{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9993998799759952,
  "eval_steps": 100,
  "global_step": 1249,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4e-08,
      "logits/chosen": -2.602003574371338,
      "logits/rejected": -2.6217830181121826,
      "logps/chosen": -260.71484375,
      "logps/rejected": -256.2999267578125,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.0000000000000003e-07,
      "logits/chosen": -2.5351831912994385,
      "logits/rejected": -2.56242036819458,
      "logps/chosen": -287.53594970703125,
      "logps/rejected": -308.8571472167969,
      "loss": 0.693,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.0010520940413698554,
      "rewards/margins": 0.0002557992411311716,
      "rewards/rejected": -0.0013078933116048574,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.000000000000001e-07,
      "logits/chosen": -2.5120716094970703,
      "logits/rejected": -2.515523672103882,
      "logps/chosen": -255.38192749023438,
      "logps/rejected": -273.31805419921875,
      "loss": 0.6934,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.0002814549079630524,
      "rewards/margins": -0.0005342562217265368,
      "rewards/rejected": 0.0002528014301788062,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2000000000000002e-06,
      "logits/chosen": -2.498893976211548,
      "logits/rejected": -2.5145130157470703,
      "logps/chosen": -276.9407958984375,
      "logps/rejected": -295.25482177734375,
      "loss": 0.693,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.0016518374904990196,
      "rewards/margins": 0.00022423597692977637,
      "rewards/rejected": -0.0018760733073577285,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6000000000000001e-06,
      "logits/chosen": -2.5344443321228027,
      "logits/rejected": -2.541226625442505,
      "logps/chosen": -251.64730834960938,
      "logps/rejected": -259.96575927734375,
      "loss": 0.6935,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.005051327869296074,
      "rewards/margins": -0.0006530196405947208,
      "rewards/rejected": -0.004398307763040066,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0000000000000003e-06,
      "logits/chosen": -2.479356288909912,
      "logits/rejected": -2.461836576461792,
      "logps/chosen": -251.314697265625,
      "logps/rejected": -256.02020263671875,
      "loss": 0.6931,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.007234510965645313,
      "rewards/margins": 0.00020712481637019664,
      "rewards/rejected": -0.0074416352435946465,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4000000000000003e-06,
      "logits/chosen": -2.5456085205078125,
      "logits/rejected": -2.5527470111846924,
      "logps/chosen": -262.48858642578125,
      "logps/rejected": -280.4261474609375,
      "loss": 0.6927,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.011781791225075722,
      "rewards/margins": 0.0009017615811899304,
      "rewards/rejected": -0.012683553621172905,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8000000000000003e-06,
      "logits/chosen": -2.545548915863037,
      "logits/rejected": -2.5333781242370605,
      "logps/chosen": -266.43798828125,
      "logps/rejected": -271.582763671875,
      "loss": 0.6935,
      "rewards/accuracies": 0.4437499940395355,
      "rewards/chosen": -0.015757283195853233,
      "rewards/margins": -0.0007508035632781684,
      "rewards/rejected": -0.015006482601165771,
      "step": 70
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.2000000000000003e-06,
      "logits/chosen": -2.5648863315582275,
      "logits/rejected": -2.5693516731262207,
      "logps/chosen": -262.5594787597656,
      "logps/rejected": -282.7106018066406,
      "loss": 0.6925,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.014923855662345886,
      "rewards/margins": 0.0014337189495563507,
      "rewards/rejected": -0.016357574611902237,
      "step": 80
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6000000000000003e-06,
      "logits/chosen": -2.565471649169922,
      "logits/rejected": -2.5509932041168213,
      "logps/chosen": -295.49615478515625,
      "logps/rejected": -300.31103515625,
      "loss": 0.6911,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.020387709140777588,
      "rewards/margins": 0.004231990315020084,
      "rewards/rejected": -0.024619702249765396,
      "step": 90
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -2.5530660152435303,
      "logits/rejected": -2.553136110305786,
      "logps/chosen": -263.5738220214844,
      "logps/rejected": -276.2574157714844,
      "loss": 0.6916,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.02410706877708435,
      "rewards/margins": 0.003276412608101964,
      "rewards/rejected": -0.027383480221033096,
      "step": 100
    },
    {
      "epoch": 0.08,
      "eval_logits/chosen": -2.508636713027954,
      "eval_logits/rejected": -2.5057852268218994,
      "eval_logps/chosen": -210.22964477539062,
      "eval_logps/rejected": -215.71873474121094,
      "eval_loss": 0.6924654245376587,
      "eval_rewards/accuracies": 0.527999997138977,
      "eval_rewards/chosen": -0.01618967391550541,
      "eval_rewards/margins": 0.0015068231150507927,
      "eval_rewards/rejected": -0.017696497961878777,
      "eval_runtime": 1581.5421,
      "eval_samples_per_second": 1.893,
      "eval_steps_per_second": 0.237,
      "step": 100
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.4e-06,
      "logits/chosen": -2.55187726020813,
      "logits/rejected": -2.5549445152282715,
      "logps/chosen": -283.41143798828125,
      "logps/rejected": -293.6915588378906,
      "loss": 0.6912,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.026097621768712997,
      "rewards/margins": 0.004130261484533548,
      "rewards/rejected": -0.030227884650230408,
      "step": 110
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.800000000000001e-06,
      "logits/chosen": -2.561483860015869,
      "logits/rejected": -2.575721263885498,
      "logps/chosen": -265.60711669921875,
      "logps/rejected": -282.2021789550781,
      "loss": 0.691,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.030240505933761597,
      "rewards/margins": 0.004798956215381622,
      "rewards/rejected": -0.03503946214914322,
      "step": 120
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.999755876225375e-06,
      "logits/chosen": -2.5272843837738037,
      "logits/rejected": -2.511995792388916,
      "logps/chosen": -279.7336120605469,
      "logps/rejected": -298.2309875488281,
      "loss": 0.6886,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.03772541880607605,
      "rewards/margins": 0.009567076340317726,
      "rewards/rejected": -0.047292497009038925,
      "step": 130
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.997803172081864e-06,
      "logits/chosen": -2.549031972885132,
      "logits/rejected": -2.550044536590576,
      "logps/chosen": -274.913818359375,
      "logps/rejected": -287.93389892578125,
      "loss": 0.6866,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.0441165529191494,
      "rewards/margins": 0.01394923496991396,
      "rewards/rejected": -0.058065783232450485,
      "step": 140
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9938992891651825e-06,
      "logits/chosen": -2.511295795440674,
      "logits/rejected": -2.5027806758880615,
      "logps/chosen": -263.2046203613281,
      "logps/rejected": -284.345703125,
      "loss": 0.6827,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.05038716644048691,
      "rewards/margins": 0.021908113732933998,
      "rewards/rejected": -0.07229527831077576,
      "step": 150
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.988047277024456e-06,
      "logits/chosen": -2.551429271697998,
      "logits/rejected": -2.5654749870300293,
      "logps/chosen": -274.80670166015625,
      "logps/rejected": -288.6151428222656,
      "loss": 0.682,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.07800472527742386,
