/rejected": -2.027235507965088,
      "logps/chosen": -264.58343505859375,
      "logps/rejected": -402.2667541503906,
      "loss": 0.0046,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.5264443159103394,
      "rewards/margins": 11.265119552612305,
      "rewards/rejected": -10.738676071166992,
      "step": 660
    },
    {
      "epoch": 2.15,
      "learning_rate": 1.5795724465558193e-07,
      "logits/chosen": -2.0862302780151367,
      "logits/rejected": -2.094032049179077,
      "logps/chosen": -219.7861785888672,
      "logps/rejected": -376.1147155761719,
      "loss": 0.0031,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6179927587509155,
      "rewards/margins": 14.43371868133545,
      "rewards/rejected": -12.815725326538086,
      "step": 670
    },
    {
      "epoch": 2.18,
      "learning_rate": 1.520190023752969e-07,
      "logits/chosen": -2.141444206237793,
      "logits/rejected": -2.1569528579711914,
      "logps/chosen": -217.67626953125,
      "logps/rejected": -390.2943115234375,
      "loss": 0.0026,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5685292482376099,
      "rewards/margins": 14.526385307312012,
      "rewards/rejected": -12.957855224609375,
      "step": 680
    },
    {
      "epoch": 2.21,
      "learning_rate": 1.4608076009501184e-07,
      "logits/chosen": -2.137134552001953,
      "logits/rejected": -2.1524899005889893,
      "logps/chosen": -213.85012817382812,
      "logps/rejected": -384.1824645996094,
      "loss": 0.0038,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9703474044799805,
      "rewards/margins": 15.583375930786133,
      "rewards/rejected": -13.613027572631836,
      "step": 690
    },
    {
      "epoch": 2.24,
      "learning_rate": 1.4014251781472683e-07,
      "logits/chosen": -2.0553553104400635,
      "logits/rejected": -2.0486502647399902,
      "logps/chosen": -226.2926788330078,
      "logps/rejected": -420.90472412109375,
      "loss": 0.0029,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.2629456520080566,
      "rewards/margins": 17.714906692504883,
      "rewards/rejected": -15.451959609985352,
      "step": 700
    },
    {
      "epoch": 2.24,
      "eval_logits/chosen": -2.091404676437378,
      "eval_logits/rejected": -2.118736982345581,
      "eval_logps/chosen": -283.0655517578125,
      "eval_logps/rejected": -303.959228515625,
      "eval_loss": 0.9730385541915894,
      "eval_rewards/accuracies": 0.5873016119003296,
      "eval_rewards/chosen": -4.612995147705078,
      "eval_rewards/margins": 0.5450049042701721,
      "eval_rewards/rejected": -5.157999515533447,
      "eval_runtime": 135.0313,
      "eval_samples_per_second": 7.406,
      "eval_steps_per_second": 0.467,
      "step": 700
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.342042755344418e-07,
      "logits/chosen": -2.106203079223633,
      "logits/rejected": -2.101205825805664,
      "logps/chosen": -209.47457885742188,
      "logps/rejected": -427.4419860839844,
      "loss": 0.0017,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.1314749717712402,
      "rewards/margins": 18.671459197998047,
      "rewards/rejected": -16.539981842041016,
      "step": 710
    },
    {
      "epoch": 2.31,
      "learning_rate": 1.2826603325415677e-07,
      "logits/chosen": -2.0930514335632324,
      "logits/rejected": -2.100776195526123,
      "logps/chosen": -232.50479125976562,
      "logps/rejected": -422.82537841796875,
      "loss": 0.0014,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.4403505325317383,
      "rewards/margins": 17.976802825927734,
      "rewards/rejected": -15.536453247070312,
      "step": 720
    },
    {
      "epoch": 2.34,
      "learning_rate": 1.2232779097387173e-07,
      "logits/chosen": -2.0570120811462402,
      "logits/rejected": -2.048424243927002,
      "logps/chosen": -220.833251953125,
      "logps/rejected": -409.28887939453125,
      "loss": 0.0009,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.0816102027893066,
      "rewards/margins": 19.023530960083008,
      "rewards/rejected": -16.941919326782227,
      "step": 730
    },
    {
      "epoch": 2.37,
      "learning_rate": 1.163895486935867e-07,
      "logits/chosen": -2.0480129718780518,
      "logits/rejected": -2.0570969581604004,
      "logps/chosen": -218.54385375976562,
      "logps/rejected": -423.27685546875,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.177452802658081,
      "rewards/margins": 19.3106689453125,
      "rewards/rejected": -17.13321876525879,
      "step": 740
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.1045130641330165e-07,
      "logits/chosen": -2.0393099784851074,
      "logits/rejected": -2.064631938934326,
      "logps/chosen": -240.68948364257812,
      "logps/rejected": -438.4137268066406,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9609296321868896,
