{
  "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": 37.91044616699219,
      "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": 43.6518,
      "r_dpo/chosen_len": 257.75,
      "r_dpo/length_delta": 47.875,
      "r_dpo/regularization_term": 4.787500381469727,
      "r_dpo/rejected_len": 209.875,
      "step": 1
    },
    {
      "epoch": 0.020942408376963352,
      "grad_norm": 34.83302307128906,
      "learning_rate": 9.375e-08,
      "logits/chosen": -0.6325575709342957,
      "logits/rejected": -0.6374133229255676,
      "logps/chosen": -296.6266174316406,
      "logps/ref_chosen": -296.63226318359375,
      "logps/ref_rejected": -258.9539489746094,
      "logps/rejected": -259.0103454589844,
      "loss": 41.9209,
      "r_dpo/chosen_len": 291.8680419921875,
      "r_dpo/length_delta": 49.76388931274414,
      "r_dpo/regularization_term": 4.976388931274414,
      "r_dpo/rejected_len": 242.1041717529297,
      "step": 10
    },
    {
      "epoch": 0.041884816753926704,
      "grad_norm": 35.58102035522461,
      "learning_rate": 1.9791666666666664e-07,
      "logits/chosen": -0.59663987159729,
      "logits/rejected": -0.627370297908783,
      "logps/chosen": -297.972412109375,
      "logps/ref_chosen": -297.9349365234375,
      "logps/ref_rejected": -256.9902648925781,
      "logps/rejected": -257.0819396972656,
      "loss": 41.1855,
      "r_dpo/chosen_len": 291.29998779296875,
      "r_dpo/length_delta": 52.89374923706055,
      "r_dpo/regularization_term": 5.289375305175781,
      "r_dpo/rejected_len": 238.40625,
      "step": 20
    },
    {
      "epoch": 0.06282722513089005,
      "grad_norm": 40.33411407470703,
      "learning_rate": 3.020833333333333e-07,
      "logits/chosen": -0.6156569719314575,
      "logits/rejected": -0.6075023412704468,
      "logps/chosen": -278.87744140625,
      "logps/ref_chosen": -278.64752197265625,
      "logps/ref_rejected": -249.309814453125,
      "logps/rejected": -249.558837890625,
      "loss": 43.5709,
      "r_dpo/chosen_len": 270.8812561035156,
      "r_dpo/length_delta": 25.228124618530273,
      "r_dpo/regularization_term": 2.522813081741333,
      "r_dpo/rejected_len": 245.6531219482422,
      "step": 30
    },
    {
      "epoch": 0.08376963350785341,
      "grad_norm": 32.58840560913086,
      "learning_rate": 4.0625e-07,
      "logits/chosen": -0.6117347478866577,
      "logits/rejected": -0.6369837522506714,
      "logps/chosen": -284.4401550292969,
      "logps/ref_chosen": -283.49981689453125,
      "logps/ref_rejected": -265.32733154296875,
      "logps/rejected": -266.5763244628906,
      "loss": 41.7009,
      "r_dpo/chosen_len": 281.43438720703125,
      "r_dpo/length_delta": 33.34375,
      "r_dpo/regularization_term": 3.3343749046325684,
      "r_dpo/rejected_len": 248.0906219482422,
      "step": 40
    },
    {
      "epoch": 0.10471204188481675,
      "grad_norm": 37.238525390625,
      "learning_rate": 4.999932966293553e-07,
      "logits/chosen": -0.6101135611534119,
      "logits/rejected": -0.651242733001709,
      "logps/chosen": -283.24017333984375,
      "logps/ref_chosen": -280.224365234375,
      "logps/ref_rejected": -274.3541259765625,
      "logps/rejected": -278.0672302246094,
      "loss": 43.8578,
      "r_dpo/chosen_len": 290.32501220703125,
      "r_dpo/length_delta": 35.11249923706055,
      "r_dpo/regularization_term": 3.5112500190734863,
      "r_dpo/rejected_len": 255.21249389648438,
      "step": 50
    },
    {
      "epoch": 0.1256544502617801,
      "grad_norm": 44.93247604370117,
      "learning_rate": 4.991893270335525e-07,
      "logits/chosen": -0.6015418767929077,
      "logits/rejected": -0.6169525384902954,
      "logps/chosen": -289.905029296875,
      "logps/ref_chosen": -281.12664794921875,
      "logps/ref_rejected": -259.86456298828125,
      "logps/rejected": -269.1077575683594,
      "loss": 46.1321,
      "r_dpo/chosen_len": 273.953125,
      "r_dpo/length_delta": 29.084375381469727,
