{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984431759211209,
  "eval_steps": 1000,
  "global_step": 481,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020757654385054488,
      "grad_norm": 7.453312579539728,
      "learning_rate": 1.020408163265306e-08,
      "logits/chosen": -2.730942726135254,
      "logits/rejected": -2.654609203338623,
      "logps/chosen": -350.489990234375,
      "logps/rejected": -325.546875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.02075765438505449,
      "grad_norm": 7.878888995232912,
      "learning_rate": 1.0204081632653061e-07,
      "logits/chosen": -2.7330236434936523,
      "logits/rejected": -2.735116720199585,
      "logps/chosen": -366.51531982421875,
      "logps/rejected": -412.2677001953125,
      "loss": 0.6931,
      "rewards/accuracies": 0.4270833432674408,
      "rewards/chosen": -0.000205132644623518,
      "rewards/margins": 4.354613702162169e-05,
      "rewards/rejected": -0.00024867875617928803,
      "step": 10
    },
    {
      "epoch": 0.04151530877010898,
      "grad_norm": 7.53294584904676,
      "learning_rate": 2.0408163265306121e-07,
      "logits/chosen": -2.7173304557800293,
      "logits/rejected": -2.693912982940674,
      "logps/chosen": -378.73748779296875,
      "logps/rejected": -404.47003173828125,
      "loss": 0.6892,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": 0.0025507560931146145,
      "rewards/margins": 0.008084132336080074,
      "rewards/rejected": -0.005533376708626747,
      "step": 20
    },
    {
      "epoch": 0.062272963155163466,
      "grad_norm": 8.007471678418003,
      "learning_rate": 3.0612244897959183e-07,
      "logits/chosen": -2.716646194458008,
      "logits/rejected": -2.700786590576172,
      "logps/chosen": -363.6639709472656,
      "logps/rejected": -390.54083251953125,
      "loss": 0.6692,
      "rewards/accuracies": 0.903124988079071,
      "rewards/chosen": 0.02478734776377678,
      "rewards/margins": 0.051134396344423294,
      "rewards/rejected": -0.026347041130065918,
      "step": 30
    },
    {
      "epoch": 0.08303061754021795,
      "grad_norm": 9.3622528637074,
      "learning_rate": 4.0816326530612243e-07,
      "logits/chosen": -2.7087109088897705,
      "logits/rejected": -2.669712543487549,
      "logps/chosen": -347.83538818359375,
      "logps/rejected": -376.85260009765625,
      "loss": 0.6044,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 0.05488457530736923,
      "rewards/margins": 0.19498120248317719,
      "rewards/rejected": -0.14009663462638855,
      "step": 40
    },
    {
      "epoch": 0.10378827192527244,
      "grad_norm": 11.969966849217528,
      "learning_rate": 4.999933894080444e-07,
      "logits/chosen": -2.7135281562805176,
      "logits/rejected": -2.6938090324401855,
      "logps/chosen": -403.2617492675781,
      "logps/rejected": -495.21270751953125,
      "loss": 0.4674,
      "rewards/accuracies": 0.871874988079071,
      "rewards/chosen": -0.37011662125587463,
      "rewards/margins": 0.6785963177680969,
      "rewards/rejected": -1.048712968826294,
      "step": 50
    },
    {
      "epoch": 0.12454592631032693,
      "grad_norm": 19.176201042712012,
      "learning_rate": 4.992005413014143e-07,
      "logits/chosen": -2.7302985191345215,
      "logits/rejected": -2.7273764610290527,
      "logps/chosen": -528.5646362304688,
      "logps/rejected": -741.4615478515625,
      "loss": 0.3523,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -1.8138965368270874,
      "rewards/margins": 1.7118844985961914,
      "rewards/rejected": -3.5257811546325684,
      "step": 60
    },
    {
      "epoch": 0.14530358069538143,
      "grad_norm": 20.310776567548704,
      "learning_rate": 4.970903776169402e-07,
      "logits/chosen": -2.7460341453552246,
      "logits/rejected": -2.7275753021240234,
      "logps/chosen": -634.8268432617188,
      "logps/rejected": -865.6439208984375,
      "loss": 0.3052,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -2.382263422012329,
      "rewards/margins": 2.2324626445770264,
      "rewards/rejected": -4.6147260665893555,
      "step": 70
    },
    {
      "epoch": 0.1660612350804359,
      "grad_norm": 19.886871444649117,
      "learning_rate": 4.936740530314087e-07,
