{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9989528795811519,
  "eval_steps": 200,
  "global_step": 477,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020942408376963353,
      "grad_norm": 286.114501953125,
      "learning_rate": 0.0,
      "logits/chosen": -0.5995081663131714,
      "logits/rejected": -0.6144353747367859,
      "logps/chosen": -267.5272216796875,
      "logps/ref_chosen": -267.5935363769531,
      "logps/ref_rejected": -204.2306671142578,
      "logps/rejected": -204.23907470703125,
      "loss": 5.5602,
      "r_dpo/chosen_len": 257.75,
      "r_dpo/length_delta": 47.875,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 209.875,
      "step": 1
    },
    {
      "epoch": 0.020942408376963352,
      "grad_norm": 265.7362365722656,
      "learning_rate": 9.375e-08,
      "logits/chosen": -0.6309213042259216,
      "logits/rejected": -0.63545161485672,
      "logps/chosen": -296.74920654296875,
      "logps/ref_chosen": -296.72576904296875,
      "logps/ref_rejected": -258.9681396484375,
      "logps/rejected": -259.0067443847656,
      "loss": 5.5477,
      "r_dpo/chosen_len": 291.8680419921875,
      "r_dpo/length_delta": 49.76388931274414,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 242.1041717529297,
      "step": 10
    },
    {
      "epoch": 0.041884816753926704,
      "grad_norm": 296.5692138671875,
      "learning_rate": 1.9791666666666664e-07,
      "logits/chosen": -0.597561240196228,
      "logits/rejected": -0.6285912990570068,
      "logps/chosen": -297.9436340332031,
      "logps/ref_chosen": -297.9349365234375,
      "logps/ref_rejected": -256.9902648925781,
      "logps/rejected": -256.98876953125,
      "loss": 5.5369,
      "r_dpo/chosen_len": 291.29998779296875,
      "r_dpo/length_delta": 52.89374923706055,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 238.40625,
      "step": 20
    },
    {
      "epoch": 0.06282722513089005,
      "grad_norm": 279.9869384765625,
      "learning_rate": 3.020833333333333e-07,
      "logits/chosen": -0.6136946678161621,
      "logits/rejected": -0.605636477470398,
      "logps/chosen": -278.4833679199219,
      "logps/ref_chosen": -278.64752197265625,
      "logps/ref_rejected": -249.309814453125,
      "logps/rejected": -249.22158813476562,
      "loss": 5.495,
      "r_dpo/chosen_len": 270.8812561035156,
      "r_dpo/length_delta": 25.228124618530273,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 245.6531219482422,
      "step": 30
    },
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 258.0906677246094,
      "learning_rate": 4.0625e-07,
      "logits/chosen": -0.6181729435920715,
      "logits/rejected": -0.6429071426391602,
      "logps/chosen": -282.7382507324219,
      "logps/ref_chosen": -283.49981689453125,
      "logps/ref_rejected": -265.32733154296875,
      "logps/rejected": -265.03875732421875,
      "loss": 5.3372,
      "r_dpo/chosen_len": 281.43438720703125,
      "r_dpo/length_delta": 33.34375,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 248.0906219482422,
      "step": 40
    },
    {
      "epoch": 0.10471204188481675,
      "grad_norm": 267.5690002441406,
      "learning_rate": 4.999932966293553e-07,
      "logits/chosen": -0.6292977333068848,
      "logits/rejected": -0.6698824763298035,
      "logps/chosen": -278.5817565917969,
      "logps/ref_chosen": -280.224365234375,
      "logps/ref_rejected": -274.3541259765625,
      "logps/rejected": -273.8267517089844,
      "loss": 5.1553,
      "r_dpo/chosen_len": 290.32501220703125,
      "r_dpo/length_delta": 35.11249923706055,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 255.21249389648438,
      "step": 50
    },
    {
      "epoch": 0.1256544502617801,
      "grad_norm": 258.2756652832031,
      "learning_rate": 4.991893270335525e-07,