      "r_dpo/regularization_term": 2.908437728881836,
      "r_dpo/rejected_len": 244.86874389648438,
      "step": 60
    },
    {
      "epoch": 0.14659685863874344,
      "grad_norm": 45.56925964355469,
      "learning_rate": 4.970496218214204e-07,
      "logits/chosen": -0.6171663403511047,
      "logits/rejected": -0.6253775954246521,
      "logps/chosen": -308.27239990234375,
      "logps/ref_chosen": -287.71063232421875,
      "logps/ref_rejected": -276.839599609375,
      "logps/rejected": -298.0381164550781,
      "loss": 42.0532,
      "r_dpo/chosen_len": 267.4937438964844,
      "r_dpo/length_delta": 14.484375,
      "r_dpo/regularization_term": 1.4484375715255737,
      "r_dpo/rejected_len": 253.00936889648438,
      "step": 70
    },
    {
      "epoch": 0.16753926701570682,
      "grad_norm": 55.37849426269531,
      "learning_rate": 4.935856505068998e-07,
      "logits/chosen": -0.5272337198257446,
      "logits/rejected": -0.5274912118911743,
      "logps/chosen": -323.0798645019531,
      "logps/ref_chosen": -280.123046875,
      "logps/ref_rejected": -258.8989562988281,
      "logps/rejected": -299.3633117675781,
      "loss": 44.0294,
      "r_dpo/chosen_len": 267.4781188964844,
      "r_dpo/length_delta": 32.46562576293945,
      "r_dpo/regularization_term": 3.2465622425079346,
      "r_dpo/rejected_len": 235.0124969482422,
      "step": 80
    },
    {
      "epoch": 0.18848167539267016,
      "grad_norm": 100.971923828125,
      "learning_rate": 4.8881598109976e-07,
      "logits/chosen": -0.46989256143569946,
      "logits/rejected": -0.482027530670166,
      "logps/chosen": -357.15936279296875,
      "logps/ref_chosen": -278.02545166015625,
      "logps/ref_rejected": -251.0922393798828,
      "logps/rejected": -327.29400634765625,
      "loss": 41.8973,
      "r_dpo/chosen_len": 274.20623779296875,
      "r_dpo/length_delta": 44.97187423706055,
      "r_dpo/regularization_term": 4.497187614440918,
      "r_dpo/rejected_len": 229.234375,
      "step": 90
    },
    {
      "epoch": 0.2094240837696335,
      "grad_norm": 248.96990966796875,
      "learning_rate": 4.827661805750437e-07,
      "logits/chosen": -0.40429097414016724,
      "logits/rejected": -0.41918930411338806,
      "logps/chosen": -423.6812438964844,
      "logps/ref_chosen": -274.0089416503906,
      "logps/ref_rejected": -274.14447021484375,
      "logps/rejected": -434.80133056640625,
      "loss": 40.9219,
      "r_dpo/chosen_len": 275.3343811035156,
      "r_dpo/length_delta": 21.912500381469727,
      "r_dpo/regularization_term": 2.1912498474121094,
      "r_dpo/rejected_len": 253.421875,
      "step": 100
    },
    {
      "epoch": 0.23036649214659685,
      "grad_norm": 1346.3997802734375,
      "learning_rate": 4.75468677825789e-07,
      "logits/chosen": -0.35518646240234375,
      "logits/rejected": -0.3851023316383362,
      "logps/chosen": -763.7197265625,
      "logps/ref_chosen": -273.23333740234375,
      "logps/ref_rejected": -263.88787841796875,
      "logps/rejected": -708.5206909179688,
      "loss": 31.9332,
      "r_dpo/chosen_len": 283.43438720703125,
      "r_dpo/length_delta": 50.34375,
      "r_dpo/regularization_term": 5.034375190734863,
      "r_dpo/rejected_len": 233.0906219482422,
      "step": 110
    },
    {
      "epoch": 0.2513089005235602,
      "grad_norm": 7287.37548828125,
      "learning_rate": 4.669625898336438e-07,
      "logits/chosen": 0.0007931709405966103,
      "logits/rejected": -0.02699187956750393,
      "logps/chosen": -2300.340576171875,
      "logps/ref_chosen": -269.77142333984375,
      "logps/ref_rejected": -272.7685546875,
      "logps/rejected": -2165.2236328125,
      "loss": 18.6284,
      "r_dpo/chosen_len": 264.7593688964844,
      "r_dpo/length_delta": 13.840624809265137,
      "r_dpo/regularization_term": 1.3840625286102295,
      "r_dpo/rejected_len": 250.9187469482422,
      "step": 120
    },
    {
