{
  "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": 28.589035034179688,
      "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": 5.5463,
      "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": 26.56291389465332,
      "learning_rate": 9.375e-08,
      "logits/chosen": -0.6324527263641357,
      "logits/rejected": -0.6372823119163513,
      "logps/chosen": -296.603759765625,
      "logps/ref_chosen": -296.63226318359375,
      "logps/ref_rejected": -258.9539489746094,
      "logps/rejected": -259.0047302246094,
      "loss": 5.5445,
      "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": 29.713520050048828,
      "learning_rate": 1.9791666666666664e-07,
      "logits/chosen": -0.5963870286941528,
      "logits/rejected": -0.6269619464874268,
      "logps/chosen": -297.92315673828125,
      "logps/ref_chosen": -297.9349365234375,
      "logps/ref_rejected": -256.9902648925781,
      "logps/rejected": -256.97802734375,
      "loss": 5.5435,
      "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": 28.98917007446289,
      "learning_rate": 3.020833333333333e-07,
      "logits/chosen": -0.6142657995223999,
      "logits/rejected": -0.6058592796325684,
      "logps/chosen": -278.4171142578125,
      "logps/ref_chosen": -278.64752197265625,
      "logps/ref_rejected": -249.309814453125,
      "logps/rejected": -249.23779296875,
      "loss": 5.5396,
      "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": 27.35612678527832,
      "learning_rate": 4.0625e-07,
      "logits/chosen": -0.6192952394485474,
      "logits/rejected": -0.644347071647644,
      "logps/chosen": -282.6344299316406,
      "logps/ref_chosen": -283.49981689453125,
      "logps/ref_rejected": -265.32733154296875,
      "logps/rejected": -265.03369140625,
      "loss": 5.521,
      "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": 27.939252853393555,
      "learning_rate": 4.999932966293553e-07,
      "logits/chosen": -0.6302677392959595,
      "logits/rejected": -0.6705285310745239,
      "logps/chosen": -278.43548583984375,
      "logps/ref_chosen": -280.224365234375,
      "logps/ref_rejected": -274.3541259765625,
      "logps/rejected": -273.73004150390625,
      "loss": 5.4954,
      "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": 27.91963005065918,
      "learning_rate": 4.991893270335525e-07,
      "logits/chosen": -0.6450083255767822,
      "logits/rejected": -0.6583200693130493,
      "logps/chosen": -278.49346923828125,
      "logps/ref_chosen": -281.12664794921875,
      "logps/ref_rejected": -259.86456298828125,
      "logps/rejected": -259.6600646972656,
      "loss": 5.4458,
      "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": 28.88052749633789,
      "learning_rate": 4.970496218214204e-07,
      "logits/chosen": -0.7053675055503845,
      "logits/rejected": -0.7107682228088379,
      "logps/chosen": -283.94683837890625,
      "logps/ref_chosen": -287.71063232421875,
      "logps/ref_rejected": -276.839599609375,
      "logps/rejected": -277.0175476074219,
      "loss": 5.3873,
      "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": 28.927474975585938,
      "learning_rate": 4.935856505068998e-07,
      "logits/chosen": -0.6918989419937134,
      "logits/rejected": -0.6877058148384094,
      "logps/chosen": -276.62353515625,
      "logps/ref_chosen": -280.123046875,
      "logps/ref_rejected": -258.8989562988281,
      "logps/rejected": -260.3608093261719,
      "loss": 5.3156,
      "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": 29.801456451416016,
      "learning_rate": 4.8881598109976e-07,
      "logits/chosen": -0.715398907661438,
      "logits/rejected": -0.7198300361633301,
      "logps/chosen": -277.6268615722656,
      "logps/ref_chosen": -278.02545166015625,
      "logps/ref_rejected": -251.0922393798828,
      "logps/rejected": -258.9493713378906,
      "loss": 5.2562,
      "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": 35.680721282958984,
      "learning_rate": 4.827661805750437e-07,
      "logits/chosen": -0.7235929369926453,
      "logits/rejected": -0.7395325303077698,
      "logps/chosen": -277.54632568359375,
      "logps/ref_chosen": -274.0089416503906,
      "logps/ref_rejected": -274.14447021484375,
      "logps/rejected": -288.9579162597656,
      "loss": 5.1804,
