{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 250,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 67.9625,
      "epoch": 0.1,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 9.97e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 67.125,
      "epoch": 0.2,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 9.936666666666667e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 58.3125,
      "epoch": 0.3,
      "grad_norm": 0.0,
      "kl": 1.6021728515625e-05,
      "learning_rate": 9.903333333333333e-07,
      "loss": 0.0,
      "reward": 0.0007361111231148243,
      "reward_std": 0.0019508397206664085,
      "rewards/DCR_reward": 0.0007361111231148243,
      "step": 30
    },
    {
      "completion_length": 81.925,
      "epoch": 0.4,
      "grad_norm": 0.0,
      "kl": 0.0004387378692626953,
      "learning_rate": 9.87e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 69.9625,
      "epoch": 0.5,
      "grad_norm": 0.0,
      "kl": 0.001148056983947754,
      "learning_rate": 9.836666666666666e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 103.45,
      "epoch": 0.6,
      "grad_norm": 0.0,
      "kl": 0.02473886013031006,
      "learning_rate": 9.803333333333332e-07,
      "loss": -0.0,
      "reward": 0.004910160228610039,
      "reward_std": 0.013888029754161835,
      "rewards/DCR_reward": 0.004910160228610039,
      "step": 60
    },
    {
      "completion_length": 74.0125,
      "epoch": 0.7,
      "grad_norm": 24.0,
      "kl": 0.000667726993560791,
      "learning_rate": 9.77e-07,
      "loss": -0.0,
      "reward": 0.052958965534344316,
      "reward_std": 0.022137212846428157,
      "rewards/DCR_reward": 0.052958965534344316,
      "step": 70
    },
    {
      "completion_length": 64.8625,
      "epoch": 0.8,
      "grad_norm": 0.0,
      "kl": 0.000375521183013916,
      "learning_rate": 9.736666666666667e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 80
    },
    {
      "completion_length": 65.25,
      "epoch": 0.9,
      "grad_norm": 0.0,
      "kl": 0.0009482383728027343,
      "learning_rate": 9.703333333333332e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 90
    },
    {
      "completion_length": 67.1875,
      "epoch": 1.0,
      "grad_norm": 0.0,
      "kl": 0.0008351325988769532,
      "learning_rate": 9.67e-07,
      "loss": 0.0,
      "reward": 0.02388598620891571,
      "reward_std": 0.026214474439620973,
      "rewards/DCR_reward": 0.02388598620891571,
      "step": 100
    },
    {
      "completion_length": 58.2125,
      "epoch": 1.1,
      "grad_norm": 0.0,
      "kl": 0.0007820606231689453,
      "learning_rate": 9.636666666666666e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 110
    },
    {
      "completion_length": 63.0375,
      "epoch": 1.2,
      "grad_norm": 0.0,
      "kl": 0.002427983283996582,
      "learning_rate": 9.603333333333333e-07,
      "loss": 0.0,
      "reward": 0.009990863502025604,
      "reward_std": 0.01850293278694153,
      "rewards/DCR_reward": 0.009990863502025604,
      "step": 120
    },
    {
      "completion_length": 98.425,
      "epoch": 1.3,
      "grad_norm": 0.0,
      "kl": 0.0005758762359619141,
      "learning_rate": 9.57e-07,
      "loss": 0.0,
      "reward": 0.00625,
      "reward_std": 0.01767766922712326,
      "rewards/DCR_reward": 0.00625,
      "step": 130
    },
    {
      "completion_length": 74.8375,
      "epoch": 1.4,
      "grad_norm": 0.0,
      "kl": 0.0006227374076843261,
      "learning_rate": 9.536666666666667e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 140
    },
    {
      "completion_length": 74.525,
      "epoch": 1.5,
      "grad_norm": 0.0,
      "kl": 0.0026155948638916016,
      "learning_rate": 9.503333333333333e-07,
      "loss": 0.0,
      "reward": 0.05994144082069397,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.05994144082069397,
      "step": 150
    },
    {
      "completion_length": 107.5625,
      "epoch": 1.6,
      "grad_norm": 0.0,
      "kl": 0.0027020096778869627,