      "epoch": 0.27225130890052357,
      "grad_norm": 2478.686767578125,
      "learning_rate": 4.5729351198915705e-07,
      "logits/chosen": -0.08189640194177628,
      "logits/rejected": -0.0916411355137825,
      "logps/chosen": -2413.971923828125,
      "logps/ref_chosen": -275.03448486328125,
      "logps/ref_rejected": -276.39862060546875,
      "logps/rejected": -2312.552734375,
      "loss": 13.7684,
      "r_dpo/chosen_len": 266.625,
      "r_dpo/length_delta": 18.668750762939453,
      "r_dpo/regularization_term": 1.8668749332427979,
      "r_dpo/rejected_len": 247.9562530517578,
      "step": 130
    },
    {
      "epoch": 0.2931937172774869,
      "grad_norm": 1929.59423828125,
      "learning_rate": 4.4651327368569684e-07,
      "logits/chosen": -0.08853980153799057,
      "logits/rejected": -0.12103728950023651,
      "logps/chosen": -2533.98876953125,
      "logps/ref_chosen": -276.0029602050781,
      "logps/ref_rejected": -255.9320526123047,
      "logps/rejected": -2369.556884765625,
      "loss": 10.996,
      "r_dpo/chosen_len": 261.46875,
      "r_dpo/length_delta": 22.375,
      "r_dpo/regularization_term": 2.237499952316284,
      "r_dpo/rejected_len": 239.09375,
      "step": 140
    },
    {
      "epoch": 0.31413612565445026,
      "grad_norm": 1273.77880859375,
      "learning_rate": 4.346796604970912e-07,
      "logits/chosen": -0.03744601085782051,
      "logits/rejected": -0.0864439308643341,
      "logps/chosen": -2928.5224609375,
      "logps/ref_chosen": -298.2093505859375,
      "logps/ref_rejected": -254.8907012939453,
      "logps/rejected": -2446.001953125,
      "loss": 8.5066,
      "r_dpo/chosen_len": 283.84375,
      "r_dpo/length_delta": 48.359375,
      "r_dpo/regularization_term": 4.8359375,
      "r_dpo/rejected_len": 235.484375,
      "step": 150
    },
    {
      "epoch": 0.33507853403141363,
      "grad_norm": 637.610595703125,
      "learning_rate": 4.218561044282098e-07,
      "logits/chosen": -0.11696837842464447,
      "logits/rejected": -0.14894640445709229,
      "logps/chosen": -2785.264892578125,
      "logps/ref_chosen": -281.94189453125,
      "logps/ref_rejected": -255.5653533935547,
      "logps/rejected": -2388.72412109375,
      "loss": 7.7028,
      "r_dpo/chosen_len": 267.828125,
      "r_dpo/length_delta": 41.368751525878906,
      "r_dpo/regularization_term": 4.136875152587891,
      "r_dpo/rejected_len": 226.45938110351562,
      "step": 160
    },
    {
      "epoch": 0.35602094240837695,
      "grad_norm": 1146.212646484375,
      "learning_rate": 4.081113438988443e-07,
      "logits/chosen": -0.19355528056621552,
      "logits/rejected": -0.21304742991924286,
      "logps/chosen": -2857.443359375,
      "logps/ref_chosen": -288.2863464355469,
      "logps/ref_rejected": -239.758056640625,
      "logps/rejected": -2423.08349609375,
      "loss": 6.7267,
      "r_dpo/chosen_len": 285.203125,
      "r_dpo/length_delta": 46.396873474121094,
      "r_dpo/regularization_term": 4.639687538146973,
      "r_dpo/rejected_len": 238.80624389648438,
      "step": 170
    },
    {
      "epoch": 0.3769633507853403,
      "grad_norm": 608.8386840820312,
      "learning_rate": 3.935190552834828e-07,
      "logits/chosen": -0.3072236180305481,
      "logits/rejected": -0.32055360078811646,
      "logps/chosen": -2654.16943359375,
      "logps/ref_chosen": -286.17889404296875,
      "logps/ref_rejected": -249.9820098876953,
      "logps/rejected": -2274.89892578125,
      "loss": 6.9642,
      "r_dpo/chosen_len": 266.09063720703125,
      "r_dpo/length_delta": 40.12812423706055,
      "r_dpo/regularization_term": 4.012812614440918,
      "r_dpo/rejected_len": 225.96249389648438,
      "step": 180
    },
    {
      "epoch": 0.39790575916230364,
      "grad_norm": 977.245361328125,
      "learning_rate": 3.781574579820464e-07,
      "logits/chosen": -0.33928078413009644,
      "logits/rejected": -0.34710997343063354,
      "logps/chosen": -2732.35009765625,