      "rewards/margins": 0.023449204862117767,
      "rewards/rejected": -0.10145392268896103,
      "step": 160
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.980251707005417e-06,
      "logits/chosen": -2.535489320755005,
      "logits/rejected": -2.511200428009033,
      "logps/chosen": -291.24853515625,
      "logps/rejected": -298.6773376464844,
      "loss": 0.6896,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.10200835764408112,
      "rewards/margins": 0.008326217532157898,
      "rewards/rejected": -0.11033459007740021,
      "step": 170
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.970518668679459e-06,
      "logits/chosen": -2.540640354156494,
      "logits/rejected": -2.5266213417053223,
      "logps/chosen": -289.3337707519531,
      "logps/rejected": -295.79278564453125,
      "loss": 0.6841,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.09090762585401535,
      "rewards/margins": 0.020125988870859146,
      "rewards/rejected": -0.1110336035490036,
      "step": 180
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.958855765086722e-06,
      "logits/chosen": -2.563791036605835,
      "logits/rejected": -2.562145233154297,
      "logps/chosen": -269.8388671875,
      "logps/rejected": -280.38037109375,
      "loss": 0.6853,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.05399826169013977,
      "rewards/margins": 0.017181050032377243,
      "rewards/rejected": -0.07117930799722672,
      "step": 190
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.945272106796919e-06,
      "logits/chosen": -2.5541298389434814,
      "logits/rejected": -2.564225673675537,
      "logps/chosen": -269.95001220703125,
      "logps/rejected": -285.06451416015625,
      "loss": 0.6855,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.06484891474246979,
      "rewards/margins": 0.017776602879166603,
      "rewards/rejected": -0.08262552320957184,
      "step": 200
    },
    {
      "epoch": 0.16,
      "eval_logits/chosen": -2.517770528793335,
      "eval_logits/rejected": -2.515197277069092,
      "eval_logps/chosen": -215.1240234375,
      "eval_logps/rejected": -221.67100524902344,
      "eval_loss": 0.6880091428756714,
      "eval_rewards/accuracies": 0.5613333582878113,
      "eval_rewards/chosen": -0.06513328850269318,
      "eval_rewards/margins": 0.012086095288395882,
      "eval_rewards/rejected": -0.07721938192844391,
      "eval_runtime": 1599.371,
      "eval_samples_per_second": 1.872,
      "eval_steps_per_second": 0.234,
      "step": 200
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.929778304792537e-06,
      "logits/chosen": -2.5422654151916504,
      "logits/rejected": -2.5487327575683594,
      "logps/chosen": -294.87188720703125,
      "logps/rejected": -299.1329040527344,
      "loss": 0.6792,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.08577840030193329,
      "rewards/margins": 0.03281060978770256,
      "rewards/rejected": -0.11858902126550674,
      "step": 210
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.912386462179987e-06,
      "logits/chosen": -2.56231951713562,
      "logits/rejected": -2.559842348098755,
      "logps/chosen": -283.178955078125,
      "logps/rejected": -308.49322509765625,
      "loss": 0.6772,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.09053866565227509,
      "rewards/margins": 0.03568854182958603,
      "rewards/rejected": -0.1262272149324417,
      "step": 220
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.893110164735167e-06,
      "logits/chosen": -2.5928561687469482,
      "logits/rejected": -2.5934693813323975,
      "logps/chosen": -288.0644226074219,
      "logps/rejected": -301.1351013183594,
      "loss": 0.6831,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.11280493438243866,
      "rewards/margins": 0.023493468761444092,
      "rewards/rejected": -0.13629841804504395,
      "step": 230
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.871964470290823e-06,
      "logits/chosen": -2.532960891723633,
      "logits/rejected": -2.55849027633667,
      "logps/chosen": -289.32354736328125,
      "logps/rejected": -309.2322692871094,
      "loss": 0.6735,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.11586654186248779,
      "rewards/margins": 0.04449574649333954,
      "rewards/rejected": -0.16036227345466614,
      "step": 240
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.848965896974006e-06,
      "logits/chosen": -2.541764259338379,
      "logits/rejected": -2.545576572418213,
      "logps/chosen": -281.2950439453125,
      "logps/rejected": -300.70416259765625,
      "loss": 0.6803,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.11654965579509735,
      "rewards/margins": 0.02992970123887062,
      "rewards/rejected": -0.14647935330867767,
      "step": 250
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8241324103028055e-06,
      "logits/chosen": -2.5795302391052246,
      "logits/rejected": -2.5413551330566406,
      "logps/chosen": -287.0888977050781,
      "logps/rejected": -298.43524169921875,
      "loss": 0.6762,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.1340574324131012,
      "rewards/margins": 0.03809996321797371,
      "rewards/rejected": -0.1721573919057846,
      "step": 260
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.797483409152438e-06,
      "logits/chosen": -2.565472364425659,
      "logits/rejected": -2.569817066192627,
      "logps/chosen": -276.9258728027344,
      "logps/rejected": -297.29931640625,
      "loss": 0.668,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.14538586139678955,
      "rewards/margins": 0.05918685719370842,
      "rewards/rejected": -0.20457272231578827,
      "step": 270
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.769039710601669e-06,
      "logits/chosen": -2.5913045406341553,
      "logits/rejected": -2.592714786529541,
      "logps/chosen": -277.0105285644531,
      "logps/rejected": -293.9090576171875,
      "loss": 0.678,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.16800367832183838,
      "rewards/margins": 0.03504541888833046,
      "rewards/rejected": -0.20304909348487854,
      "step": 280
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.738823533671383e-06,
      "logits/chosen": -2.59189510345459,
      "logits/rejected": -2.5966787338256836,
      "logps/chosen": -300.8684387207031,
      "logps/rejected": -316.1703186035156,
      "loss": 0.677,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.18113958835601807,
      "rewards/margins": 0.0392177477478981,
      "rewards/rejected": -0.22035734355449677,
      "step": 290
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.706858481968017e-06,
      "logits/chosen": -2.5426743030548096,
      "logits/rejected": -2.546731472015381,
      "logps/chosen": -288.5105895996094,
      "logps/rejected": -297.8955993652344,
      "loss": 0.6825,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.22026526927947998,
      "rewards/margins": 0.027653247117996216,
      "rewards/rejected": -0.2479185312986374,
      "step": 300
    },
    {
      "epoch": 0.24,
      "eval_logits/chosen": -2.519216537475586,
      "eval_logits/rejected": -2.517465591430664,
      "eval_logps/chosen": -227.3457489013672,
      "eval_logps/rejected": -234.75460815429688,
      "eval_loss": 0.6853793859481812,
      "eval_rewards/accuracies": 0.5473333597183228,
      "eval_rewards/chosen": -0.18735063076019287,
      "eval_rewards/margins": 0.02070464752614498,
      "eval_rewards/rejected": -0.2080552726984024,
      "eval_runtime": 1584.577,
      "eval_samples_per_second": 1.889,
      "eval_steps_per_second": 0.237,
      "step": 300
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.673169525245416e-06,
      "logits/chosen": -2.534403085708618,
      "logits/rejected": -2.514641284942627,
      "logps/chosen": -286.3177185058594,
      "logps/rejected": -313.84521484375,
