{
  "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": 7.453095565847237,
      "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": 7.86433669667092,
      "learning_rate": 1.0204081632653061e-07,
      "logits/chosen": -2.7328553199768066,
      "logits/rejected": -2.734957695007324,
      "logps/chosen": -366.4744567871094,
      "logps/rejected": -412.2707824707031,
      "loss": 0.693,
      "rewards/accuracies": 0.4583333432674408,
      "rewards/chosen": 0.0002039541577687487,
      "rewards/margins": 0.00048357987543568015,
      "rewards/rejected": -0.00027962555759586394,
      "step": 10
    },
    {
      "epoch": 0.04151530877010898,
      "grad_norm": 7.528284552671321,
      "learning_rate": 2.0408163265306121e-07,
      "logits/chosen": -2.717175006866455,
      "logits/rejected": -2.6938576698303223,
      "logps/chosen": -378.77813720703125,
      "logps/rejected": -404.4546813964844,
      "loss": 0.6892,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 0.002144110854715109,
      "rewards/margins": 0.007523530628532171,
      "rewards/rejected": -0.00537942023947835,
      "step": 20
    },
    {
      "epoch": 0.062272963155163466,
      "grad_norm": 8.005217747452576,
      "learning_rate": 3.0612244897959183e-07,
      "logits/chosen": -2.7168049812316895,
      "logits/rejected": -2.7009923458099365,
      "logps/chosen": -363.69952392578125,
      "logps/rejected": -390.5137939453125,
      "loss": 0.6692,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": 0.024431603029370308,
      "rewards/margins": 0.05050837993621826,
      "rewards/rejected": -0.026076778769493103,
      "step": 30
    },
    {
      "epoch": 0.08303061754021795,
      "grad_norm": 9.363164050327798,
      "learning_rate": 4.0816326530612243e-07,
      "logits/chosen": -2.70872163772583,
      "logits/rejected": -2.6695823669433594,
      "logps/chosen": -347.7822265625,
      "logps/rejected": -376.8358459472656,
      "loss": 0.6045,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 0.05541626736521721,
      "rewards/margins": 0.19534507393836975,
      "rewards/rejected": -0.13992881774902344,
      "step": 40
    },
    {
      "epoch": 0.10378827192527244,
      "grad_norm": 12.348348618840042,
      "learning_rate": 4.999933894080444e-07,
      "logits/chosen": -2.714632511138916,
      "logits/rejected": -2.6948368549346924,
      "logps/chosen": -402.96624755859375,
      "logps/rejected": -494.892578125,
      "loss": 0.4675,
      "rewards/accuracies": 0.8656250238418579,
      "rewards/chosen": -0.3671617805957794,
      "rewards/margins": 0.6783493161201477,
      "rewards/rejected": -1.04551100730896,
      "step": 50
    },
    {
      "epoch": 0.12454592631032693,
      "grad_norm": 17.768413475155672,
      "learning_rate": 4.992005413014143e-07,
      "logits/chosen": -2.7408804893493652,
      "logits/rejected": -2.736759662628174,
      "logps/chosen": -526.3796997070312,
      "logps/rejected": -740.1470947265625,
      "loss": 0.3505,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": -1.7920459508895874,
      "rewards/margins": 1.7205908298492432,
      "rewards/rejected": -3.512636661529541,
      "step": 60
    },
    {
      "epoch": 0.14530358069538143,
      "grad_norm": 20.91483447209899,
      "learning_rate": 4.970903776169402e-07,
      "logits/chosen": -2.7555935382843018,
      "logits/rejected": -2.731161117553711,
      "logps/chosen": -641.5057373046875,
      "logps/rejected": -879.7139892578125,
      "loss": 0.3023,
      "rewards/accuracies": 0.846875011920929,
      "rewards/chosen": -2.449052333831787,
      "rewards/margins": 2.306373119354248,
      "rewards/rejected": -4.755425453186035,
      "step": 70
    },
    {
      "epoch": 0.1660612350804359,
      "grad_norm": 18.23581038330664,
      "learning_rate": 4.936740530314087e-07,
      "logits/chosen": -2.339261054992676,
      "logits/rejected": -2.113029956817627,
      "logps/chosen": -601.9942626953125,
      "logps/rejected": -914.1705932617188,
      "loss": 0.2471,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -2.1097259521484375,
      "rewards/margins": 3.008650064468384,