      "logps/ref_chosen": -280.9278259277344,
      "logps/ref_rejected": -254.3533477783203,
      "logps/rejected": -2360.5947265625,
      "loss": 6.7039,
      "r_dpo/chosen_len": 276.33123779296875,
      "r_dpo/length_delta": 41.993751525878906,
      "r_dpo/regularization_term": 4.199374675750732,
      "r_dpo/rejected_len": 234.33749389648438,
      "step": 190
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 616.4851684570312,
      "learning_rate": 3.621088951385353e-07,
      "logits/chosen": -0.2376505434513092,
      "logits/rejected": -0.25424009561538696,
      "logps/chosen": -2470.09521484375,
      "logps/ref_chosen": -253.1712188720703,
      "logps/ref_rejected": -241.90478515625,
      "logps/rejected": -2249.572021484375,
      "loss": 6.4185,
      "r_dpo/chosen_len": 248.0749969482422,
      "r_dpo/length_delta": 28.131250381469727,
      "r_dpo/regularization_term": 2.8131251335144043,
      "r_dpo/rejected_len": 219.94375610351562,
      "step": 200
    },
    {
      "epoch": 0.418848167539267,
      "eval_logits/chosen": -0.22856707870960236,
      "eval_logits/rejected": -0.23529842495918274,
      "eval_logps/chosen": -2812.3984375,
      "eval_logps/ref_chosen": -288.6414794921875,
      "eval_logps/ref_rejected": -265.96160888671875,
      "eval_logps/rejected": -2464.037109375,
      "eval_loss": 0.7758205533027649,
      "eval_r_dpo/chosen_len": 286.97601318359375,
      "eval_r_dpo/length_delta": 40.88800048828125,
      "eval_r_dpo/regularization_term": 4.088800430297852,
      "eval_r_dpo/rejected_len": 246.08799743652344,
      "eval_runtime": 209.8859,
      "eval_samples_per_second": 9.529,
      "eval_steps_per_second": 1.191,
      "step": 200
    },
    {
      "epoch": 0.4397905759162304,
      "grad_norm": 846.9074096679688,
      "learning_rate": 3.454593922550693e-07,
      "logits/chosen": -0.3170786499977112,
      "logits/rejected": -0.32814693450927734,
      "logps/chosen": -2884.038330078125,
      "logps/ref_chosen": -287.9228210449219,
      "logps/ref_rejected": -263.35595703125,
      "logps/rejected": -2537.31689453125,
      "loss": 6.545,
      "r_dpo/chosen_len": 280.3125,
      "r_dpo/length_delta": 36.68437576293945,
      "r_dpo/regularization_term": 3.6684374809265137,
      "r_dpo/rejected_len": 243.6281280517578,
      "step": 210
    },
    {
      "epoch": 0.4607329842931937,
      "grad_norm": 546.0598754882812,
      "learning_rate": 3.2829819606729477e-07,
      "logits/chosen": -0.3490475118160248,
      "logits/rejected": -0.35787373781204224,
      "logps/chosen": -2633.925048828125,
      "logps/ref_chosen": -282.3331604003906,
      "logps/ref_rejected": -272.5645446777344,
      "logps/rejected": -2502.832763671875,
      "loss": 6.0607,
      "r_dpo/chosen_len": 261.359375,
      "r_dpo/length_delta": 17.865625381469727,
      "r_dpo/regularization_term": 1.7865626811981201,
      "r_dpo/rejected_len": 243.49374389648438,
      "step": 220
    },
    {
      "epoch": 0.4816753926701571,
      "grad_norm": 405.8363037109375,
      "learning_rate": 3.1071729615293424e-07,
      "logits/chosen": -0.36600062251091003,
      "logits/rejected": -0.3698672950267792,
      "logps/chosen": -2655.687255859375,
      "logps/ref_chosen": -276.1485595703125,
      "logps/ref_rejected": -252.81198120117188,
      "logps/rejected": -2387.78125,
      "loss": 5.9948,
      "r_dpo/chosen_len": 264.43438720703125,
      "r_dpo/length_delta": 31.256250381469727,
      "r_dpo/regularization_term": 3.1256251335144043,
      "r_dpo/rejected_len": 233.17813110351562,
      "step": 230
    },
    {
      "epoch": 0.5026178010471204,
      "grad_norm": 502.72357177734375,
      "learning_rate": 2.9281093183781403e-07,
      "logits/chosen": -0.3056614398956299,
      "logits/rejected": -0.3123835325241089,
      "logps/chosen": -2676.06982421875,
      "logps/ref_chosen": -270.52520751953125,
      "logps/ref_rejected": -254.83334350585938,