      "loss": 0.6698,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.19986188411712646,
      "rewards/margins": 0.05483303219079971,
      "rewards/rejected": -0.2546949088573456,
      "step": 310
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.63778297989952e-06,
      "logits/chosen": -2.5624048709869385,
      "logits/rejected": -2.54682993888855,
      "logps/chosen": -292.1237487792969,
      "logps/rejected": -307.5107116699219,
      "loss": 0.6719,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2259463369846344,
      "rewards/margins": 0.05093250423669815,
      "rewards/rejected": -0.27687883377075195,
      "step": 320
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.60072648841109e-06,
      "logits/chosen": -2.540750741958618,
      "logits/rejected": -2.536940574645996,
      "logps/chosen": -302.9012756347656,
      "logps/rejected": -322.0403747558594,
      "loss": 0.6561,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.25178760290145874,
      "rewards/margins": 0.08651792258024216,
      "rewards/rejected": -0.3383055329322815,
      "step": 330
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.562028997752574e-06,
      "logits/chosen": -2.5020062923431396,
      "logits/rejected": -2.525498867034912,
      "logps/chosen": -299.34295654296875,
      "logps/rejected": -324.51239013671875,
      "loss": 0.6697,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.22524788975715637,
      "rewards/margins": 0.05736035853624344,
      "rewards/rejected": -0.2826082408428192,
      "step": 340
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.521720736775947e-06,
      "logits/chosen": -2.551377534866333,
      "logits/rejected": -2.549631118774414,
      "logps/chosen": -312.30010986328125,
      "logps/rejected": -318.33203125,
      "loss": 0.6713,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.31575754284858704,
      "rewards/margins": 0.05324278026819229,
      "rewards/rejected": -0.36900031566619873,
      "step": 350
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.479833192599198e-06,
      "logits/chosen": -2.6122488975524902,
      "logits/rejected": -2.597867488861084,
      "logps/chosen": -310.14825439453125,
      "logps/rejected": -329.4725646972656,
      "loss": 0.6601,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.26400071382522583,
      "rewards/margins": 0.07782687991857529,
      "rewards/rejected": -0.3418275713920593,
      "step": 360
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.436399086009928e-06,
      "logits/chosen": -2.592072010040283,
      "logits/rejected": -2.582606554031372,
      "logps/chosen": -298.0146789550781,
      "logps/rejected": -312.88323974609375,
      "loss": 0.6639,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.24674613773822784,
      "rewards/margins": 0.06771402060985565,
      "rewards/rejected": -0.3144601583480835,
      "step": 370
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.391452345905239e-06,
      "logits/chosen": -2.501129150390625,
      "logits/rejected": -2.5144436359405518,
      "logps/chosen": -302.98785400390625,
      "logps/rejected": -315.341552734375,
      "loss": 0.6643,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2700803875923157,
      "rewards/margins": 0.07336559146642685,
      "rewards/rejected": -0.3434459865093231,
      "step": 380
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3450280827879125e-06,
      "logits/chosen": -2.5604705810546875,
      "logits/rejected": -2.5804944038391113,
      "logps/chosen": -301.7215881347656,
      "logps/rejected": -316.622314453125,
      "loss": 0.6673,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.2707824110984802,
      "rewards/margins": 0.06462465226650238,
      "rewards/rejected": -0.3354070484638214,
      "step": 390
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297162561339554e-06,
      "logits/chosen": -2.5028789043426514,
      "logits/rejected": -2.5059926509857178,
      "logps/chosen": -308.93524169921875,
      "logps/rejected": -328.21649169921875,
      "loss": 0.6676,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.30648913979530334,
      "rewards/margins": 0.0709182545542717,
      "rewards/rejected": -0.37740740180015564,
      "step": 400
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -2.536813497543335,
      "eval_logits/rejected": -2.534651756286621,
      "eval_logps/chosen": -237.7042236328125,
      "eval_logps/rejected": -246.16815185546875,
      "eval_loss": 0.68265300989151,
      "eval_rewards/accuracies": 0.5476666688919067,
      "eval_rewards/chosen": -0.2909356355667114,
      "eval_rewards/margins": 0.031255174428224564,
      "eval_rewards/rejected": -0.3221907913684845,
      "eval_runtime": 1589.7925,
      "eval_samples_per_second": 1.883,
      "eval_steps_per_second": 0.236,
      "step": 400
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.247893172092157e-06,
      "logits/chosen": -2.531207799911499,
      "logits/rejected": -2.548520565032959,
      "logps/chosen": -297.398681640625,
      "logps/rejected": -326.1677551269531,
      "loss": 0.6617,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.313217431306839,
      "rewards/margins": 0.08246447145938873,
      "rewards/rejected": -0.3956819176673889,
      "step": 410
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.197258402220187e-06,
      "logits/chosen": -2.5356483459472656,
      "logits/rejected": -2.544675350189209,
      "logps/chosen": -305.0682373046875,
      "logps/rejected": -318.51068115234375,
      "loss": 0.6603,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.31003984808921814,
      "rewards/margins": 0.08686444163322449,
      "rewards/rejected": -0.39690425992012024,
      "step": 420
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.145297805476023e-06,
      "logits/chosen": -2.5800435543060303,
      "logits/rejected": -2.580294132232666,
      "logps/chosen": -302.9384765625,
      "logps/rejected": -323.58026123046875,
      "loss": 0.6537,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.31787100434303284,
      "rewards/margins": 0.09609754383563995,
      "rewards/rejected": -0.413968563079834,
      "step": 430
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.092051971292228e-06,
      "logits/chosen": -2.5852980613708496,
      "logits/rejected": -2.595292806625366,
      "logps/chosen": -300.8592224121094,
      "logps/rejected": -319.3016662597656,
      "loss": 0.6657,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.3366487920284271,
      "rewards/margins": 0.07542940974235535,
      "rewards/rejected": -0.41207820177078247,
      "step": 440
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.037562493074792e-06,
      "logits/chosen": -2.5786662101745605,
      "logits/rejected": -2.5908896923065186,
      "logps/chosen": -317.9658203125,
      "logps/rejected": -327.48614501953125,
      "loss": 0.6589,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.33508211374282837,
      "rewards/margins": 0.08991469442844391,
      "rewards/rejected": -0.4249967634677887,
      "step": 450
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.981871935712112e-06,
      "logits/chosen": -2.6687827110290527,
      "logits/rejected": -2.6695172786712646,
      "logps/chosen": -296.9199523925781,
      "logps/rejected": -318.85943603515625,
      "loss": 0.666,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.2961326241493225,
      "rewards/margins": 0.07453998178243637,
      "rewards/rejected": -0.3706726133823395,
      "step": 460
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.925023802325094e-06,
      "logits/chosen": -2.6613316535949707,
      "logits/rejected": -2.6546504497528076,
      "logps/chosen": -305.7056884765625,
      "logps/rejected": -325.8073425292969,
      "loss": 0.6564,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.2978198826313019,
      "rewards/margins": 0.09947486966848373,
