{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.5625e-07,
      "logits/chosen": 0.184401273727417,
      "logits/rejected": 0.45644012093544006,
      "logps/chosen": -431.40997314453125,
      "logps/rejected": -374.7772521972656,
      "loss": 0.1668,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5625e-06,
      "logits/chosen": 0.19659145176410675,
      "logits/rejected": 0.16923505067825317,
      "logps/chosen": -399.8372802734375,
      "logps/rejected": -407.80474853515625,
      "loss": 0.2135,
      "rewards/accuracies": 0.3055555522441864,
      "rewards/chosen": -0.001171681797131896,
      "rewards/margins": -4.31987973570358e-05,
      "rewards/rejected": -0.001128483098000288,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": 0.19973915815353394,
      "logits/rejected": 0.2067337930202484,
      "logps/chosen": -425.3497619628906,
      "logps/rejected": -433.87921142578125,
      "loss": 0.2124,
      "rewards/accuracies": 0.38749998807907104,
      "rewards/chosen": -0.0007560320082120597,
      "rewards/margins": 8.589267963543534e-06,
      "rewards/rejected": -0.0007646213052794337,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.6875000000000004e-06,
      "logits/chosen": 0.18633976578712463,
      "logits/rejected": 0.30335497856140137,
      "logps/chosen": -459.7779235839844,
      "logps/rejected": -473.452392578125,
      "loss": 0.1993,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": -0.0007412955746985972,
      "rewards/margins": 0.00032791445846669376,
      "rewards/rejected": -0.0010692101204767823,
      "step": 30
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": 0.13782885670661926,
      "logits/rejected": 0.31285157799720764,
      "logps/chosen": -400.82794189453125,
      "logps/rejected": -394.58905029296875,
      "loss": 0.2101,
      "rewards/accuracies": 0.4312500059604645,
      "rewards/chosen": -0.0014155835378915071,
      "rewards/margins": 0.0005916984518989921,
      "rewards/rejected": -0.0020072818733751774,
      "step": 40
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.949188496058089e-06,
      "logits/chosen": 0.1621820479631424,
      "logits/rejected": 0.3057764172554016,
      "logps/chosen": -386.84515380859375,
      "logps/rejected": -372.78460693359375,
      "loss": 0.1983,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.003419263754040003,
      "rewards/margins": 0.0011660968884825706,
      "rewards/rejected": -0.0045853606425225735,
      "step": 50
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": 0.12111353874206543,
      "logits/rejected": 0.23442435264587402,
      "logps/chosen": -416.0478515625,
      "logps/rejected": -421.8002014160156,
      "loss": 0.2125,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0044263326562941074,
      "rewards/margins": 0.0019439695170149207,
      "rewards/rejected": -0.006370303221046925,
      "step": 60
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.7761938666470405e-06,
      "logits/chosen": 0.18056654930114746,
      "logits/rejected": 0.24711184203624725,
      "logps/chosen": -412.12152099609375,
      "logps/rejected": -416.8878479003906,
      "loss": 0.2155,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.0073882960714399815,
      "rewards/margins": 0.0029027846176177263,
      "rewards/rejected": -0.010291079990565777,
      "step": 70
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": 0.11649110168218613,
      "logits/rejected": 0.18730294704437256,
      "logps/chosen": -455.03399658203125,
      "logps/rejected": -423.7317810058594,
      "loss": 0.2059,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.01960514858365059,
      "rewards/margins": 0.0047067212872207165,
      "rewards/rejected": -0.024311870336532593,
      "step": 80
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.4890613722044526e-06,
      "logits/chosen": 0.13214334845542908,
      "logits/rejected": 0.1627044379711151,
      "logps/chosen": -420.0687561035156,
      "logps/rejected": -419.1974182128906,
      "loss": 0.1974,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.036356501281261444,
      "rewards/margins": 0.009195582941174507,
      "rewards/rejected": -0.0455520860850811,
      "step": 90
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": 0.09840121865272522,
      "logits/rejected": 0.19065344333648682,
      "logps/chosen": -573.3145751953125,
      "logps/rejected": -588.6065063476562,
      "loss": 0.1903,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.07631657272577286,
      "rewards/margins": 0.014788051135838032,
      "rewards/rejected": -0.09110463410615921,
      "step": 100
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.102189034962561e-06,