      "rewards/rejected": -5.1183762550354,
      "step": 80
    },
    {
      "epoch": 0.1868188894654904,
      "grad_norm": 22.34513227915026,
      "learning_rate": 4.889696268057348e-07,
      "logits/chosen": -1.7139612436294556,
      "logits/rejected": -0.9158167839050293,
      "logps/chosen": -591.7190551757812,
      "logps/rejected": -955.1848754882812,
      "loss": 0.2266,
      "rewards/accuracies": 0.909375011920929,
      "rewards/chosen": -2.2645881175994873,
      "rewards/margins": 3.4328598976135254,
      "rewards/rejected": -5.697447776794434,
      "step": 90
    },
    {
      "epoch": 0.2075765438505449,
      "grad_norm": 15.36420864345506,
      "learning_rate": 4.830019673206996e-07,
      "logits/chosen": -1.521246075630188,
      "logits/rejected": -0.20611262321472168,
      "logps/chosen": -612.281005859375,
      "logps/rejected": -1068.5550537109375,
      "loss": 0.2001,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -2.3677589893341064,
      "rewards/margins": 4.30899715423584,
      "rewards/rejected": -6.676755428314209,
      "step": 100
    },
    {
      "epoch": 0.2283341982355994,
      "grad_norm": 21.617954842362128,
      "learning_rate": 4.7580262061854606e-07,
      "logits/chosen": -1.2374727725982666,
      "logits/rejected": 0.18440499901771545,
      "logps/chosen": -599.9720458984375,
      "logps/rejected": -1089.302001953125,
      "loss": 0.2019,
      "rewards/accuracies": 0.940625011920929,
      "rewards/chosen": -2.2848434448242188,
      "rewards/margins": 4.657064914703369,
      "rewards/rejected": -6.9419074058532715,
      "step": 110
    },
    {
      "epoch": 0.24909185262065386,
      "grad_norm": 14.742905110037716,
      "learning_rate": 4.674096436453447e-07,
      "logits/chosen": -0.7879268527030945,
      "logits/rejected": 0.3564757704734802,
      "logps/chosen": -665.6008911132812,
      "logps/rejected": -1112.988037109375,
      "loss": 0.1942,
      "rewards/accuracies": 0.909375011920929,
      "rewards/chosen": -2.7063541412353516,
      "rewards/margins": 4.290424346923828,
      "rewards/rejected": -6.9967780113220215,
      "step": 120
    },
    {
      "epoch": 0.26984950700570837,
      "grad_norm": 11.973289583871454,
      "learning_rate": 4.578674030756363e-07,
      "logits/chosen": -1.418628454208374,
      "logits/rejected": 0.030070018023252487,
      "logps/chosen": -627.1656494140625,
      "logps/rejected": -1140.0579833984375,
      "loss": 0.1772,
      "rewards/accuracies": 0.890625,
      "rewards/chosen": -2.4757330417633057,
      "rewards/margins": 4.854559898376465,
      "rewards/rejected": -7.330292701721191,
      "step": 130
    },
    {
      "epoch": 0.29060716139076287,
      "grad_norm": 15.153812063781148,
      "learning_rate": 4.4722634078279865e-07,
      "logits/chosen": -1.3328105211257935,
      "logits/rejected": 0.4199509620666504,
      "logps/chosen": -610.9807739257812,
      "logps/rejected": -1104.3985595703125,
      "loss": 0.1988,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -2.4051623344421387,
      "rewards/margins": 4.705986976623535,
      "rewards/rejected": -7.111149787902832,
      "step": 140
    },
    {
      "epoch": 0.3113648157758173,
      "grad_norm": 17.491121999774048,
      "learning_rate": 4.355427071949004e-07,
      "logits/chosen": -1.5610778331756592,
      "logits/rejected": 0.6799826622009277,
      "logps/chosen": -596.3429565429688,
      "logps/rejected": -1119.742431640625,
      "loss": 0.1647,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -2.3613154888153076,
      "rewards/margins": 5.037424564361572,
      "rewards/rejected": -7.398741245269775,
      "step": 150
    },
    {
      "epoch": 0.3321224701608718,
      "grad_norm": 19.441381636430275,
      "learning_rate": 4.228782639455674e-07,
      "logits/chosen": -1.1263091564178467,
      "logits/rejected": 1.3055658340454102,
      "logps/chosen": -651.0065307617188,
      "logps/rejected": -1252.4947509765625,
      "loss": 0.169,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -2.671025276184082,
      "rewards/margins": 5.759725570678711,
      "rewards/rejected": -8.430750846862793,