      "rewards/rejected": -0.3972947299480438,
      "step": 470
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.867062500284342e-06,
      "logits/chosen": -2.6371190547943115,
      "logits/rejected": -2.631121873855591,
      "logps/chosen": -297.20257568359375,
      "logps/rejected": -324.0657653808594,
      "loss": 0.6543,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.36376848816871643,
      "rewards/margins": 0.09993237257003784,
      "rewards/rejected": -0.4637008607387543,
      "step": 480
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8080333065209885e-06,
      "logits/chosen": -2.6479365825653076,
      "logits/rejected": -2.646528720855713,
      "logps/chosen": -306.7449035644531,
      "logps/rejected": -303.7771301269531,
      "loss": 0.6912,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.41228657960891724,
      "rewards/margins": 0.029567983001470566,
      "rewards/rejected": -0.4418545663356781,
      "step": 490
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7479823321582624e-06,
      "logits/chosen": -2.632742404937744,
      "logits/rejected": -2.633615016937256,
      "logps/chosen": -306.31195068359375,
      "logps/rejected": -336.17388916015625,
      "loss": 0.6458,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.34735265374183655,
      "rewards/margins": 0.11751687526702881,
      "rewards/rejected": -0.46486955881118774,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_logits/chosen": -2.6363847255706787,
      "eval_logits/rejected": -2.6328208446502686,
      "eval_logps/chosen": -245.54345703125,
      "eval_logps/rejected": -254.98521423339844,
      "eval_loss": 0.6804653406143188,
      "eval_rewards/accuracies": 0.5566666722297668,
      "eval_rewards/chosen": -0.3693278133869171,
      "eval_rewards/margins": 0.0410337969660759,
      "eval_rewards/rejected": -0.41036155819892883,
      "eval_runtime": 1591.6944,
      "eval_samples_per_second": 1.881,
      "eval_steps_per_second": 0.236,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.686956486491419e-06,
      "logits/chosen": -2.6169426441192627,
      "logits/rejected": -2.6043953895568848,
      "logps/chosen": -318.00299072265625,
      "logps/rejected": -336.43048095703125,
      "loss": 0.6546,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.37818408012390137,
      "rewards/margins": 0.10718680918216705,
      "rewards/rejected": -0.485370934009552,
      "step": 510
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.625003440344166e-06,
      "logits/chosen": -2.638087034225464,
      "logits/rejected": -2.6651127338409424,
      "logps/chosen": -298.8196105957031,
      "logps/rejected": -309.6526794433594,
      "loss": 0.6669,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.3525494933128357,
      "rewards/margins": 0.06792327761650085,
      "rewards/rejected": -0.42047280073165894,
      "step": 520
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.562171588830231e-06,
      "logits/chosen": -2.682291269302368,
      "logits/rejected": -2.690966844558716,
      "logps/chosen": -308.031005859375,
      "logps/rejected": -333.34600830078125,
      "loss": 0.664,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.3355502486228943,
      "rewards/margins": 0.08054343611001968,
      "rewards/rejected": -0.416093647480011,
      "step": 530
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.4985100135491245e-06,
      "logits/chosen": -2.6738035678863525,
      "logits/rejected": -2.665532112121582,
      "logps/chosen": -317.4346008300781,
      "logps/rejected": -350.239990234375,
      "loss": 0.6401,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.3944160044193268,
      "rewards/margins": 0.13535484671592712,
      "rewards/rejected": -0.5297707915306091,
      "step": 540
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4340684442456673e-06,
      "logits/chosen": -2.634321689605713,
      "logits/rejected": -2.6402688026428223,
      "logps/chosen": -321.502197265625,
      "logps/rejected": -339.1592712402344,
      "loss": 0.6613,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.47532662749290466,
      "rewards/margins": 0.08404591679573059,
      "rewards/rejected": -0.5593725442886353,
      "step": 550
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3688972199631974e-06,
      "logits/chosen": -2.6393518447875977,
      "logits/rejected": -2.6311287879943848,
      "logps/chosen": -318.1193542480469,
      "logps/rejected": -333.4119873046875,
      "loss": 0.6462,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.45164427161216736,
      "rewards/margins": 0.13988950848579407,
      "rewards/rejected": -0.5915338397026062,
      "step": 560
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3030472497208354e-06,
      "logits/chosen": -2.6581737995147705,
      "logits/rejected": -2.6506381034851074,
      "logps/chosen": -315.7310485839844,
      "logps/rejected": -369.3522033691406,
      "loss": 0.6318,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.5240504741668701,
      "rewards/margins": 0.17177370190620422,
      "rewards/rejected": -0.695824146270752,
      "step": 570
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.236569972745492e-06,
      "logits/chosen": -2.6991753578186035,
      "logits/rejected": -2.6973986625671387,
      "logps/chosen": -313.09747314453125,
      "logps/rejected": -330.6873779296875,
      "loss": 0.6638,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.5288428068161011,
      "rewards/margins": 0.08699531853199005,
      "rewards/rejected": -0.6158381700515747,
      "step": 580
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1695173182897126e-06,
      "logits/chosen": -2.622006893157959,
      "logits/rejected": -2.6231589317321777,
      "logps/chosen": -323.0411376953125,
      "logps/rejected": -351.798583984375,
      "loss": 0.657,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.5726518034934998,
      "rewards/margins": 0.10712896287441254,
      "rewards/rejected": -0.6797807812690735,
      "step": 590
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.10194166506673e-06,
      "logits/chosen": -2.637044906616211,
      "logits/rejected": -2.6247122287750244,
      "logps/chosen": -323.566650390625,
      "logps/rejected": -354.9677429199219,
      "loss": 0.6592,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5322225093841553,
      "rewards/margins": 0.10880555957555771,
      "rewards/rejected": -0.6410280466079712,
      "step": 600
    },
    {
      "epoch": 0.48,
      "eval_logits/chosen": -2.6844964027404785,
      "eval_logits/rejected": -2.6804616451263428,
      "eval_logps/chosen": -268.708740234375,
      "eval_logps/rejected": -279.2277526855469,
      "eval_loss": 0.6789478063583374,
      "eval_rewards/accuracies": 0.5559999942779541,
      "eval_rewards/chosen": -0.6009806990623474,
      "eval_rewards/margins": 0.05180613696575165,
      "eval_rewards/rejected": -0.6527868509292603,
      "eval_runtime": 1598.2559,
      "eval_samples_per_second": 1.873,
      "eval_steps_per_second": 0.235,
      "step": 600
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0338958003344115e-06,
      "logits/chosen": -2.703444004058838,
      "logits/rejected": -2.6941075325012207,
      "logps/chosen": -320.3475341796875,
      "logps/rejected": -349.988037109375,
      "loss": 0.6418,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.5905410051345825,
      "rewards/margins": 0.1995875984430313,
      "rewards/rejected": -0.7901285886764526,
      "step": 610
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9654328786600823e-06,
      "logits/chosen": -2.657546281814575,
      "logits/rejected": -2.6613922119140625,
      "logps/chosen": -323.674560546875,
      "logps/rejected": -357.7910461425781,
      "loss": 0.6488,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.6175565123558044,
      "rewards/margins": 0.11701612174510956,
      "rewards/rejected": -0.7345726490020752,
