{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 212,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 305.40234375,
      "epoch": 0.009512485136741973,
      "grad_norm": 9.479197648905142,
      "kl": 1.3113021850585938e-05,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 2.353515625,
      "reward_std": 0.5666993586346507,
      "rewards/accuracy_reward": 0.46484375,
      "rewards/format_reward": 0.9375,
      "rewards/influence_reward": 0.3828125,
      "rewards/len_reward": 0.568359375,
      "step": 1
    },
    {
      "completion_length": 303.59375,
      "epoch": 0.04756242568370987,
      "grad_norm": 5.559375419637804,
      "kl": 0.00265657901763916,
      "learning_rate": 1.818181818181818e-07,
      "loss": 0.0001,
      "reward": 2.5146484375,
      "reward_std": 0.530997786205262,
      "rewards/accuracy_reward": 0.54736328125,
      "rewards/format_reward": 0.9443359375,
      "rewards/influence_reward": 0.42724609375,
      "rewards/len_reward": 0.595703125,
      "step": 5
    },
    {
      "completion_length": 301.755859375,
      "epoch": 0.09512485136741974,
      "grad_norm": 7.509497472122595,
      "kl": 0.010486793518066407,
      "learning_rate": 4.090909090909091e-07,
      "loss": 0.0004,
      "reward": 2.534765625,
      "reward_std": 0.5271333329379558,
      "rewards/accuracy_reward": 0.542578125,
      "rewards/format_reward": 0.95234375,
      "rewards/influence_reward": 0.441015625,
      "rewards/len_reward": 0.598828125,
      "step": 10
    },
    {
      "completion_length": 308.83046875,
      "epoch": 0.1426872770511296,
      "grad_norm": 9.5559930417524,
      "kl": 0.3596527099609375,
      "learning_rate": 6.363636363636363e-07,
      "loss": 0.0144,
      "reward": 2.4265625,
      "reward_std": 0.4884017549455166,
      "rewards/accuracy_reward": 0.512109375,
      "rewards/format_reward": 0.95625,
      "rewards/influence_reward": 0.419140625,
      "rewards/len_reward": 0.5390625,
      "step": 15
    },
    {
      "completion_length": 309.508984375,
      "epoch": 0.1902497027348395,
      "grad_norm": 6.575100009019049,
      "kl": 1.8339599609375,
      "learning_rate": 8.636363636363636e-07,
      "loss": 0.0734,
      "reward": 2.483984375,
      "reward_std": 0.4793000495061278,
      "rewards/accuracy_reward": 0.534765625,
      "rewards/format_reward": 0.971875,
      "rewards/influence_reward": 0.444921875,
      "rewards/len_reward": 0.532421875,
      "step": 20
    },
    {
      "completion_length": 302.12109375,
      "epoch": 0.23781212841854935,
      "grad_norm": 3.8082898220241357,
      "kl": 1.6882080078125,
      "learning_rate": 9.99726628670463e-07,
      "loss": 0.0675,
      "reward": 2.571484375,
      "reward_std": 0.46187512911856177,
      "rewards/accuracy_reward": 0.56015625,
      "rewards/format_reward": 0.97578125,
      "rewards/influence_reward": 0.478125,
      "rewards/len_reward": 0.557421875,
      "step": 25
    },
    {
      "completion_length": 285.939453125,
      "epoch": 0.2853745541022592,
      "grad_norm": 5.417820371147177,
      "kl": 1.641796875,
      "learning_rate": 9.966546331768192e-07,
      "loss": 0.0657,
      "reward": 2.583984375,
      "reward_std": 0.48238000813871623,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.985546875,
      "rewards/influence_reward": 0.465625,
      "rewards/len_reward": 0.5703125,
      "step": 30
    },
    {
      "completion_length": 295.471875,
      "epoch": 0.3329369797859691,
      "grad_norm": 3.3934355489448493,
      "kl": 1.362841796875,