      "rewards/margins": 19.048954010009766,
      "rewards/rejected": -17.088024139404297,
      "step": 750
    },
    {
      "epoch": 2.44,
      "learning_rate": 1.0451306413301662e-07,
      "logits/chosen": -2.078218698501587,
      "logits/rejected": -2.09151554107666,
      "logps/chosen": -238.7852783203125,
      "logps/rejected": -427.560791015625,
      "loss": 0.0023,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3349409103393555,
      "rewards/margins": 17.530380249023438,
      "rewards/rejected": -16.195438385009766,
      "step": 760
    },
    {
      "epoch": 2.47,
      "learning_rate": 9.857482185273158e-08,
      "logits/chosen": -2.0929994583129883,
      "logits/rejected": -2.1046335697174072,
      "logps/chosen": -257.6733703613281,
      "logps/rejected": -464.05523681640625,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.210766315460205,
      "rewards/margins": 18.859294891357422,
      "rewards/rejected": -16.648527145385742,
      "step": 770
    },
    {
      "epoch": 2.5,
      "learning_rate": 9.263657957244655e-08,
      "logits/chosen": -2.113415241241455,
      "logits/rejected": -2.1267588138580322,
      "logps/chosen": -249.73681640625,
      "logps/rejected": -431.0743713378906,
      "loss": 0.002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.9247767329216003,
      "rewards/margins": 16.176433563232422,
      "rewards/rejected": -15.251657485961914,
      "step": 780
    },
    {
      "epoch": 2.53,
      "learning_rate": 8.669833729216151e-08,
      "logits/chosen": -2.074481725692749,
      "logits/rejected": -2.0785088539123535,
      "logps/chosen": -227.53369140625,
      "logps/rejected": -384.003173828125,
      "loss": 0.0013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.8787251710891724,
      "rewards/margins": 15.583755493164062,
      "rewards/rejected": -14.705029487609863,
      "step": 790
    },
    {
      "epoch": 2.56,
      "learning_rate": 8.076009501187649e-08,
      "logits/chosen": -2.1291556358337402,
      "logits/rejected": -2.144925594329834,
      "logps/chosen": -226.36203002929688,
      "logps/rejected": -399.88360595703125,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.0979222059249878,
      "rewards/margins": 16.30962562561035,
      "rewards/rejected": -15.211705207824707,
      "step": 800
    },
    {
      "epoch": 2.56,
      "eval_logits/chosen": -2.076233148574829,
      "eval_logits/rejected": -2.10378098487854,
      "eval_logps/chosen": -289.43609619140625,
      "eval_logps/rejected": -310.7235412597656,
      "eval_loss": 1.015881061553955,
      "eval_rewards/accuracies": 0.5873016119003296,
      "eval_rewards/chosen": -5.250052452087402,
      "eval_rewards/margins": 0.5843777656555176,
      "eval_rewards/rejected": -5.834429740905762,
      "eval_runtime": 135.0469,
      "eval_samples_per_second": 7.405,
      "eval_steps_per_second": 0.467,
      "step": 800
    },
    {
      "epoch": 2.6,
      "learning_rate": 7.482185273159145e-08,
      "logits/chosen": -2.072613477706909,
      "logits/rejected": -2.0949771404266357,
      "logps/chosen": -245.9056854248047,
      "logps/rejected": -431.50079345703125,
      "loss": 0.0015,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3889291286468506,
      "rewards/margins": 18.065763473510742,
      "rewards/rejected": -16.676836013793945,
      "step": 810
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.88836104513064e-08,
      "logits/chosen": -2.1127004623413086,
      "logits/rejected": -2.095151424407959,
      "logps/chosen": -224.3485565185547,
      "logps/rejected": -392.89373779296875,
      "loss": 0.0016,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.0925308465957642,
      "rewards/margins": 16.016094207763672,
      "rewards/rejected": -14.923563957214355,
      "step": 820
    },
    {
      "epoch": 2.66,
      "learning_rate": 6.294536817102138e-08,
      "logits/chosen": -2.043457508087158,
      "logits/rejected": -2.0708208084106445,
      "logps/chosen": -245.05282592773438,
      "logps/rejected": -444.1641540527344,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6503627300262451,
      "rewards/margins": 18.23184585571289,
      "rewards/rejected": -16.58148193359375,
      "step": 830
    },
    {
      "epoch": 2.69,
      "learning_rate": 5.700712589073634e-08,
      "logits/chosen": -2.0662128925323486,
      "logits/rejected": -2.0693325996398926,
      "logps/chosen": -219.8564910888672,
      "logps/rejected": -358.3108215332031,
      "loss": 0.002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.03243887424468994,
      "rewards/margins": 13.410202026367188,
      "rewards/rejected": -13.442642211914062,
      "step": 840
    },
    {
      "epoch": 2.72,
      "learning_rate": 5.10688836104513e-08,
      "logits/chosen": -2.0619301795959473,
