{
  "best_metric": 2.0116560459136963,
  "best_model_checkpoint": "./checkpoints/ultrafeedback_binarized/phi-2-ultrafeedback_binarized-lambda0.15-ORPO-1-14-26/checkpoint-4491",
  "epoch": 1.5,
  "eval_steps": 500,
  "global_step": 4491,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016700066800267203,
      "grad_norm": 120.0,
      "learning_rate": 1e-07,
      "loss": 3.6046,
      "step": 50
    },
    {
      "epoch": 0.033400133600534405,
      "grad_norm": 42.5,
      "learning_rate": 2e-07,
      "loss": 3.5223,
      "step": 100
    },
    {
      "epoch": 0.050100200400801605,
      "grad_norm": 34.0,
      "learning_rate": 3e-07,
      "loss": 3.4339,
      "step": 150
    },
    {
      "epoch": 0.06680026720106881,
      "grad_norm": 38.0,
      "learning_rate": 4e-07,
      "loss": 3.4468,
      "step": 200
    },
    {
      "epoch": 0.08350033400133601,
      "grad_norm": 39.25,
      "learning_rate": 5e-07,
      "loss": 3.3893,
      "step": 250
    },
    {
      "epoch": 0.10020040080160321,
      "grad_norm": 49.75,
      "learning_rate": 6e-07,
      "loss": 3.3252,
      "step": 300
    },
    {
      "epoch": 0.11690046760187041,
      "grad_norm": 111.5,
      "learning_rate": 7e-07,
      "loss": 3.2598,
      "step": 350
    },
    {
      "epoch": 0.13360053440213762,
      "grad_norm": 31.0,
      "learning_rate": 8e-07,
      "loss": 3.2579,
      "step": 400
    },
    {
      "epoch": 0.15030060120240482,
      "grad_norm": 111.0,
      "learning_rate": 9e-07,
      "loss": 3.2124,
      "step": 450
    },
    {
      "epoch": 0.16700066800267202,
      "grad_norm": 356.0,
      "learning_rate": 1e-06,
      "loss": 3.1667,
      "step": 500
    },
    {
      "epoch": 0.18370073480293922,
      "grad_norm": 22.875,
      "learning_rate": 9.996127778044774e-07,
      "loss": 3.1074,
      "step": 550
    },
    {
      "epoch": 0.20040080160320642,
      "grad_norm": 154.0,
      "learning_rate": 9.984517109820244e-07,
      "loss": 3.0856,
      "step": 600
    },
    {
      "epoch": 0.21710086840347362,
      "grad_norm": 197.0,
      "learning_rate": 9.965185978960174e-07,
      "loss": 3.0459,
      "step": 650
    },
    {
      "epoch": 0.23380093520374082,
      "grad_norm": 83.0,
      "learning_rate": 9.9381643272363e-07,
      "loss": 2.9253,
      "step": 700
    },
    {
      "epoch": 0.250501002004008,
      "grad_norm": 125.0,
      "learning_rate": 9.90349400818185e-07,
      "loss": 2.9436,
      "step": 750
    },
    {
      "epoch": 0.26720106880427524,
      "grad_norm": 64.5,
      "learning_rate": 9.86122872226508e-07,
      "loss": 2.9872,
      "step": 800
    },
    {
      "epoch": 0.28390113560454244,
      "grad_norm": 37.5,
      "learning_rate": 9.811433933713217e-07,
      "loss": 2.9256,
      "step": 850
    },
    {
      "epoch": 0.30060120240480964,
      "grad_norm": 51.0,
      "learning_rate": 9.754186769115654e-07,
      "loss": 2.8895,
      "step": 900
    },
    {
      "epoch": 0.31730126920507684,
      "grad_norm": 46.0,
      "learning_rate": 9.689575897963448e-07,
      "loss": 2.8348,
      "step": 950
    },
    {
      "epoch": 0.33400133600534404,
      "grad_norm": 115.0,
      "learning_rate": 9.617701395310122e-07,
      "loss": 2.83,
      "step": 1000
    },
    {
      "epoch": 0.35070140280561124,
      "grad_norm": 29.625,
      "learning_rate": 9.538674586766557e-07,
      "loss": 2.7793,
      "step": 1050
    },
    {
      "epoch": 0.36740146960587844,
      "grad_norm": 286.0,
      "learning_rate": 9.45261787606999e-07,
      "loss": 2.7292,
      "step": 1100
    },
    {
      "epoch": 0.38410153640614564,
      "grad_norm": 64.5,
      "learning_rate": 9.359664555494242e-07,