      "logits/chosen": 0.03982686251401901,
      "logits/rejected": 0.06170179322361946,
      "logps/chosen": -483.71844482421875,
      "logps/rejected": -537.0131225585938,
      "loss": 0.1974,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.09070506691932678,
      "rewards/margins": 0.014559125527739525,
      "rewards/rejected": -0.10526418685913086,
      "step": 110
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": 0.018964925780892372,
      "logits/rejected": 0.08643153309822083,
      "logps/chosen": -517.4832763671875,
      "logps/rejected": -530.7005615234375,
      "loss": 0.2021,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.1091010719537735,
      "rewards/margins": 0.02261071279644966,
      "rewards/rejected": -0.13171178102493286,
      "step": 120
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.634976249348867e-06,
      "logits/chosen": 0.055875927209854126,
      "logits/rejected": 0.0218794047832489,
      "logps/chosen": -516.7933959960938,
      "logps/rejected": -543.1007690429688,
      "loss": 0.1878,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.1324981302022934,
      "rewards/margins": 0.03664865717291832,
      "rewards/rejected": -0.1691468060016632,
      "step": 130
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -0.0752568170428276,
      "logits/rejected": 0.07607444375753403,
      "logps/chosen": -510.17987060546875,
      "logps/rejected": -535.81494140625,
      "loss": 0.1899,
      "rewards/accuracies": 0.4437499940395355,
      "rewards/chosen": -0.13363127410411835,
      "rewards/margins": 0.040403980761766434,
      "rewards/rejected": -0.1740352362394333,
      "step": 140
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.1108510153447352e-06,
      "logits/chosen": -0.08575725555419922,
      "logits/rejected": -0.02042851224541664,
      "logps/chosen": -584.4342041015625,
      "logps/rejected": -560.9251708984375,
      "loss": 0.2106,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": -0.163564532995224,
      "rewards/margins": 0.027438053861260414,
      "rewards/rejected": -0.19100257754325867,
      "step": 150
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": 0.0028429715894162655,
      "logits/rejected": 0.10017697513103485,
      "logps/chosen": -546.78515625,
      "logps/rejected": -583.113525390625,
      "loss": 0.2005,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.13426996767520905,
      "rewards/margins": 0.034793704748153687,
      "rewards/rejected": -0.16906365752220154,
      "step": 160
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.556095160739513e-06,
      "logits/chosen": -0.0652703195810318,
      "logits/rejected": 0.08785363286733627,
      "logps/chosen": -474.89013671875,
      "logps/rejected": -519.8175048828125,
      "loss": 0.1939,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": -0.11705396324396133,
      "rewards/margins": 0.03326694667339325,
      "rewards/rejected": -0.150320902466774,
      "step": 170
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -0.06097496673464775,
      "logits/rejected": -0.00452041020616889,
      "logps/chosen": -511.9466247558594,
      "logps/rejected": -552.163330078125,
      "loss": 0.1901,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.12096891552209854,
      "rewards/margins": 0.03578288108110428,
      "rewards/rejected": -0.15675179660320282,
      "step": 180
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9985264605418185e-06,
      "logits/chosen": -0.005833833012729883,
      "logits/rejected": 0.08312544971704483,
      "logps/chosen": -529.5341796875,
      "logps/rejected": -528.4331665039062,
      "loss": 0.1788,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.12104544788599014,
      "rewards/margins": 0.03279378265142441,
      "rewards/rejected": -0.15383923053741455,
      "step": 190
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -0.053509682416915894,
      "logits/rejected": -0.0027845636941492558,
      "logps/chosen": -601.9525146484375,
      "logps/rejected": -663.5382690429688,
      "loss": 0.1869,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.1414756029844284,
      "rewards/margins": 0.056746870279312134,
      "rewards/rejected": -0.19822247326374054,
      "step": 200
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.466103737583699e-06,
      "logits/chosen": -0.005456715822219849,
      "logits/rejected": -0.13148556649684906,
      "logps/chosen": -496.91949462890625,
      "logps/rejected": -558.6058959960938,
      "loss": 0.1844,
      "rewards/accuracies": 0.4437499940395355,
      "rewards/chosen": -0.13160409033298492,
      "rewards/margins": 0.049931664019823074,