      "learning_rate": 9.901899829374047e-07,
      "loss": 0.0545,
      "reward": 2.55234375,
      "reward_std": 0.48458676002919676,
      "rewards/accuracy_reward": 0.516796875,
      "rewards/format_reward": 0.9796875,
      "rewards/influence_reward": 0.429296875,
      "rewards/len_reward": 0.6265625,
      "step": 35
    },
    {
      "completion_length": 295.90390625,
      "epoch": 0.380499405469679,
      "grad_norm": 4.1900410232582335,
      "kl": 1.5516845703125,
      "learning_rate": 9.803768380684242e-07,
      "loss": 0.0621,
      "reward": 2.50546875,
      "reward_std": 0.46826295778155325,
      "rewards/accuracy_reward": 0.50703125,
      "rewards/format_reward": 0.97578125,
      "rewards/influence_reward": 0.428125,
      "rewards/len_reward": 0.59453125,
      "step": 40
    },
    {
      "completion_length": 292.028515625,
      "epoch": 0.4280618311533888,
      "grad_norm": 7.946193406732366,
      "kl": 1.88642578125,
      "learning_rate": 9.672822322997304e-07,
      "loss": 0.0754,
      "reward": 2.47421875,
      "reward_std": 0.45962891932576894,
      "rewards/accuracy_reward": 0.49296875,
      "rewards/format_reward": 0.982421875,
      "rewards/influence_reward": 0.411328125,
      "rewards/len_reward": 0.5875,
      "step": 45
    },
    {
      "completion_length": 292.873046875,
      "epoch": 0.4756242568370987,
      "grad_norm": 7.492938961448867,
      "kl": 1.96015625,
      "learning_rate": 9.509956150664795e-07,
      "loss": 0.0784,
      "reward": 2.596484375,
      "reward_std": 0.460917086713016,
      "rewards/accuracy_reward": 0.52578125,
      "rewards/format_reward": 0.982421875,
      "rewards/influence_reward": 0.4453125,
      "rewards/len_reward": 0.64296875,
      "step": 50
    },
    {
      "completion_length": 286.85859375,
      "epoch": 0.5231866825208086,
      "grad_norm": 3.5519428103835224,
      "kl": 1.998779296875,
      "learning_rate": 9.316282404787869e-07,
      "loss": 0.0799,
      "reward": 2.530078125,
      "reward_std": 0.43998180609196424,
      "rewards/accuracy_reward": 0.506640625,
      "rewards/format_reward": 0.973828125,
      "rewards/influence_reward": 0.43046875,
      "rewards/len_reward": 0.619140625,
      "step": 55
    },
    {
      "completion_length": 284.312890625,
      "epoch": 0.5707491082045184,
      "grad_norm": 3.7968869284274014,
      "kl": 2.133349609375,
      "learning_rate": 9.093124073433462e-07,
      "loss": 0.0854,
      "reward": 2.521875,
      "reward_std": 0.5203634534031153,
      "rewards/accuracy_reward": 0.506640625,
      "rewards/format_reward": 0.979296875,
      "rewards/influence_reward": 0.419140625,
      "rewards/len_reward": 0.616796875,
      "step": 60
    },
    {
      "completion_length": 275.38203125,
      "epoch": 0.6183115338882283,
      "grad_norm": 4.3717453671021165,
      "kl": 2.27578125,
      "learning_rate": 8.842005554284295e-07,
      "loss": 0.091,
      "reward": 2.609765625,
      "reward_std": 0.4997987896203995,
      "rewards/accuracy_reward": 0.539453125,
      "rewards/format_reward": 0.983203125,
      "rewards/influence_reward": 0.4578125,
      "rewards/len_reward": 0.629296875,
      "step": 65
    },
    {
      "completion_length": 271.71171875,
      "epoch": 0.6658739595719382,
      "grad_norm": 5.274292147592828,
      "kl": 2.325927734375,
      "learning_rate": 8.564642241456986e-07,
      "loss": 0.093,
      "reward": 2.628515625,
      "reward_std": 0.47792479060590265,
      "rewards/accuracy_reward": 0.543359375,
