{
  "best_metric": 0.15087111294269562,
  "best_model_checkpoint": "./checkpoints/ultrafeedback_binarized/phi-2-ultrafeedback_binarized-lambda0.25-ORPO-17-4-15/checkpoint-5988",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 5988,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016700066800267203,
      "grad_norm": 36.0,
      "learning_rate": 2.0000000000000002e-07,
      "loss": 0.5754,
      "step": 50
    },
    {
      "epoch": 0.033400133600534405,
      "grad_norm": 9.0,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 0.3356,
      "step": 100
    },
    {
      "epoch": 0.050100200400801605,
      "grad_norm": 8.1875,
      "learning_rate": 6.000000000000001e-07,
      "loss": 0.2953,
      "step": 150
    },
    {
      "epoch": 0.06680026720106881,
      "grad_norm": 9.8125,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.2954,
      "step": 200
    },
    {
      "epoch": 0.08350033400133601,
      "grad_norm": 66.5,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.2946,
      "step": 250
    },
    {
      "epoch": 0.10020040080160321,
      "grad_norm": 15.5625,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.2982,
      "step": 300
    },
    {
      "epoch": 0.11690046760187041,
      "grad_norm": 9.0,
      "learning_rate": 1.4000000000000001e-06,
      "loss": 0.2912,
      "step": 350
    },
    {
      "epoch": 0.13360053440213762,
      "grad_norm": 18.125,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 0.2994,
      "step": 400
    },
    {
      "epoch": 0.15030060120240482,
      "grad_norm": 14.75,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.2869,
      "step": 450
    },
    {
      "epoch": 0.16700066800267202,
      "grad_norm": 7.4375,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.2779,
      "step": 500
    },
    {
      "epoch": 0.18370073480293922,
      "grad_norm": 6.3125,
      "learning_rate": 2.2e-06,
      "loss": 0.2781,
      "step": 550
    },
    {
      "epoch": 0.20040080160320642,
      "grad_norm": 10.625,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.2836,
      "step": 600
    },
    {
      "epoch": 0.21710086840347362,
      "grad_norm": 18.25,
      "learning_rate": 2.6e-06,
      "loss": 0.2825,
      "step": 650
    },
    {
      "epoch": 0.23380093520374082,
      "grad_norm": 9.875,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.2932,
      "step": 700
    },
    {
      "epoch": 0.250501002004008,
      "grad_norm": 20.125,
      "learning_rate": 3e-06,
      "loss": 0.2681,
      "step": 750
    },
    {
      "epoch": 0.26720106880427524,
      "grad_norm": 9.8125,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.2639,
      "step": 800
    },
    {
      "epoch": 0.28390113560454244,
      "grad_norm": 6.65625,
      "learning_rate": 3.4000000000000005e-06,
      "loss": 0.2706,
      "step": 850
    },
    {
      "epoch": 0.30060120240480964,
      "grad_norm": 15.875,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 0.2742,
      "step": 900
    },
    {
      "epoch": 0.31730126920507684,
      "grad_norm": 8.375,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.2729,
      "step": 950
    },
    {
      "epoch": 0.33400133600534404,
      "grad_norm": 6.65625,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.2595,
      "step": 1000
    },
    {
      "epoch": 0.35070140280561124,
      "grad_norm": 7.3125,
      "learning_rate": 4.2000000000000004e-06,
      "loss": 0.2596,
      "step": 1050
    },
    {
      "epoch": 0.36740146960587844,
      "grad_norm": 7.6875,
      "learning_rate": 4.4e-06,
      "loss": 0.2615,
      "step": 1100
    },
    {
      "epoch": 0.38410153640614564,
      "grad_norm": 12.125,
      "learning_rate": 4.600000000000001e-06,
      "loss": 0.2614,
      "step": 1150
    },
    {