      "learning_rate": 9.469999999999999e-07,
      "loss": 0.0,
      "reward": 0.028785842657089233,
      "reward_std": 0.024412646889686584,
      "rewards/DCR_reward": 0.028785842657089233,
      "step": 160
    },
    {
      "completion_length": 92.0,
      "epoch": 1.7,
      "grad_norm": 0.0,
      "kl": 0.001540231704711914,
      "learning_rate": 9.436666666666667e-07,
      "loss": -0.0,
      "reward": 0.005954481801018119,
      "reward_std": 0.01657787673175335,
      "rewards/DCR_reward": 0.005954481801018119,
      "step": 170
    },
    {
      "completion_length": 89.55,
      "epoch": 1.8,
      "grad_norm": 0.0,
      "kl": 0.00039608478546142577,
      "learning_rate": 9.403333333333333e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 180
    },
    {
      "completion_length": 66.075,
      "epoch": 1.9,
      "grad_norm": 0.0,
      "kl": 0.0017781257629394531,
      "learning_rate": 9.37e-07,
      "loss": 0.0,
      "reward": 0.0013888888992369176,
      "reward_std": 0.003928371146321297,
      "rewards/DCR_reward": 0.0013888888992369176,
      "step": 190
    },
    {
      "completion_length": 57.3625,
      "epoch": 2.0,
      "grad_norm": 0.0,
      "kl": 0.0024953842163085937,
      "learning_rate": 9.336666666666666e-07,
      "loss": 0.0,
      "reward": 0.00016590238083153964,
      "reward_std": 6.703471299260855e-05,
      "rewards/DCR_reward": 0.00016590238083153964,
      "step": 200
    },
    {
      "completion_length": 112.0625,
      "epoch": 2.1,
      "grad_norm": 0.0,
      "kl": 0.004532432556152344,
      "learning_rate": 9.303333333333333e-07,
      "loss": -0.0,
      "reward": 0.005078960955142975,
      "reward_std": 0.014365470409393311,
      "rewards/DCR_reward": 0.005078960955142975,
      "step": 210
    },
    {
      "completion_length": 87.725,
      "epoch": 2.2,
      "grad_norm": 0.0,
      "kl": 0.009652423858642577,
      "learning_rate": 9.27e-07,
      "loss": -0.0,
      "reward": 0.03418266177177429,
      "reward_std": 0.021894259750843047,
      "rewards/DCR_reward": 0.03418266177177429,
      "step": 220
    },
    {
      "completion_length": 99.2375,
      "epoch": 2.3,
      "grad_norm": 0.0,
      "kl": 0.001990985870361328,
      "learning_rate": 9.236666666666666e-07,
      "loss": 0.0,
      "reward": 0.006875000149011612,
      "reward_std": 0.01944543719291687,
      "rewards/DCR_reward": 0.006875000149011612,
      "step": 230
    },
    {
      "completion_length": 87.775,
      "epoch": 2.4,
      "grad_norm": 0.0,
      "kl": 0.0026398658752441405,
      "learning_rate": 9.203333333333333e-07,
      "loss": 0.0,
      "reward": 0.002777777798473835,
      "reward_std": 0.005143444985151291,
      "rewards/DCR_reward": 0.002777777798473835,
      "step": 240
    },
    {
      "completion_length": 74.575,
      "epoch": 2.5,
      "grad_norm": 0.0,
      "kl": 0.01164557933807373,
      "learning_rate": 9.17e-07,
      "loss": 0.0,
      "reward": 0.00416666679084301,
      "reward_std": 0.0025717224925756454,
      "rewards/DCR_reward": 0.00416666679084301,
      "step": 250
    },
    {
      "epoch": 2.5,
      "eval_completion_length": 84.36375,
      "eval_kl": 0.006733891963958741,
      "eval_loss": -2.638111595842929e-07,
      "eval_reward": 0.016272501772618853,
      "eval_reward_std": 0.014607391188983741,
      "eval_rewards/DCR_reward": 0.016272501772618853,
      "eval_runtime": 478.3232,
      "eval_samples_per_second": 0.209,
      "eval_steps_per_second": 0.027,
      "step": 250
    },
    {
      "completion_length": 89.6375,
      "epoch": 2.6,
      "grad_norm": 0.0,
      "kl": 0.009255027770996094,
      "learning_rate": 9.136666666666666e-07,
      "loss": 0.0,
      "reward": 0.018863770365715026,
      "reward_std": 0.025783935189247133,
      "rewards/DCR_reward": 0.018863770365715026,
      "step": 260
    },
    {
      "completion_length": 60.0,
      "epoch": 2.7,
      "grad_norm": 0.0,
      "kl": 0.005229568481445313,