      "rewards/format_reward": 0.9765625,
      "rewards/influence_reward": 0.4640625,
      "rewards/len_reward": 0.64453125,
      "step": 70
    },
    {
      "completion_length": 272.53828125,
      "epoch": 0.713436385255648,
      "grad_norm": 4.422045761319486,
      "kl": 2.280224609375,
      "learning_rate": 8.262928807620843e-07,
      "loss": 0.0912,
      "reward": 2.548046875,
      "reward_std": 0.504490308649838,
      "rewards/accuracy_reward": 0.511328125,
      "rewards/format_reward": 0.98046875,
      "rewards/influence_reward": 0.433203125,
      "rewards/len_reward": 0.623046875,
      "step": 75
    },
    {
      "completion_length": 274.9234375,
      "epoch": 0.760998810939358,
      "grad_norm": 3.788180136247375,
      "kl": 2.3419921875,
      "learning_rate": 7.938926261462365e-07,
      "loss": 0.0937,
      "reward": 2.491796875,
      "reward_std": 0.47606104165315627,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.980859375,
      "rewards/influence_reward": 0.40546875,
      "rewards/len_reward": 0.62109375,
      "step": 80
    },
    {
      "completion_length": 284.03671875,
      "epoch": 0.8085612366230678,
      "grad_norm": 2.828523061717255,
      "kl": 2.372216796875,
      "learning_rate": 7.594847868906076e-07,
      "loss": 0.0949,
      "reward": 2.5171875,
      "reward_std": 0.4772877097129822,
      "rewards/accuracy_reward": 0.4734375,
      "rewards/format_reward": 0.98046875,
      "rewards/influence_reward": 0.40390625,
      "rewards/len_reward": 0.659375,
      "step": 85
    },
    {
      "completion_length": 280.916796875,
      "epoch": 0.8561236623067776,
      "grad_norm": 607.3521812300567,
      "kl": 2.50087890625,
      "learning_rate": 7.233044034264033e-07,
      "loss": 0.1001,
      "reward": 2.506640625,
      "reward_std": 0.46796532850712536,
      "rewards/accuracy_reward": 0.495703125,
      "rewards/format_reward": 0.9734375,
      "rewards/influence_reward": 0.42109375,
      "rewards/len_reward": 0.61640625,
      "step": 90
    },
    {
      "completion_length": 289.612890625,
      "epoch": 0.9036860879904876,
      "grad_norm": 2.4261586696137574,
      "kl": 2.483837890625,
      "learning_rate": 6.855986244591103e-07,
      "loss": 0.0994,
      "reward": 2.565625,
      "reward_std": 0.4893400952219963,
      "rewards/accuracy_reward": 0.525,
      "rewards/format_reward": 0.977734375,
      "rewards/influence_reward": 0.4375,
      "rewards/len_reward": 0.625390625,
      "step": 95
    },
    {
      "completion_length": 286.48984375,
      "epoch": 0.9512485136741974,
      "grad_norm": 4.005939428029324,
      "kl": 2.36318359375,
      "learning_rate": 6.466250186922324e-07,
      "loss": 0.0945,
      "reward": 2.48828125,
      "reward_std": 0.45276672914624216,
      "rewards/accuracy_reward": 0.488671875,
      "rewards/format_reward": 0.978515625,
      "rewards/influence_reward": 0.403515625,
      "rewards/len_reward": 0.617578125,
      "step": 100
    },
    {
      "completion_length": 288.877734375,
      "epoch": 0.9988109393579072,
      "grad_norm": 3.7990751409015475,
      "kl": 2.427783203125,
      "learning_rate": 6.066498153718734e-07,
      "loss": 0.0971,
      "reward": 2.457421875,
      "reward_std": 0.5042316474020481,
      "rewards/accuracy_reward": 0.47734375,
      "rewards/format_reward": 0.9734375,
      "rewards/influence_reward": 0.380859375,
      "rewards/len_reward": 0.62578125,
      "step": 105
    },
    {
      "completion_length": 274.04876893939394,
      "epoch": 1.0380499405469679,