      "logits/chosen": -0.6408439874649048,
      "logits/rejected": -0.6542555093765259,
      "logps/chosen": -278.864501953125,
      "logps/ref_chosen": -281.12664794921875,
      "logps/ref_rejected": -259.86456298828125,
      "logps/rejected": -259.77728271484375,
      "loss": 4.9283,
      "r_dpo/chosen_len": 273.953125,
      "r_dpo/length_delta": 29.084375381469727,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 244.86874389648438,
      "step": 60
    },
    {
      "epoch": 0.14659685863874344,
      "grad_norm": 233.6326446533203,
      "learning_rate": 4.970496218214204e-07,
      "logits/chosen": -0.7015193700790405,
      "logits/rejected": -0.7072200179100037,
      "logps/chosen": -284.4013977050781,
      "logps/ref_chosen": -287.71063232421875,
      "logps/ref_rejected": -276.839599609375,
      "logps/rejected": -276.4761047363281,
      "loss": 4.76,
      "r_dpo/chosen_len": 267.4937438964844,
      "r_dpo/length_delta": 14.484375,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 253.00936889648438,
      "step": 70
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 234.1904296875,
      "learning_rate": 4.935856505068998e-07,
      "logits/chosen": -0.6775287389755249,
      "logits/rejected": -0.6742324829101562,
      "logps/chosen": -275.77288818359375,
      "logps/ref_chosen": -280.123046875,
      "logps/ref_rejected": -258.8989562988281,
      "logps/rejected": -258.0046691894531,
      "loss": 4.6236,
      "r_dpo/chosen_len": 267.4781188964844,
      "r_dpo/length_delta": 32.46562576293945,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 235.0124969482422,
      "step": 80
    },
    {
      "epoch": 0.18848167539267016,
      "grad_norm": 249.3949737548828,
      "learning_rate": 4.8881598109976e-07,
      "logits/chosen": -0.6876357197761536,
      "logits/rejected": -0.694031834602356,
      "logps/chosen": -273.43939208984375,
      "logps/ref_chosen": -278.02545166015625,
      "logps/ref_rejected": -251.0922393798828,
      "logps/rejected": -251.4302978515625,
      "loss": 4.5727,
      "r_dpo/chosen_len": 274.20623779296875,
      "r_dpo/length_delta": 44.97187423706055,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 229.234375,
      "step": 90
    },
    {
      "epoch": 0.2094240837696335,
      "grad_norm": 247.84217834472656,
      "learning_rate": 4.827661805750437e-07,
      "logits/chosen": -0.6776241660118103,
      "logits/rejected": -0.6956905126571655,
      "logps/chosen": -269.8210754394531,
      "logps/ref_chosen": -274.0089416503906,
      "logps/ref_rejected": -274.14447021484375,
      "logps/rejected": -275.0433349609375,
      "loss": 4.6252,
      "r_dpo/chosen_len": 275.3343811035156,
      "r_dpo/length_delta": 21.912500381469727,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 253.421875,
      "step": 100
    },
    {
      "epoch": 0.23036649214659685,
      "grad_norm": 227.6590576171875,
      "learning_rate": 4.75468677825789e-07,
      "logits/chosen": -0.701612114906311,
      "logits/rejected": -0.7261066436767578,
      "logps/chosen": -269.59173583984375,
      "logps/ref_chosen": -273.23333740234375,
      "logps/ref_rejected": -263.88787841796875,
      "logps/rejected": -266.48272705078125,
      "loss": 4.2996,
      "r_dpo/chosen_len": 283.43438720703125,
      "r_dpo/length_delta": 50.34375,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 233.0906219482422,
      "step": 110
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 243.6549835205078,
      "learning_rate": 4.669625898336438e-07,
      "logits/chosen": -0.7124736905097961,
      "logits/rejected": -0.7152305841445923,
      "logps/chosen": -267.23974609375,
      "logps/ref_chosen": -269.54656982421875,
      "logps/ref_rejected": -272.7981872558594,
      "logps/rejected": -275.8104553222656,
