{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "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-08,
      "logits/chosen": -3.15263032913208,
      "logits/rejected": -3.167269468307495,
      "logps/chosen": -410.4931945800781,
      "logps/rejected": -418.35675048828125,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5624999999999999e-07,
      "logits/chosen": -3.1613829135894775,
      "logits/rejected": -3.1396090984344482,
      "logps/chosen": -395.172607421875,
      "logps/rejected": -445.29937744140625,
      "loss": 0.6929,
      "rewards/accuracies": 0.4791666567325592,
      "rewards/chosen": 0.0008217308204621077,
      "rewards/margins": 0.0008943447028286755,
      "rewards/rejected": -7.261387509061024e-05,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -3.176825761795044,
      "logits/rejected": -3.1581790447235107,
      "logps/chosen": -407.73675537109375,
      "logps/rejected": -476.7193298339844,
      "loss": 0.6907,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.006854791194200516,
      "rewards/margins": 0.0052978708408772945,
      "rewards/rejected": -0.012152662500739098,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.6874999999999996e-07,
      "logits/chosen": -3.1266303062438965,
      "logits/rejected": -3.104175567626953,
      "logps/chosen": -379.49029541015625,
      "logps/rejected": -453.5487365722656,
      "loss": 0.6765,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.059107840061187744,
      "rewards/margins": 0.03545823693275452,
      "rewards/rejected": -0.09456606954336166,
      "step": 30
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988097e-07,
      "logits/chosen": -3.2789883613586426,
      "logits/rejected": -3.2879974842071533,
      "logps/chosen": -407.4563903808594,
      "logps/rejected": -472.73638916015625,
      "loss": 0.649,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.3158263564109802,
      "rewards/margins": 0.08948425948619843,
      "rewards/rejected": -0.40531063079833984,
      "step": 40
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.949188496058089e-07,
      "logits/chosen": -3.3798394203186035,
      "logits/rejected": -3.365309238433838,
      "logps/chosen": -461.53118896484375,
      "logps/rejected": -543.4600830078125,
      "loss": 0.6241,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.7056063413619995,
      "rewards/margins": 0.2066904753446579,
      "rewards/rejected": -0.9122966527938843,
      "step": 50
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.877641290737883e-07,
      "logits/chosen": -3.391582489013672,
      "logits/rejected": -3.3562042713165283,
      "logps/chosen": -509.7789611816406,
      "logps/rejected": -615.9661254882812,
      "loss": 0.5795,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.1751911640167236,
      "rewards/margins": 0.28747081756591797,
      "rewards/rejected": -1.4626619815826416,
      "step": 60
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.776193866647039e-07,
      "logits/chosen": -3.2612807750701904,
      "logits/rejected": -3.2683074474334717,
      "logps/chosen": -575.8006591796875,
      "logps/rejected": -681.9743041992188,
      "loss": 0.5936,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.586014747619629,
      "rewards/margins": 0.3645917773246765,
      "rewards/rejected": -1.9506065845489502,
      "step": 70
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004665e-07,
      "logits/chosen": -3.149127960205078,
      "logits/rejected": -3.1329290866851807,
      "logps/chosen": -516.2041625976562,
      "logps/rejected": -626.0014038085938,
      "loss": 0.5641,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.2160618305206299,
      "rewards/margins": 0.4495382308959961,
      "rewards/rejected": -1.6655998229980469,
      "step": 80
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.489061372204452e-07,
      "logits/chosen": -3.1035568714141846,
      "logits/rejected": -3.1130125522613525,
      "logps/chosen": -568.3084106445312,
      "logps/rejected": -665.2039184570312,
      "loss": 0.5624,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -1.4589087963104248,
      "rewards/margins": 0.38649308681488037,
      "rewards/rejected": -1.8454017639160156,
      "step": 90
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684787e-07,
      "logits/chosen": -3.124558210372925,
      "logits/rejected": -3.1174521446228027,
      "logps/chosen": -525.4788818359375,
      "logps/rejected": -628.5894775390625,
      "loss": 0.5687,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.3210557699203491,
      "rewards/margins": 0.48030009865760803,
      "rewards/rejected": -1.8013557195663452,
      "step": 100
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -3.141606092453003,
      "eval_logits/rejected": -3.1296393871307373,
      "eval_logps/chosen": -548.8582763671875,
      "eval_logps/rejected": -571.9306030273438,
      "eval_loss": 0.7266306281089783,
      "eval_rewards/accuracies": 0.53515625,
      "eval_rewards/chosen": -1.4744807481765747,
      "eval_rewards/margins": 0.06807918846607208,
      "eval_rewards/rejected": -1.54256010055542,
      "eval_runtime": 126.328,
      "eval_samples_per_second": 7.916,
      "eval_steps_per_second": 0.253,
      "step": 100
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.10218903496256e-07,