      "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": 34.81735610961914,
      "learning_rate": 4.75468677825789e-07,
      "logits/chosen": -0.7712054252624512,
      "logits/rejected": -0.7870631814002991,
      "logps/chosen": -280.66912841796875,
      "logps/ref_chosen": -273.23333740234375,
      "logps/ref_rejected": -263.88787841796875,
      "logps/rejected": -287.0477600097656,
      "loss": 5.0027,
      "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": 41.90164566040039,
      "learning_rate": 4.669625898336438e-07,
      "logits/chosen": -0.8202114105224609,
      "logits/rejected": -0.8147541284561157,
      "logps/chosen": -291.4042663574219,
      "logps/ref_chosen": -269.77142333984375,
      "logps/ref_rejected": -272.7685546875,
      "logps/rejected": -311.06072998046875,
      "loss": 4.9989,
      "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": 57.423763275146484,
      "learning_rate": 4.5729351198915705e-07,
      "logits/chosen": -0.8498390316963196,
      "logits/rejected": -0.8324364423751831,
      "logps/chosen": -301.84613037109375,
      "logps/ref_chosen": -275.03448486328125,
      "logps/ref_rejected": -276.39862060546875,
      "logps/rejected": -325.33062744140625,
      "loss": 4.8776,
      "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": 60.88969039916992,
      "learning_rate": 4.4651327368569684e-07,
      "logits/chosen": -0.8470001220703125,
      "logits/rejected": -0.8457162976264954,
      "logps/chosen": -308.84027099609375,
      "logps/ref_chosen": -276.0029602050781,
      "logps/ref_rejected": -255.9320526123047,
      "logps/rejected": -314.761962890625,
      "loss": 4.8439,
      "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": 59.7264518737793,
      "learning_rate": 4.346796604970912e-07,
      "logits/chosen": -0.8876619338989258,
      "logits/rejected": -0.8721216320991516,
      "logps/chosen": -330.7905578613281,
      "logps/ref_chosen": -298.2093505859375,
      "logps/ref_rejected": -254.8907012939453,
      "logps/rejected": -320.9139709472656,
      "loss": 4.7236,
      "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": 58.573604583740234,
      "learning_rate": 4.218561044282098e-07,
      "logits/chosen": -0.8934988975524902,
      "logits/rejected": -0.8782498240470886,
      "logps/chosen": -337.43865966796875,
      "logps/ref_chosen": -281.94189453125,
      "logps/ref_rejected": -255.5653533935547,
      "logps/rejected": -353.12567138671875,
      "loss": 4.4456,
      "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": 92.63309478759766,
      "learning_rate": 4.081113438988443e-07,
      "logits/chosen": -0.851898193359375,
      "logits/rejected": -0.8330786824226379,
      "logps/chosen": -346.3147888183594,
      "logps/ref_chosen": -288.2863464355469,
      "logps/ref_rejected": -239.758056640625,
      "logps/rejected": -337.37396240234375,
      "loss": 4.4733,
      "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": 93.2479019165039,
      "learning_rate": 3.935190552834828e-07,
      "logits/chosen": -0.8184630274772644,
      "logits/rejected": -0.8205466270446777,
      "logps/chosen": -341.13372802734375,
      "logps/ref_chosen": -286.17889404296875,
      "logps/ref_rejected": -249.9820098876953,
      "logps/rejected": -348.2437438964844,
      "loss": 4.512,
      "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": 80.11067962646484,
      "learning_rate": 3.781574579820464e-07,
      "logits/chosen": -0.859279453754425,
      "logits/rejected": -0.8603144884109497,
      "logps/chosen": -355.4273376464844,
      "logps/ref_chosen": -280.9278259277344,
      "logps/ref_rejected": -254.3533477783203,
      "logps/rejected": -383.27703857421875,
      "loss": 4.3425,
      "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": 117.87115478515625,
      "learning_rate": 3.621088951385353e-07,
      "logits/chosen": -0.8809002041816711,
      "logits/rejected": -0.8806599378585815,
      "logps/chosen": -334.10260009765625,
      "logps/ref_chosen": -253.1712188720703,
      "logps/ref_rejected": -241.90478515625,
      "logps/rejected": -369.25811767578125,
      "loss": 4.4576,
      "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": -0.8859605193138123,