      "logps/rejected": -2375.5478515625,
      "loss": 6.2228,
      "r_dpo/chosen_len": 271.81561279296875,
      "r_dpo/length_delta": 37.099998474121094,
      "r_dpo/regularization_term": 3.7100002765655518,
      "r_dpo/rejected_len": 234.7156219482422,
      "step": 240
    },
    {
      "epoch": 0.5235602094240838,
      "grad_norm": 619.77783203125,
      "learning_rate": 2.7467508704251135e-07,
      "logits/chosen": -0.33823755383491516,
      "logits/rejected": -0.33679407835006714,
      "logps/chosen": -2791.576171875,
      "logps/ref_chosen": -289.6054992675781,
      "logps/ref_rejected": -265.0482482910156,
      "logps/rejected": -2412.063720703125,
      "loss": 6.0215,
      "r_dpo/chosen_len": 277.50311279296875,
      "r_dpo/length_delta": 41.103126525878906,
      "r_dpo/regularization_term": 4.110312461853027,
      "r_dpo/rejected_len": 236.39999389648438,
      "step": 250
    },
    {
      "epoch": 0.5445026178010471,
      "grad_norm": 730.1420288085938,
      "learning_rate": 2.5640697577740815e-07,
      "logits/chosen": -0.3466644883155823,
      "logits/rejected": -0.34878239035606384,
      "logps/chosen": -2722.85693359375,
      "logps/ref_chosen": -288.6393737792969,
      "logps/ref_rejected": -265.315673828125,
      "logps/rejected": -2532.46484375,
      "loss": 5.8667,
      "r_dpo/chosen_len": 271.48126220703125,
      "r_dpo/length_delta": 24.390625,
      "r_dpo/regularization_term": 2.4390625953674316,
      "r_dpo/rejected_len": 247.0906219482422,
      "step": 260
    },
    {
      "epoch": 0.5654450261780105,
      "grad_norm": 586.3690795898438,
      "learning_rate": 2.381045210440644e-07,
      "logits/chosen": -0.3886115252971649,
      "logits/rejected": -0.3948022127151489,
      "logps/chosen": -2752.974853515625,
      "logps/ref_chosen": -280.1373596191406,
      "logps/ref_rejected": -264.84295654296875,
      "logps/rejected": -2619.281494140625,
      "loss": 5.7602,
      "r_dpo/chosen_len": 272.2875061035156,
      "r_dpo/length_delta": 19.956249237060547,
      "r_dpo/regularization_term": 1.9956251382827759,
      "r_dpo/rejected_len": 252.3312530517578,
      "step": 270
    },
    {
      "epoch": 0.5863874345549738,
      "grad_norm": 440.32672119140625,
      "learning_rate": 2.1986582993616925e-07,
      "logits/chosen": -0.38749268651008606,
      "logits/rejected": -0.3799517750740051,
      "logps/chosen": -2869.620849609375,
      "logps/ref_chosen": -301.7547912597656,
      "logps/ref_rejected": -254.6543731689453,
      "logps/rejected": -2384.27294921875,
      "loss": 5.5612,
      "r_dpo/chosen_len": 285.44061279296875,
      "r_dpo/length_delta": 52.962501525878906,
      "r_dpo/regularization_term": 5.296250343322754,
      "r_dpo/rejected_len": 232.47811889648438,
      "step": 280
    },
    {
      "epoch": 0.6073298429319371,
      "grad_norm": 649.3321533203125,
      "learning_rate": 2.0178866775369774e-07,
      "logits/chosen": -0.39530712366104126,
      "logits/rejected": -0.3784952759742737,
      "logps/chosen": -2955.660400390625,
      "logps/ref_chosen": -302.79217529296875,
      "logps/ref_rejected": -292.9220275878906,
      "logps/rejected": -2789.33447265625,
      "loss": 5.8331,
      "r_dpo/chosen_len": 294.90625,
      "r_dpo/length_delta": 20.774999618530273,
      "r_dpo/regularization_term": 2.077500104904175,
      "r_dpo/rejected_len": 274.1312561035156,
      "step": 290
    },
    {
      "epoch": 0.6282722513089005,
      "grad_norm": 495.6314697265625,
      "learning_rate": 1.839699339491937e-07,
      "logits/chosen": -0.3710184097290039,
      "logits/rejected": -0.36620840430259705,
      "logps/chosen": -2728.482421875,
      "logps/ref_chosen": -275.8238220214844,
      "logps/ref_rejected": -264.05743408203125,
      "logps/rejected": -2575.94580078125,
      "loss": 5.8295,
      "r_dpo/chosen_len": 266.859375,
      "r_dpo/length_delta": 20.734375,