      "loss": 4.5034,
      "r_dpo/chosen_len": 264.7593688964844,
      "r_dpo/length_delta": 13.840624809265137,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 250.9187469482422,
      "step": 120
    },
    {
      "epoch": 0.27225130890052357,
      "grad_norm": 274.58111572265625,
      "learning_rate": 4.5729351198915705e-07,
      "logits/chosen": -0.7257631421089172,
      "logits/rejected": -0.7154949903488159,
      "logps/chosen": -272.1502990722656,
      "logps/ref_chosen": -275.03448486328125,
      "logps/ref_rejected": -276.39862060546875,
      "logps/rejected": -279.466796875,
      "loss": 4.445,
      "r_dpo/chosen_len": 266.625,
      "r_dpo/length_delta": 18.668750762939453,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 247.9562530517578,
      "step": 130
    },
    {
      "epoch": 0.2931937172774869,
      "grad_norm": 247.52589416503906,
      "learning_rate": 4.4651327368569684e-07,
      "logits/chosen": -0.7425049543380737,
      "logits/rejected": -0.7479813098907471,
      "logps/chosen": -273.00714111328125,
      "logps/ref_chosen": -276.0029602050781,
      "logps/ref_rejected": -255.9320526123047,
      "logps/rejected": -259.60784912109375,
      "loss": 4.4958,
      "r_dpo/chosen_len": 261.46875,
      "r_dpo/length_delta": 22.375,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 239.09375,
      "step": 140
    },
    {
      "epoch": 0.31413612565445026,
      "grad_norm": 264.3229675292969,
      "learning_rate": 4.346796604970912e-07,
      "logits/chosen": -0.7595945596694946,
      "logits/rejected": -0.7660830616950989,
      "logps/chosen": -294.779541015625,
      "logps/ref_chosen": -298.2093505859375,
      "logps/ref_rejected": -254.8907012939453,
      "logps/rejected": -258.84454345703125,
      "loss": 4.2489,
      "r_dpo/chosen_len": 283.84375,
      "r_dpo/length_delta": 48.359375,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 235.484375,
      "step": 150
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 212.1317138671875,
      "learning_rate": 4.218561044282098e-07,
      "logits/chosen": -0.7484012246131897,
      "logits/rejected": -0.7548493146896362,
      "logps/chosen": -280.19281005859375,
      "logps/ref_chosen": -281.94189453125,
      "logps/ref_rejected": -255.5653533935547,
      "logps/rejected": -261.9532470703125,
      "loss": 4.0494,
      "r_dpo/chosen_len": 267.828125,
      "r_dpo/length_delta": 41.368751525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 226.45938110351562,
      "step": 160
    },
    {
      "epoch": 0.35602094240837695,
      "grad_norm": 263.6576843261719,
      "learning_rate": 4.081113438988443e-07,
      "logits/chosen": -0.7331063151359558,
      "logits/rejected": -0.7357569932937622,
      "logps/chosen": -287.6741638183594,
      "logps/ref_chosen": -288.32232666015625,
      "logps/ref_rejected": -239.85415649414062,
      "logps/rejected": -246.4853973388672,
      "loss": 4.1229,
      "r_dpo/chosen_len": 285.203125,
      "r_dpo/length_delta": 46.396873474121094,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 238.80624389648438,
      "step": 170
    },
    {
      "epoch": 0.3769633507853403,
      "grad_norm": 204.8552703857422,
      "learning_rate": 3.935190552834828e-07,
      "logits/chosen": -0.7205232977867126,
      "logits/rejected": -0.7394840121269226,
      "logps/chosen": -285.03363037109375,
      "logps/ref_chosen": -286.17889404296875,
      "logps/ref_rejected": -249.9820098876953,
      "logps/rejected": -256.336669921875,
      "loss": 4.2693,
      "r_dpo/chosen_len": 266.09063720703125,
      "r_dpo/length_delta": 40.12812423706055,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 225.96249389648438,
      "step": 180
    },
    {
      "epoch": 0.39790575916230364,