      "eval_logits/rejected": -0.8661972880363464,
      "eval_logps/chosen": -391.96575927734375,
      "eval_logps/ref_chosen": -288.6414794921875,
      "eval_logps/ref_rejected": -265.96160888671875,
      "eval_logps/rejected": -416.974365234375,
      "eval_loss": 0.5648660659790039,
      "eval_r_dpo/chosen_len": 286.97601318359375,
      "eval_r_dpo/length_delta": 40.88800048828125,
      "eval_r_dpo/regularization_term": 0.0,
      "eval_r_dpo/rejected_len": 246.08799743652344,
      "eval_runtime": 78.8271,
      "eval_samples_per_second": 25.372,
      "eval_steps_per_second": 3.171,
      "step": 200
    },
    {
      "epoch": 0.4397905759162304,
      "grad_norm": 102.8453140258789,
      "learning_rate": 3.454593922550693e-07,
      "logits/chosen": -0.8247052431106567,
      "logits/rejected": -0.8323475122451782,
      "logps/chosen": -390.46563720703125,
      "logps/ref_chosen": -287.9228210449219,
      "logps/ref_rejected": -263.35595703125,
      "logps/rejected": -411.89306640625,
      "loss": 4.5528,
      "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": 84.93110656738281,
      "learning_rate": 3.2829819606729477e-07,
      "logits/chosen": -0.8513854742050171,
      "logits/rejected": -0.8432670831680298,
      "logps/chosen": -361.94427490234375,
      "logps/ref_chosen": -282.3331604003906,
      "logps/ref_rejected": -272.5645446777344,
      "logps/rejected": -407.8734436035156,
      "loss": 4.3287,
      "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": 88.449951171875,
      "learning_rate": 3.1071729615293424e-07,
      "logits/chosen": -0.8409557342529297,
      "logits/rejected": -0.8231566548347473,
      "logps/chosen": -375.97259521484375,
      "logps/ref_chosen": -276.1485595703125,
      "logps/ref_rejected": -252.81198120117188,
      "logps/rejected": -408.45989990234375,
      "loss": 4.2955,
      "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": 87.3523941040039,
      "learning_rate": 2.9281093183781403e-07,
      "logits/chosen": -0.8152298927307129,
      "logits/rejected": -0.8264015316963196,
      "logps/chosen": -361.360595703125,
      "logps/ref_chosen": -270.52520751953125,
      "logps/ref_rejected": -254.83334350585938,
      "logps/rejected": -398.59173583984375,
      "loss": 4.3402,
      "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": 88.13154602050781,
      "learning_rate": 2.7467508704251135e-07,
      "logits/chosen": -0.845689594745636,
      "logits/rejected": -0.8341258764266968,
      "logps/chosen": -376.0411682128906,
      "logps/ref_chosen": -289.6054992675781,
      "logps/ref_rejected": -265.0482482910156,
      "logps/rejected": -409.2091369628906,
      "loss": 4.3706,
      "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": 99.26053619384766,
      "learning_rate": 2.5640697577740815e-07,
      "logits/chosen": -0.8479117155075073,
      "logits/rejected": -0.8312094807624817,
      "logps/chosen": -401.3951110839844,
      "logps/ref_chosen": -288.6393737792969,
      "logps/ref_rejected": -265.315673828125,
      "logps/rejected": -437.06854248046875,
      "loss": 4.3553,
      "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": 88.68135070800781,
      "learning_rate": 2.381045210440644e-07,
      "logits/chosen": -0.8226224184036255,
      "logits/rejected": -0.8202828168869019,
      "logps/chosen": -395.2716064453125,
      "logps/ref_chosen": -280.1373596191406,
      "logps/ref_rejected": -264.84295654296875,
      "logps/rejected": -442.5419921875,
      "loss": 4.228,
      "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": 84.24311828613281,
      "learning_rate": 2.1986582993616925e-07,
      "logits/chosen": -0.8553133010864258,
      "logits/rejected": -0.8398975133895874,
      "logps/chosen": -408.2679748535156,
      "logps/ref_chosen": -301.7547912597656,
      "logps/ref_rejected": -254.6543731689453,
      "logps/rejected": -426.813720703125,
      "loss": 4.2273,
      "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": 103.96916198730469,
      "learning_rate": 2.0178866775369774e-07,
      "logits/chosen": -0.8476747274398804,
      "logits/rejected": -0.8177559971809387,
      "logps/chosen": -426.84906005859375,
      "logps/ref_chosen": -302.79217529296875,
      "logps/ref_rejected": -292.9220275878906,