      "r_dpo/regularization_term": 2.073437452316284,
      "r_dpo/rejected_len": 246.125,
      "step": 300
    },
    {
      "epoch": 0.6492146596858639,
      "grad_norm": 494.1045837402344,
      "learning_rate": 1.6650514271527465e-07,
      "logits/chosen": -0.3966061472892761,
      "logits/rejected": -0.3880545496940613,
      "logps/chosen": -3014.114013671875,
      "logps/ref_chosen": -296.6716003417969,
      "logps/ref_rejected": -278.68426513671875,
      "logps/rejected": -2734.437255859375,
      "loss": 5.8626,
      "r_dpo/chosen_len": 292.91876220703125,
      "r_dpo/length_delta": 32.55937576293945,
      "r_dpo/regularization_term": 3.2559375762939453,
      "r_dpo/rejected_len": 260.359375,
      "step": 310
    },
    {
      "epoch": 0.6701570680628273,
      "grad_norm": 792.8230590820312,
      "learning_rate": 1.4948791099758052e-07,
      "logits/chosen": -0.39138588309288025,
      "logits/rejected": -0.3824613690376282,
      "logps/chosen": -2841.270751953125,
      "logps/ref_chosen": -284.1717529296875,
      "logps/ref_rejected": -261.2606506347656,
      "logps/rejected": -2460.640869140625,
      "loss": 5.3257,
      "r_dpo/chosen_len": 279.90313720703125,
      "r_dpo/length_delta": 44.537498474121094,
      "r_dpo/regularization_term": 4.453750133514404,
      "r_dpo/rejected_len": 235.36563110351562,
      "step": 320
    },
    {
      "epoch": 0.6910994764397905,
      "grad_norm": 460.28485107421875,
      "learning_rate": 1.3300945667758012e-07,
      "logits/chosen": -0.40215617418289185,
      "logits/rejected": -0.40834659337997437,
      "logps/chosen": -2728.718017578125,
      "logps/ref_chosen": -283.40338134765625,
      "logps/ref_rejected": -271.27569580078125,
      "logps/rejected": -2660.712890625,
      "loss": 5.4139,
      "r_dpo/chosen_len": 267.67498779296875,
      "r_dpo/length_delta": 13.015625,
      "r_dpo/regularization_term": 1.3015625476837158,
      "r_dpo/rejected_len": 254.6593780517578,
      "step": 330
    },
    {
      "epoch": 0.7120418848167539,
      "grad_norm": 641.794189453125,
      "learning_rate": 1.1715810961514072e-07,
      "logits/chosen": -0.38855430483818054,
      "logits/rejected": -0.37820714712142944,
      "logps/chosen": -2629.27197265625,
      "logps/ref_chosen": -259.7261962890625,
      "logps/ref_rejected": -243.4088897705078,
      "logps/rejected": -2364.59521484375,
      "loss": 5.6332,
      "r_dpo/chosen_len": 256.11248779296875,
      "r_dpo/length_delta": 32.546875,
      "r_dpo/regularization_term": 3.254687547683716,
      "r_dpo/rejected_len": 223.5656280517578,
      "step": 340
    },
    {
      "epoch": 0.7329842931937173,
      "grad_norm": 610.6142578125,
      "learning_rate": 1.0201883817182949e-07,
      "logits/chosen": -0.39518895745277405,
      "logits/rejected": -0.3816191256046295,
      "logps/chosen": -2879.398681640625,
      "logps/ref_chosen": -298.24725341796875,
      "logps/ref_rejected": -272.657958984375,
      "logps/rejected": -2458.894775390625,
      "loss": 5.4022,
      "r_dpo/chosen_len": 281.4624938964844,
      "r_dpo/length_delta": 45.275001525878906,
      "r_dpo/regularization_term": 4.527500152587891,
      "r_dpo/rejected_len": 236.1875,
      "step": 350
    },
    {
      "epoch": 0.7539267015706806,
      "grad_norm": 533.6195678710938,
      "learning_rate": 8.76727937529367e-08,
      "logits/chosen": -0.36005398631095886,
      "logits/rejected": -0.3565328121185303,
      "logps/chosen": -2746.4658203125,
      "logps/ref_chosen": -281.881103515625,
      "logps/ref_rejected": -265.4746398925781,
      "logps/rejected": -2505.47509765625,
      "loss": 5.5109,
      "r_dpo/chosen_len": 272.64373779296875,
      "r_dpo/length_delta": 30.071874618530273,
      "r_dpo/regularization_term": 3.0071873664855957,
      "r_dpo/rejected_len": 242.57186889648438,
      "step": 360
    },
    {
      "epoch": 0.774869109947644,