      "logits/rejected": -2.0707077980041504,
      "logps/chosen": -218.51416015625,
      "logps/rejected": -394.4737243652344,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.7091328501701355,
      "rewards/margins": 15.42210865020752,
      "rewards/rejected": -14.712974548339844,
      "step": 850
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.5130641330166267e-08,
      "logits/chosen": -2.098428964614868,
      "logits/rejected": -2.1084647178649902,
      "logps/chosen": -253.61538696289062,
      "logps/rejected": -429.78759765625,
      "loss": 0.002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4912437200546265,
      "rewards/margins": 17.59982681274414,
      "rewards/rejected": -16.108583450317383,
      "step": 860
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.919239904988123e-08,
      "logits/chosen": -2.1191351413726807,
      "logits/rejected": -2.1331193447113037,
      "logps/chosen": -219.34622192382812,
      "logps/rejected": -370.738037109375,
      "loss": 0.0105,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.8942230939865112,
      "rewards/margins": 14.622076034545898,
      "rewards/rejected": -13.727853775024414,
      "step": 870
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.32541567695962e-08,
      "logits/chosen": -2.104769229888916,
      "logits/rejected": -2.111499309539795,
      "logps/chosen": -249.5308380126953,
      "logps/rejected": -430.3035583496094,
      "loss": 0.0016,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.38554805517196655,
      "rewards/margins": 14.73956298828125,
      "rewards/rejected": -14.354013442993164,
      "step": 880
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.7315914489311164e-08,
      "logits/chosen": -2.1363842487335205,
      "logits/rejected": -2.141021490097046,
      "logps/chosen": -232.86355590820312,
      "logps/rejected": -376.73626708984375,
      "loss": 0.0025,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.6885503530502319,
      "rewards/margins": 13.440515518188477,
      "rewards/rejected": -12.75196361541748,
      "step": 890
    },
    {
      "epoch": 2.88,
      "learning_rate": 2.1377672209026125e-08,
      "logits/chosen": -2.136543035507202,
      "logits/rejected": -2.1440882682800293,
      "logps/chosen": -252.4918975830078,
      "logps/rejected": -385.3263854980469,
      "loss": 0.0128,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.040738582611084,
      "rewards/margins": 14.618490219116211,
      "rewards/rejected": -13.577753067016602,
      "step": 900
    },
    {
      "epoch": 2.88,
      "eval_logits/chosen": -2.1040961742401123,
      "eval_logits/rejected": -2.13144588470459,
      "eval_logps/chosen": -289.4458312988281,
      "eval_logps/rejected": -310.62359619140625,
      "eval_loss": 1.0216755867004395,
      "eval_rewards/accuracies": 0.5753968358039856,
      "eval_rewards/chosen": -5.25102424621582,
      "eval_rewards/margins": 0.5734108090400696,
      "eval_rewards/rejected": -5.824434757232666,
      "eval_runtime": 135.1223,
      "eval_samples_per_second": 7.401,
      "eval_steps_per_second": 0.466,
      "step": 900
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.5439429928741092e-08,
      "logits/chosen": -2.087221622467041,
      "logits/rejected": -2.0887157917022705,
      "logps/chosen": -234.03524780273438,
      "logps/rejected": -412.8173828125,
      "loss": 0.0013,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.5808283090591431,
      "rewards/margins": 14.590052604675293,
      "rewards/rejected": -14.009223937988281,
      "step": 910
    },
    {
      "epoch": 2.95,
      "learning_rate": 9.501187648456057e-09,
      "logits/chosen": -2.142925262451172,
      "logits/rejected": -2.151294231414795,
      "logps/chosen": -233.02587890625,
      "logps/rejected": -392.322998046875,
      "loss": 0.0016,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.5956318974494934,
      "rewards/margins": 14.147176742553711,
      "rewards/rejected": -13.551542282104492,
      "step": 920
    },
    {
      "epoch": 2.98,
      "learning_rate": 3.562945368171021e-09,
      "logits/chosen": -2.0257132053375244,
      "logits/rejected": -2.0395452976226807,
      "logps/chosen": -239.80917358398438,
      "logps/rejected": -391.6462097167969,
      "loss": 0.0017,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.49750471115112305,
      "rewards/margins": 14.218188285827637,
      "rewards/rejected": -13.720685005187988,
      "step": 930
    },
    {
      "epoch": 3.0,
      "step": 936,
      "total_flos": 0.0,
      "train_loss": 0.23198359412724215,
      "train_runtime": 18849.1473,
      "train_samples_per_second": 3.176,
      "train_steps_per_second": 0.05
    }
  ],
  "logging_steps": 10,
  "max_steps": 936,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}