      "learning_rate": 9.103333333333333e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 270
    },
    {
      "completion_length": 89.1875,
      "epoch": 2.8,
      "grad_norm": 0.0,
      "kl": 0.005951988697052002,
      "learning_rate": 9.07e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 280
    },
    {
      "completion_length": 76.9125,
      "epoch": 2.9,
      "grad_norm": 0.0,
      "kl": 0.008041763305664062,
      "learning_rate": 9.036666666666666e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 290
    },
    {
      "completion_length": 64.475,
      "epoch": 3.0,
      "grad_norm": 0.0,
      "kl": 0.005487966537475586,
      "learning_rate": 9.003333333333333e-07,
      "loss": 0.0,
      "reward": 0.06013102883007378,
      "reward_std": 3.2332184218830664e-08,
      "rewards/DCR_reward": 0.06013102883007378,
      "step": 300
    },
    {
      "completion_length": 97.425,
      "epoch": 3.1,
      "grad_norm": 0.0,
      "kl": 0.00860748291015625,
      "learning_rate": 8.969999999999999e-07,
      "loss": 0.0,
      "reward": 0.041901327669620514,
      "reward_std": 0.027376230992376804,
      "rewards/DCR_reward": 0.041901327669620514,
      "step": 310
    },
    {
      "completion_length": 96.8125,
      "epoch": 3.2,
      "grad_norm": 0.0,
      "kl": 0.01550760269165039,
      "learning_rate": 8.936666666666667e-07,
      "loss": 0.0,
      "reward": 0.00040142778307199477,
      "reward_std": 0.001058001583442092,
      "rewards/DCR_reward": 0.00040142778307199477,
      "step": 320
    },
    {
      "completion_length": 90.7875,
      "epoch": 3.3,
      "grad_norm": 0.0,
      "kl": 0.0132171630859375,
      "learning_rate": 8.903333333333333e-07,
      "loss": 0.0,
      "reward": 0.05994144082069397,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.05994144082069397,
      "step": 330
    },
    {
      "completion_length": 68.4875,
      "epoch": 3.4,
      "grad_norm": 0.0,
      "kl": 0.01154632568359375,
      "learning_rate": 8.869999999999999e-07,
      "loss": 0.0,
      "reward": 0.005439520999789238,
      "reward_std": 0.015385289490222932,
      "rewards/DCR_reward": 0.005439520999789238,
      "step": 340
    },
    {
      "completion_length": 98.8875,
      "epoch": 3.5,
      "grad_norm": 0.0,
      "kl": 0.00932455062866211,
      "learning_rate": 8.836666666666667e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/DCR_reward": 0.0,
      "step": 350
    },
    {
      "completion_length": 89.1875,
      "epoch": 3.6,
      "grad_norm": 0.0,
      "kl": 0.009013175964355469,
      "learning_rate": 8.803333333333333e-07,
      "loss": -0.0,
      "reward": 0.025011462631664472,
      "reward_std": 0.04632342683034949,
      "rewards/DCR_reward": 0.025011462631664472,
      "step": 360
    },
    {
      "completion_length": 72.875,
      "epoch": 3.7,
      "grad_norm": 0.0,
      "kl": 0.01702561378479004,
      "learning_rate": 8.769999999999999e-07,
      "loss": 0.0,
      "reward": 0.009234594414010644,
      "reward_std": 0.0059265575400786474,
      "rewards/DCR_reward": 0.009234594414010644,
      "step": 370
    },
    {
      "completion_length": 113.2875,
      "epoch": 3.8,
      "grad_norm": 0.0,
      "kl": 0.02369537353515625,
      "learning_rate": 8.736666666666667e-07,
      "loss": 0.0,
      "reward": 0.021757069602608682,
      "reward_std": 0.045767249166965486,
      "rewards/DCR_reward": 0.021757069602608682,
      "step": 380
    },
    {
      "completion_length": 93.05,
      "epoch": 3.9,
      "grad_norm": 0.0,
      "kl": 0.017023229598999025,
      "learning_rate": 8.703333333333333e-07,
      "loss": 0.0,
      "reward": 0.024103129375725986,
      "reward_std": 0.0580611415207386,
      "rewards/DCR_reward": 0.024103129375725986,
      "step": 390
    },
    {
      "completion_length": 92.2875,
      "epoch": 4.0,
      "grad_norm": 3.109375,
      "kl": 0.0402587890625,
      "learning_rate": 8.669999999999999e-07,
      "loss": 0.0,
      "reward": 0.06285573422210292,