      "grad_norm": 227.1675567626953,
      "learning_rate": 3.781574579820464e-07,
      "logits/chosen": -0.7226096987724304,
      "logits/rejected": -0.7422462701797485,
      "logps/chosen": -279.1224365234375,
      "logps/ref_chosen": -280.9278259277344,
      "logps/ref_rejected": -254.3533477783203,
      "logps/rejected": -262.07611083984375,
      "loss": 4.0622,
      "r_dpo/chosen_len": 276.33123779296875,
      "r_dpo/length_delta": 41.993751525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 234.33749389648438,
      "step": 190
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 257.41168212890625,
      "learning_rate": 3.621088951385353e-07,
      "logits/chosen": -0.741692841053009,
      "logits/rejected": -0.7598770260810852,
      "logps/chosen": -250.05776977539062,
      "logps/ref_chosen": -253.1712188720703,
      "logps/ref_rejected": -241.90478515625,
      "logps/rejected": -246.9326629638672,
      "loss": 4.178,
      "r_dpo/chosen_len": 248.0749969482422,
      "r_dpo/length_delta": 28.131250381469727,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 219.94375610351562,
      "step": 200
    },
    {
      "epoch": 0.418848167539267,
      "eval_logits/chosen": -0.7478832006454468,
      "eval_logits/rejected": -0.7490243911743164,
      "eval_logps/chosen": -287.2392883300781,
      "eval_logps/ref_chosen": -289.1346435546875,
      "eval_logps/ref_rejected": -264.7782287597656,
      "eval_logps/rejected": -271.315185546875,
      "eval_loss": 0.5266098380088806,
      "eval_r_dpo/chosen_len": 291.2619934082031,
      "eval_r_dpo/length_delta": 42.86600112915039,
      "eval_r_dpo/regularization_term": 0.0,
      "eval_r_dpo/rejected_len": 248.39599609375,
      "eval_runtime": 81.9324,
      "eval_samples_per_second": 24.41,
      "eval_steps_per_second": 1.526,
      "step": 200
    },
    {
      "epoch": 0.4397905759162304,
      "grad_norm": 272.2992858886719,
      "learning_rate": 3.454593922550693e-07,
      "logits/chosen": -0.7123077511787415,
      "logits/rejected": -0.7341128587722778,
      "logps/chosen": -285.9898376464844,
      "logps/ref_chosen": -287.9228210449219,
      "logps/ref_rejected": -263.35595703125,
      "logps/rejected": -270.154541015625,
      "loss": 4.2219,
      "r_dpo/chosen_len": 280.3125,
      "r_dpo/length_delta": 36.68437576293945,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 243.6281280517578,
      "step": 210
    },
    {
      "epoch": 0.4607329842931937,
      "grad_norm": 244.5552520751953,
      "learning_rate": 3.2829819606729477e-07,
      "logits/chosen": -0.7547723054885864,
      "logits/rejected": -0.7528045773506165,
      "logps/chosen": -280.8878479003906,
      "logps/ref_chosen": -282.3331604003906,
      "logps/ref_rejected": -272.5645446777344,
      "logps/rejected": -279.7627868652344,
      "loss": 4.0545,
      "r_dpo/chosen_len": 261.359375,
      "r_dpo/length_delta": 17.865625381469727,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 243.49374389648438,
      "step": 220
    },
    {
      "epoch": 0.4816753926701571,
      "grad_norm": 290.22955322265625,
      "learning_rate": 3.1071729615293424e-07,
      "logits/chosen": -0.7329878211021423,
      "logits/rejected": -0.7301102876663208,
      "logps/chosen": -276.18194580078125,
      "logps/ref_chosen": -276.1485595703125,
      "logps/ref_rejected": -252.81198120117188,
      "logps/rejected": -262.63323974609375,
      "loss": 3.8953,
      "r_dpo/chosen_len": 264.43438720703125,
      "r_dpo/length_delta": 31.256250381469727,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 233.17813110351562,
      "step": 230
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 242.89169311523438,
      "learning_rate": 2.9281093183781403e-07,
      "logits/chosen": -0.7076805233955383,