      "step": 620
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.896606380398402e-06,
      "logits/chosen": -2.6770036220550537,
      "logits/rejected": -2.697390079498291,
      "logps/chosen": -333.11932373046875,
      "logps/rejected": -353.01739501953125,
      "loss": 0.6521,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.6429452896118164,
      "rewards/margins": 0.12399964034557343,
      "rewards/rejected": -0.766944944858551,
      "step": 630
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.827470069914772e-06,
      "logits/chosen": -2.674020528793335,
      "logits/rejected": -2.660238265991211,
      "logps/chosen": -336.7355041503906,
      "logps/rejected": -360.3841247558594,
      "loss": 0.66,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.6630152463912964,
      "rewards/margins": 0.10159214586019516,
      "rewards/rejected": -0.7646073698997498,
      "step": 640
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7580779535868675e-06,
      "logits/chosen": -2.641157388687134,
      "logits/rejected": -2.65820574760437,
      "logps/chosen": -333.5285339355469,
      "logps/rejected": -352.81988525390625,
      "loss": 0.6624,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.617414116859436,
      "rewards/margins": 0.1025906577706337,
      "rewards/rejected": -0.7200047373771667,
      "step": 650
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.688484237617129e-06,
      "logits/chosen": -2.6740407943725586,
      "logits/rejected": -2.6733462810516357,
      "logps/chosen": -334.4920654296875,
      "logps/rejected": -359.688720703125,
      "loss": 0.6463,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.6058083772659302,
      "rewards/margins": 0.14193257689476013,
      "rewards/rejected": -0.7477409243583679,
      "step": 660
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6187432856891585e-06,
      "logits/chosen": -2.677578926086426,
      "logits/rejected": -2.694331169128418,
      "logps/chosen": -342.4553527832031,
      "logps/rejected": -371.22515869140625,
      "loss": 0.6558,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.6532286405563354,
      "rewards/margins": 0.11075172573328018,
      "rewards/rejected": -0.7639803290367126,
      "step": 670
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.548909576501096e-06,
      "logits/chosen": -2.7143263816833496,
      "logits/rejected": -2.7066876888275146,
      "logps/chosen": -340.72137451171875,
      "logps/rejected": -362.9776306152344,
      "loss": 0.6509,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.6533491015434265,
      "rewards/margins": 0.12043479830026627,
      "rewards/rejected": -0.7737839818000793,
      "step": 680
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4790376612091503e-06,
      "logits/chosen": -2.726318359375,
      "logits/rejected": -2.720698833465576,
      "logps/chosen": -359.000732421875,
      "logps/rejected": -382.2019958496094,
      "loss": 0.6319,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.7622326612472534,
      "rewards/margins": 0.1633436232805252,
      "rewards/rejected": -0.9255763292312622,
      "step": 690
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.40918212081453e-06,
      "logits/chosen": -2.7099740505218506,
      "logits/rejected": -2.6980385780334473,
      "logps/chosen": -330.88818359375,
      "logps/rejected": -372.46075439453125,
      "loss": 0.6107,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.739021897315979,
      "rewards/margins": 0.22892406582832336,
      "rewards/rejected": -0.9679458737373352,
      "step": 700
    },
    {
      "epoch": 0.56,
      "eval_logits/chosen": -2.696856737136841,
      "eval_logits/rejected": -2.691408634185791,
      "eval_logps/chosen": -290.1963806152344,
      "eval_logps/rejected": -301.8047180175781,
      "eval_loss": 0.6785151362419128,
      "eval_rewards/accuracies": 0.5550000071525574,
      "eval_rewards/chosen": -0.8158571124076843,
      "eval_rewards/margins": 0.06269912421703339,
      "eval_rewards/rejected": -0.8785561323165894,
      "eval_runtime": 1591.1417,
      "eval_samples_per_second": 1.882,
      "eval_steps_per_second": 0.236,
      "step": 700
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3393975235270654e-06,
      "logits/chosen": -2.684086322784424,
      "logits/rejected": -2.68058443069458,
      "logps/chosen": -360.47100830078125,
      "logps/rejected": -390.83441162109375,
      "loss": 0.6495,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.7877064943313599,
      "rewards/margins": 0.1392781138420105,
      "rewards/rejected": -0.9269846677780151,
      "step": 710
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2697383821388153e-06,
      "logits/chosen": -2.7011210918426514,
      "logits/rejected": -2.688472270965576,
      "logps/chosen": -351.14263916015625,
      "logps/rejected": -368.4684143066406,
      "loss": 0.6541,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.7785929441452026,
      "rewards/margins": 0.14113947749137878,
      "rewards/rejected": -0.919732391834259,
      "step": 720
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2002591114409657e-06,
      "logits/chosen": -2.693358898162842,
      "logits/rejected": -2.6756577491760254,
      "logps/chosen": -362.7990417480469,
      "logps/rejected": -383.33843994140625,
      "loss": 0.6601,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.8580299615859985,
      "rewards/margins": 0.10724015533924103,
      "rewards/rejected": -0.9652700424194336,
      "step": 730
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.131013985717285e-06,
      "logits/chosen": -2.7266604900360107,
      "logits/rejected": -2.6952764987945557,
      "logps/chosen": -370.53106689453125,
      "logps/rejected": -411.46875,
      "loss": 0.6317,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.8816037178039551,
      "rewards/margins": 0.18020646274089813,
      "rewards/rejected": -1.0618102550506592,
      "step": 740
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.062057096347338e-06,
      "logits/chosen": -2.737968921661377,
      "logits/rejected": -2.733065605163574,
      "logps/chosen": -361.11993408203125,
      "logps/rejected": -372.26513671875,
      "loss": 0.6619,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.8483070135116577,
      "rewards/margins": 0.1061801165342331,
      "rewards/rejected": -0.954487144947052,
      "step": 750
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9934423095525733e-06,
      "logits/chosen": -2.691333055496216,
      "logits/rejected": -2.6811554431915283,
      "logps/chosen": -362.3067932128906,
      "logps/rejected": -375.4438171386719,
      "loss": 0.6545,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.7950974702835083,
      "rewards/margins": 0.12981590628623962,
      "rewards/rejected": -0.9249132871627808,
      "step": 760
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9252232243182986e-06,
      "logits/chosen": -2.7668118476867676,
      "logits/rejected": -2.7533440589904785,
      "logps/chosen": -333.151611328125,
      "logps/rejected": -369.8766784667969,
      "loss": 0.6235,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.7605150938034058,
      "rewards/margins": 0.20025065541267395,
      "rewards/rejected": -0.9607658386230469,
      "step": 770
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8574531305244043e-06,
      "logits/chosen": -2.650963068008423,
      "logits/rejected": -2.6607377529144287,
      "logps/chosen": -363.749267578125,
      "logps/rejected": -396.4261169433594,
      "loss": 0.6247,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.829413115978241,
      "rewards/margins": 0.19826601445674896,
      "rewards/rejected": -1.0276790857315063,
      "step": 780
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.7901849673175559e-06,
      "logits/chosen": -2.719726085662842,