      "epoch": 0.40080160320641284,
      "grad_norm": 20.375,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.2537,
      "step": 1200
    },
    {
      "epoch": 0.41750167000668004,
      "grad_norm": 8.75,
      "learning_rate": 5e-06,
      "loss": 0.2646,
      "step": 1250
    },
    {
      "epoch": 0.43420173680694724,
      "grad_norm": 14.9375,
      "learning_rate": 5.2e-06,
      "loss": 0.2609,
      "step": 1300
    },
    {
      "epoch": 0.45090180360721444,
      "grad_norm": 24.375,
      "learning_rate": 5.400000000000001e-06,
      "loss": 0.2541,
      "step": 1350
    },
    {
      "epoch": 0.46760187040748163,
      "grad_norm": 30.25,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.2584,
      "step": 1400
    },
    {
      "epoch": 0.48430193720774883,
      "grad_norm": 12.1875,
      "learning_rate": 5.8e-06,
      "loss": 0.2572,
      "step": 1450
    },
    {
      "epoch": 0.501002004008016,
      "grad_norm": 9.0625,
      "learning_rate": 6e-06,
      "loss": 0.2511,
      "step": 1500
    },
    {
      "epoch": 0.5177020708082832,
      "grad_norm": 26.25,
      "learning_rate": 6.200000000000001e-06,
      "loss": 0.2595,
      "step": 1550
    },
    {
      "epoch": 0.5344021376085505,
      "grad_norm": 12.1875,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.2555,
      "step": 1600
    },
    {
      "epoch": 0.5511022044088176,
      "grad_norm": 14.375,
      "learning_rate": 6.600000000000001e-06,
      "loss": 0.2283,
      "step": 1650
    },
    {
      "epoch": 0.5678022712090849,
      "grad_norm": 15.6875,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.2375,
      "step": 1700
    },
    {
      "epoch": 0.584502338009352,
      "grad_norm": 7.3125,
      "learning_rate": 7e-06,
      "loss": 0.2378,
      "step": 1750
    },
    {
      "epoch": 0.6012024048096193,
      "grad_norm": 8.25,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.231,
      "step": 1800
    },
    {
      "epoch": 0.6179024716098864,
      "grad_norm": 9.4375,
      "learning_rate": 7.4e-06,
      "loss": 0.2211,
      "step": 1850
    },
    {
      "epoch": 0.6346025384101537,
      "grad_norm": 10.1875,
      "learning_rate": 7.600000000000001e-06,
      "loss": 0.2427,
      "step": 1900
    },
    {
      "epoch": 0.6513026052104208,
      "grad_norm": 21.375,
      "learning_rate": 7.800000000000002e-06,
      "loss": 0.23,
      "step": 1950
    },
    {
      "epoch": 0.6680026720106881,
      "grad_norm": 6.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.2312,
      "step": 2000
    },
    {
      "epoch": 0.6847027388109552,
      "grad_norm": 12.1875,
      "learning_rate": 8.2e-06,
      "loss": 0.2334,
      "step": 2050
    },
    {
      "epoch": 0.7014028056112225,
      "grad_norm": 13.875,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.2316,
      "step": 2100
    },
    {
      "epoch": 0.7181028724114896,
      "grad_norm": 15.1875,
      "learning_rate": 8.6e-06,
      "loss": 0.2337,
      "step": 2150
    },
    {
      "epoch": 0.7348029392117569,
      "grad_norm": 9.0,
      "learning_rate": 8.8e-06,
      "loss": 0.2315,
      "step": 2200
    },
    {
      "epoch": 0.751503006012024,
      "grad_norm": 10.9375,
      "learning_rate": 9e-06,
      "loss": 0.2182,
      "step": 2250
    },
    {
      "epoch": 0.7682030728122913,
      "grad_norm": 10.4375,
      "learning_rate": 9.200000000000002e-06,
      "loss": 0.2326,
      "step": 2300
    },
    {
      "epoch": 0.7849031396125584,
      "grad_norm": 37.0,
      "learning_rate": 9.4e-06,
      "loss": 0.2186,
      "step": 2350
    },
    {
      "epoch": 0.8016032064128257,
      "grad_norm": 8.5,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.2195,
      "step": 2400
    },
    {
      "epoch": 0.8183032732130928,