      "grad_norm": 22.164672338850448,
      "kl": 2.5658735795454546,
      "learning_rate": 5.659460856710345e-07,
      "loss": 0.1004,
      "reward": 2.5634469696969697,
      "reward_std": 0.5097437008763804,
      "rewards/accuracy_reward": 0.5350378787878788,
      "rewards/format_reward": 0.9772727272727273,
      "rewards/influence_reward": 0.4308712121212121,
      "rewards/len_reward": 0.6202651515151515,
      "step": 110
    },
    {
      "completion_length": 275.455078125,
      "epoch": 1.0856123662306778,
      "grad_norm": 3.2333169625918923,
      "kl": 2.7962890625,
      "learning_rate": 5.247918773366111e-07,
      "loss": 0.1119,
      "reward": 2.575390625,
      "reward_std": 0.4654921619221568,
      "rewards/accuracy_reward": 0.5125,
      "rewards/format_reward": 0.984375,
      "rewards/influence_reward": 0.426953125,
      "rewards/len_reward": 0.6515625,
      "step": 115
    },
    {
      "completion_length": 289.22265625,
      "epoch": 1.1331747919143877,
      "grad_norm": 6.3823521689302085,
      "kl": 2.52705078125,
      "learning_rate": 4.834683153413459e-07,
      "loss": 0.1011,
      "reward": 2.573046875,
      "reward_std": 0.4989847050979733,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.971875,
      "rewards/influence_reward": 0.41875,
      "rewards/len_reward": 0.666796875,
      "step": 120
    },
    {
      "completion_length": 269.36796875,
      "epoch": 1.1807372175980975,
      "grad_norm": 9.314483672694436,
      "kl": 2.718115234375,
      "learning_rate": 4.4225768151520694e-07,
      "loss": 0.1087,
      "reward": 2.595703125,
      "reward_std": 0.4847894934937358,
      "rewards/accuracy_reward": 0.530078125,
      "rewards/format_reward": 0.975,
      "rewards/influence_reward": 0.440625,
      "rewards/len_reward": 0.65,
      "step": 125
    },
    {
      "completion_length": 272.870703125,
      "epoch": 1.2282996432818074,
      "grad_norm": 9.964810062893706,
      "kl": 2.692333984375,
      "learning_rate": 4.0144148627425986e-07,
      "loss": 0.1077,
      "reward": 2.548828125,
      "reward_std": 0.4715102421119809,
      "rewards/accuracy_reward": 0.50546875,
      "rewards/format_reward": 0.979296875,
      "rewards/influence_reward": 0.421875,
      "rewards/len_reward": 0.6421875,
      "step": 130
    },
    {
      "completion_length": 271.344921875,
      "epoch": 1.2758620689655173,
      "grad_norm": 4.882054754352994,
      "kl": 2.486865234375,
      "learning_rate": 3.612985456190778e-07,
      "loss": 0.0995,
      "reward": 2.56328125,
      "reward_std": 0.47647856548428535,
      "rewards/accuracy_reward": 0.51328125,
      "rewards/format_reward": 0.978515625,
      "rewards/influence_reward": 0.422265625,
      "rewards/len_reward": 0.64921875,
      "step": 135
    },
    {
      "completion_length": 274.757421875,
      "epoch": 1.323424494649227,
      "grad_norm": 4.091037506852225,
      "kl": 2.571435546875,
      "learning_rate": 3.221030765387417e-07,
      "loss": 0.1029,
      "reward": 2.5890625,
      "reward_std": 0.47138190008699893,
      "rewards/accuracy_reward": 0.512109375,
      "rewards/format_reward": 0.978515625,
      "rewards/influence_reward": 0.434765625,
      "rewards/len_reward": 0.663671875,
      "step": 140
    },
    {
      "completion_length": 273.21875,
      "epoch": 1.370986920332937,
      "grad_norm": 3.923917161371445,
      "kl": 2.58369140625,
      "learning_rate": 2.841228238307536e-07,
      "loss": 0.1033,
      "reward": 2.52421875,