      "logps/rejected": -473.33697509765625,
      "loss": 4.4579,
      "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": 112.53483581542969,
      "learning_rate": 1.839699339491937e-07,
      "logits/chosen": -0.8564668893814087,
      "logits/rejected": -0.8317262530326843,
      "logps/chosen": -385.36322021484375,
      "logps/ref_chosen": -275.8238220214844,
      "logps/ref_rejected": -264.05743408203125,
      "logps/rejected": -432.60552978515625,
      "loss": 4.251,
      "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": 88.61668395996094,
      "learning_rate": 1.6650514271527465e-07,
      "logits/chosen": -0.8322170376777649,
      "logits/rejected": -0.8294069170951843,
      "logps/chosen": -419.35638427734375,
      "logps/ref_chosen": -296.6716003417969,
      "logps/ref_rejected": -278.68426513671875,
      "logps/rejected": -460.2979431152344,
      "loss": 4.1383,
      "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": 88.22819519042969,
      "learning_rate": 1.4948791099758052e-07,
      "logits/chosen": -0.8486505746841431,
      "logits/rejected": -0.8500319719314575,
      "logps/chosen": -415.5774841308594,
      "logps/ref_chosen": -284.1717529296875,
      "logps/ref_rejected": -261.2606506347656,
      "logps/rejected": -457.5267639160156,
      "loss": 4.095,
      "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": 103.7956771850586,
      "learning_rate": 1.3300945667758012e-07,
      "logits/chosen": -0.8600236773490906,
      "logits/rejected": -0.8557920455932617,
      "logps/chosen": -416.3182678222656,
      "logps/ref_chosen": -283.40338134765625,
      "logps/ref_rejected": -271.27569580078125,
      "logps/rejected": -467.2439880371094,
      "loss": 4.2369,
      "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": 103.91631317138672,
      "learning_rate": 1.1715810961514072e-07,
      "logits/chosen": -0.8616652488708496,
      "logits/rejected": -0.845625102519989,
      "logps/chosen": -396.5005187988281,
      "logps/ref_chosen": -259.7261962890625,
      "logps/ref_rejected": -243.4088897705078,
      "logps/rejected": -445.0213317871094,
      "loss": 4.2243,
      "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": 87.64006805419922,
      "learning_rate": 1.0201883817182949e-07,
      "logits/chosen": -0.8792071342468262,
      "logits/rejected": -0.869489312171936,
      "logps/chosen": -430.8462829589844,
      "logps/ref_chosen": -298.24725341796875,
      "logps/ref_rejected": -272.657958984375,
      "logps/rejected": -468.8990173339844,
      "loss": 4.3118,
      "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": 109.85051727294922,
      "learning_rate": 8.76727937529367e-08,
      "logits/chosen": -0.8390272855758667,
      "logits/rejected": -0.83982914686203,
      "logps/chosen": -407.44158935546875,
      "logps/ref_chosen": -281.881103515625,
      "logps/ref_rejected": -265.4746398925781,
      "logps/rejected": -459.4778747558594,
      "loss": 4.3315,
      "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": 75.16207885742188,
      "learning_rate": 7.419687580962222e-08,
      "logits/chosen": -0.8644768595695496,
      "logits/rejected": -0.8538848161697388,
      "logps/chosen": -422.214111328125,
      "logps/ref_chosen": -302.17822265625,
      "logps/ref_rejected": -265.92877197265625,
      "logps/rejected": -455.0023498535156,
      "loss": 4.0813,
      "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": 131.6292266845703,
      "learning_rate": 6.166331963291519e-08,
      "logits/chosen": -0.8290479779243469,
      "logits/rejected": -0.8196717500686646,
      "logps/chosen": -419.83135986328125,
      "logps/ref_chosen": -301.2120361328125,
      "logps/ref_rejected": -266.4872741699219,
      "logps/rejected": -442.3086853027344,
      "loss": 4.2961,
      "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": 74.49371337890625,
      "learning_rate": 5.013930914912476e-08,
      "logits/chosen": -0.84967440366745,
      "logits/rejected": -0.8341225385665894,
      "logps/chosen": -414.8948669433594,
      "logps/ref_chosen": -296.6472473144531,
      "logps/ref_rejected": -278.953857421875,
      "logps/rejected": -463.76885986328125,
      "loss": 4.1709,
      "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": 97.45182037353516,