      "grad_norm": 33.0,
      "learning_rate": 9.800000000000001e-06,
      "loss": 0.2154,
      "step": 2450
    },
    {
      "epoch": 0.8350033400133601,
      "grad_norm": 8.375,
      "learning_rate": 1e-05,
      "loss": 0.2301,
      "step": 2500
    },
    {
      "epoch": 0.8517034068136272,
      "grad_norm": 11.625,
      "learning_rate": 1.02e-05,
      "loss": 0.235,
      "step": 2550
    },
    {
      "epoch": 0.8684034736138945,
      "grad_norm": 5.25,
      "learning_rate": 1.04e-05,
      "loss": 0.2086,
      "step": 2600
    },
    {
      "epoch": 0.8851035404141616,
      "grad_norm": 10.0,
      "learning_rate": 1.0600000000000002e-05,
      "loss": 0.2183,
      "step": 2650
    },
    {
      "epoch": 0.9018036072144289,
      "grad_norm": 5.25,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 0.2047,
      "step": 2700
    },
    {
      "epoch": 0.918503674014696,
      "grad_norm": 44.5,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.206,
      "step": 2750
    },
    {
      "epoch": 0.9352037408149633,
      "grad_norm": 19.5,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.2095,
      "step": 2800
    },
    {
      "epoch": 0.9519038076152304,
      "grad_norm": 9.5625,
      "learning_rate": 1.14e-05,
      "loss": 0.2039,
      "step": 2850
    },
    {
      "epoch": 0.9686038744154977,
      "grad_norm": 11.375,
      "learning_rate": 1.16e-05,
      "loss": 0.2096,
      "step": 2900
    },
    {
      "epoch": 0.9853039412157648,
      "grad_norm": 5.09375,
      "learning_rate": 1.18e-05,
      "loss": 0.2087,
      "step": 2950
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.2021823674440384,
      "eval_runtime": 181.6869,
      "eval_samples_per_second": 8.872,
      "eval_steps_per_second": 0.556,
      "step": 2994
    },
    {
      "epoch": 1.002004008016032,
      "grad_norm": 19.0,
      "learning_rate": 1.2e-05,
      "loss": 0.2132,
      "step": 3000
    },
    {
      "epoch": 1.0187040748162992,
      "grad_norm": 7.46875,
      "learning_rate": 1.22e-05,
      "loss": 0.2114,
      "step": 3050
    },
    {
      "epoch": 1.0354041416165665,
      "grad_norm": 6.75,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 0.2034,
      "step": 3100
    },
    {
      "epoch": 1.0521042084168337,
      "grad_norm": 6.46875,
      "learning_rate": 1.2600000000000001e-05,
      "loss": 0.1949,
      "step": 3150
    },
    {
      "epoch": 1.0688042752171008,
      "grad_norm": 5.625,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.2028,
      "step": 3200
    },
    {
      "epoch": 1.085504342017368,
      "grad_norm": 5.34375,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.2062,
      "step": 3250
    },
    {
      "epoch": 1.1022044088176353,
      "grad_norm": 7.4375,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 0.1937,
      "step": 3300
    },
    {
      "epoch": 1.1189044756179025,
      "grad_norm": 44.25,
      "learning_rate": 1.3400000000000002e-05,
      "loss": 0.2108,
      "step": 3350
    },
    {
      "epoch": 1.1356045424181698,
      "grad_norm": 4.75,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.201,
      "step": 3400
    },
    {
      "epoch": 1.1523046092184368,
      "grad_norm": 8.1875,
      "learning_rate": 1.38e-05,
      "loss": 0.2053,
      "step": 3450
    },
    {
      "epoch": 1.169004676018704,
      "grad_norm": 7.5625,
      "learning_rate": 1.4e-05,
      "loss": 0.1996,
      "step": 3500
    },
    {
      "epoch": 1.1857047428189713,
      "grad_norm": 7.125,
      "learning_rate": 1.4200000000000001e-05,
      "loss": 0.2019,
      "step": 3550
    },
    {
      "epoch": 1.2024048096192386,
      "grad_norm": 6.40625,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.2104,
      "step": 3600
    },