      "logits/rejected": -0.729147732257843,
      "logps/chosen": -270.7227783203125,
      "logps/ref_chosen": -270.52520751953125,
      "logps/ref_rejected": -254.83334350585938,
      "logps/rejected": -263.5503234863281,
      "loss": 4.1412,
      "r_dpo/chosen_len": 271.81561279296875,
      "r_dpo/length_delta": 37.099998474121094,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 234.7156219482422,
      "step": 240
    },
    {
      "epoch": 0.5235602094240838,
      "grad_norm": 281.21990966796875,
      "learning_rate": 2.7467508704251135e-07,
      "logits/chosen": -0.7120658159255981,
      "logits/rejected": -0.7247544527053833,
      "logps/chosen": -288.3058166503906,
      "logps/ref_chosen": -288.9503173828125,
      "logps/ref_rejected": -265.0694580078125,
      "logps/rejected": -273.70220947265625,
      "loss": 4.2125,
      "r_dpo/chosen_len": 277.50311279296875,
      "r_dpo/length_delta": 41.103126525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 236.39999389648438,
      "step": 250
    },
    {
      "epoch": 0.5445026178010471,
      "grad_norm": 305.6956787109375,
      "learning_rate": 2.5640697577740815e-07,
      "logits/chosen": -0.7282508611679077,
      "logits/rejected": -0.7263582944869995,
      "logps/chosen": -287.1412353515625,
      "logps/ref_chosen": -288.6393737792969,
      "logps/ref_rejected": -265.315673828125,
      "logps/rejected": -273.0039367675781,
      "loss": 4.111,
      "r_dpo/chosen_len": 271.48126220703125,
      "r_dpo/length_delta": 24.390625,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 247.0906219482422,
      "step": 260
    },
    {
      "epoch": 0.5654450261780105,
      "grad_norm": 248.96470642089844,
      "learning_rate": 2.381045210440644e-07,
      "logits/chosen": -0.7059388756752014,
      "logits/rejected": -0.7142220735549927,
      "logps/chosen": -279.7881774902344,
      "logps/ref_chosen": -280.1373596191406,
      "logps/ref_rejected": -264.84295654296875,
      "logps/rejected": -273.27484130859375,
      "loss": 4.011,
      "r_dpo/chosen_len": 272.2875061035156,
      "r_dpo/length_delta": 19.956249237060547,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 252.3312530517578,
      "step": 270
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 216.84811401367188,
      "learning_rate": 2.1986582993616925e-07,
      "logits/chosen": -0.7371554374694824,
      "logits/rejected": -0.7457467317581177,
      "logps/chosen": -300.7189636230469,
      "logps/ref_chosen": -301.8018798828125,
      "logps/ref_rejected": -254.75112915039062,
      "logps/rejected": -263.86175537109375,
      "loss": 3.9179,
      "r_dpo/chosen_len": 285.44061279296875,
      "r_dpo/length_delta": 52.962501525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 232.47811889648438,
      "step": 280
    },
    {
      "epoch": 0.6073298429319371,
      "grad_norm": 258.0394592285156,
      "learning_rate": 2.0178866775369774e-07,
      "logits/chosen": -0.7233304977416992,
      "logits/rejected": -0.7041932344436646,
      "logps/chosen": -303.56158447265625,
      "logps/ref_chosen": -302.79217529296875,
      "logps/ref_rejected": -292.9220275878906,
      "logps/rejected": -301.3799743652344,
      "loss": 4.2767,
      "r_dpo/chosen_len": 294.90625,
      "r_dpo/length_delta": 20.774999618530273,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 274.1312561035156,
      "step": 290
    },
    {
      "epoch": 0.6282722513089005,
      "grad_norm": 259.0774230957031,
      "learning_rate": 1.839699339491937e-07,
      "logits/chosen": -0.7432774305343628,
      "logits/rejected": -0.7294503450393677,
      "logps/chosen": -273.68316650390625,
      "logps/ref_chosen": -275.8238220214844,
      "logps/ref_rejected": -264.05743408203125,