      "learning_rate": 3.968661679220467e-08,
      "logits/chosen": -0.8492805361747742,
      "logits/rejected": -0.8483866453170776,
      "logps/chosen": -420.40008544921875,
      "logps/ref_chosen": -296.6556091308594,
      "logps/ref_rejected": -256.9266662597656,
      "logps/rejected": -440.20458984375,
      "loss": 4.2579,
      "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": -0.8584261536598206,
      "eval_logits/rejected": -0.8411309719085693,
      "eval_logps/chosen": -414.447509765625,
      "eval_logps/ref_chosen": -288.6414794921875,
      "eval_logps/ref_rejected": -265.96160888671875,
      "eval_logps/rejected": -451.4491882324219,
      "eval_loss": 0.5327035188674927,
      "eval_r_dpo/chosen_len": 286.97601318359375,
      "eval_r_dpo/length_delta": 40.88800048828125,
      "eval_r_dpo/regularization_term": 0.0,
      "eval_r_dpo/rejected_len": 246.08799743652344,
      "eval_runtime": 78.3571,
      "eval_samples_per_second": 25.524,
      "eval_steps_per_second": 3.191,
      "step": 400
    },
    {
      "epoch": 0.8586387434554974,
      "grad_norm": 85.77227020263672,
      "learning_rate": 3.036127238347164e-08,
      "logits/chosen": -0.830724835395813,
      "logits/rejected": -0.8160354495048523,
      "logps/chosen": -421.79754638671875,
      "logps/ref_chosen": -289.9568786621094,
      "logps/ref_rejected": -272.4674377441406,
      "logps/rejected": -459.7171936035156,
      "loss": 4.1428,
      "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": 128.62335205078125,
      "learning_rate": 2.2213262793589482e-08,
      "logits/chosen": -0.8531728982925415,
      "logits/rejected": -0.8363476991653442,
      "logps/chosen": -435.88006591796875,
      "logps/ref_chosen": -307.40240478515625,
      "logps/ref_rejected": -279.85760498046875,
      "logps/rejected": -469.5859375,
      "loss": 4.1314,
      "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": 97.44547271728516,
      "learning_rate": 1.5286263996730026e-08,
      "logits/chosen": -0.842852771282196,
      "logits/rejected": -0.8340854644775391,
      "logps/chosen": -426.78375244140625,
      "logps/ref_chosen": -297.7133483886719,
      "logps/ref_rejected": -266.862060546875,
      "logps/rejected": -465.3374938964844,
      "loss": 4.0648,
      "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": 95.01736450195312,
      "learning_rate": 9.617406953185136e-09,
      "logits/chosen": -0.8415233492851257,
      "logits/rejected": -0.8436342477798462,
      "logps/chosen": -426.42633056640625,
      "logps/ref_chosen": -293.67095947265625,
      "logps/ref_rejected": -289.4698791503906,
      "logps/rejected": -479.0205993652344,
      "loss": 4.2677,
      "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": 91.81388092041016,
      "learning_rate": 5.2370785753763356e-09,
      "logits/chosen": -0.8524943590164185,
      "logits/rejected": -0.8484461903572083,
      "logps/chosen": -422.54278564453125,
      "logps/ref_chosen": -296.9415283203125,
      "logps/ref_rejected": -262.6710510253906,
      "logps/rejected": -453.1717834472656,
      "loss": 4.0939,
      "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": 85.33897399902344,
      "learning_rate": 2.168758844148272e-09,
      "logits/chosen": -0.866470992565155,
      "logits/rejected": -0.8637819290161133,
      "logps/chosen": -444.849609375,
      "logps/ref_chosen": -312.42291259765625,
      "logps/ref_rejected": -278.7356262207031,
      "logps/rejected": -475.77838134765625,
      "loss": 4.2438,
      "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": 81.34379577636719,
      "learning_rate": 4.288949484559934e-10,
      "logits/chosen": -0.832243800163269,
      "logits/rejected": -0.8189598321914673,
      "logps/chosen": -404.83941650390625,
      "logps/ref_chosen": -278.0654602050781,
      "logps/ref_rejected": -256.5596618652344,
      "logps/rejected": -447.73919677734375,
      "loss": 4.1491,
      "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": 4.583878276233153,
      "train_runtime": 6810.0393,
      "train_samples_per_second": 8.977,
      "train_steps_per_second": 0.07
    }
  ],
  "logging_steps": 10,
  "max_steps": 477,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}