      "step": 160
    },
    {
      "epoch": 0.3528801245459263,
      "grad_norm": 15.454380512498973,
      "learning_rate": 4.092999573916971e-07,
      "logits/chosen": -0.8963528871536255,
      "logits/rejected": 1.6645536422729492,
      "logps/chosen": -615.8960571289062,
      "logps/rejected": -1188.300048828125,
      "loss": 0.1723,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -2.5705718994140625,
      "rewards/margins": 5.532119274139404,
      "rewards/rejected": -8.102691650390625,
      "step": 170
    },
    {
      "epoch": 0.3736377789309808,
      "grad_norm": 18.19445786127511,
      "learning_rate": 3.948795647238637e-07,
      "logits/chosen": -0.9939937591552734,
      "logits/rejected": 1.5464706420898438,
      "logps/chosen": -629.0621948242188,
      "logps/rejected": -1243.3023681640625,
      "loss": 0.175,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -2.624224901199341,
      "rewards/margins": 5.946156978607178,
      "rewards/rejected": -8.570383071899414,
      "step": 180
    },
    {
      "epoch": 0.39439543331603527,
      "grad_norm": 16.368013896038917,
      "learning_rate": 3.796933145401304e-07,
      "logits/chosen": -1.174119472503662,
      "logits/rejected": 1.0995179414749146,
      "logps/chosen": -662.320556640625,
      "logps/rejected": -1293.4354248046875,
      "loss": 0.1671,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -2.7964344024658203,
      "rewards/margins": 5.946679592132568,
      "rewards/rejected": -8.743115425109863,
      "step": 190
    },
    {
      "epoch": 0.4151530877010898,
      "grad_norm": 14.042046955172085,
      "learning_rate": 3.638214838889801e-07,
      "logits/chosen": -0.7083699107170105,
      "logits/rejected": 1.8419784307479858,
      "logps/chosen": -618.2698364257812,
      "logps/rejected": -1210.1761474609375,
      "loss": 0.1636,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -2.5069775581359863,
      "rewards/margins": 5.630901336669922,
      "rewards/rejected": -8.137879371643066,
      "step": 200
    },
    {
      "epoch": 0.4359107420861443,
      "grad_norm": 16.47603854480112,
      "learning_rate": 3.4734797391146383e-07,
      "logits/chosen": -0.4444239139556885,
      "logits/rejected": 2.3249332904815674,
      "logps/chosen": -619.8510131835938,
      "logps/rejected": -1215.9896240234375,
      "loss": 0.1522,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -2.5075623989105225,
      "rewards/margins": 5.768418312072754,
      "rewards/rejected": -8.275980949401855,
      "step": 210
    },
    {
      "epoch": 0.4566683964711988,
      "grad_norm": 29.46403047682205,
      "learning_rate": 3.3035986632579036e-07,
      "logits/chosen": 0.16329942643642426,
      "logits/rejected": 2.8997693061828613,
      "logps/chosen": -659.2125854492188,
      "logps/rejected": -1382.5194091796875,
      "loss": 0.1615,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -2.939960479736328,
      "rewards/margins": 7.053119659423828,
      "rewards/rejected": -9.993081092834473,
      "step": 220
    },
    {
      "epoch": 0.4774260508562532,
      "grad_norm": 21.839461051416258,
      "learning_rate": 3.1294696309885716e-07,
      "logits/chosen": 0.3789651095867157,
      "logits/rejected": 3.142848253250122,
      "logps/chosen": -676.8601684570312,
      "logps/rejected": -1436.1219482421875,
      "loss": 0.1628,
      "rewards/accuracies": 0.9156249761581421,
      "rewards/chosen": -3.1580305099487305,
      "rewards/margins": 7.436465263366699,
      "rewards/rejected": -10.59449577331543,
      "step": 230
    },
    {
      "epoch": 0.49818370524130773,
      "grad_norm": 16.48504711480412,
      "learning_rate": 2.952013117380913e-07,
      "logits/chosen": -0.004539084620773792,
      "logits/rejected": 3.229100465774536,
      "logps/chosen": -651.2918701171875,
      "logps/rejected": -1440.683837890625,
      "loss": 0.1534,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -2.9313418865203857,
      "rewards/margins": 7.655765533447266,
      "rewards/rejected": -10.587106704711914,
      "step": 240
    },
    {
      "epoch": 0.5189413596263622,
      "grad_norm": 18.156585865508323,