      "logits/rejected": -2.7071306705474854,
      "logps/chosen": -357.9602355957031,
      "logps/rejected": -386.6010437011719,
      "loss": 0.6424,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.7654613852500916,
      "rewards/margins": 0.14790861308574677,
      "rewards/rejected": -0.9133701324462891,
      "step": 790
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7234712817573555e-06,
      "logits/chosen": -2.7113006114959717,
      "logits/rejected": -2.6781070232391357,
      "logps/chosen": -382.4976806640625,
      "logps/rejected": -398.0575256347656,
      "loss": 0.6475,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.8575699925422668,
      "rewards/margins": 0.13993313908576965,
      "rewards/rejected": -0.9975031614303589,
      "step": 800
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.729461669921875,
      "eval_logits/rejected": -2.723686933517456,
      "eval_logps/chosen": -297.0627136230469,
      "eval_logps/rejected": -309.38665771484375,
      "eval_loss": 0.6770467162132263,
      "eval_rewards/accuracies": 0.5609999895095825,
      "eval_rewards/chosen": -0.8845203518867493,
      "eval_rewards/margins": 0.06985554099082947,
      "eval_rewards/rejected": -0.9543759226799011,
      "eval_runtime": 1591.3996,
      "eval_samples_per_second": 1.881,
      "eval_steps_per_second": 0.236,
      "step": 800
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6573641877687936e-06,
      "logits/chosen": -2.712907314300537,
      "logits/rejected": -2.712637424468994,
      "logps/chosen": -359.29473876953125,
      "logps/rejected": -395.03173828125,
      "loss": 0.6372,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.8719032406806946,
      "rewards/margins": 0.15319783985614777,
      "rewards/rejected": -1.0251010656356812,
      "step": 810
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.591915325433034e-06,
      "logits/chosen": -2.771751880645752,
      "logits/rejected": -2.764472246170044,
      "logps/chosen": -348.4260559082031,
      "logps/rejected": -368.5625305175781,
      "loss": 0.6422,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.8010624647140503,
      "rewards/margins": 0.15961696207523346,
      "rewards/rejected": -0.9606794118881226,
      "step": 820
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5271758206483664e-06,
      "logits/chosen": -2.740337371826172,
      "logits/rejected": -2.732513666152954,
      "logps/chosen": -366.95660400390625,
      "logps/rejected": -389.9357604980469,
      "loss": 0.6581,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.8896878361701965,
      "rewards/margins": 0.12456034123897552,
      "rewards/rejected": -1.014248251914978,
      "step": 830
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4631962451927966e-06,
      "logits/chosen": -2.6755502223968506,
      "logits/rejected": -2.6818313598632812,
      "logps/chosen": -362.2518005371094,
      "logps/rejected": -391.2507019042969,
      "loss": 0.6378,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.8435665369033813,
      "rewards/margins": 0.16070519387722015,
      "rewards/rejected": -1.0042717456817627,
      "step": 840
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.4000265772195032e-06,
      "logits/chosen": -2.792430877685547,
      "logits/rejected": -2.7686753273010254,
      "logps/chosen": -358.8890075683594,
      "logps/rejected": -395.0196838378906,
      "loss": 0.6195,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.8025849461555481,
      "rewards/margins": 0.2058737725019455,
      "rewards/rejected": -1.0084587335586548,
      "step": 850
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3377161622160077e-06,
      "logits/chosen": -2.709834337234497,
      "logits/rejected": -2.7066776752471924,
      "logps/chosen": -356.4862365722656,
      "logps/rejected": -385.40625,
      "loss": 0.6288,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.8100967407226562,
      "rewards/margins": 0.16853158175945282,
      "rewards/rejected": -0.9786283373832703,
      "step": 860
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.276313674457553e-06,
      "logits/chosen": -2.737943172454834,
      "logits/rejected": -2.74971079826355,
      "logps/chosen": -342.73614501953125,
      "logps/rejected": -379.23126220703125,
      "loss": 0.6226,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.8370000123977661,
      "rewards/margins": 0.18809911608695984,
      "rewards/rejected": -1.0250990390777588,
      "step": 870
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2158670789848095e-06,
      "logits/chosen": -2.713247537612915,
      "logits/rejected": -2.6928553581237793,
      "logps/chosen": -373.70361328125,
      "logps/rejected": -397.2867431640625,
      "loss": 0.6407,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.9539483189582825,
      "rewards/margins": 0.16210444271564484,
      "rewards/rejected": -1.1160527467727661,
      "step": 880
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1564235941356016e-06,
      "logits/chosen": -2.7370991706848145,
      "logits/rejected": -2.7364439964294434,
      "logps/chosen": -360.29534912109375,
      "logps/rejected": -403.6412048339844,
      "loss": 0.6203,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.97996985912323,
      "rewards/margins": 0.23149892687797546,
      "rewards/rejected": -1.2114689350128174,
      "step": 890
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.0980296546599254e-06,
      "logits/chosen": -2.6951401233673096,
      "logits/rejected": -2.6780078411102295,
      "logps/chosen": -381.99420166015625,
      "logps/rejected": -393.02215576171875,
      "loss": 0.6639,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.9827243685722351,
      "rewards/margins": 0.15567822754383087,
      "rewards/rejected": -1.1384025812149048,
      "step": 900
    },
    {
      "epoch": 0.72,
      "eval_logits/chosen": -2.752532482147217,
      "eval_logits/rejected": -2.7463557720184326,
      "eval_logps/chosen": -305.6557922363281,
      "eval_logps/rejected": -318.45068359375,
      "eval_loss": 0.6765736937522888,
      "eval_rewards/accuracies": 0.5666666626930237,
      "eval_rewards/chosen": -0.9704511761665344,
      "eval_rewards/margins": 0.07456483691930771,
      "eval_rewards/rejected": -1.0450160503387451,
      "eval_runtime": 1591.8909,
      "eval_samples_per_second": 1.881,
      "eval_steps_per_second": 0.236,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.040730875447083e-06,
      "logits/chosen": -2.7219176292419434,
      "logits/rejected": -2.7300069332122803,
      "logps/chosen": -374.46649169921875,
      "logps/rejected": -392.5831604003906,
      "loss": 0.6604,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.9455657005310059,
      "rewards/margins": 0.11010564863681793,
      "rewards/rejected": -1.0556714534759521,
      "step": 910
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.845720158932414e-07,
      "logits/chosen": -2.6920173168182373,
      "logits/rejected": -2.69575834274292,
      "logps/chosen": -352.3219299316406,
      "logps/rejected": -367.9543762207031,
      "loss": 0.6388,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.9019610285758972,
      "rewards/margins": 0.1525803804397583,
      "rewards/rejected": -1.0545413494110107,
      "step": 920
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.295969449372796e-07,
      "logits/chosen": -2.768012523651123,
      "logits/rejected": -2.773406505584717,
      "logps/chosen": -356.3282165527344,
      "logps/rejected": -381.96478271484375,
      "loss": 0.6444,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.873378574848175,
      "rewards/margins": 0.13878589868545532,
      "rewards/rejected": -1.0121644735336304,
      "step": 930
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.758486067922176e-07,
      "logits/chosen": -2.7590746879577637,