      "logps/rejected": -269.7857666015625,
      "loss": 4.1741,
      "r_dpo/chosen_len": 266.859375,
      "r_dpo/length_delta": 20.734375,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 246.125,
      "step": 300
    },
    {
      "epoch": 0.6492146596858639,
      "grad_norm": 330.3496398925781,
      "learning_rate": 1.6650514271527465e-07,
      "logits/chosen": -0.7020508646965027,
      "logits/rejected": -0.7177096605300903,
      "logps/chosen": -294.5872802734375,
      "logps/ref_chosen": -296.6716003417969,
      "logps/ref_rejected": -278.68426513671875,
      "logps/rejected": -285.25396728515625,
      "loss": 4.015,
      "r_dpo/chosen_len": 292.91876220703125,
      "r_dpo/length_delta": 32.55937576293945,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 260.359375,
      "step": 310
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 258.3315734863281,
      "learning_rate": 1.4948791099758052e-07,
      "logits/chosen": -0.7137759923934937,
      "logits/rejected": -0.7293487191200256,
      "logps/chosen": -283.85919189453125,
      "logps/ref_chosen": -284.1717529296875,
      "logps/ref_rejected": -261.2606506347656,
      "logps/rejected": -270.0918884277344,
      "loss": 4.014,
      "r_dpo/chosen_len": 279.90313720703125,
      "r_dpo/length_delta": 44.537498474121094,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 235.36563110351562,
      "step": 320
    },
    {
      "epoch": 0.6910994764397905,
      "grad_norm": 268.5577087402344,
      "learning_rate": 1.3300945667758012e-07,
      "logits/chosen": -0.7268847227096558,
      "logits/rejected": -0.7360986471176147,
      "logps/chosen": -284.22406005859375,
      "logps/ref_chosen": -283.40338134765625,
      "logps/ref_rejected": -271.27569580078125,
      "logps/rejected": -280.9209899902344,
      "loss": 4.0598,
      "r_dpo/chosen_len": 267.67498779296875,
      "r_dpo/length_delta": 13.015625,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 254.6593780517578,
      "step": 330
    },
    {
      "epoch": 0.7120418848167539,
      "grad_norm": 247.73794555664062,
      "learning_rate": 1.1715810961514072e-07,
      "logits/chosen": -0.7288721203804016,
      "logits/rejected": -0.7343642115592957,
      "logps/chosen": -263.1542053222656,
      "logps/ref_chosen": -259.82537841796875,
      "logps/ref_rejected": -243.50222778320312,
      "logps/rejected": -255.48574829101562,
      "loss": 4.0065,
      "r_dpo/chosen_len": 256.11248779296875,
      "r_dpo/length_delta": 32.546875,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 223.5656280517578,
      "step": 340
    },
    {
      "epoch": 0.7329842931937173,
      "grad_norm": 270.27557373046875,
      "learning_rate": 1.0201883817182949e-07,
      "logits/chosen": -0.7445515990257263,
      "logits/rejected": -0.7549839615821838,
      "logps/chosen": -299.8614501953125,
      "logps/ref_chosen": -298.24725341796875,
      "logps/ref_rejected": -272.657958984375,
      "logps/rejected": -283.16290283203125,
      "loss": 4.0568,
      "r_dpo/chosen_len": 281.4624938964844,
      "r_dpo/length_delta": 45.275001525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 236.1875,
      "step": 350
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 239.69332885742188,
      "learning_rate": 8.76727937529367e-08,
      "logits/chosen": -0.7207680344581604,
      "logits/rejected": -0.7342051863670349,
      "logps/chosen": -282.36883544921875,
      "logps/ref_chosen": -281.881103515625,
      "logps/ref_rejected": -265.4746398925781,
      "logps/rejected": -275.1003112792969,
      "loss": 4.1996,
      "r_dpo/chosen_len": 272.64373779296875,
      "r_dpo/length_delta": 30.071874618530273,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 242.57186889648438,
      "step": 360
    },
    {
