{
  "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": 28.59265899658203,
      "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": 26.517539978027344,
      "learning_rate": 9.375e-08,
      "logits/chosen": -0.6324708461761475,
      "logits/rejected": -0.6374377012252808,
      "logps/chosen": -296.6641845703125,
      "logps/ref_chosen": -296.63226318359375,
      "logps/ref_rejected": -258.9539489746094,
      "logps/rejected": -258.9976501464844,
      "loss": 5.5453,
      "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": 29.77713966369629,
      "learning_rate": 1.9791666666666664e-07,
      "logits/chosen": -0.5966575145721436,
      "logits/rejected": -0.6271787881851196,
      "logps/chosen": -297.93768310546875,
      "logps/ref_chosen": -297.9349365234375,
      "logps/ref_rejected": -256.9902648925781,
      "logps/rejected": -257.028564453125,
      "loss": 5.543,
      "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": 28.96179962158203,
      "learning_rate": 3.020833333333333e-07,
      "logits/chosen": -0.6152126789093018,
      "logits/rejected": -0.6072013974189758,
      "logps/chosen": -278.40167236328125,
      "logps/ref_chosen": -278.64752197265625,
      "logps/ref_rejected": -249.309814453125,
      "logps/rejected": -249.2447052001953,
      "loss": 5.5384,
      "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": 27.361820220947266,
      "learning_rate": 4.0625e-07,
      "logits/chosen": -0.6175118088722229,
      "logits/rejected": -0.6425970196723938,
      "logps/chosen": -282.63714599609375,
      "logps/ref_chosen": -283.49981689453125,
      "logps/ref_rejected": -265.32733154296875,
      "logps/rejected": -265.02880859375,
      "loss": 5.5221,
      "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": 27.90146827697754,
      "learning_rate": 4.999932966293553e-07,
      "logits/chosen": -0.6317179799079895,
      "logits/rejected": -0.6722968220710754,
      "logps/chosen": -278.4624938964844,
      "logps/ref_chosen": -280.224365234375,
      "logps/ref_rejected": -274.3541259765625,
      "logps/rejected": -273.77716064453125,
      "loss": 5.4963,
      "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": 28.248376846313477,
      "learning_rate": 4.991893270335525e-07,
      "logits/chosen": -0.6456497311592102,
      "logits/rejected": -0.6589030027389526,
      "logps/chosen": -278.4886474609375,
      "logps/ref_chosen": -281.12664794921875,
      "logps/ref_rejected": -259.86456298828125,
      "logps/rejected": -259.68756103515625,
      "loss": 5.4449,
      "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": 28.587810516357422,
      "learning_rate": 4.970496218214204e-07,
      "logits/chosen": -0.7054015398025513,
      "logits/rejected": -0.710257887840271,
      "logps/chosen": -283.9915466308594,
      "logps/ref_chosen": -287.71063232421875,
      "logps/ref_rejected": -276.839599609375,
      "logps/rejected": -277.05084228515625,
      "loss": 5.3872,
      "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": 28.91149139404297,
      "learning_rate": 4.935856505068998e-07,
      "logits/chosen": -0.6922556757926941,
      "logits/rejected": -0.6884378790855408,
      "logps/chosen": -276.6068115234375,
      "logps/ref_chosen": -280.123046875,
      "logps/ref_rejected": -258.8989562988281,
      "logps/rejected": -260.3894958496094,
      "loss": 5.316,
      "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": 29.948001861572266,
      "learning_rate": 4.8881598109976e-07,
      "logits/chosen": -0.7148661017417908,
      "logits/rejected": -0.7195515632629395,
      "logps/chosen": -277.6590270996094,
      "logps/ref_chosen": -278.02545166015625,
      "logps/ref_rejected": -251.0922393798828,
      "logps/rejected": -258.99139404296875,
      "loss": 5.256,
      "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": 35.5301628112793,
      "learning_rate": 4.827661805750437e-07,
      "logits/chosen": -0.7245436906814575,
      "logits/rejected": -0.740323543548584,
      "logps/chosen": -277.5360412597656,
      "logps/ref_chosen": -274.0089416503906,
      "logps/ref_rejected": -274.14447021484375,
      "logps/rejected": -289.0398254394531,