      "grad_norm": 462.19488525390625,
      "learning_rate": 7.419687580962222e-08,
      "logits/chosen": -0.4140191674232483,
      "logits/rejected": -0.4048995077610016,
      "logps/chosen": -2749.48779296875,
      "logps/ref_chosen": -302.17822265625,
      "logps/ref_rejected": -265.92877197265625,
      "logps/rejected": -2479.15966796875,
      "loss": 5.1316,
      "r_dpo/chosen_len": 273.88751220703125,
      "r_dpo/length_delta": 33.759376525878906,
      "r_dpo/regularization_term": 3.3759372234344482,
      "r_dpo/rejected_len": 240.1281280517578,
      "step": 370
    },
    {
      "epoch": 0.7958115183246073,
      "grad_norm": 304.5794372558594,
      "learning_rate": 6.166331963291519e-08,
      "logits/chosen": -0.3827962875366211,
      "logits/rejected": -0.3733735680580139,
      "logps/chosen": -2883.859619140625,
      "logps/ref_chosen": -301.2120361328125,
      "logps/ref_rejected": -266.4872741699219,
      "logps/rejected": -2585.99365234375,
      "loss": 5.3235,
      "r_dpo/chosen_len": 286.75311279296875,
      "r_dpo/length_delta": 33.53125,
      "r_dpo/regularization_term": 3.3531250953674316,
      "r_dpo/rejected_len": 253.2218780517578,
      "step": 380
    },
    {
      "epoch": 0.8167539267015707,
      "grad_norm": 505.1428527832031,
      "learning_rate": 5.013930914912476e-08,
      "logits/chosen": -0.39441806077957153,
      "logits/rejected": -0.3891170024871826,
      "logps/chosen": -2894.778076171875,
      "logps/ref_chosen": -296.6472473144531,
      "logps/ref_rejected": -278.953857421875,
      "logps/rejected": -2663.129150390625,
      "loss": 5.7398,
      "r_dpo/chosen_len": 287.91876220703125,
      "r_dpo/length_delta": 30.081249237060547,
      "r_dpo/regularization_term": 3.008124828338623,
      "r_dpo/rejected_len": 257.8374938964844,
      "step": 390
    },
    {
      "epoch": 0.837696335078534,
      "grad_norm": 815.798828125,
      "learning_rate": 3.968661679220467e-08,
      "logits/chosen": -0.3783096373081207,
      "logits/rejected": -0.37295296788215637,
      "logps/chosen": -2797.49951171875,
      "logps/ref_chosen": -296.6556091308594,
      "logps/ref_rejected": -256.9266662597656,
      "logps/rejected": -2443.60888671875,
      "loss": 5.4191,
      "r_dpo/chosen_len": 278.96875,
      "r_dpo/length_delta": 39.60625076293945,
      "r_dpo/regularization_term": 3.960625171661377,
      "r_dpo/rejected_len": 239.3625030517578,
      "step": 400
    },
    {
      "epoch": 0.837696335078534,
      "eval_logits/chosen": -0.3396916091442108,
      "eval_logits/rejected": -0.3239772915840149,
      "eval_logps/chosen": -2847.308349609375,
      "eval_logps/ref_chosen": -288.6414794921875,
      "eval_logps/ref_rejected": -265.96160888671875,
      "eval_logps/rejected": -2499.736328125,
      "eval_loss": 0.6649385690689087,
      "eval_r_dpo/chosen_len": 286.97601318359375,
      "eval_r_dpo/length_delta": 40.88800048828125,
      "eval_r_dpo/regularization_term": 4.088800430297852,
      "eval_r_dpo/rejected_len": 246.08799743652344,
      "eval_runtime": 211.0607,
      "eval_samples_per_second": 9.476,
      "eval_steps_per_second": 1.184,
      "step": 400
    },
    {
      "epoch": 0.8586387434554974,
      "grad_norm": 450.2701721191406,
      "learning_rate": 3.036127238347164e-08,
      "logits/chosen": -0.38624387979507446,
      "logits/rejected": -0.3744073510169983,
      "logps/chosen": -2840.36572265625,
      "logps/ref_chosen": -289.9568786621094,
      "logps/ref_rejected": -272.4674377441406,
      "logps/rejected": -2619.307861328125,
      "loss": 5.3372,
      "r_dpo/chosen_len": 282.40625,
      "r_dpo/length_delta": 26.265625,
      "r_dpo/regularization_term": 2.6265625953674316,
      "r_dpo/rejected_len": 256.140625,
      "step": 410
    },
    {
      "epoch": 0.8795811518324608,
      "grad_norm": 457.4356384277344,
      "learning_rate": 2.2213262793589482e-08,
      "logits/chosen": -0.3804797828197479,