      "loss": 2.7686,
      "step": 1150
    },
    {
      "epoch": 0.40080160320641284,
      "grad_norm": 117.5,
      "learning_rate": 9.259958599394812e-07,
      "loss": 2.7036,
      "step": 1200
    },
    {
      "epoch": 0.41750167000668004,
      "grad_norm": 27.25,
      "learning_rate": 9.153654441208611e-07,
      "loss": 2.6712,
      "step": 1250
    },
    {
      "epoch": 0.43420173680694724,
      "grad_norm": 99.5,
      "learning_rate": 9.040916734253739e-07,
      "loss": 2.6642,
      "step": 1300
    },
    {
      "epoch": 0.45090180360721444,
      "grad_norm": 255.0,
      "learning_rate": 8.921920096699821e-07,
      "loss": 2.6795,
      "step": 1350
    },
    {
      "epoch": 0.46760187040748163,
      "grad_norm": 143.0,
      "learning_rate": 8.796848841103867e-07,
      "loss": 2.603,
      "step": 1400
    },
    {
      "epoch": 0.48430193720774883,
      "grad_norm": 286.0,
      "learning_rate": 8.665896688930638e-07,
      "loss": 2.6033,
      "step": 1450
    },
    {
      "epoch": 0.501002004008016,
      "grad_norm": 25.25,
      "learning_rate": 8.529266470499618e-07,
      "loss": 2.5379,
      "step": 1500
    },
    {
      "epoch": 0.5177020708082832,
      "grad_norm": 564.0,
      "learning_rate": 8.387169810823436e-07,
      "loss": 2.5032,
      "step": 1550
    },
    {
      "epoch": 0.5344021376085505,
      "grad_norm": 45.25,
      "learning_rate": 8.239826801824232e-07,
      "loss": 2.4736,
      "step": 1600
    },
    {
      "epoch": 0.5511022044088176,
      "grad_norm": 53.5,
      "learning_rate": 8.087465661435767e-07,
      "loss": 2.5829,
      "step": 1650
    },
    {
      "epoch": 0.5678022712090849,
      "grad_norm": 66.0,
      "learning_rate": 7.930322380119213e-07,
      "loss": 2.4656,
      "step": 1700
    },
    {
      "epoch": 0.584502338009352,
      "grad_norm": 22.125,
      "learning_rate": 7.768640355340184e-07,
      "loss": 2.4649,
      "step": 1750
    },
    {
      "epoch": 0.6012024048096193,
      "grad_norm": 84.0,
      "learning_rate": 7.602670014573127e-07,
      "loss": 2.3869,
      "step": 1800
    },
    {
      "epoch": 0.6179024716098864,
      "grad_norm": 26.5,
      "learning_rate": 7.432668427417013e-07,
      "loss": 2.4197,
      "step": 1850
    },
    {
      "epoch": 0.6346025384101537,
      "grad_norm": 68.0,
      "learning_rate": 7.258898907423128e-07,
      "loss": 2.38,
      "step": 1900
    },
    {
      "epoch": 0.6513026052104208,
      "grad_norm": 75.0,
      "learning_rate": 7.081630604251657e-07,
      "loss": 2.4333,
      "step": 1950
    },
    {
      "epoch": 0.6680026720106881,
      "grad_norm": 29.5,
      "learning_rate": 6.901138086788807e-07,
      "loss": 2.3658,
      "step": 2000
    },
    {
      "epoch": 0.6847027388109552,
      "grad_norm": 27.375,
      "learning_rate": 6.717700917870123e-07,
      "loss": 2.3418,
      "step": 2050
    },
    {
      "epoch": 0.7014028056112225,
      "grad_norm": 20.125,
      "learning_rate": 6.531603221268764e-07,
      "loss": 2.3305,
      "step": 2100
    },
    {
      "epoch": 0.7181028724114896,
      "grad_norm": 82.0,
      "learning_rate": 6.343133241619368e-07,
      "loss": 2.2667,
      "step": 2150
    },
    {
      "epoch": 0.7348029392117569,
      "grad_norm": 63.25,
      "learning_rate": 6.152582897959174e-07,
      "loss": 2.2144,
      "step": 2200
    },
    {
      "epoch": 0.751503006012024,
      "grad_norm": 91.0,
      "learning_rate": 5.960247331577903e-07,
      "loss": 2.3461,
      "step": 2250
    },
    {
      "epoch": 0.7682030728122913,
      "grad_norm": 40.5,
      "learning_rate": 5.766424448876717e-07,
      "loss": 2.3372,
      "step": 2300
    },