      "reward_std": 0.07154790845233946,
      "rewards/DCR_reward": 0.06285573422210292,
      "step": 400
    },
    {
      "completion_length": 66.1,
      "epoch": 4.1,
      "grad_norm": 0.0,
      "kl": 0.022429752349853515,
      "learning_rate": 8.636666666666667e-07,
      "loss": 0.0,
      "reward": 0.0001454122830182314,
      "reward_std": 0.0003706513671204448,
      "rewards/DCR_reward": 0.0001454122830182314,
      "step": 410
    },
    {
      "completion_length": 137.125,
      "epoch": 4.2,
      "grad_norm": 0.0,
      "kl": 0.024456501007080078,
      "learning_rate": 8.603333333333332e-07,
      "loss": 0.0,
      "reward": 0.05007606785511598,
      "reward_std": 0.06857064368668944,
      "rewards/DCR_reward": 0.05007606785511598,
      "step": 420
    },
    {
      "completion_length": 92.9625,
      "epoch": 4.3,
      "grad_norm": 0.0,
      "kl": 0.023699188232421876,
      "learning_rate": 8.569999999999999e-07,
      "loss": 0.0,
      "reward": 0.0036682719714008273,
      "reward_std": 0.0028008831664919852,
      "rewards/DCR_reward": 0.0036682719714008273,
      "step": 430
    },
    {
      "completion_length": 95.85,
      "epoch": 4.4,
      "grad_norm": 0.0,
      "kl": 0.03151016235351563,
      "learning_rate": 8.536666666666667e-07,
      "loss": -0.0,
      "reward": 0.05042804731056094,
      "reward_std": 0.05373804932460189,
      "rewards/DCR_reward": 0.05042804731056094,
      "step": 440
    },
    {
      "completion_length": 118.325,
      "epoch": 4.5,
      "grad_norm": 7.3125,
      "kl": 0.03622512817382813,
      "learning_rate": 8.503333333333333e-07,
      "loss": 0.0,
      "reward": 0.011024426942458376,
      "reward_std": 0.02974587368662469,
      "rewards/DCR_reward": 0.011024426942458376,
      "step": 450
    },
    {
      "completion_length": 73.175,
      "epoch": 4.6,
      "grad_norm": 1.015625,
      "kl": 0.03716583251953125,
      "learning_rate": 8.469999999999999e-07,
      "loss": 0.0,
      "reward": 0.07448566257953644,
      "reward_std": 0.023564168593065916,
      "rewards/DCR_reward": 0.07448566257953644,
      "step": 460
    },
    {
      "completion_length": 81.0875,
      "epoch": 4.7,
      "grad_norm": 0.0,
      "kl": 0.052947998046875,
      "learning_rate": 8.436666666666667e-07,
      "loss": 0.0,
      "reward": 0.09821241516910958,
      "reward_std": 0.0942218255950138,
      "rewards/DCR_reward": 0.09821241516910958,
      "step": 470
    },
    {
      "completion_length": 99.0375,
      "epoch": 4.8,
      "grad_norm": 9.8125,
      "kl": 0.03128204345703125,
      "learning_rate": 8.403333333333333e-07,
      "loss": -0.0,
      "reward": 0.04295953951077536,
      "reward_std": 0.01739810509607196,
      "rewards/DCR_reward": 0.04295953951077536,
      "step": 480
    },
    {
      "completion_length": 98.1,
      "epoch": 4.9,
      "grad_norm": 28.5,
      "kl": 0.051171875,
      "learning_rate": 8.369999999999999e-07,
      "loss": 0.0,
      "reward": 0.08375828897114843,
      "reward_std": 0.1000140183372423,
      "rewards/DCR_reward": 0.08375828897114843,
      "step": 490
    },
    {
      "completion_length": 117.9375,
      "epoch": 5.0,
      "grad_norm": 0.0,
      "kl": 0.039456844329833984,
      "learning_rate": 8.336666666666667e-07,
      "loss": -0.0,
      "reward": 0.006845223042182625,
      "reward_std": 0.017963543720543384,
      "rewards/DCR_reward": 0.006845223042182625,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_completion_length": 102.48375,
      "eval_kl": 0.06146286010742188,
      "eval_loss": 1.9775862369897368e-07,
      "eval_reward": 0.08125734120461857,
      "eval_reward_std": 0.05342855209446043,
      "eval_rewards/DCR_reward": 0.08125734120461857,
      "eval_runtime": 1460.9247,
      "eval_samples_per_second": 0.068,
      "eval_steps_per_second": 0.009,
      "step": 500
    },
    {
      "completion_length": 105.3125,
      "epoch": 5.1,
      "grad_norm": 14.875,
      "kl": 0.073431396484375,
      "learning_rate": 8.303333333333333e-07,