      "reward_std": 0.48247870467603204,
      "rewards/accuracy_reward": 0.496875,
      "rewards/format_reward": 0.973046875,
      "rewards/influence_reward": 0.407421875,
      "rewards/len_reward": 0.646875,
      "step": 145
    },
    {
      "completion_length": 270.95703125,
      "epoch": 1.418549346016647,
      "grad_norm": 3.288471428585341,
      "kl": 2.6228515625,
      "learning_rate": 2.476172311325783e-07,
      "loss": 0.1049,
      "reward": 2.5703125,
      "reward_std": 0.530765401944518,
      "rewards/accuracy_reward": 0.52265625,
      "rewards/format_reward": 0.975390625,
      "rewards/influence_reward": 0.416796875,
      "rewards/len_reward": 0.65546875,
      "step": 150
    },
    {
      "completion_length": 276.080859375,
      "epoch": 1.4661117717003567,
      "grad_norm": 5.406976549541725,
      "kl": 3.496875,
      "learning_rate": 2.128356686585282e-07,
      "loss": 0.1399,
      "reward": 2.603515625,
      "reward_std": 0.5087036734446884,
      "rewards/accuracy_reward": 0.539453125,
      "rewards/format_reward": 0.976171875,
      "rewards/influence_reward": 0.444140625,
      "rewards/len_reward": 0.64375,
      "step": 155
    },
    {
      "completion_length": 283.583203125,
      "epoch": 1.5136741973840666,
      "grad_norm": 2.4719135321159986,
      "kl": 2.696875,
      "learning_rate": 1.8001572974834168e-07,
      "loss": 0.1079,
      "reward": 2.537109375,
      "reward_std": 0.4965396413579583,
      "rewards/accuracy_reward": 0.495703125,
      "rewards/format_reward": 0.97890625,
      "rewards/influence_reward": 0.4078125,
      "rewards/len_reward": 0.6546875,
      "step": 160
    },
    {
      "completion_length": 281.8515625,
      "epoch": 1.5612366230677766,
      "grad_norm": 6.845144411654133,
      "kl": 2.631396484375,
      "learning_rate": 1.493816078637557e-07,
      "loss": 0.1052,
      "reward": 2.54296875,
      "reward_std": 0.48724669627845285,
      "rewards/accuracy_reward": 0.501171875,
      "rewards/format_reward": 0.97421875,
      "rewards/influence_reward": 0.41015625,
      "rewards/len_reward": 0.657421875,
      "step": 165
    },
    {
      "completion_length": 276.32578125,
      "epoch": 1.6087990487514863,
      "grad_norm": 2.820753870764231,
      "kl": 2.634814453125,
      "learning_rate": 1.2114256511983274e-07,
      "loss": 0.1054,
      "reward": 2.5515625,
      "reward_std": 0.492490841075778,
      "rewards/accuracy_reward": 0.501171875,
      "rewards/format_reward": 0.978125,
      "rewards/influence_reward": 0.408203125,
      "rewards/len_reward": 0.6640625,
      "step": 170
    },
    {
      "completion_length": 282.89296875,
      "epoch": 1.6563614744351962,
      "grad_norm": 3.065824553875067,
      "kl": 2.65234375,
      "learning_rate": 9.549150281252632e-08,
      "loss": 0.1061,
      "reward": 2.530078125,
      "reward_std": 0.4955192942172289,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.973046875,
      "rewards/influence_reward": 0.4046875,
      "rewards/len_reward": 0.65234375,
      "step": 175
    },
    {
      "completion_length": 274.28359375,
      "epoch": 1.7039239001189062,
      "grad_norm": 3.281579185573704,
      "kl": 2.66640625,
      "learning_rate": 7.260364370723043e-08,
      "loss": 0.1066,
      "reward": 2.540625,
      "reward_std": 0.45585995763540266,
      "rewards/accuracy_reward": 0.504296875,
      "rewards/format_reward": 0.98203125,
      "rewards/influence_reward": 0.420703125,
      "rewards/len_reward": 0.63359375,
      "step": 180
    },
    {