      "logits/chosen": -2.3413853645324707,
      "logits/rejected": -2.102804660797119,
      "logps/chosen": -591.3840942382812,
      "logps/rejected": -896.90625,
      "loss": 0.25,
      "rewards/accuracies": 0.903124988079071,
      "rewards/chosen": -2.003624200820923,
      "rewards/margins": 2.942108631134033,
      "rewards/rejected": -4.945733070373535,
      "step": 80
    },
    {
      "epoch": 0.1868188894654904,
      "grad_norm": 16.07417524586734,
      "learning_rate": 4.889696268057348e-07,
      "logits/chosen": -1.8468377590179443,
      "logits/rejected": -1.1863611936569214,
      "logps/chosen": -567.8921508789062,
      "logps/rejected": -934.1871337890625,
      "loss": 0.2254,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -2.0263195037841797,
      "rewards/margins": 3.4611504077911377,
      "rewards/rejected": -5.487469673156738,
      "step": 90
    },
    {
      "epoch": 0.2075765438505449,
      "grad_norm": 15.952464662940557,
      "learning_rate": 4.830019673206996e-07,
      "logits/chosen": -1.3128455877304077,
      "logits/rejected": -0.37191733717918396,
      "logps/chosen": -637.7520751953125,
      "logps/rejected": -1123.6968994140625,
      "loss": 0.2034,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -2.622469425201416,
      "rewards/margins": 4.605704307556152,
      "rewards/rejected": -7.22817325592041,
      "step": 100
    },
    {
      "epoch": 0.2283341982355994,
      "grad_norm": 20.61911089038355,
      "learning_rate": 4.7580262061854606e-07,
      "logits/chosen": -0.8984780311584473,
      "logits/rejected": 0.00587611785158515,
      "logps/chosen": -629.09521484375,
      "logps/rejected": -1123.9976806640625,
      "loss": 0.2043,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -2.5760746002197266,
      "rewards/margins": 4.712790012359619,
      "rewards/rejected": -7.288865089416504,
      "step": 110
    },
    {
      "epoch": 0.24909185262065386,
      "grad_norm": 13.519087980837655,
      "learning_rate": 4.674096436453447e-07,
      "logits/chosen": -0.8746647834777832,
      "logits/rejected": 0.03333142027258873,
      "logps/chosen": -662.9947509765625,
      "logps/rejected": -1107.119140625,
      "loss": 0.1947,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -2.6802916526794434,
      "rewards/margins": 4.257796287536621,
      "rewards/rejected": -6.938088417053223,
      "step": 120
    },
    {
      "epoch": 0.26984950700570837,
      "grad_norm": 15.179032922942401,
      "learning_rate": 4.578674030756363e-07,
      "logits/chosen": -0.5216516256332397,
      "logits/rejected": 0.797188401222229,
      "logps/chosen": -672.4591064453125,
      "logps/rejected": -1201.9649658203125,
      "loss": 0.1766,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -2.92866849899292,
      "rewards/margins": 5.020692348480225,
      "rewards/rejected": -7.949362277984619,
      "step": 130
    },
    {
      "epoch": 0.29060716139076287,
      "grad_norm": 19.394840025974254,
      "learning_rate": 4.4722634078279865e-07,
      "logits/chosen": 0.05672640725970268,
      "logits/rejected": 1.2894923686981201,
      "logps/chosen": -632.3123779296875,
      "logps/rejected": -1134.365478515625,
      "loss": 0.1989,
      "rewards/accuracies": 0.8843749761581421,
      "rewards/chosen": -2.6184778213500977,
      "rewards/margins": 4.792341709136963,
      "rewards/rejected": -7.410820007324219,
      "step": 140
    },
    {
      "epoch": 0.3113648157758173,
      "grad_norm": 15.545837682982413,
      "learning_rate": 4.355427071949004e-07,
      "logits/chosen": -0.034926723688840866,
      "logits/rejected": 1.3178081512451172,
      "logps/chosen": -625.6254272460938,
      "logps/rejected": -1133.8699951171875,
      "loss": 0.1657,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -2.654139757156372,
      "rewards/margins": 4.885876655578613,
      "rewards/rejected": -7.540017127990723,
      "step": 150
    },
    {
      "epoch": 0.3321224701608718,
      "grad_norm": 21.099719863638594,
      "learning_rate": 4.228782639455674e-07,
      "logits/chosen": -0.2264009416103363,
      "logits/rejected": 1.3679448366165161,
      "logps/chosen": -684.0534057617188,