      "loss": 5.1787,
      "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": 34.80721664428711,
      "learning_rate": 4.75468677825789e-07,
      "logits/chosen": -0.7716718912124634,
      "logits/rejected": -0.787286102771759,
      "logps/chosen": -280.61236572265625,
      "logps/ref_chosen": -273.23333740234375,
      "logps/ref_rejected": -263.88787841796875,
      "logps/rejected": -287.048583984375,
      "loss": 5.0023,
      "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": 42.601844787597656,
      "learning_rate": 4.669625898336438e-07,
      "logits/chosen": -0.8156879544258118,
      "logits/rejected": -0.8101061582565308,
      "logps/chosen": -291.2214660644531,
      "logps/ref_chosen": -269.77142333984375,
      "logps/ref_rejected": -272.7685546875,
      "logps/rejected": -310.86175537109375,
      "loss": 4.9998,
      "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": 55.5117301940918,
      "learning_rate": 4.5729351198915705e-07,
      "logits/chosen": -0.8489806056022644,
      "logits/rejected": -0.8312622904777527,
      "logps/chosen": -301.61578369140625,
      "logps/ref_chosen": -275.03448486328125,
      "logps/ref_rejected": -276.39862060546875,
      "logps/rejected": -325.17193603515625,
      "loss": 4.8754,
      "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": 54.120121002197266,
      "learning_rate": 4.4651327368569684e-07,
      "logits/chosen": -0.8442527651786804,
      "logits/rejected": -0.842776894569397,
      "logps/chosen": -307.4860534667969,
      "logps/ref_chosen": -276.0029602050781,
      "logps/ref_rejected": -255.9320526123047,
      "logps/rejected": -313.87664794921875,
      "loss": 4.8385,
      "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": 64.18093872070312,
      "learning_rate": 4.346796604970912e-07,
      "logits/chosen": -0.8901342153549194,
      "logits/rejected": -0.8738521337509155,
      "logps/chosen": -332.3876037597656,
      "logps/ref_chosen": -298.2093505859375,
      "logps/ref_rejected": -254.8907012939453,
      "logps/rejected": -322.8716125488281,
      "loss": 4.7144,
      "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": 61.34012985229492,
      "learning_rate": 4.218561044282098e-07,
      "logits/chosen": -0.8931280374526978,
      "logits/rejected": -0.8777166604995728,
      "logps/chosen": -336.1731262207031,
      "logps/ref_chosen": -281.94189453125,
      "logps/ref_rejected": -255.5653533935547,
      "logps/rejected": -352.1561584472656,
      "loss": 4.4414,
      "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": 99.01383209228516,
      "learning_rate": 4.081113438988443e-07,
      "logits/chosen": -0.8554754257202148,
      "logits/rejected": -0.836098313331604,
      "logps/chosen": -345.37591552734375,
      "logps/ref_chosen": -288.2863464355469,
      "logps/ref_rejected": -239.758056640625,
      "logps/rejected": -336.4610595703125,
      "loss": 4.4666,
      "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": 119.2037353515625,
      "learning_rate": 3.935190552834828e-07,
      "logits/chosen": -0.8174031376838684,
      "logits/rejected": -0.8205310702323914,
      "logps/chosen": -338.1886291503906,
      "logps/ref_chosen": -286.17889404296875,
      "logps/ref_rejected": -249.9820098876953,
      "logps/rejected": -345.1977233886719,
      "loss": 4.5145,
      "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": 115.86841583251953,
      "learning_rate": 3.781574579820464e-07,
      "logits/chosen": -0.8579525947570801,
      "logits/rejected": -0.8599478602409363,
      "logps/chosen": -350.3121032714844,
      "logps/ref_chosen": -280.9278259277344,
      "logps/ref_rejected": -254.3533477783203,
      "logps/rejected": -377.564208984375,
      "loss": 4.3438,
      "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": 108.07948303222656,
      "learning_rate": 3.621088951385353e-07,
      "logits/chosen": -0.8816668391227722,
      "logits/rejected": -0.8807156682014465,
      "logps/chosen": -321.8008728027344,
      "logps/ref_chosen": -253.1712188720703,
      "logps/ref_rejected": -241.90478515625,
      "logps/rejected": -354.93450927734375,