    {
      "epoch": 0.7849031396125584,
      "grad_norm": 112.5,
      "learning_rate": 5.571414459944348e-07,
      "loss": 2.2002,
      "step": 2350
    },
    {
      "epoch": 0.8016032064128257,
      "grad_norm": 30.375,
      "learning_rate": 5.375519413565045e-07,
      "loss": 2.3136,
      "step": 2400
    },
    {
      "epoch": 0.8183032732130928,
      "grad_norm": 98.0,
      "learning_rate": 5.179042729378615e-07,
      "loss": 2.2905,
      "step": 2450
    },
    {
      "epoch": 0.8350033400133601,
      "grad_norm": 34.0,
      "learning_rate": 4.982288727917136e-07,
      "loss": 2.2557,
      "step": 2500
    },
    {
      "epoch": 0.8517034068136272,
      "grad_norm": 35.25,
      "learning_rate": 4.785562159246301e-07,
      "loss": 2.2342,
      "step": 2550
    },
    {
      "epoch": 0.8684034736138945,
      "grad_norm": 80.0,
      "learning_rate": 4.5891677309414674e-07,
      "loss": 2.2967,
      "step": 2600
    },
    {
      "epoch": 0.8851035404141616,
      "grad_norm": 162.0,
      "learning_rate": 4.3934096361294974e-07,
      "loss": 2.2667,
      "step": 2650
    },
    {
      "epoch": 0.9018036072144289,
      "grad_norm": 208.0,
      "learning_rate": 4.198591082327453e-07,
      "loss": 2.2909,
      "step": 2700
    },
    {
      "epoch": 0.918503674014696,
      "grad_norm": 149.0,
      "learning_rate": 4.0050138218078553e-07,
      "loss": 2.2519,
      "step": 2750
    },
    {
      "epoch": 0.9352037408149633,
      "grad_norm": 53.25,
      "learning_rate": 3.812977684217997e-07,
      "loss": 2.1819,
      "step": 2800
    },
    {
      "epoch": 0.9519038076152304,
      "grad_norm": 21.875,
      "learning_rate": 3.622780112177145e-07,
      "loss": 2.1872,
      "step": 2850
    },
    {
      "epoch": 0.9686038744154977,
      "grad_norm": 79.5,
      "learning_rate": 3.4347157005710127e-07,
      "loss": 2.254,
      "step": 2900
    },
    {
      "epoch": 0.9853039412157648,
      "grad_norm": 73.5,
      "learning_rate": 3.249075740257048e-07,
      "loss": 2.1445,
      "step": 2950
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.1521201133728027,
      "eval_runtime": 175.2067,
      "eval_samples_per_second": 9.201,
      "eval_steps_per_second": 1.153,
      "step": 2994
    },
    {
      "epoch": 1.002004008016032,
      "grad_norm": 115.5,
      "learning_rate": 3.066147766887289e-07,
      "loss": 2.1825,
      "step": 3000
    },
    {
      "epoch": 1.0187040748162992,
      "grad_norm": 52.5,
      "learning_rate": 2.886215115547621e-07,
      "loss": 2.1662,
      "step": 3050
    },
    {
      "epoch": 1.0354041416165665,
      "grad_norm": 34.75,
      "learning_rate": 2.7095564819032335e-07,
      "loss": 2.2043,
      "step": 3100
    },
    {
      "epoch": 1.0521042084168337,
      "grad_norm": 87.5,
      "learning_rate": 2.5364454905300353e-07,
      "loss": 2.1481,
      "step": 3150
    },
    {
      "epoch": 1.0688042752171008,
      "grad_norm": 76.5,
      "learning_rate": 2.3671502711006264e-07,
      "loss": 2.0757,
      "step": 3200
    },
    {
      "epoch": 1.085504342017368,
      "grad_norm": 176.0,
      "learning_rate": 2.201933043081239e-07,
      "loss": 2.1454,
      "step": 3250
    },
    {
      "epoch": 1.1022044088176353,
      "grad_norm": 32.75,
      "learning_rate": 2.0410497095829616e-07,
      "loss": 2.1603,
      "step": 3300
    },
    {
      "epoch": 1.1189044756179025,
      "grad_norm": 576.0,
      "learning_rate": 1.8847494609962726e-07,
      "loss": 2.1754,
      "step": 3350
    },
    {
      "epoch": 1.1356045424181698,
      "grad_norm": 154.0,
      "learning_rate": 1.7332743890228485e-07,
      "loss": 2.2309,
      "step": 3400
    },
    {