      "logps/rejected": -1281.2177734375,
      "loss": 0.1738,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -3.0014939308166504,
      "rewards/margins": 5.716488838195801,
      "rewards/rejected": -8.717982292175293,
      "step": 160
    },
    {
      "epoch": 0.3528801245459263,
      "grad_norm": 14.4755316075222,
      "learning_rate": 4.092999573916971e-07,
      "logits/chosen": 0.14696760475635529,
      "logits/rejected": 1.6602694988250732,
      "logps/chosen": -664.4630126953125,
      "logps/rejected": -1225.433837890625,
      "loss": 0.1804,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -3.056241273880005,
      "rewards/margins": 5.417787551879883,
      "rewards/rejected": -8.474028587341309,
      "step": 170
    },
    {
      "epoch": 0.3736377789309808,
      "grad_norm": 15.213191213812825,
      "learning_rate": 3.948795647238637e-07,
      "logits/chosen": -0.7323606014251709,
      "logits/rejected": 1.1011723279953003,
      "logps/chosen": -630.0668334960938,
      "logps/rejected": -1234.099609375,
      "loss": 0.1783,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -2.6342709064483643,
      "rewards/margins": 5.844083786010742,
      "rewards/rejected": -8.478353500366211,
      "step": 180
    },
    {
      "epoch": 0.39439543331603527,
      "grad_norm": 16.709219865079046,
      "learning_rate": 3.796933145401304e-07,
      "logits/chosen": -0.12861236929893494,
      "logits/rejected": 1.5260117053985596,
      "logps/chosen": -731.4463500976562,
      "logps/rejected": -1386.452880859375,
      "loss": 0.1646,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -3.487692356109619,
      "rewards/margins": 6.185595989227295,
      "rewards/rejected": -9.673288345336914,
      "step": 190
    },
    {
      "epoch": 0.4151530877010898,
      "grad_norm": 16.46330948268556,
      "learning_rate": 3.638214838889801e-07,
      "logits/chosen": 0.014425823464989662,
      "logits/rejected": 1.5777640342712402,
      "logps/chosen": -647.3234252929688,
      "logps/rejected": -1219.0548095703125,
      "loss": 0.1669,
      "rewards/accuracies": 0.909375011920929,
      "rewards/chosen": -2.797513484954834,
      "rewards/margins": 5.429152011871338,
      "rewards/rejected": -8.226665496826172,
      "step": 200
    },
    {
      "epoch": 0.4359107420861443,
      "grad_norm": 19.39279809291309,
      "learning_rate": 3.4734797391146383e-07,
      "logits/chosen": -0.12084762752056122,
      "logits/rejected": 1.6122562885284424,
      "logps/chosen": -656.5711059570312,
      "logps/rejected": -1250.8482666015625,
      "loss": 0.1594,
      "rewards/accuracies": 0.9281250238418579,
      "rewards/chosen": -2.874763011932373,
      "rewards/margins": 5.74980354309082,
      "rewards/rejected": -8.624567031860352,
      "step": 210
    },
    {
      "epoch": 0.4566683964711988,
      "grad_norm": 23.517592451341034,
      "learning_rate": 3.3035986632579036e-07,
      "logits/chosen": -1.0772771835327148,
      "logits/rejected": 0.6209205389022827,
      "logps/chosen": -622.2130126953125,
      "logps/rejected": -1293.0042724609375,
      "loss": 0.1678,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -2.5699634552001953,
      "rewards/margins": 6.527965545654297,
      "rewards/rejected": -9.097929000854492,
      "step": 220
    },
    {
      "epoch": 0.4774260508562532,
      "grad_norm": 20.580951166094668,
      "learning_rate": 3.1294696309885716e-07,
      "logits/chosen": -1.1179264783859253,
      "logits/rejected": 0.7691652178764343,
      "logps/chosen": -666.9544677734375,
      "logps/rejected": -1416.962158203125,
      "loss": 0.1673,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -3.058974027633667,
      "rewards/margins": 7.343924522399902,
      "rewards/rejected": -10.402898788452148,
      "step": 230
    },
    {
      "epoch": 0.49818370524130773,
      "grad_norm": 14.600883148928759,
      "learning_rate": 2.952013117380913e-07,
      "logits/chosen": -0.9207614660263062,
      "logits/rejected": 1.1953575611114502,
      "logps/chosen": -662.4710693359375,
      "logps/rejected": -1418.23291015625,
      "loss": 0.1614,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -3.0431346893310547,