      "loss": -0.0,
      "reward": 0.09785518775461241,
      "reward_std": 0.08415243490599096,
      "rewards/DCR_reward": 0.09785518775461241,
      "step": 510
    },
    {
      "completion_length": 92.6375,
      "epoch": 5.2,
      "grad_norm": 0.0,
      "kl": 0.05964393615722656,
      "learning_rate": 8.269999999999999e-07,
      "loss": 0.0,
      "reward": 0.0921564630290959,
      "reward_std": 0.04181258587050252,
      "rewards/DCR_reward": 0.0921564630290959,
      "step": 520
    },
    {
      "completion_length": 91.2875,
      "epoch": 5.3,
      "grad_norm": 19.5,
      "kl": 0.08701934814453124,
      "learning_rate": 8.236666666666666e-07,
      "loss": 0.0,
      "reward": 0.16571124200709164,
      "reward_std": 0.04422773125115782,
      "rewards/DCR_reward": 0.16571124200709164,
      "step": 530
    },
    {
      "completion_length": 118.0875,
      "epoch": 5.4,
      "grad_norm": 4.90625,
      "kl": 0.09522705078125,
      "learning_rate": 8.203333333333333e-07,
      "loss": -0.0,
      "reward": 0.12790518356487154,
      "reward_std": 0.04112411521346075,
      "rewards/DCR_reward": 0.12790518356487154,
      "step": 540
    },
    {
      "completion_length": 92.725,
      "epoch": 5.5,
      "grad_norm": 29.25,
      "kl": 0.08489990234375,
      "learning_rate": 8.169999999999999e-07,
      "loss": -0.0,
      "reward": 0.09454966578632593,
      "reward_std": 0.04570461367693497,
      "rewards/DCR_reward": 0.09454966578632593,
      "step": 550
    },
    {
      "completion_length": 83.2375,
      "epoch": 5.6,
      "grad_norm": 0.0,
      "kl": 0.0607513427734375,
      "learning_rate": 8.136666666666666e-07,
      "loss": 0.0,
      "reward": 0.07220683824270964,
      "reward_std": 0.050523467175662515,
      "rewards/DCR_reward": 0.07220683824270964,
      "step": 560
    },
    {
      "completion_length": 99.6875,
      "epoch": 5.7,
      "grad_norm": 0.0,
      "kl": 0.068798828125,
      "learning_rate": 8.103333333333333e-07,
      "loss": -0.0,
      "reward": 0.096117812365992,
      "reward_std": 0.05572115568793379,
      "rewards/DCR_reward": 0.096117812365992,
      "step": 570
    },
    {
      "completion_length": 61.5,
      "epoch": 5.8,
      "grad_norm": 0.0,
      "kl": 0.07020721435546876,
      "learning_rate": 8.070000000000001e-07,
      "loss": -0.0,
      "reward": 0.047906511649489406,
      "reward_std": 4.001859270204022e-05,
      "rewards/DCR_reward": 0.047906511649489406,
      "step": 580
    },
    {
      "completion_length": 99.3125,
      "epoch": 5.9,
      "grad_norm": 0.0,
      "kl": 0.123101806640625,
      "learning_rate": 8.036666666666666e-07,
      "loss": -0.0,
      "reward": 0.13380602395627647,
      "reward_std": 0.10035560713149608,
      "rewards/DCR_reward": 0.13380602395627647,
      "step": 590
    },
    {
      "completion_length": 103.025,
      "epoch": 6.0,
      "grad_norm": 0.0,
      "kl": 0.081585693359375,
      "learning_rate": 8.003333333333333e-07,
      "loss": -0.0,
      "reward": 0.061570275388658044,
      "reward_std": 0.04306753019336611,
      "rewards/DCR_reward": 0.061570275388658044,
      "step": 600
    },
    {
      "completion_length": 97.9125,
      "epoch": 6.1,
      "grad_norm": 0.0,
      "kl": 0.11351318359375,
      "learning_rate": 7.970000000000001e-07,
      "loss": 0.0,
      "reward": 0.09666887713829056,
      "reward_std": 0.059037036258087025,
      "rewards/DCR_reward": 0.09666887713829056,
      "step": 610
    },
    {
      "completion_length": 91.7,
      "epoch": 6.2,
      "grad_norm": 0.0,
      "kl": 0.055133056640625,
      "learning_rate": 7.936666666666666e-07,
      "loss": 0.0,
      "reward": 0.13362326713686343,
      "reward_std": 0.07874106459098584,
      "rewards/DCR_reward": 0.13362326713686343,
      "step": 620
    },
    {
      "completion_length": 107.4125,
      "epoch": 6.3,
      "grad_norm": 9.25,
      "kl": 0.1246337890625,
      "learning_rate": 7.903333333333333e-07,