      "loss": 4.4915,
      "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": -0.9071072340011597,
      "eval_logits/rejected": -0.884462296962738,
      "eval_logps/chosen": -383.4085998535156,
      "eval_logps/ref_chosen": -288.6414794921875,
      "eval_logps/ref_rejected": -265.96160888671875,
      "eval_logps/rejected": -408.76361083984375,
      "eval_loss": 0.5614430904388428,
      "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": 78.5514,
      "eval_samples_per_second": 25.461,
      "eval_steps_per_second": 3.183,
      "step": 200
    },
    {
      "epoch": 0.4397905759162304,
      "grad_norm": 98.39983367919922,
      "learning_rate": 3.454593922550693e-07,
      "logits/chosen": -0.844610869884491,
      "logits/rejected": -0.8483532667160034,
      "logps/chosen": -389.2551574707031,
      "logps/ref_chosen": -287.9228210449219,
      "logps/ref_rejected": -263.35595703125,
      "logps/rejected": -412.36932373046875,
      "loss": 4.5587,
      "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": 88.01953125,
      "learning_rate": 3.2829819606729477e-07,
      "logits/chosen": -0.860802948474884,
      "logits/rejected": -0.8526037335395813,
      "logps/chosen": -362.9411315917969,
      "logps/ref_chosen": -282.3331604003906,
      "logps/ref_rejected": -272.5645446777344,
      "logps/rejected": -409.726318359375,
      "loss": 4.3191,
      "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": 80.2542953491211,
      "learning_rate": 3.1071729615293424e-07,
      "logits/chosen": -0.8298615217208862,
      "logits/rejected": -0.8124361038208008,
      "logps/chosen": -361.07958984375,
      "logps/ref_chosen": -276.1485595703125,
      "logps/ref_rejected": -252.81198120117188,
      "logps/rejected": -392.0174255371094,
      "loss": 4.2918,
      "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": 102.94511413574219,
      "learning_rate": 2.9281093183781403e-07,
      "logits/chosen": -0.8141033053398132,
      "logits/rejected": -0.8255136609077454,
      "logps/chosen": -350.80035400390625,
      "logps/ref_chosen": -270.52520751953125,
      "logps/ref_rejected": -254.83334350585938,
      "logps/rejected": -386.9250183105469,
      "loss": 4.3624,
      "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": 99.23353576660156,
      "learning_rate": 2.7467508704251135e-07,
      "logits/chosen": -0.8481950759887695,
      "logits/rejected": -0.8352023959159851,
      "logps/chosen": -369.6294860839844,
      "logps/ref_chosen": -289.6054992675781,
      "logps/ref_rejected": -265.0482482910156,
      "logps/rejected": -404.7892761230469,
      "loss": 4.3335,
      "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": 103.33648681640625,
      "learning_rate": 2.5640697577740815e-07,
      "logits/chosen": -0.8530448079109192,
      "logits/rejected": -0.8350385427474976,
      "logps/chosen": -389.0526123046875,
      "logps/ref_chosen": -288.6393737792969,
      "logps/ref_rejected": -265.315673828125,
      "logps/rejected": -423.25390625,
      "loss": 4.3842,
      "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": 79.62167358398438,
      "learning_rate": 2.381045210440644e-07,
      "logits/chosen": -0.8272823095321655,
      "logits/rejected": -0.8247928619384766,
      "logps/chosen": -396.57269287109375,
      "logps/ref_chosen": -280.1373596191406,
      "logps/ref_rejected": -264.84295654296875,
      "logps/rejected": -442.4415588378906,
      "loss": 4.2579,
      "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": 73.61250305175781,
      "learning_rate": 2.1986582993616925e-07,
      "logits/chosen": -0.8542989492416382,
      "logits/rejected": -0.8387068510055542,
      "logps/chosen": -406.976806640625,
      "logps/ref_chosen": -301.7547912597656,
      "logps/ref_rejected": -254.6543731689453,
      "logps/rejected": -424.46075439453125,
      "loss": 4.2315,
      "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": 103.07135009765625,
      "learning_rate": 2.0178866775369774e-07,
      "logits/chosen": -0.8455541729927063,
      "logits/rejected": -0.8156248927116394,
      "logps/chosen": -421.5577087402344,
      "logps/ref_chosen": -302.79217529296875,