      "logits/chosen": -3.023545742034912,
      "logits/rejected": -3.021019220352173,
      "logps/chosen": -570.5040283203125,
      "logps/rejected": -650.7171630859375,
      "loss": 0.5714,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.4776172637939453,
      "rewards/margins": 0.3569051921367645,
      "rewards/rejected": -1.8345226049423218,
      "step": 110
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.877242453630256e-07,
      "logits/chosen": -2.912734031677246,
      "logits/rejected": -2.893033266067505,
      "logps/chosen": -547.8372802734375,
      "logps/rejected": -656.3632202148438,
      "loss": 0.557,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.3632628917694092,
      "rewards/margins": 0.49201005697250366,
      "rewards/rejected": -1.8552730083465576,
      "step": 120
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.634976249348867e-07,
      "logits/chosen": -2.885218620300293,
      "logits/rejected": -2.895531415939331,
      "logps/chosen": -534.7477416992188,
      "logps/rejected": -645.1211547851562,
      "loss": 0.5572,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -1.410983681678772,
      "rewards/margins": 0.4554961621761322,
      "rewards/rejected": -1.8664801120758057,
      "step": 130
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.378437060203357e-07,
      "logits/chosen": -2.7777891159057617,
      "logits/rejected": -2.754775047302246,
      "logps/chosen": -589.5593872070312,
      "logps/rejected": -722.3430786132812,
      "loss": 0.5561,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -1.5671552419662476,
      "rewards/margins": 0.4903177320957184,
      "rewards/rejected": -2.0574729442596436,
      "step": 140
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.110851015344735e-07,
      "logits/chosen": -2.9534642696380615,
      "logits/rejected": -2.938105344772339,
      "logps/chosen": -572.8906860351562,
      "logps/rejected": -684.935302734375,
      "loss": 0.5776,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.72029709815979,
      "rewards/margins": 0.4341273307800293,
      "rewards/rejected": -2.1544244289398193,
      "step": 150
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.8355831645441387e-07,
      "logits/chosen": -2.9231104850769043,
      "logits/rejected": -2.9094436168670654,
      "logps/chosen": -572.6766967773438,
      "logps/rejected": -667.0675048828125,
      "loss": 0.5856,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.6497344970703125,
      "rewards/margins": 0.4040594696998596,
      "rewards/rejected": -2.0537939071655273,
      "step": 160
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.5560951607395126e-07,
      "logits/chosen": -3.066063165664673,
      "logits/rejected": -3.0717461109161377,
      "logps/chosen": -541.1444091796875,
      "logps/rejected": -630.5333862304688,
      "loss": 0.5867,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.3789122104644775,
      "rewards/margins": 0.38145893812179565,
      "rewards/rejected": -1.760371208190918,
      "step": 170
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414164e-07,
      "logits/chosen": -3.058684825897217,
      "logits/rejected": -3.044579029083252,
      "logps/chosen": -521.42431640625,
      "logps/rejected": -601.6608276367188,
      "loss": 0.5895,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.1656461954116821,
      "rewards/margins": 0.37542057037353516,
      "rewards/rejected": -1.5410667657852173,
      "step": 180
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.998526460541818e-07,
      "logits/chosen": -3.0458412170410156,
      "logits/rejected": -3.019465446472168,
      "logps/chosen": -549.1454467773438,
      "logps/rejected": -631.4022216796875,
      "loss": 0.5923,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.2936315536499023,
      "rewards/margins": 0.31920185685157776,
      "rewards/rejected": -1.6128332614898682,
      "step": 190
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626315e-07,
      "logits/chosen": -3.0824837684631348,
      "logits/rejected": -3.063490390777588,
      "logps/chosen": -519.8092651367188,
      "logps/rejected": -580.396484375,
      "loss": 0.5901,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.2900115251541138,
      "rewards/margins": 0.2603384256362915,
      "rewards/rejected": -1.5503495931625366,
      "step": 200
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -3.1155648231506348,
      "eval_logits/rejected": -3.1042416095733643,
      "eval_logps/chosen": -542.4268798828125,
      "eval_logps/rejected": -566.1749267578125,
      "eval_loss": 0.7176706790924072,
      "eval_rewards/accuracies": 0.546875,
      "eval_rewards/chosen": -1.4101661443710327,
      "eval_rewards/margins": 0.07483740895986557,
      "eval_rewards/rejected": -1.4850035905838013,
      "eval_runtime": 126.7957,
      "eval_samples_per_second": 7.887,
      "eval_steps_per_second": 0.252,
      "step": 200
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4661037375836987e-07,
      "logits/chosen": -3.0473384857177734,
      "logits/rejected": -3.034618377685547,
      "logps/chosen": -558.7247924804688,
      "logps/rejected": -649.9381103515625,
      "loss": 0.5889,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.4881097078323364,
      "rewards/margins": 0.40473729372024536,
      "rewards/rejected": -1.8928470611572266,