      "epoch": 0.774869109947644,
      "grad_norm": 198.85562133789062,
      "learning_rate": 7.419687580962222e-08,
      "logits/chosen": -0.7548901438713074,
      "logits/rejected": -0.7558341026306152,
      "logps/chosen": -302.01898193359375,
      "logps/ref_chosen": -302.17822265625,
      "logps/ref_rejected": -265.92877197265625,
      "logps/rejected": -275.6526794433594,
      "loss": 3.8979,
      "r_dpo/chosen_len": 273.88751220703125,
      "r_dpo/length_delta": 33.759376525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 240.1281280517578,
      "step": 370
    },
    {
      "epoch": 0.7958115183246073,
      "grad_norm": 286.9093933105469,
      "learning_rate": 6.166331963291519e-08,
      "logits/chosen": -0.7173658013343811,
      "logits/rejected": -0.7228809595108032,
      "logps/chosen": -300.5129699707031,
      "logps/ref_chosen": -301.07965087890625,
      "logps/ref_rejected": -266.7089538574219,
      "logps/rejected": -274.7568359375,
      "loss": 4.1149,
      "r_dpo/chosen_len": 286.75311279296875,
      "r_dpo/length_delta": 33.53125,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 253.2218780517578,
      "step": 380
    },
    {
      "epoch": 0.8167539267015707,
      "grad_norm": 256.02996826171875,
      "learning_rate": 5.013930914912476e-08,
      "logits/chosen": -0.7370638251304626,
      "logits/rejected": -0.7369734048843384,
      "logps/chosen": -295.3925476074219,
      "logps/ref_chosen": -296.6472473144531,
      "logps/ref_rejected": -278.953857421875,
      "logps/rejected": -287.25274658203125,
      "loss": 4.0032,
      "r_dpo/chosen_len": 287.91876220703125,
      "r_dpo/length_delta": 30.081249237060547,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 257.8374938964844,
      "step": 390
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 229.88453674316406,
      "learning_rate": 3.968661679220467e-08,
      "logits/chosen": -0.726976752281189,
      "logits/rejected": -0.7373378872871399,
      "logps/chosen": -295.95574951171875,
      "logps/ref_chosen": -296.6556091308594,
      "logps/ref_rejected": -256.9266662597656,
      "logps/rejected": -264.715576171875,
      "loss": 4.0423,
      "r_dpo/chosen_len": 278.96875,
      "r_dpo/length_delta": 39.60625076293945,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 239.3625030517578,
      "step": 400
    },
    {
      "epoch": 0.837696335078534,
      "eval_logits/chosen": -0.7442260980606079,
      "eval_logits/rejected": -0.7460020780563354,
      "eval_logps/chosen": -288.06787109375,
      "eval_logps/ref_chosen": -289.1346435546875,
      "eval_logps/ref_rejected": -264.7782287597656,
      "eval_logps/rejected": -272.9751281738281,
      "eval_loss": 0.5080092549324036,
      "eval_r_dpo/chosen_len": 291.2619934082031,
      "eval_r_dpo/length_delta": 42.86600112915039,
      "eval_r_dpo/regularization_term": 0.0,
      "eval_r_dpo/rejected_len": 248.39599609375,
      "eval_runtime": 82.0059,
      "eval_samples_per_second": 24.388,
      "eval_steps_per_second": 1.524,
      "step": 400
    },
    {
      "epoch": 0.8586387434554974,
      "grad_norm": 257.00933837890625,
      "learning_rate": 3.036127238347164e-08,
      "logits/chosen": -0.7183164358139038,
      "logits/rejected": -0.7075697183609009,
      "logps/chosen": -290.19342041015625,
      "logps/ref_chosen": -290.08154296875,
      "logps/ref_rejected": -272.56988525390625,
      "logps/rejected": -280.45965576171875,
      "loss": 4.0132,
      "r_dpo/chosen_len": 282.40625,
      "r_dpo/length_delta": 26.265625,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 256.140625,
      "step": 410
    },
    {
      "epoch": 0.8795811518324608,
      "grad_norm": 258.52166748046875,
      "learning_rate": 2.2213262793589482e-08,
      "logits/chosen": -0.7189906239509583,