      "epoch": 1.1523046092184368,
      "grad_norm": 23.5,
      "learning_rate": 1.586859111702436e-07,
      "loss": 2.0781,
      "step": 3450
    },
    {
      "epoch": 1.169004676018704,
      "grad_norm": 43.0,
      "learning_rate": 1.445730410015602e-07,
      "loss": 2.0762,
      "step": 3500
    },
    {
      "epoch": 1.1857047428189713,
      "grad_norm": 28.375,
      "learning_rate": 1.310106876625218e-07,
      "loss": 2.0587,
      "step": 3550
    },
    {
      "epoch": 1.2024048096192386,
      "grad_norm": 70.5,
      "learning_rate": 1.180198577300745e-07,
      "loss": 2.0736,
      "step": 3600
    },
    {
      "epoch": 1.2191048764195056,
      "grad_norm": 104.5,
      "learning_rate": 1.0562067255497037e-07,
      "loss": 2.0893,
      "step": 3650
    },
    {
      "epoch": 1.2358049432197729,
      "grad_norm": 181.0,
      "learning_rate": 9.383233709603422e-08,
      "loss": 2.1176,
      "step": 3700
    },
    {
      "epoch": 1.25250501002004,
      "grad_norm": 114.0,
      "learning_rate": 8.267311017381778e-08,
      "loss": 2.1095,
      "step": 3750
    },
    {
      "epoch": 1.2692050768203074,
      "grad_norm": 217.0,
      "learning_rate": 7.216027618971804e-08,
      "loss": 2.0956,
      "step": 3800
    },
    {
      "epoch": 1.2859051436205746,
      "grad_norm": 404.0,
      "learning_rate": 6.231011835436073e-08,
      "loss": 2.0772,
      "step": 3850
    },
    {
      "epoch": 1.3026052104208417,
      "grad_norm": 144.0,
      "learning_rate": 5.3137893466718706e-08,
      "loss": 2.1183,
      "step": 3900
    },
    {
      "epoch": 1.319305277221109,
      "grad_norm": 29.625,
      "learning_rate": 4.465780828302756e-08,
      "loss": 2.0784,
      "step": 3950
    },
    {
      "epoch": 1.3360053440213762,
      "grad_norm": 278.0,
      "learning_rate": 3.6882997512099334e-08,
      "loss": 2.0097,
      "step": 4000
    },
    {
      "epoch": 1.3527054108216432,
      "grad_norm": 39.0,
      "learning_rate": 2.982550347111995e-08,
      "loss": 2.0364,
      "step": 4050
    },
    {
      "epoch": 1.3694054776219104,
      "grad_norm": 124.0,
      "learning_rate": 2.3496257433439193e-08,
      "loss": 2.007,
      "step": 4100
    },
    {
      "epoch": 1.3861055444221777,
      "grad_norm": 44.5,
      "learning_rate": 1.7905062697243956e-08,
      "loss": 2.0192,
      "step": 4150
    },
    {
      "epoch": 1.402805611222445,
      "grad_norm": 29.25,
      "learning_rate": 1.3060579401339644e-08,
      "loss": 2.0723,
      "step": 4200
    },
    {
      "epoch": 1.4195056780227122,
      "grad_norm": 103.0,
      "learning_rate": 8.970311111558215e-09,
      "loss": 2.0223,
      "step": 4250
    },
    {
      "epoch": 1.4362057448229792,
      "grad_norm": 246.0,
      "learning_rate": 5.64059319856941e-09,
      "loss": 2.0725,
      "step": 4300
    },
    {
      "epoch": 1.4529058116232465,
      "grad_norm": 43.75,
      "learning_rate": 3.0765830250963354e-09,
      "loss": 2.0345,
      "step": 4350
    },
    {
      "epoch": 1.4696058784235138,
      "grad_norm": 136.0,
      "learning_rate": 1.2822519577337932e-09,
      "loss": 2.0416,
      "step": 4400
    },
    {
      "epoch": 1.4863059452237808,
      "grad_norm": 126.5,
      "learning_rate": 2.6037921574328935e-10,
      "loss": 1.9981,
      "step": 4450
    },
    {
      "epoch": 1.5,
      "eval_loss": 2.0116560459136963,
      "eval_runtime": 175.2242,
      "eval_samples_per_second": 9.2,
      "eval_steps_per_second": 1.153,
      "step": 4491
    }
  ],
  "logging_steps": 50,
  "max_steps": 4491,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1692852540499558e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}