      "step": 210
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2177518064852348e-07,
      "logits/chosen": -3.0184223651885986,
      "logits/rejected": -3.0191338062286377,
      "logps/chosen": -550.5185546875,
      "logps/rejected": -641.130859375,
      "loss": 0.5914,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -1.4187710285186768,
      "rewards/margins": 0.4211079478263855,
      "rewards/rejected": -1.839879035949707,
      "step": 220
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.855248903979505e-08,
      "logits/chosen": -3.091451406478882,
      "logits/rejected": -3.0760319232940674,
      "logps/chosen": -546.5443115234375,
      "logps/rejected": -596.8983764648438,
      "loss": 0.6215,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.3793073892593384,
      "rewards/margins": 0.22927220165729523,
      "rewards/rejected": -1.6085796356201172,
      "step": 230
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328384e-08,
      "logits/chosen": -3.0834617614746094,
      "logits/rejected": -3.0846691131591797,
      "logps/chosen": -537.6251831054688,
      "logps/rejected": -614.9900512695312,
      "loss": 0.6082,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.3861608505249023,
      "rewards/margins": 0.38685861229896545,
      "rewards/rejected": -1.7730194330215454,
      "step": 240
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.808881491049722e-08,
      "logits/chosen": -3.067990779876709,
      "logits/rejected": -3.0540151596069336,
      "logps/chosen": -529.740234375,
      "logps/rejected": -618.4736938476562,
      "loss": 0.6129,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.354187250137329,
      "rewards/margins": 0.431583970785141,
      "rewards/rejected": -1.785771369934082,
      "step": 250
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674335e-08,
      "logits/chosen": -3.0292751789093018,
      "logits/rejected": -3.0266265869140625,
      "logps/chosen": -606.5236206054688,
      "logps/rejected": -676.3375854492188,
      "loss": 0.6213,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.5674151182174683,
      "rewards/margins": 0.31385400891304016,
      "rewards/rejected": -1.8812694549560547,
      "step": 260
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.724836895290805e-08,
      "logits/chosen": -3.087231159210205,
      "logits/rejected": -3.0835347175598145,
      "logps/chosen": -523.4954223632812,
      "logps/rejected": -572.1464233398438,
      "loss": 0.6095,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.3111555576324463,
      "rewards/margins": 0.2772643268108368,
      "rewards/rejected": -1.588419795036316,
      "step": 270
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.5941282340065697e-08,
      "logits/chosen": -3.0525717735290527,
      "logits/rejected": -3.0471031665802,
      "logps/chosen": -573.8348388671875,
      "logps/rejected": -643.480224609375,
      "loss": 0.6231,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.4671975374221802,
      "rewards/margins": 0.28212133049964905,
      "rewards/rejected": -1.7493188381195068,
      "step": 280
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.577619905828281e-09,
      "logits/chosen": -3.091033458709717,
      "logits/rejected": -3.096426248550415,
      "logps/chosen": -537.3826293945312,
      "logps/rejected": -589.7948608398438,
      "loss": 0.6389,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.4276825189590454,
      "rewards/margins": 0.31839436292648315,
      "rewards/rejected": -1.7460769414901733,
      "step": 290
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.2625595580163247e-09,
      "logits/chosen": -3.074458360671997,
      "logits/rejected": -3.0356478691101074,
      "logps/chosen": -585.68505859375,
      "logps/rejected": -669.7423095703125,
      "loss": 0.6278,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.4561302661895752,
      "rewards/margins": 0.38690048456192017,
      "rewards/rejected": -1.8430306911468506,
      "step": 300
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -3.164099931716919,
      "eval_logits/rejected": -3.1529414653778076,
      "eval_logps/chosen": -543.5320434570312,
      "eval_logps/rejected": -567.779052734375,
      "eval_loss": 0.7142006754875183,
      "eval_rewards/accuracies": 0.5546875,
      "eval_rewards/chosen": -1.4212186336517334,
      "eval_rewards/margins": 0.07982616126537323,
      "eval_rewards/rejected": -1.5010446310043335,
      "eval_runtime": 125.3843,
      "eval_samples_per_second": 7.975,
      "eval_steps_per_second": 0.255,
      "step": 300
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.294126437336733e-11,
      "logits/chosen": -3.101105213165283,
      "logits/rejected": -3.063551664352417,
      "logps/chosen": -558.6819458007812,
      "logps/rejected": -620.3117065429688,
      "loss": 0.6361,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.3710179328918457,
      "rewards/margins": 0.27891039848327637,
      "rewards/rejected": -1.649928331375122,
      "step": 310
    },
    {
      "epoch": 1.0,
      "step": 312,
      "total_flos": 0.0,
      "train_loss": 0.6047722502396657,
      "train_runtime": 5332.3567,
      "train_samples_per_second": 3.75,
      "train_steps_per_second": 0.059
    }
  ],
  "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
}