      "logps/ref_rejected": -292.9220275878906,
      "logps/rejected": -465.0528259277344,
      "loss": 4.4571,
      "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": 91.10374450683594,
      "learning_rate": 1.839699339491937e-07,
      "logits/chosen": -0.8590185046195984,
      "logits/rejected": -0.8340511322021484,
      "logps/chosen": -382.68487548828125,
      "logps/ref_chosen": -275.8238220214844,
      "logps/ref_rejected": -264.05743408203125,
      "logps/rejected": -428.8130798339844,
      "loss": 4.2559,
      "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": 82.53608703613281,
      "learning_rate": 1.6650514271527465e-07,
      "logits/chosen": -0.8300539255142212,
      "logits/rejected": -0.8278457522392273,
      "logps/chosen": -413.59539794921875,
      "logps/ref_chosen": -296.6716003417969,
      "logps/ref_rejected": -278.68426513671875,
      "logps/rejected": -453.2350158691406,
      "loss": 4.1651,
      "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": 90.0842056274414,
      "learning_rate": 1.4948791099758052e-07,
      "logits/chosen": -0.847748875617981,
      "logits/rejected": -0.849408745765686,
      "logps/chosen": -404.92718505859375,
      "logps/ref_chosen": -284.1717529296875,
      "logps/ref_rejected": -261.2606506347656,
      "logps/rejected": -445.50567626953125,
      "loss": 4.1314,
      "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": 117.30803680419922,
      "learning_rate": 1.3300945667758012e-07,
      "logits/chosen": -0.8595107793807983,
      "logits/rejected": -0.8555922508239746,
      "logps/chosen": -411.070556640625,
      "logps/ref_chosen": -283.40338134765625,
      "logps/ref_rejected": -271.27569580078125,
      "logps/rejected": -461.5215759277344,
      "loss": 4.2619,
      "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": 119.9719009399414,
      "learning_rate": 1.1715810961514072e-07,
      "logits/chosen": -0.8573703765869141,
      "logits/rejected": -0.841023325920105,
      "logps/chosen": -396.1806945800781,
      "logps/ref_chosen": -259.7261962890625,
      "logps/ref_rejected": -243.4088897705078,
      "logps/rejected": -444.3870544433594,
      "loss": 4.2292,
      "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": 92.5477066040039,
      "learning_rate": 1.0201883817182949e-07,
      "logits/chosen": -0.8849331140518188,
      "logits/rejected": -0.8748366236686707,
      "logps/chosen": -426.97198486328125,
      "logps/ref_chosen": -298.24725341796875,
      "logps/ref_rejected": -272.657958984375,
      "logps/rejected": -465.6683044433594,
      "loss": 4.2977,
      "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": 116.8237533569336,
      "learning_rate": 8.76727937529367e-08,
      "logits/chosen": -0.8409330248832703,
      "logits/rejected": -0.8409973382949829,
      "logps/chosen": -399.4936828613281,
      "logps/ref_chosen": -281.881103515625,
      "logps/ref_rejected": -265.4746398925781,
      "logps/rejected": -450.76092529296875,
      "loss": 4.3479,
      "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": 76.65868377685547,
      "learning_rate": 7.419687580962222e-08,
      "logits/chosen": -0.8691250681877136,
      "logits/rejected": -0.8575793504714966,
      "logps/chosen": -415.9898376464844,
      "logps/ref_chosen": -302.17822265625,
      "logps/ref_rejected": -265.92877197265625,
      "logps/rejected": -448.15936279296875,
      "loss": 4.0997,
      "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": 115.7716293334961,
      "learning_rate": 6.166331963291519e-08,
      "logits/chosen": -0.8324747085571289,
      "logits/rejected": -0.8221855163574219,
      "logps/chosen": -414.8014221191406,
      "logps/ref_chosen": -301.2120361328125,
      "logps/ref_rejected": -266.4872741699219,
      "logps/rejected": -437.1184997558594,
      "loss": 4.3057,
      "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": 72.40222930908203,
      "learning_rate": 5.013930914912476e-08,
      "logits/chosen": -0.8503654599189758,
      "logits/rejected": -0.8345378041267395,
      "logps/chosen": -410.3568420410156,
      "logps/ref_chosen": -296.6472473144531,