      "rewards/margins": 7.319464206695557,
      "rewards/rejected": -10.362597465515137,
      "step": 240
    },
    {
      "epoch": 0.5189413596263622,
      "grad_norm": 11.446046359523642,
      "learning_rate": 2.7721671871299114e-07,
      "logits/chosen": -0.7357327938079834,
      "logits/rejected": 1.461576223373413,
      "logps/chosen": -672.9193725585938,
      "logps/rejected": -1338.949462890625,
      "loss": 0.1602,
      "rewards/accuracies": 0.940625011920929,
      "rewards/chosen": -2.9312338829040527,
      "rewards/margins": 6.449375152587891,
      "rewards/rejected": -9.380608558654785,
      "step": 250
    },
    {
      "epoch": 0.5396990140114167,
      "grad_norm": 14.757287503572078,
      "learning_rate": 2.5908825357849993e-07,
      "logits/chosen": -0.8231679797172546,
      "logits/rejected": 1.1155385971069336,
      "logps/chosen": -656.1690063476562,
      "logps/rejected": -1271.033447265625,
      "loss": 0.1622,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.8299801349639893,
      "rewards/margins": 5.925788402557373,
      "rewards/rejected": -8.755769729614258,
      "step": 260
    },
    {
      "epoch": 0.5604566683964712,
      "grad_norm": 17.896435569322648,
      "learning_rate": 2.409117464215001e-07,
      "logits/chosen": -0.4632663130760193,
      "logits/rejected": 1.819011926651001,
      "logps/chosen": -664.986572265625,
      "logps/rejected": -1455.3441162109375,
      "loss": 0.1471,
      "rewards/accuracies": 0.940625011920929,
      "rewards/chosen": -3.113548517227173,
      "rewards/margins": 7.589502811431885,
      "rewards/rejected": -10.70305061340332,
      "step": 270
    },
    {
      "epoch": 0.5812143227815257,
      "grad_norm": 18.08668716790038,
      "learning_rate": 2.227832812870089e-07,
      "logits/chosen": -0.542155385017395,
      "logits/rejected": 1.9537347555160522,
      "logps/chosen": -686.6569213867188,
      "logps/rejected": -1460.84619140625,
      "loss": 0.1557,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -3.106727123260498,
      "rewards/margins": 7.513753414154053,
      "rewards/rejected": -10.620479583740234,
      "step": 280
    },
    {
      "epoch": 0.6019719771665801,
      "grad_norm": 20.414881165009998,
      "learning_rate": 2.0479868826190871e-07,
      "logits/chosen": -0.437448650598526,
      "logits/rejected": 1.7888593673706055,
      "logps/chosen": -709.882568359375,
      "logps/rejected": -1392.702392578125,
      "loss": 0.1617,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.35121488571167,
      "rewards/margins": 6.753846645355225,
      "rewards/rejected": -10.105062484741211,
      "step": 290
    },
    {
      "epoch": 0.6227296315516346,
      "grad_norm": 16.484968508125828,
      "learning_rate": 1.8705303690114287e-07,
      "logits/chosen": -0.2719939947128296,
      "logits/rejected": 1.811428427696228,
      "logps/chosen": -710.2088012695312,
      "logps/rejected": -1419.313720703125,
      "loss": 0.1478,
      "rewards/accuracies": 0.9468749761581421,
      "rewards/chosen": -3.3416149616241455,
      "rewards/margins": 6.840612888336182,
      "rewards/rejected": -10.182229042053223,
      "step": 300
    },
    {
      "epoch": 0.6434872859366891,
      "grad_norm": 15.309368142287557,
      "learning_rate": 1.6964013367420965e-07,
      "logits/chosen": -0.341867595911026,
      "logits/rejected": 1.7422988414764404,
      "logps/chosen": -686.5745849609375,
      "logps/rejected": -1366.7637939453125,
      "loss": 0.1534,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.167562961578369,
      "rewards/margins": 6.726889133453369,
      "rewards/rejected": -9.894452095031738,
      "step": 310
    },
    {
      "epoch": 0.6642449403217436,
      "grad_norm": 11.4821386278306,
      "learning_rate": 1.5265202608853628e-07,
      "logits/chosen": -0.17296895384788513,
      "logits/rejected": 1.9412486553192139,
      "logps/chosen": -665.36376953125,
      "logps/rejected": -1430.9146728515625,
      "loss": 0.148,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -3.0024094581604004,
      "rewards/margins": 7.437635898590088,
      "rewards/rejected": -10.440046310424805,
      "step": 320