      "logits/rejected": -0.7180750966072083,
      "logps/chosen": -306.213623046875,
      "logps/ref_chosen": -307.40240478515625,
      "logps/ref_rejected": -279.85760498046875,
      "logps/rejected": -287.7927551269531,
      "loss": 3.9865,
      "r_dpo/chosen_len": 296.8343811035156,
      "r_dpo/length_delta": 37.165626525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 259.66876220703125,
      "step": 420
    },
    {
      "epoch": 0.900523560209424,
      "grad_norm": 225.66854858398438,
      "learning_rate": 1.5286263996730026e-08,
      "logits/chosen": -0.7201143503189087,
      "logits/rejected": -0.7314451932907104,
      "logps/chosen": -297.2125244140625,
      "logps/ref_chosen": -297.7133483886719,
      "logps/ref_rejected": -266.862060546875,
      "logps/rejected": -276.1941223144531,
      "loss": 4.0823,
      "r_dpo/chosen_len": 290.703125,
      "r_dpo/length_delta": 48.868751525878906,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 241.83438110351562,
      "step": 430
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 296.7327880859375,
      "learning_rate": 9.617406953185136e-09,
      "logits/chosen": -0.7148677706718445,
      "logits/rejected": -0.7244837880134583,
      "logps/chosen": -293.4369201660156,
      "logps/ref_chosen": -293.84521484375,
      "logps/ref_rejected": -289.6698303222656,
      "logps/rejected": -297.15863037109375,
      "loss": 4.2486,
      "r_dpo/chosen_len": 285.3656311035156,
      "r_dpo/length_delta": 12.868749618530273,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 272.49688720703125,
      "step": 440
    },
    {
      "epoch": 0.9424083769633508,
      "grad_norm": 255.14761352539062,
      "learning_rate": 5.2370785753763356e-09,
      "logits/chosen": -0.735411524772644,
      "logits/rejected": -0.742519199848175,
      "logps/chosen": -295.8509826660156,
      "logps/ref_chosen": -297.1932373046875,
      "logps/ref_rejected": -262.9091796875,
      "logps/rejected": -271.23907470703125,
      "loss": 3.8827,
      "r_dpo/chosen_len": 282.81561279296875,
      "r_dpo/length_delta": 40.631248474121094,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 242.1843719482422,
      "step": 450
    },
    {
      "epoch": 0.9633507853403142,
      "grad_norm": 254.89694213867188,
      "learning_rate": 2.168758844148272e-09,
      "logits/chosen": -0.7343727946281433,
      "logits/rejected": -0.7490428686141968,
      "logps/chosen": -311.859619140625,
      "logps/ref_chosen": -312.42291259765625,
      "logps/ref_rejected": -278.7356262207031,
      "logps/rejected": -286.7757263183594,
      "loss": 4.1604,
      "r_dpo/chosen_len": 288.9125061035156,
      "r_dpo/length_delta": 43.23125076293945,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 245.68124389648438,
      "step": 460
    },
    {
      "epoch": 0.9842931937172775,
      "grad_norm": 231.78506469726562,
      "learning_rate": 4.288949484559934e-10,
      "logits/chosen": -0.7186932563781738,
      "logits/rejected": -0.7156038284301758,
      "logps/chosen": -276.20965576171875,
      "logps/ref_chosen": -278.13385009765625,
      "logps/ref_rejected": -256.70648193359375,
      "logps/rejected": -264.2013244628906,
      "loss": 3.9555,
      "r_dpo/chosen_len": 268.8687438964844,
      "r_dpo/length_delta": 26.262500762939453,
      "r_dpo/regularization_term": 0.0,
      "r_dpo/rejected_len": 242.6062469482422,
      "step": 470
    },
    {
      "epoch": 0.9989528795811519,
      "step": 477,
      "total_flos": 0.0,
      "train_loss": 4.32008807264284,
      "train_runtime": 6099.6278,
      "train_samples_per_second": 10.023,
      "train_steps_per_second": 0.078
    }
  ],
  "logging_steps": 10,
  "max_steps": 477,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}