    {
      "epoch": 1.2191048764195056,
      "grad_norm": 6.09375,
      "learning_rate": 1.46e-05,
      "loss": 0.1958,
      "step": 3650
    },
    {
      "epoch": 1.2358049432197729,
      "grad_norm": 10.6875,
      "learning_rate": 1.48e-05,
      "loss": 0.2022,
      "step": 3700
    },
    {
      "epoch": 1.25250501002004,
      "grad_norm": 7.71875,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.2028,
      "step": 3750
    },
    {
      "epoch": 1.2692050768203074,
      "grad_norm": 10.5,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.1977,
      "step": 3800
    },
    {
      "epoch": 1.2859051436205746,
      "grad_norm": 8.125,
      "learning_rate": 1.54e-05,
      "loss": 0.2056,
      "step": 3850
    },
    {
      "epoch": 1.3026052104208417,
      "grad_norm": 5.0625,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 0.2007,
      "step": 3900
    },
    {
      "epoch": 1.319305277221109,
      "grad_norm": 9.3125,
      "learning_rate": 1.58e-05,
      "loss": 0.1993,
      "step": 3950
    },
    {
      "epoch": 1.3360053440213762,
      "grad_norm": 6.6875,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2112,
      "step": 4000
    },
    {
      "epoch": 1.3527054108216432,
      "grad_norm": 7.65625,
      "learning_rate": 1.62e-05,
      "loss": 0.2082,
      "step": 4050
    },
    {
      "epoch": 1.3694054776219104,
      "grad_norm": 4.875,
      "learning_rate": 1.64e-05,
      "loss": 0.2062,
      "step": 4100
    },
    {
      "epoch": 1.3861055444221777,
      "grad_norm": 6.0625,
      "learning_rate": 1.66e-05,
      "loss": 0.1875,
      "step": 4150
    },
    {
      "epoch": 1.402805611222445,
      "grad_norm": 6.71875,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.2083,
      "step": 4200
    },
    {
      "epoch": 1.4195056780227122,
      "grad_norm": 5.21875,
      "learning_rate": 1.7e-05,
      "loss": 0.2101,
      "step": 4250
    },
    {
      "epoch": 1.4362057448229792,
      "grad_norm": 9.375,
      "learning_rate": 1.72e-05,
      "loss": 0.2163,
      "step": 4300
    },
    {
      "epoch": 1.4529058116232465,
      "grad_norm": 6.75,
      "learning_rate": 1.7400000000000003e-05,
      "loss": 0.2048,
      "step": 4350
    },
    {
      "epoch": 1.4696058784235138,
      "grad_norm": 6.75,
      "learning_rate": 1.76e-05,
      "loss": 0.2035,
      "step": 4400
    },
    {
      "epoch": 1.4863059452237808,
      "grad_norm": 10.75,
      "learning_rate": 1.7800000000000002e-05,
      "loss": 0.2062,
      "step": 4450
    },
    {
      "epoch": 1.503006012024048,
      "grad_norm": 6.71875,
      "learning_rate": 1.8e-05,
      "loss": 0.1976,
      "step": 4500
    },
    {
      "epoch": 1.5197060788243153,
      "grad_norm": 8.6875,
      "learning_rate": 1.8200000000000002e-05,
      "loss": 0.1941,
      "step": 4550
    },
    {
      "epoch": 1.5364061456245826,
      "grad_norm": 14.0,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.1868,
      "step": 4600
    },
    {
      "epoch": 1.5531062124248498,
      "grad_norm": 4.875,
      "learning_rate": 1.86e-05,
      "loss": 0.2056,
      "step": 4650
    },
    {
      "epoch": 1.569806279225117,
      "grad_norm": 13.75,
      "learning_rate": 1.88e-05,
      "loss": 0.1937,
      "step": 4700
    },
    {
      "epoch": 1.586506346025384,
      "grad_norm": 9.25,
      "learning_rate": 1.9e-05,
      "loss": 0.2051,
      "step": 4750
    },
    {
      "epoch": 1.6032064128256514,
      "grad_norm": 16.375,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.21,
      "step": 4800
    },
    {
      "epoch": 1.6199064796259184,
      "grad_norm": 10.125,
      "learning_rate": 1.94e-05,
      "loss": 0.1873,
      "step": 4850
    },
    {
      "epoch": 1.6366065464261856,