    },
    {
      "epoch": 0.6850025947067981,
      "grad_norm": 10.552793459289758,
      "learning_rate": 1.3617851611101993e-07,
      "logits/chosen": -0.5047305226325989,
      "logits/rejected": 1.5836106538772583,
      "logps/chosen": -686.6402587890625,
      "logps/rejected": -1443.398193359375,
      "loss": 0.1519,
      "rewards/accuracies": 0.8968750238418579,
      "rewards/chosen": -3.1413583755493164,
      "rewards/margins": 7.2722015380859375,
      "rewards/rejected": -10.413559913635254,
      "step": 330
    },
    {
      "epoch": 0.7057602490918526,
      "grad_norm": 17.65296980495948,
      "learning_rate": 1.2030668545986958e-07,
      "logits/chosen": -0.569928765296936,
      "logits/rejected": 1.6948425769805908,
      "logps/chosen": -716.9683837890625,
      "logps/rejected": -1518.8372802734375,
      "loss": 0.1463,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -3.4350712299346924,
      "rewards/margins": 7.720976829528809,
      "rewards/rejected": -11.156047821044922,
      "step": 340
    },
    {
      "epoch": 0.7265179034769071,
      "grad_norm": 14.715250965523067,
      "learning_rate": 1.0512043527613623e-07,
      "logits/chosen": -0.7549006342887878,
      "logits/rejected": 1.3407833576202393,
      "logps/chosen": -695.8525390625,
      "logps/rejected": -1478.35546875,
      "loss": 0.1559,
      "rewards/accuracies": 0.903124988079071,
      "rewards/chosen": -3.347827911376953,
      "rewards/margins": 7.52248477935791,
      "rewards/rejected": -10.870311737060547,
      "step": 350
    },
    {
      "epoch": 0.7472755578619616,
      "grad_norm": 18.014896980938854,
      "learning_rate": 9.070004260830294e-08,
      "logits/chosen": -0.9223737716674805,
      "logits/rejected": 1.4507310390472412,
      "logps/chosen": -692.382080078125,
      "logps/rejected": -1453.919921875,
      "loss": 0.1437,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -3.2790751457214355,
      "rewards/margins": 7.526673316955566,
      "rewards/rejected": -10.80574893951416,
      "step": 360
    },
    {
      "epoch": 0.768033212247016,
      "grad_norm": 14.401106731580096,
      "learning_rate": 7.712173605443267e-08,
      "logits/chosen": -0.9376351237297058,
      "logits/rejected": 1.583579421043396,
      "logps/chosen": -700.8660888671875,
      "logps/rejected": -1505.467529296875,
      "loss": 0.1494,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -3.340554714202881,
      "rewards/margins": 7.916815280914307,
      "rewards/rejected": -11.257369995117188,
      "step": 370
    },
    {
      "epoch": 0.7887908666320705,
      "grad_norm": 11.438125899827915,
      "learning_rate": 6.445729280509957e-08,
      "logits/chosen": -0.8953694105148315,
      "logits/rejected": 1.4685465097427368,
      "logps/chosen": -690.8338623046875,
      "logps/rejected": -1588.8245849609375,
      "loss": 0.1468,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -3.3841090202331543,
      "rewards/margins": 8.392255783081055,
      "rewards/rejected": -11.77636432647705,
      "step": 380
    },
    {
      "epoch": 0.809548521017125,
      "grad_norm": 13.207322793003062,
      "learning_rate": 5.2773659217201364e-08,
      "logits/chosen": -0.9152933359146118,
      "logits/rejected": 1.6746854782104492,
      "logps/chosen": -707.5098266601562,
      "logps/rejected": -1445.887939453125,
      "loss": 0.1555,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -3.3184616565704346,
      "rewards/margins": 7.17882776260376,
      "rewards/rejected": -10.497289657592773,
      "step": 390
    },
    {
      "epoch": 0.8303061754021795,
      "grad_norm": 11.811171239173802,
      "learning_rate": 4.213259692436366e-08,
      "logits/chosen": -0.807452380657196,
      "logits/rejected": 1.5794246196746826,
      "logps/chosen": -692.3187255859375,
      "logps/rejected": -1477.223388671875,
      "loss": 0.1455,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -3.2713589668273926,
      "rewards/margins": 7.557108402252197,
      "rewards/rejected": -10.82846736907959,
      "step": 400
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 13.900287224699019,
      "learning_rate": 3.259035635465529e-08,