      "learning_rate": 2.7721671871299114e-07,
      "logits/chosen": -0.13780581951141357,
      "logits/rejected": 2.874418258666992,
      "logps/chosen": -682.3836669921875,
      "logps/rejected": -1423.32861328125,
      "loss": 0.1577,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -3.025876522064209,
      "rewards/margins": 7.198524475097656,
      "rewards/rejected": -10.224401473999023,
      "step": 250
    },
    {
      "epoch": 0.5396990140114167,
      "grad_norm": 17.237094148806953,
      "learning_rate": 2.5908825357849993e-07,
      "logits/chosen": -0.7950791716575623,
      "logits/rejected": 2.015465497970581,
      "logps/chosen": -645.5744018554688,
      "logps/rejected": -1297.5784912109375,
      "loss": 0.1566,
      "rewards/accuracies": 0.9156249761581421,
      "rewards/chosen": -2.724034309387207,
      "rewards/margins": 6.29718542098999,
      "rewards/rejected": -9.021219253540039,
      "step": 260
    },
    {
      "epoch": 0.5604566683964712,
      "grad_norm": 17.46783320885481,
      "learning_rate": 2.409117464215001e-07,
      "logits/chosen": 0.14433838427066803,
      "logits/rejected": 3.1040682792663574,
      "logps/chosen": -684.5079345703125,
      "logps/rejected": -1501.366943359375,
      "loss": 0.1413,
      "rewards/accuracies": 0.9468749761581421,
      "rewards/chosen": -3.3087615966796875,
      "rewards/margins": 7.854517936706543,
      "rewards/rejected": -11.16327953338623,
      "step": 270
    },
    {
      "epoch": 0.5812143227815257,
      "grad_norm": 17.63149610037019,
      "learning_rate": 2.227832812870089e-07,
      "logits/chosen": 0.3872329294681549,
      "logits/rejected": 3.5623180866241455,
      "logps/chosen": -775.79736328125,
      "logps/rejected": -1631.552490234375,
      "loss": 0.1516,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -3.998131275177002,
      "rewards/margins": 8.329412460327148,
      "rewards/rejected": -12.327543258666992,
      "step": 280
    },
    {
      "epoch": 0.6019719771665801,
      "grad_norm": 21.86816160236415,
      "learning_rate": 2.0479868826190871e-07,
      "logits/chosen": -0.06945541501045227,
      "logits/rejected": 2.7963638305664062,
      "logps/chosen": -767.2408447265625,
      "logps/rejected": -1517.348876953125,
      "loss": 0.1596,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.9247982501983643,
      "rewards/margins": 7.42672872543335,
      "rewards/rejected": -11.351527214050293,
      "step": 290
    },
    {
      "epoch": 0.6227296315516346,
      "grad_norm": 14.515183994248952,
      "learning_rate": 1.8705303690114287e-07,
      "logits/chosen": 0.07333996891975403,
      "logits/rejected": 2.5020909309387207,
      "logps/chosen": -785.2969360351562,
      "logps/rejected": -1529.5716552734375,
      "loss": 0.1421,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -4.092496395111084,
      "rewards/margins": 7.192312717437744,
      "rewards/rejected": -11.284809112548828,
      "step": 300
    },
    {
      "epoch": 0.6434872859366891,
      "grad_norm": 17.979677522182545,
      "learning_rate": 1.6964013367420965e-07,
      "logits/chosen": -0.13572996854782104,
      "logits/rejected": 2.765768527984619,
      "logps/chosen": -797.8541870117188,
      "logps/rejected": -1586.290771484375,
      "loss": 0.1457,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -4.280360221862793,
      "rewards/margins": 7.809360504150391,
      "rewards/rejected": -12.0897216796875,
      "step": 310
    },
    {
      "epoch": 0.6642449403217436,
      "grad_norm": 12.321699055408487,
      "learning_rate": 1.5265202608853628e-07,
      "logits/chosen": -0.27073314785957336,
      "logits/rejected": 2.6812171936035156,
      "logps/chosen": -769.6759643554688,
      "logps/rejected": -1649.497802734375,
      "loss": 0.1453,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -4.045531272888184,
      "rewards/margins": 8.580347061157227,
      "rewards/rejected": -12.625877380371094,
      "step": 320
    },
    {
      "epoch": 0.6850025947067981,
      "grad_norm": 11.097171754192187,
      "learning_rate": 1.3617851611101993e-07,