      "completion_length": 282.63828125,
      "epoch": 1.7514863258026159,
      "grad_norm": 7.835917126431856,
      "kl": 2.7076171875,
      "learning_rate": 5.263533508961826e-08,
      "loss": 0.1083,
      "reward": 2.524609375,
      "reward_std": 0.49459295999258757,
      "rewards/accuracy_reward": 0.486328125,
      "rewards/format_reward": 0.97578125,
      "rewards/influence_reward": 0.405859375,
      "rewards/len_reward": 0.656640625,
      "step": 185
    },
    {
      "completion_length": 279.878515625,
      "epoch": 1.7990487514863258,
      "grad_norm": 3.407510264782874,
      "kl": 2.6724609375,
      "learning_rate": 3.572298075514652e-08,
      "loss": 0.1069,
      "reward": 2.572265625,
      "reward_std": 0.48079199306666853,
      "rewards/accuracy_reward": 0.5140625,
      "rewards/format_reward": 0.980859375,
      "rewards/influence_reward": 0.436328125,
      "rewards/len_reward": 0.641015625,
      "step": 190
    },
    {
      "completion_length": 280.059375,
      "epoch": 1.8466111771700358,
      "grad_norm": 20.679733097388006,
      "kl": 2.6583984375,
      "learning_rate": 2.1982109232821176e-08,
      "loss": 0.1063,
      "reward": 2.49453125,
      "reward_std": 0.4927243089303374,
      "rewards/accuracy_reward": 0.498828125,
      "rewards/format_reward": 0.97734375,
      "rewards/influence_reward": 0.403125,
      "rewards/len_reward": 0.615234375,
      "step": 195
    },
    {
      "completion_length": 278.7625,
      "epoch": 1.8941736028537455,
      "grad_norm": 2.48795349080197,
      "kl": 2.6826171875,
      "learning_rate": 1.1506584608200364e-08,
      "loss": 0.1073,
      "reward": 2.60546875,
      "reward_std": 0.4754039028659463,
      "rewards/accuracy_reward": 0.536328125,
      "rewards/format_reward": 0.981640625,
      "rewards/influence_reward": 0.446875,
      "rewards/len_reward": 0.640625,
      "step": 200
    },
    {
      "completion_length": 275.9921875,
      "epoch": 1.9417360285374554,
      "grad_norm": 2.346124453455794,
      "kl": 2.770361328125,
      "learning_rate": 4.367965336512403e-09,
      "loss": 0.1108,
      "reward": 2.6203125,
      "reward_std": 0.5024769959971309,
      "rewards/accuracy_reward": 0.5328125,
      "rewards/format_reward": 0.9796875,
      "rewards/influence_reward": 0.45,
      "rewards/len_reward": 0.6578125,
      "step": 205
    },
    {
      "completion_length": 280.690234375,
      "epoch": 1.9892984542211654,
      "grad_norm": 2.0191058844209615,
      "kl": 2.69677734375,
      "learning_rate": 6.150154258476314e-10,
      "loss": 0.1079,
      "reward": 2.544140625,
      "reward_std": 0.5300922216847539,
      "rewards/accuracy_reward": 0.4984375,
      "rewards/format_reward": 0.980078125,
      "rewards/influence_reward": 0.40703125,
      "rewards/len_reward": 0.65859375,
      "step": 210
    },
    {
      "completion_length": 299.63368055555554,
      "epoch": 2.0,
      "kl": 2.6888020833333335,
      "reward": 2.5711805555555554,
      "reward_std": 0.5023611783981323,
      "rewards/accuracy_reward": 0.5399305555555556,
      "rewards/format_reward": 0.9652777777777778,
      "rewards/influence_reward": 0.4131944444444444,
      "rewards/len_reward": 0.6527777777777778,
      "step": 212,
      "total_flos": 0.0,
      "train_loss": 0.0899008925090421,
      "train_runtime": 180643.6048,
      "train_samples_per_second": 0.149,
      "train_steps_per_second": 0.001
    }
  ],
  "logging_steps": 5,
  "max_steps": 212,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}