      "rewards/rejected": -0.1815357506275177,
      "step": 210
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -0.046624261885881424,
      "logits/rejected": -0.01640903949737549,
      "logps/chosen": -458.24212646484375,
      "logps/rejected": -496.15985107421875,
      "loss": 0.195,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.11825968325138092,
      "rewards/margins": 0.04362802952528,
      "rewards/rejected": -0.16188772022724152,
      "step": 220
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.855248903979505e-07,
      "logits/chosen": -0.058581192046403885,
      "logits/rejected": 0.12358301877975464,
      "logps/chosen": -599.2974853515625,
      "logps/rejected": -605.1329345703125,
      "loss": 0.1807,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.14849095046520233,
      "rewards/margins": 0.043377481400966644,
      "rewards/rejected": -0.19186842441558838,
      "step": 230
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -0.04961230233311653,
      "logits/rejected": 0.06744717061519623,
      "logps/chosen": -500.963134765625,
      "logps/rejected": -533.319580078125,
      "loss": 0.1874,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.1264745444059372,
      "rewards/margins": 0.0465153343975544,
      "rewards/rejected": -0.1729898750782013,
      "step": 240
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.808881491049723e-07,
      "logits/chosen": -0.012030300684273243,
      "logits/rejected": -0.08321252465248108,
      "logps/chosen": -509.8011169433594,
      "logps/rejected": -570.6807861328125,
      "loss": 0.1886,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.1262587606906891,
      "rewards/margins": 0.05674763396382332,
      "rewards/rejected": -0.1830063760280609,
      "step": 250
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -0.07445530593395233,
      "logits/rejected": 0.04640640690922737,
      "logps/chosen": -479.8208923339844,
      "logps/rejected": -571.5008544921875,
      "loss": 0.192,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.1300186812877655,
      "rewards/margins": 0.055131178349256516,
      "rewards/rejected": -0.18514983355998993,
      "step": 260
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.7248368952908055e-07,
      "logits/chosen": -0.024127716198563576,
      "logits/rejected": 0.048486463725566864,
      "logps/chosen": -492.6990661621094,
      "logps/rejected": -528.9088134765625,
      "loss": 0.1872,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.1340349167585373,
      "rewards/margins": 0.04594132676720619,
      "rewards/rejected": -0.17997623980045319,
      "step": 270
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -0.0358465239405632,
      "logits/rejected": 0.059707850217819214,
      "logps/chosen": -528.1822509765625,
      "logps/rejected": -566.1701049804688,
      "loss": 0.1907,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": -0.1274920403957367,
      "rewards/margins": 0.05404983088374138,
      "rewards/rejected": -0.18154187500476837,
      "step": 280
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.577619905828281e-08,
      "logits/chosen": -0.04293031617999077,
      "logits/rejected": -0.02192228101193905,
      "logps/chosen": -562.0809936523438,
      "logps/rejected": -616.6372680664062,
      "loss": 0.1825,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.13138429820537567,
      "rewards/margins": 0.05358911678195,
      "rewards/rejected": -0.18497340381145477,
      "step": 290
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": 0.016219165176153183,
      "logits/rejected": -0.03756276145577431,
      "logps/chosen": -530.8041381835938,
      "logps/rejected": -577.8040161132812,
      "loss": 0.1984,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.12510040402412415,
      "rewards/margins": 0.048407431691884995,
      "rewards/rejected": -0.17350783944129944,
      "step": 300
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.294126437336734e-10,
      "logits/chosen": -0.057995982468128204,
      "logits/rejected": -0.08621449768543243,
      "logps/chosen": -543.5442504882812,
      "logps/rejected": -563.89599609375,
      "loss": 0.1927,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.13217976689338684,
      "rewards/margins": 0.04179350286722183,
      "rewards/rejected": -0.17397327721118927,
      "step": 310
    },
    {
      "epoch": 1.0,
      "step": 312,
      "total_flos": 0.0,
      "train_loss": 0.19580436278230104,
      "train_runtime": 2618.6186,
      "train_samples_per_second": 3.819,
      "train_steps_per_second": 0.119
    }
  ],
  "logging_steps": 10,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}