      "grad_norm": 9.625,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.1909,
      "step": 4900
    },
    {
      "epoch": 1.653306613226453,
      "grad_norm": 11.0,
      "learning_rate": 1.98e-05,
      "loss": 0.1844,
      "step": 4950
    },
    {
      "epoch": 1.6700066800267201,
      "grad_norm": 5.65625,
      "learning_rate": 2e-05,
      "loss": 0.198,
      "step": 5000
    },
    {
      "epoch": 1.6867067468269874,
      "grad_norm": 4.625,
      "learning_rate": 1.9873880897766597e-05,
      "loss": 0.1871,
      "step": 5050
    },
    {
      "epoch": 1.7034068136272547,
      "grad_norm": 57.5,
      "learning_rate": 1.949870479665602e-05,
      "loss": 0.181,
      "step": 5100
    },
    {
      "epoch": 1.7201068804275217,
      "grad_norm": 16.75,
      "learning_rate": 1.888393507127856e-05,
      "loss": 0.166,
      "step": 5150
    },
    {
      "epoch": 1.736806947227789,
      "grad_norm": 502.0,
      "learning_rate": 1.8045078562803203e-05,
      "loss": 0.3851,
      "step": 5200
    },
    {
      "epoch": 1.753507014028056,
      "grad_norm": 135.0,
      "learning_rate": 1.7003294437180254e-05,
      "loss": 0.2137,
      "step": 5250
    },
    {
      "epoch": 1.7702070808283232,
      "grad_norm": 62.5,
      "learning_rate": 1.5784860470138633e-05,
      "loss": 0.1884,
      "step": 5300
    },
    {
      "epoch": 1.7869071476285905,
      "grad_norm": 5.53125,
      "learning_rate": 1.4420510221289137e-05,
      "loss": 0.1631,
      "step": 5350
    },
    {
      "epoch": 1.8036072144288577,
      "grad_norm": 175.0,
      "learning_rate": 1.2944657816335124e-05,
      "loss": 0.2435,
      "step": 5400
    },
    {
      "epoch": 1.820307281229125,
      "grad_norm": 109.0,
      "learning_rate": 1.139452989134496e-05,
      "loss": 0.2224,
      "step": 5450
    },
    {
      "epoch": 1.8370073480293923,
      "grad_norm": 140.0,
      "learning_rate": 9.809226594767979e-06,
      "loss": 0.1926,
      "step": 5500
    },
    {
      "epoch": 1.8537074148296593,
      "grad_norm": 129.0,
      "learning_rate": 8.228735332310575e-06,
      "loss": 0.1689,
      "step": 5550
    },
    {
      "epoch": 1.8704074816299265,
      "grad_norm": 30.25,
      "learning_rate": 6.692922131794517e-06,
      "loss": 0.143,
      "step": 5600
    },
    {
      "epoch": 1.8871075484301936,
      "grad_norm": 74.0,
      "learning_rate": 5.240526069629265e-06,
      "loss": 0.1634,
      "step": 5650
    },
    {
      "epoch": 1.9038076152304608,
      "grad_norm": 43.0,
      "learning_rate": 3.908182123304344e-06,
      "loss": 0.1372,
      "step": 5700
    },
    {
      "epoch": 1.920507682030728,
      "grad_norm": 42.5,
      "learning_rate": 2.729497097295075e-06,
      "loss": 0.1331,
      "step": 5750
    },
    {
      "epoch": 1.9372077488309953,
      "grad_norm": 448.0,
      "learning_rate": 1.7342019310607062e-06,
      "loss": 0.137,
      "step": 5800
    },
    {
      "epoch": 1.9539078156312626,
      "grad_norm": 44.5,
      "learning_rate": 9.474017711657835e-07,
      "loss": 0.1754,
      "step": 5850
    },
    {
      "epoch": 1.9706078824315298,
      "grad_norm": 64.0,
      "learning_rate": 3.889427235709153e-07,
      "loss": 0.1797,
      "step": 5900
    },
    {
      "epoch": 1.9873079492317969,
      "grad_norm": 241.0,
      "learning_rate": 7.291125901946027e-08,
      "loss": 0.1781,
      "step": 5950
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.15087111294269562,
      "eval_runtime": 181.739,
      "eval_samples_per_second": 8.87,
      "eval_steps_per_second": 0.556,
      "step": 5988
    }
  ],
  "logging_steps": 50,
  "max_steps": 5988,
  "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": 1.5590253079717478e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}