      "logits/rejected": -2.735861301422119,
      "logps/chosen": -354.44818115234375,
      "logps/rejected": -391.2872619628906,
      "loss": 0.6291,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.9065491557121277,
      "rewards/margins": 0.18637417256832123,
      "rewards/rejected": -1.0929234027862549,
      "step": 940
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.233689873990006e-07,
      "logits/chosen": -2.7534782886505127,
      "logits/rejected": -2.7382559776306152,
      "logps/chosen": -355.0606384277344,
      "logps/rejected": -398.70513916015625,
      "loss": 0.6269,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9113260507583618,
      "rewards/margins": 0.26781052350997925,
      "rewards/rejected": -1.1791365146636963,
      "step": 950
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.721990816288555e-07,
      "logits/chosen": -2.7826993465423584,
      "logits/rejected": -2.763814687728882,
      "logps/chosen": -336.1342468261719,
      "logps/rejected": -366.10809326171875,
      "loss": 0.63,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.8676854372024536,
      "rewards/margins": 0.1765536516904831,
      "rewards/rejected": -1.0442390441894531,
      "step": 960
    },
    {
      "epoch": 0.78,
      "learning_rate": 7.223788612598148e-07,
      "logits/chosen": -2.7675793170928955,
      "logits/rejected": -2.776650905609131,
      "logps/chosen": -366.38525390625,
      "logps/rejected": -391.9957580566406,
      "loss": 0.6576,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.886553943157196,
      "rewards/margins": 0.10182847827672958,
      "rewards/rejected": -0.9883824586868286,
      "step": 970
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.73947243752448e-07,
      "logits/chosen": -2.7559187412261963,
      "logits/rejected": -2.751338481903076,
      "logps/chosen": -376.8429260253906,
      "logps/rejected": -383.2920837402344,
      "loss": 0.6617,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.9805536270141602,
      "rewards/margins": 0.13034404814243317,
      "rewards/rejected": -1.1108976602554321,
      "step": 980
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.269420618491759e-07,
      "logits/chosen": -2.759143352508545,
      "logits/rejected": -2.74459171295166,
      "logps/chosen": -351.4375305175781,
      "logps/rejected": -376.80633544921875,
      "loss": 0.6654,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.9471473693847656,
      "rewards/margins": 0.10330625623464584,
      "rewards/rejected": -1.0504536628723145,
      "step": 990
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.814000340209267e-07,
      "logits/chosen": -2.746699810028076,
      "logits/rejected": -2.7150912284851074,
      "logps/chosen": -363.809814453125,
      "logps/rejected": -401.50927734375,
      "loss": 0.6305,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.9182988405227661,
      "rewards/margins": 0.19320932030677795,
      "rewards/rejected": -1.1115081310272217,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -2.7606282234191895,
      "eval_logits/rejected": -2.7542688846588135,
      "eval_logps/chosen": -307.0535583496094,
      "eval_logps/rejected": -319.9799499511719,
      "eval_loss": 0.6764061450958252,
      "eval_rewards/accuracies": 0.5680000185966492,
      "eval_rewards/chosen": -0.9844286441802979,
      "eval_rewards/margins": 0.07588021457195282,
      "eval_rewards/rejected": -1.060308814048767,
      "eval_runtime": 1582.573,
      "eval_samples_per_second": 1.892,
      "eval_steps_per_second": 0.237,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.373567357842111e-07,
      "logits/chosen": -2.703906536102295,
      "logits/rejected": -2.7083990573883057,
      "logps/chosen": -357.48388671875,
      "logps/rejected": -387.49078369140625,
      "loss": 0.6159,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.8543448448181152,
      "rewards/margins": 0.21335168182849884,
      "rewards/rejected": -1.0676965713500977,
      "step": 1010
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.948465719110226e-07,
      "logits/chosen": -2.7659733295440674,
      "logits/rejected": -2.7676990032196045,
      "logps/chosen": -350.13482666015625,
      "logps/rejected": -371.6733703613281,
      "loss": 0.6451,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.8929713368415833,
      "rewards/margins": 0.14709730446338654,
      "rewards/rejected": -1.0400686264038086,
      "step": 1020
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.539027495532766e-07,
      "logits/chosen": -2.715686798095703,
      "logits/rejected": -2.7308743000030518,
      "logps/chosen": -346.8404235839844,
      "logps/rejected": -375.3435363769531,
      "loss": 0.6263,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.8631433248519897,
      "rewards/margins": 0.18629541993141174,
      "rewards/rejected": -1.049438714981079,
      "step": 1030
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.14557252302783e-07,
      "logits/chosen": -2.7563765048980713,
      "logits/rejected": -2.741762399673462,
      "logps/chosen": -368.244140625,
      "logps/rejected": -396.12298583984375,
      "loss": 0.6673,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.0178955793380737,
      "rewards/margins": 0.13189546763896942,
      "rewards/rejected": -1.1497912406921387,
      "step": 1040
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7684081520700884e-07,
      "logits/chosen": -2.719506025314331,
      "logits/rejected": -2.712674140930176,
      "logps/chosen": -373.14007568359375,
      "logps/rejected": -384.85870361328125,
      "loss": 0.6379,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.888410210609436,
      "rewards/margins": 0.16670987010002136,
      "rewards/rejected": -1.0551201105117798,
      "step": 1050
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.407829007601507e-07,
      "logits/chosen": -2.723992109298706,
      "logits/rejected": -2.7108616828918457,
      "logps/chosen": -362.71112060546875,
      "logps/rejected": -396.30413818359375,
      "loss": 0.6307,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.9032829403877258,
      "rewards/margins": 0.18060991168022156,
      "rewards/rejected": -1.0838927030563354,
      "step": 1060
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.064116758882724e-07,
      "logits/chosen": -2.7330386638641357,
      "logits/rejected": -2.700990676879883,
      "logps/chosen": -375.42425537109375,
      "logps/rejected": -414.94635009765625,
      "loss": 0.6324,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.9860410690307617,
      "rewards/margins": 0.20865873992443085,
      "rewards/rejected": -1.1946996450424194,
      "step": 1070
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.737539899464908e-07,
      "logits/chosen": -2.7359976768493652,
      "logits/rejected": -2.7335023880004883,
      "logps/chosen": -345.5082092285156,
      "logps/rejected": -376.73382568359375,
      "loss": 0.633,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.9359216690063477,
      "rewards/margins": 0.19209426641464233,
      "rewards/rejected": -1.1280159950256348,
      "step": 1080
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4283535374538645e-07,
      "logits/chosen": -2.7470860481262207,
      "logits/rejected": -2.759739398956299,
      "logps/chosen": -367.7596130371094,
      "logps/rejected": -393.4772644042969,
      "loss": 0.6363,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.8926227688789368,
      "rewards/margins": 0.18235111236572266,
      "rewards/rejected": -1.0749738216400146,
      "step": 1090
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1367991962303298e-07,
      "logits/chosen": -2.701397657394409,
      "logits/rejected": -2.7229931354522705,
      "logps/chosen": -348.02447509765625,
      "logps/rejected": -368.96197509765625,
      "loss": 0.6754,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.9437961578369141,