      "logits/chosen": -0.5093133449554443,
      "logits/rejected": 2.036428928375244,
      "logps/chosen": -759.6971435546875,
      "logps/rejected": -1584.39892578125,
      "loss": 0.1446,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -3.8719277381896973,
      "rewards/margins": 7.951642036437988,
      "rewards/rejected": -11.823568344116211,
      "step": 330
    },
    {
      "epoch": 0.7057602490918526,
      "grad_norm": 14.623856678041504,
      "learning_rate": 1.2030668545986958e-07,
      "logits/chosen": -0.3804703652858734,
      "logits/rejected": 2.350010871887207,
      "logps/chosen": -824.4983520507812,
      "logps/rejected": -1692.0347900390625,
      "loss": 0.137,
      "rewards/accuracies": 0.940625011920929,
      "rewards/chosen": -4.51037073135376,
      "rewards/margins": 8.377652168273926,
      "rewards/rejected": -12.888023376464844,
      "step": 340
    },
    {
      "epoch": 0.7265179034769071,
      "grad_norm": 17.122700499170215,
      "learning_rate": 1.0512043527613623e-07,
      "logits/chosen": -0.2618183493614197,
      "logits/rejected": 2.3642690181732178,
      "logps/chosen": -845.6104736328125,
      "logps/rejected": -1714.4976806640625,
      "loss": 0.1494,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -4.845407009124756,
      "rewards/margins": 8.386329650878906,
      "rewards/rejected": -13.23173713684082,
      "step": 350
    },
    {
      "epoch": 0.7472755578619616,
      "grad_norm": 19.444982747452297,
      "learning_rate": 9.070004260830294e-08,
      "logits/chosen": -0.7990102767944336,
      "logits/rejected": 2.3197755813598633,
      "logps/chosen": -802.9656982421875,
      "logps/rejected": -1649.383544921875,
      "loss": 0.1393,
      "rewards/accuracies": 0.934374988079071,
      "rewards/chosen": -4.384912490844727,
      "rewards/margins": 8.375471115112305,
      "rewards/rejected": -12.760382652282715,
      "step": 360
    },
    {
      "epoch": 0.768033212247016,
      "grad_norm": 17.93208817126597,
      "learning_rate": 7.712173605443267e-08,
      "logits/chosen": -0.9170758128166199,
      "logits/rejected": 2.2540900707244873,
      "logps/chosen": -794.91015625,
      "logps/rejected": -1676.653564453125,
      "loss": 0.1463,
      "rewards/accuracies": 0.9156249761581421,
      "rewards/chosen": -4.2809953689575195,
      "rewards/margins": 8.688231468200684,
      "rewards/rejected": -12.96922779083252,
      "step": 370
    },
    {
      "epoch": 0.7887908666320705,
      "grad_norm": 11.992323551908767,
      "learning_rate": 6.445729280509957e-08,
      "logits/chosen": -0.5615069270133972,
      "logits/rejected": 2.490211009979248,
      "logps/chosen": -826.1986083984375,
      "logps/rejected": -1815.263916015625,
      "loss": 0.1361,
      "rewards/accuracies": 0.9281250238418579,
      "rewards/chosen": -4.737757682800293,
      "rewards/margins": 9.302997589111328,
      "rewards/rejected": -14.040754318237305,
      "step": 380
    },
    {
      "epoch": 0.809548521017125,
      "grad_norm": 14.867121986591556,
      "learning_rate": 5.2773659217201364e-08,
      "logits/chosen": -0.4665653109550476,
      "logits/rejected": 2.803576946258545,
      "logps/chosen": -860.3473510742188,
      "logps/rejected": -1715.7301025390625,
      "loss": 0.1459,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -4.846836566925049,
      "rewards/margins": 8.348875045776367,
      "rewards/rejected": -13.195712089538574,
      "step": 390
    },
    {
      "epoch": 0.8303061754021795,
      "grad_norm": 14.137593981426674,
      "learning_rate": 4.213259692436366e-08,
      "logits/chosen": -0.40155744552612305,
      "logits/rejected": 2.635136604309082,
      "logps/chosen": -858.0413818359375,
      "logps/rejected": -1749.977294921875,
      "loss": 0.1388,
      "rewards/accuracies": 0.940625011920929,
      "rewards/chosen": -4.928585529327393,
      "rewards/margins": 8.627421379089355,
      "rewards/rejected": -13.556007385253906,
      "step": 400
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 18.46721900012293,
      "learning_rate": 3.259035635465529e-08,
      "logits/chosen": -0.4056881070137024,