      "loss": 0.0,
      "reward": 0.08847815722692758,
      "reward_std": 0.06451865802846442,
      "rewards/DCR_reward": 0.08847815722692758,
      "step": 630
    },
    {
      "completion_length": 77.0875,
      "epoch": 6.4,
      "grad_norm": 25.0,
      "kl": 0.15296096801757814,
      "learning_rate": 7.87e-07,
      "loss": -0.0,
      "reward": 0.1423336612060666,
      "reward_std": 0.055696507578250024,
      "rewards/DCR_reward": 0.1423336612060666,
      "step": 640
    },
    {
      "completion_length": 99.6375,
      "epoch": 6.5,
      "grad_norm": 0.0,
      "kl": 0.08688135147094726,
      "learning_rate": 7.836666666666666e-07,
      "loss": -0.0,
      "reward": 0.173234991542995,
      "reward_std": 0.06990011496527586,
      "rewards/DCR_reward": 0.173234991542995,
      "step": 650
    },
    {
      "completion_length": 95.975,
      "epoch": 6.6,
      "grad_norm": 12.9375,
      "kl": 0.14229736328125,
      "learning_rate": 7.803333333333333e-07,
      "loss": 0.0,
      "reward": 0.19796406209934503,
      "reward_std": 0.08495658059261757,
      "rewards/DCR_reward": 0.19796406209934503,
      "step": 660
    },
    {
      "completion_length": 51.9875,
      "epoch": 6.7,
      "grad_norm": 26.0,
      "kl": 0.19248046875,
      "learning_rate": 7.77e-07,
      "loss": 0.0,
      "reward": 0.12773155540926381,
      "reward_std": 0.018970384920248762,
      "rewards/DCR_reward": 0.12773155540926381,
      "step": 670
    },
    {
      "completion_length": 81.6875,
      "epoch": 6.8,
      "grad_norm": 10.0625,
      "kl": 0.09774169921875,
      "learning_rate": 7.736666666666666e-07,
      "loss": -0.0,
      "reward": 0.09495227632578462,
      "reward_std": 0.05938452887904759,
      "rewards/DCR_reward": 0.09495227632578462,
      "step": 680
    },
    {
      "completion_length": 100.675,
      "epoch": 6.9,
      "grad_norm": 19.75,
      "kl": 0.1118896484375,
      "learning_rate": 7.703333333333333e-07,
      "loss": 0.0,
      "reward": 0.09238526365661529,
      "reward_std": 0.04735179884301033,
      "rewards/DCR_reward": 0.09238526365661529,
      "step": 690
    },
    {
      "completion_length": 63.1625,
      "epoch": 7.0,
      "grad_norm": 17.625,
      "kl": 0.0974609375,
      "learning_rate": 7.67e-07,
      "loss": 0.0,
      "reward": 0.15320108719170095,
      "reward_std": 0.04531068232899997,
      "rewards/DCR_reward": 0.15320108719170095,
      "step": 700
    },
    {
      "completion_length": 93.4125,
      "epoch": 7.1,
      "grad_norm": 18.5,
      "kl": 0.101190185546875,
      "learning_rate": 7.636666666666667e-07,
      "loss": 0.0,
      "reward": 0.02544752674875781,
      "reward_std": 0.05906593499239534,
      "rewards/DCR_reward": 0.02544752674875781,
      "step": 710
    },
    {
      "completion_length": 98.0875,
      "epoch": 7.2,
      "grad_norm": 0.0,
      "kl": 0.12484130859375,
      "learning_rate": 7.603333333333332e-07,
      "loss": -0.0,
      "reward": 0.17555389162153007,
      "reward_std": 0.06005739986721892,
      "rewards/DCR_reward": 0.17555389162153007,
      "step": 720
    },
    {
      "completion_length": 85.3,
      "epoch": 7.3,
      "grad_norm": 0.03076171875,
      "kl": 0.06852807998657226,
      "learning_rate": 7.57e-07,
      "loss": 0.0,
      "reward": 0.09980509513407014,
      "reward_std": 0.022045876948665465,
      "rewards/DCR_reward": 0.09980509513407014,
      "step": 730
    },
    {
      "completion_length": 60.525,
      "epoch": 7.4,
      "grad_norm": 0.0,
      "kl": 0.146484375,
      "learning_rate": 7.536666666666667e-07,
      "loss": 0.0,
      "reward": 0.17459992747753858,
      "reward_std": 0.04686348429240752,
      "rewards/DCR_reward": 0.17459992747753858,
      "step": 740
    },
    {
      "completion_length": 146.6375,
      "epoch": 7.5,
      "grad_norm": 5.125,
      "kl": 0.13475341796875,
      "learning_rate": 7.503333333333332e-07,
      "loss": -0.0,
      "reward": 0.12549885590560733,
      "reward_std": 0.003405572484291497,