      "logits/rejected": -0.36465829610824585,
      "logps/chosen": -3006.153564453125,
      "logps/ref_chosen": -307.40240478515625,
      "logps/ref_rejected": -279.85760498046875,
      "logps/rejected": -2696.711669921875,
      "loss": 5.0272,
      "r_dpo/chosen_len": 296.8343811035156,
      "r_dpo/length_delta": 37.165626525878906,
      "r_dpo/regularization_term": 3.7165627479553223,
      "r_dpo/rejected_len": 259.66876220703125,
      "step": 420
    },
    {
      "epoch": 0.900523560209424,
      "grad_norm": 489.8297119140625,
      "learning_rate": 1.5286263996730026e-08,
      "logits/chosen": -0.3834610879421234,
      "logits/rejected": -0.37544751167297363,
      "logps/chosen": -2919.646484375,
      "logps/ref_chosen": -297.7133483886719,
      "logps/ref_rejected": -266.862060546875,
      "logps/rejected": -2492.005859375,
      "loss": 5.2819,
      "r_dpo/chosen_len": 290.703125,
      "r_dpo/length_delta": 48.868751525878906,
      "r_dpo/regularization_term": 4.886875152587891,
      "r_dpo/rejected_len": 241.83438110351562,
      "step": 430
    },
    {
      "epoch": 0.9214659685863874,
      "grad_norm": 458.5052795410156,
      "learning_rate": 9.617406953185136e-09,
      "logits/chosen": -0.38718608021736145,
      "logits/rejected": -0.3853945732116699,
      "logps/chosen": -2841.03955078125,
      "logps/ref_chosen": -293.67095947265625,
      "logps/ref_rejected": -289.4698791503906,
      "logps/rejected": -2790.7744140625,
      "loss": 5.4764,
      "r_dpo/chosen_len": 285.3656311035156,
      "r_dpo/length_delta": 12.868749618530273,
      "r_dpo/regularization_term": 1.286874771118164,
      "r_dpo/rejected_len": 272.49688720703125,
      "step": 440
    },
    {
      "epoch": 0.9424083769633508,
      "grad_norm": 631.5553588867188,
      "learning_rate": 5.2370785753763356e-09,
      "logits/chosen": -0.39063310623168945,
      "logits/rejected": -0.3791298270225525,
      "logps/chosen": -2851.560546875,
      "logps/ref_chosen": -296.9415283203125,
      "logps/ref_rejected": -262.6710510253906,
      "logps/rejected": -2496.45947265625,
      "loss": 5.2111,
      "r_dpo/chosen_len": 282.81561279296875,
      "r_dpo/length_delta": 40.631248474121094,
      "r_dpo/regularization_term": 4.063125133514404,
      "r_dpo/rejected_len": 242.1843719482422,
      "step": 450
    },
    {
      "epoch": 0.9633507853403142,
      "grad_norm": 656.00634765625,
      "learning_rate": 2.168758844148272e-09,
      "logits/chosen": -0.3994686007499695,
      "logits/rejected": -0.3830736577510834,
      "logps/chosen": -2940.84423828125,
      "logps/ref_chosen": -312.42291259765625,
      "logps/ref_rejected": -278.7356262207031,
      "logps/rejected": -2538.37939453125,
      "loss": 5.6979,
      "r_dpo/chosen_len": 288.9125061035156,
      "r_dpo/length_delta": 43.23125076293945,
      "r_dpo/regularization_term": 4.323124885559082,
      "r_dpo/rejected_len": 245.68124389648438,
      "step": 460
    },
    {
      "epoch": 0.9842931937172775,
      "grad_norm": 382.99267578125,
      "learning_rate": 4.288949484559934e-10,
      "logits/chosen": -0.3687112331390381,
      "logits/rejected": -0.35538098216056824,
      "logps/chosen": -2719.51953125,
      "logps/ref_chosen": -278.0654602050781,
      "logps/ref_rejected": -256.5596618652344,
      "logps/rejected": -2522.361083984375,
      "loss": 5.2482,
      "r_dpo/chosen_len": 268.8687438964844,
      "r_dpo/length_delta": 26.262500762939453,
      "r_dpo/regularization_term": 2.6262500286102295,
      "r_dpo/rejected_len": 242.6062469482422,
      "step": 470
    },
    {
      "epoch": 0.9989528795811519,
      "step": 477,
      "total_flos": 0.0,
      "train_loss": 14.700552830406204,
      "train_runtime": 18661.0672,
      "train_samples_per_second": 3.276,
      "train_steps_per_second": 0.026
    }
  ],
  "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
}