      "logits/chosen": -0.6530941128730774,
      "logits/rejected": 1.565045714378357,
      "logps/chosen": -704.8355712890625,
      "logps/rejected": -1375.3914794921875,
      "loss": 0.1474,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -3.393775463104248,
      "rewards/margins": 6.558309078216553,
      "rewards/rejected": -9.952085494995117,
      "step": 410
    },
    {
      "epoch": 0.8718214841722886,
      "grad_norm": 25.604242461790278,
      "learning_rate": 2.4197379381453942e-08,
      "logits/chosen": -0.6059257388114929,
      "logits/rejected": 1.747667670249939,
      "logps/chosen": -710.49365234375,
      "logps/rejected": -1475.5013427734375,
      "loss": 0.1551,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.3643269538879395,
      "rewards/margins": 7.329138278961182,
      "rewards/rejected": -10.693464279174805,
      "step": 420
    },
    {
      "epoch": 0.892579138557343,
      "grad_norm": 17.02837493916288,
      "learning_rate": 1.699803267930039e-08,
      "logits/chosen": -0.6381738781929016,
      "logits/rejected": 1.581946611404419,
      "logps/chosen": -700.8256225585938,
      "logps/rejected": -1424.5767822265625,
      "loss": 0.14,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.278334140777588,
      "rewards/margins": 7.011561393737793,
      "rewards/rejected": -10.289896965026855,
      "step": 430
    },
    {
      "epoch": 0.9133367929423976,
      "grad_norm": 12.115208573188879,
      "learning_rate": 1.1030373194265114e-08,
      "logits/chosen": -0.6104884743690491,
      "logits/rejected": 1.8680555820465088,
      "logps/chosen": -696.623046875,
      "logps/rejected": -1484.712646484375,
      "loss": 0.1555,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -3.214094638824463,
      "rewards/margins": 7.734494686126709,
      "rewards/rejected": -10.948590278625488,
      "step": 440
    },
    {
      "epoch": 0.934094447327452,
      "grad_norm": 16.53240755993655,
      "learning_rate": 6.325946968591317e-09,
      "logits/chosen": -0.5625468492507935,
      "logits/rejected": 2.0248143672943115,
      "logps/chosen": -701.9017333984375,
      "logps/rejected": -1457.185791015625,
      "loss": 0.1506,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -3.3476765155792236,
      "rewards/margins": 7.459791660308838,
      "rewards/rejected": -10.807466506958008,
      "step": 450
    },
    {
      "epoch": 0.9548521017125065,
      "grad_norm": 13.032337936414088,
      "learning_rate": 2.909622383059834e-09,
      "logits/chosen": -0.5611749291419983,
      "logits/rejected": 1.8944803476333618,
      "logps/chosen": -692.7264404296875,
      "logps/rejected": -1460.9708251953125,
      "loss": 0.1466,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -3.228968858718872,
      "rewards/margins": 7.5441155433654785,
      "rewards/rejected": -10.77308464050293,
      "step": 460
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 17.252869051917436,
      "learning_rate": 7.994586985856089e-10,
      "logits/chosen": -0.6213638782501221,
      "logits/rejected": 1.7617113590240479,
      "logps/chosen": -707.3765869140625,
      "logps/rejected": -1481.257080078125,
      "loss": 0.1374,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.3621768951416016,
      "rewards/margins": 7.499251365661621,
      "rewards/rejected": -10.861429214477539,
      "step": 470
    },
    {
      "epoch": 0.9963674104826155,
      "grad_norm": 12.988129191271089,
      "learning_rate": 6.610591955641398e-12,
      "logits/chosen": -0.5171535015106201,
      "logits/rejected": 1.7204986810684204,
      "logps/chosen": -687.4800415039062,
      "logps/rejected": -1466.64501953125,
      "loss": 0.1487,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -3.355544328689575,
      "rewards/margins": 7.49337911605835,
      "rewards/rejected": -10.848923683166504,
      "step": 480
    },
    {
      "epoch": 0.9984431759211209,
      "step": 481,
      "total_flos": 0.0,
      "train_loss": 0.21933725711709495,
      "train_runtime": 13808.8776,
      "train_samples_per_second": 8.93,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 10,
  "max_steps": 481,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}