      "logps/ref_rejected": -278.953857421875,
      "logps/rejected": -457.876220703125,
      "loss": 4.1891,
      "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": 100.89118957519531,
      "learning_rate": 3.968661679220467e-08,
      "logits/chosen": -0.8497657775878906,
      "logits/rejected": -0.8488407135009766,
      "logps/chosen": -412.8147888183594,
      "logps/ref_chosen": -296.6556091308594,
      "logps/ref_rejected": -256.9266662597656,
      "logps/rejected": -432.18450927734375,
      "loss": 4.2712,
      "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": -0.8653351068496704,
      "eval_logits/rejected": -0.8488379716873169,
      "eval_logps/chosen": -406.7590026855469,
      "eval_logps/ref_chosen": -288.6414794921875,
      "eval_logps/ref_rejected": -265.96160888671875,
      "eval_logps/rejected": -443.36309814453125,
      "eval_loss": 0.5338261723518372,
      "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": 78.6291,
      "eval_samples_per_second": 25.436,
      "eval_steps_per_second": 3.179,
      "step": 400
    },
    {
      "epoch": 0.8586387434554974,
      "grad_norm": 83.20862579345703,
      "learning_rate": 3.036127238347164e-08,
      "logits/chosen": -0.8332167863845825,
      "logits/rejected": -0.8190832138061523,
      "logps/chosen": -414.7737731933594,
      "logps/ref_chosen": -289.9568786621094,
      "logps/ref_rejected": -272.4674377441406,
      "logps/rejected": -451.9229431152344,
      "loss": 4.1574,
      "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": 116.7166976928711,
      "learning_rate": 2.2213262793589482e-08,
      "logits/chosen": -0.8510443568229675,
      "logits/rejected": -0.8338598012924194,
      "logps/chosen": -428.4990234375,
      "logps/ref_chosen": -307.40240478515625,
      "logps/ref_rejected": -279.85760498046875,
      "logps/rejected": -461.937255859375,
      "loss": 4.1325,
      "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": 93.6956787109375,
      "learning_rate": 1.5286263996730026e-08,
      "logits/chosen": -0.8407491445541382,
      "logits/rejected": -0.8321961164474487,
      "logps/chosen": -418.92791748046875,
      "logps/ref_chosen": -297.7133483886719,
      "logps/ref_rejected": -266.862060546875,
      "logps/rejected": -456.95721435546875,
      "loss": 4.0672,
      "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": 90.70089721679688,
      "learning_rate": 9.617406953185136e-09,
      "logits/chosen": -0.8413969278335571,
      "logits/rejected": -0.8441200256347656,
      "logps/chosen": -419.41925048828125,
      "logps/ref_chosen": -293.67095947265625,
      "logps/ref_rejected": -289.4698791503906,
      "logps/rejected": -471.10052490234375,
      "loss": 4.2715,
      "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": 93.14322662353516,
      "learning_rate": 5.2370785753763356e-09,
      "logits/chosen": -0.8536258935928345,
      "logits/rejected": -0.8501097559928894,
      "logps/chosen": -415.6376037597656,
      "logps/ref_chosen": -296.9415283203125,
      "logps/ref_rejected": -262.6710510253906,
      "logps/rejected": -445.570556640625,
      "loss": 4.0987,
      "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": 86.69686126708984,
      "learning_rate": 2.168758844148272e-09,
      "logits/chosen": -0.8644816279411316,
      "logits/rejected": -0.8618124723434448,
      "logps/chosen": -437.73199462890625,
      "logps/ref_chosen": -312.42291259765625,
      "logps/ref_rejected": -278.7356262207031,
      "logps/rejected": -466.789306640625,
      "loss": 4.2414,
      "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": 72.70208740234375,
      "learning_rate": 4.288949484559934e-10,
      "logits/chosen": -0.8316763043403625,
      "logits/rejected": -0.8185717463493347,
      "logps/chosen": -396.99578857421875,
      "logps/ref_chosen": -278.0654602050781,
      "logps/ref_rejected": -256.5596618652344,
      "logps/rejected": -439.91888427734375,
      "loss": 4.1528,
      "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": 4.588983364824979,
      "train_runtime": 6140.251,
      "train_samples_per_second": 9.956,
      "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
}