      "rewards/DCR_reward": 0.12549885590560733,
      "step": 750
    },
    {
      "epoch": 7.5,
      "eval_completion_length": 85.11125,
      "eval_kl": 0.133615665435791,
      "eval_loss": -8.860533853294328e-07,
      "eval_reward": 0.13431876484770328,
      "eval_reward_std": 0.05950616509797925,
      "eval_rewards/DCR_reward": 0.13431876484770328,
      "eval_runtime": 2234.9394,
      "eval_samples_per_second": 0.045,
      "eval_steps_per_second": 0.006,
      "step": 750
    },
    {
      "completion_length": 79.35,
      "epoch": 7.6,
      "grad_norm": 17.875,
      "kl": 0.166937255859375,
      "learning_rate": 7.47e-07,
      "loss": -0.0,
      "reward": 0.04600536972284317,
      "reward_std": 0.030051297834233992,
      "rewards/DCR_reward": 0.04600536972284317,
      "step": 760
    },
    {
      "completion_length": 59.425,
      "epoch": 7.7,
      "grad_norm": 18.875,
      "kl": 0.144036865234375,
      "learning_rate": 7.436666666666667e-07,
      "loss": -0.0,
      "reward": 0.25486029861494897,
      "reward_std": 0.06726857685171125,
      "rewards/DCR_reward": 0.25486029861494897,
      "step": 770
    },
    {
      "completion_length": 59.25,
      "epoch": 7.8,
      "grad_norm": 13.875,
      "kl": 0.1705810546875,
      "learning_rate": 7.403333333333332e-07,
      "loss": 0.0,
      "reward": 0.18612688397988678,
      "reward_std": 0.060024100821465254,
      "rewards/DCR_reward": 0.18612688397988678,
      "step": 780
    },
    {
      "completion_length": 91.675,
      "epoch": 7.9,
      "grad_norm": 8.3125,
      "kl": 0.1150146484375,
      "learning_rate": 7.37e-07,
      "loss": -0.0,
      "reward": 0.22636549319140614,
      "reward_std": 0.09487768108156161,
      "rewards/DCR_reward": 0.22636549319140614,
      "step": 790
    },
    {
      "completion_length": 89.375,
      "epoch": 8.0,
      "grad_norm": 13.25,
      "kl": 0.1668212890625,
      "learning_rate": 7.336666666666667e-07,
      "loss": -0.0,
      "reward": 0.11243776695337146,
      "reward_std": 0.0434065388621093,
      "rewards/DCR_reward": 0.11243776695337146,
      "step": 800
    },
    {
      "completion_length": 60.225,
      "epoch": 8.1,
      "grad_norm": 0.0,
      "kl": 0.144354248046875,
      "learning_rate": 7.303333333333332e-07,
      "loss": 0.0,
      "reward": 0.17428544777212665,
      "reward_std": 0.03378924725689103,
      "rewards/DCR_reward": 0.17428544777212665,
      "step": 810
    },
    {
      "completion_length": 67.4375,
      "epoch": 8.2,
      "grad_norm": 29.625,
      "kl": 0.2034912109375,
      "learning_rate": 7.27e-07,
      "loss": -0.0,
      "reward": 0.09552836455404759,
      "reward_std": 0.0525470721883039,
      "rewards/DCR_reward": 0.09552836455404759,
      "step": 820
    },
    {
      "completion_length": 78.75,
      "epoch": 8.3,
      "grad_norm": 0.0,
      "kl": 0.185302734375,
      "learning_rate": 7.236666666666666e-07,
      "loss": -0.0,
      "reward": 0.10288139216136187,
      "reward_std": 0.07548805264668772,
      "rewards/DCR_reward": 0.10288139216136187,
      "step": 830
    },
    {
      "completion_length": 104.2875,
      "epoch": 8.4,
      "grad_norm": 0.0,
      "kl": 0.1417724609375,
      "learning_rate": 7.203333333333333e-07,
      "loss": -0.0,
      "reward": 0.12569777632597834,
      "reward_std": 0.0003777303310926072,
      "rewards/DCR_reward": 0.12569777632597834,
      "step": 840
    },
    {
      "completion_length": 89.9625,
      "epoch": 8.5,
      "grad_norm": 0.0,
      "kl": 0.124444580078125,
      "learning_rate": 7.17e-07,
      "loss": 0.0,
      "reward": 0.08354408431332558,
      "reward_std": 0.016342163346146778,
      "rewards/DCR_reward": 0.08354408431332558,
      "step": 850
    },
    {
      "completion_length": 126.1625,
      "epoch": 8.6,
      "grad_norm": 5.75,
      "kl": 0.15697021484375,
      "learning_rate": 7.136666666666666e-07,
      "loss": -0.0,
      "reward": 0.18000225534196942,