      "rewards/margins": 0.08066528290510178,
      "rewards/rejected": -1.0244615077972412,
      "step": 1100
    },
    {
      "epoch": 0.88,
      "eval_logits/chosen": -2.7602105140686035,
      "eval_logits/rejected": -2.7538294792175293,
      "eval_logps/chosen": -307.4263916015625,
      "eval_logps/rejected": -320.4283447265625,
      "eval_loss": 0.6762885451316833,
      "eval_rewards/accuracies": 0.5686666369438171,
      "eval_rewards/chosen": -0.9881570935249329,
      "eval_rewards/margins": 0.07663587480783463,
      "eval_rewards/rejected": -1.0647929906845093,
      "eval_runtime": 1581.3137,
      "eval_samples_per_second": 1.893,
      "eval_steps_per_second": 0.237,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.8631046257820278e-07,
      "logits/chosen": -2.747465133666992,
      "logits/rejected": -2.7517151832580566,
      "logps/chosen": -368.2339172363281,
      "logps/rejected": -389.92803955078125,
      "loss": 0.6368,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.9324411153793335,
      "rewards/margins": 0.17461669445037842,
      "rewards/rejected": -1.1070579290390015,
      "step": 1110
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6074836247950143e-07,
      "logits/chosen": -2.7465901374816895,
      "logits/rejected": -2.7471508979797363,
      "logps/chosen": -372.44866943359375,
      "logps/rejected": -390.7837829589844,
      "loss": 0.6522,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.8873372077941895,
      "rewards/margins": 0.12168030440807343,
      "rewards/rejected": -1.0090174674987793,
      "step": 1120
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.370135873643097e-07,
      "logits/chosen": -2.7472376823425293,
      "logits/rejected": -2.7466180324554443,
      "logps/chosen": -373.3634338378906,
      "logps/rejected": -377.52569580078125,
      "loss": 0.6504,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.9632595777511597,
      "rewards/margins": 0.13923458755016327,
      "rewards/rejected": -1.102494239807129,
      "step": 1130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1512467784059372e-07,
      "logits/chosen": -2.7725350856781006,
      "logits/rejected": -2.748663902282715,
      "logps/chosen": -343.6497802734375,
      "logps/rejected": -369.282470703125,
      "loss": 0.6371,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.9474760293960571,
      "rewards/margins": 0.1754283607006073,
      "rewards/rejected": -1.1229044198989868,
      "step": 1140
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.509873260376251e-08,
      "logits/chosen": -2.727968692779541,
      "logits/rejected": -2.703981876373291,
      "logps/chosen": -363.1959228515625,
      "logps/rejected": -414.9847106933594,
      "loss": 0.6061,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.9189531207084656,
      "rewards/margins": 0.2506302297115326,
      "rewards/rejected": -1.1695834398269653,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.695139507988559e-08,
      "logits/chosen": -2.7512032985687256,
      "logits/rejected": -2.7607004642486572,
      "logps/chosen": -376.4388427734375,
      "logps/rejected": -397.3370056152344,
      "loss": 0.6506,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.925561785697937,
      "rewards/margins": 0.1294144093990326,
      "rewards/rejected": -1.054976224899292,
      "step": 1160
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.069684120570684e-08,
      "logits/chosen": -2.715827703475952,
      "logits/rejected": -2.679953098297119,
      "logps/chosen": -362.00079345703125,
      "logps/rejected": -398.26141357421875,
      "loss": 0.6424,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.9743494987487793,
      "rewards/margins": 0.17094501852989197,
      "rewards/rejected": -1.145294427871704,
      "step": 1170
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.634776835499871e-08,
      "logits/chosen": -2.679133415222168,
      "logits/rejected": -2.6611809730529785,
      "logps/chosen": -350.9591369628906,
      "logps/rejected": -379.08416748046875,
      "loss": 0.648,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.944057822227478,
      "rewards/margins": 0.15136626362800598,
      "rewards/rejected": -1.0954240560531616,
      "step": 1180
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.3915385419908964e-08,
      "logits/chosen": -2.730663299560547,
      "logits/rejected": -2.7344131469726562,
      "logps/chosen": -362.0574951171875,
      "logps/rejected": -382.64788818359375,
      "loss": 0.645,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9124767184257507,
      "rewards/margins": 0.15995648503303528,
      "rewards/rejected": -1.0724332332611084,
      "step": 1190
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.3409404055043938e-08,
      "logits/chosen": -2.7462306022644043,
      "logits/rejected": -2.7565507888793945,
      "logps/chosen": -369.1392517089844,
      "logps/rejected": -390.51190185546875,
      "loss": 0.6577,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.9762216806411743,
      "rewards/margins": 0.11748246848583221,
      "rewards/rejected": -1.093704104423523,
      "step": 1200
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -2.7601613998413086,
      "eval_logits/rejected": -2.7538154125213623,
      "eval_logps/chosen": -307.4615478515625,
      "eval_logps/rejected": -320.441162109375,
      "eval_loss": 0.6764407753944397,
      "eval_rewards/accuracies": 0.5663333535194397,
      "eval_rewards/chosen": -0.9885084629058838,
      "eval_rewards/margins": 0.07641255110502243,
      "eval_rewards/rejected": -1.0649209022521973,
      "eval_runtime": 1656.3369,
      "eval_samples_per_second": 1.808,
      "eval_steps_per_second": 0.226,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4838031091134186e-08,
      "logits/chosen": -2.711413860321045,
      "logits/rejected": -2.700925350189209,
      "logps/chosen": -341.56524658203125,
      "logps/rejected": -391.4617614746094,
      "loss": 0.6092,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.8985296487808228,
      "rewards/margins": 0.250393271446228,
      "rewards/rejected": -1.1489229202270508,
      "step": 1210
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.207962124201774e-09,
      "logits/chosen": -2.766512393951416,
      "logits/rejected": -2.751699924468994,
      "logps/chosen": -370.7591857910156,
      "logps/rejected": -398.38201904296875,
      "loss": 0.6128,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.873375415802002,
      "rewards/margins": 0.22166454792022705,
      "rewards/rejected": -1.0950400829315186,
      "step": 1220
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.5243762852441023e-09,
      "logits/chosen": -2.7122433185577393,
      "logits/rejected": -2.7001757621765137,
      "logps/chosen": -358.870849609375,
      "logps/rejected": -391.4342956542969,
      "loss": 0.6408,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.9104388952255249,
      "rewards/margins": 0.1689816415309906,
      "rewards/rejected": -1.0794206857681274,
      "step": 1230
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.909321945129278e-10,
      "logits/chosen": -2.7206053733825684,
      "logits/rejected": -2.7036147117614746,
      "logps/chosen": -379.60906982421875,
      "logps/rejected": -411.9668884277344,
      "loss": 0.6327,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9332448840141296,
      "rewards/margins": 0.1873912662267685,
      "rewards/rejected": -1.120636224746704,
      "step": 1240
    },
    {
      "epoch": 1.0,
      "step": 1249,
      "total_flos": 0.0,
      "train_loss": 0.6566078573155155,
      "train_runtime": 38377.8314,
      "train_samples_per_second": 0.521,
      "train_steps_per_second": 0.033
    }
  ],
  "logging_steps": 10,
  "max_steps": 1249,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}