      "logits/rejected": 2.437743902206421,
      "logps/chosen": -871.8626098632812,
      "logps/rejected": -1658.709716796875,
      "loss": 0.1416,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -5.064044952392578,
      "rewards/margins": 7.721222877502441,
      "rewards/rejected": -12.785266876220703,
      "step": 410
    },
    {
      "epoch": 0.8718214841722886,
      "grad_norm": 18.74063979970814,
      "learning_rate": 2.4197379381453942e-08,
      "logits/chosen": -0.33163413405418396,
      "logits/rejected": 2.5784339904785156,
      "logps/chosen": -880.70654296875,
      "logps/rejected": -1740.9976806640625,
      "loss": 0.1443,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -5.066455364227295,
      "rewards/margins": 8.281970024108887,
      "rewards/rejected": -13.348426818847656,
      "step": 420
    },
    {
      "epoch": 0.892579138557343,
      "grad_norm": 17.16551956889699,
      "learning_rate": 1.699803267930039e-08,
      "logits/chosen": -0.3569720387458801,
      "logits/rejected": 2.325591564178467,
      "logps/chosen": -865.0040283203125,
      "logps/rejected": -1691.733642578125,
      "loss": 0.1333,
      "rewards/accuracies": 0.921875,
      "rewards/chosen": -4.9201178550720215,
      "rewards/margins": 8.041346549987793,
      "rewards/rejected": -12.961465835571289,
      "step": 430
    },
    {
      "epoch": 0.9133367929423976,
      "grad_norm": 15.033949759220667,
      "learning_rate": 1.1030373194265114e-08,
      "logits/chosen": -0.3576432466506958,
      "logits/rejected": 2.751772403717041,
      "logps/chosen": -866.8192138671875,
      "logps/rejected": -1763.2193603515625,
      "loss": 0.1491,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -4.9160566329956055,
      "rewards/margins": 8.817602157592773,
      "rewards/rejected": -13.733657836914062,
      "step": 440
    },
    {
      "epoch": 0.934094447327452,
      "grad_norm": 17.88061807294248,
      "learning_rate": 6.325946968591317e-09,
      "logits/chosen": -0.2799724042415619,
      "logits/rejected": 2.8802428245544434,
      "logps/chosen": -871.8025512695312,
      "logps/rejected": -1729.011962890625,
      "loss": 0.1411,
      "rewards/accuracies": 0.9281250238418579,
      "rewards/chosen": -5.046684265136719,
      "rewards/margins": 8.479043960571289,
      "rewards/rejected": -13.525728225708008,
      "step": 450
    },
    {
      "epoch": 0.9548521017125065,
      "grad_norm": 16.510600900187168,
      "learning_rate": 2.909622383059834e-09,
      "logits/chosen": -0.2751966118812561,
      "logits/rejected": 2.7767112255096436,
      "logps/chosen": -868.7591552734375,
      "logps/rejected": -1734.588134765625,
      "loss": 0.1428,
      "rewards/accuracies": 0.940625011920929,
      "rewards/chosen": -4.989295959472656,
      "rewards/margins": 8.519960403442383,
      "rewards/rejected": -13.509256362915039,
      "step": 460
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 19.07149083473138,
      "learning_rate": 7.994586985856089e-10,
      "logits/chosen": -0.4134143888950348,
      "logits/rejected": 2.6374351978302,
      "logps/chosen": -877.0185546875,
      "logps/rejected": -1758.679443359375,
      "loss": 0.1319,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -5.058596611022949,
      "rewards/margins": 8.577057838439941,
      "rewards/rejected": -13.635653495788574,
      "step": 470
    },
    {
      "epoch": 0.9963674104826155,
      "grad_norm": 12.263362857750089,
      "learning_rate": 6.610591955641398e-12,
      "logits/chosen": -0.18534007668495178,
      "logits/rejected": 2.617680311203003,
      "logps/chosen": -867.3248291015625,
      "logps/rejected": -1753.2431640625,
      "loss": 0.1436,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -5.15399169921875,
      "rewards/margins": 8.56091594696045,
      "rewards/rejected": -13.7149076461792,
      "step": 480
    },
    {
      "epoch": 0.9984431759211209,
      "step": 481,
      "total_flos": 0.0,
      "train_loss": 0.21509297628908297,
      "train_runtime": 13791.8998,
      "train_samples_per_second": 8.941,
      "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
}