      "reward_std": 0.11962818971369416,
      "rewards/DCR_reward": 0.18000225534196942,
      "step": 860
    },
    {
      "completion_length": 66.125,
      "epoch": 8.7,
      "grad_norm": 15.8125,
      "kl": 0.131787109375,
      "learning_rate": 7.103333333333333e-07,
      "loss": 0.0,
      "reward": 0.14025002100970596,
      "reward_std": 0.07934097726297296,
      "rewards/DCR_reward": 0.14025002100970596,
      "step": 870
    },
    {
      "completion_length": 85.55,
      "epoch": 8.8,
      "grad_norm": 30.875,
      "kl": 0.14227294921875,
      "learning_rate": 7.07e-07,
      "loss": 0.0,
      "reward": 0.32092891409993174,
      "reward_std": 0.12094202971202321,
      "rewards/DCR_reward": 0.32092891409993174,
      "step": 880
    },
    {
      "completion_length": 80.625,
      "epoch": 8.9,
      "grad_norm": 0.0,
      "kl": 0.11814746856689454,
      "learning_rate": 7.036666666666666e-07,
      "loss": 0.0,
      "reward": 0.09874425530433655,
      "reward_std": 0.020111887441453292,
      "rewards/DCR_reward": 0.09874425530433655,
      "step": 890
    },
    {
      "completion_length": 74.525,
      "epoch": 9.0,
      "grad_norm": 0.0,
      "kl": 0.107318115234375,
      "learning_rate": 7.003333333333333e-07,
      "loss": -0.0,
      "reward": 0.17968311300501227,
      "reward_std": 0.03413876986596733,
      "rewards/DCR_reward": 0.17968311300501227,
      "step": 900
    },
    {
      "completion_length": 57.0625,
      "epoch": 9.1,
      "grad_norm": 17.625,
      "kl": 0.158935546875,
      "learning_rate": 6.97e-07,
      "loss": 0.0,
      "reward": 0.1742305759107694,
      "reward_std": 0.07847042060523109,
      "rewards/DCR_reward": 0.1742305759107694,
      "step": 910
    },
    {
      "completion_length": 64.8625,
      "epoch": 9.2,
      "grad_norm": 0.0,
      "kl": 0.17060546875,
      "learning_rate": 6.936666666666666e-07,
      "loss": -0.0,
      "reward": 0.18635233133099974,
      "reward_std": 0.07983016533326008,
      "rewards/DCR_reward": 0.18635233133099974,
      "step": 920
    },
    {
      "completion_length": 77.625,
      "epoch": 9.3,
      "grad_norm": 23.25,
      "kl": 0.218511962890625,
      "learning_rate": 6.903333333333333e-07,
      "loss": 0.0,
      "reward": 0.19767590372357519,
      "reward_std": 0.043497299042064695,
      "rewards/DCR_reward": 0.19767590372357519,
      "step": 930
    },
    {
      "completion_length": 47.325,
      "epoch": 9.4,
      "grad_norm": 21.375,
      "kl": 0.189459228515625,
      "learning_rate": 6.87e-07,
      "loss": 0.0,
      "reward": 0.22449640462873505,
      "reward_std": 0.0027420094997694378,
      "rewards/DCR_reward": 0.22449640462873505,
      "step": 940
    },
    {
      "completion_length": 74.6,
      "epoch": 9.5,
      "grad_norm": 0.0,
      "kl": 0.14364013671875,
      "learning_rate": 6.836666666666666e-07,
      "loss": -0.0,
      "reward": 0.1431873946392443,
      "reward_std": 0.001265111715017042,
      "rewards/DCR_reward": 0.1431873946392443,
      "step": 950
    },
    {
      "completion_length": 95.7625,
      "epoch": 9.6,
      "grad_norm": 0.0,
      "kl": 0.1601806640625,
      "learning_rate": 6.803333333333333e-07,
      "loss": -0.0,
      "reward": 0.1265127849765122,
      "reward_std": 0.06392315039120149,
      "rewards/DCR_reward": 0.1265127849765122,
      "step": 960
    },
    {
      "completion_length": 79.325,
      "epoch": 9.7,
      "grad_norm": 0.019287109375,
      "kl": 0.1714111328125,
      "learning_rate": 6.77e-07,
      "loss": 0.0,
      "reward": 0.1620770814595744,
      "reward_std": 0.07664856179035269,
      "rewards/DCR_reward": 0.1620770814595744,
      "step": 970
    },
    {
      "completion_length": 113.475,
      "epoch": 9.8,
      "grad_norm": 40.5,
      "kl": 0.09014892578125,
      "learning_rate": 6.736666666666666e-07,
      "loss": -0.0,
      "reward": 0.12962240994675084,
      "reward_std": 0.08673616686003242,
      "rewards/DCR_reward": 0.12962240994675084,
      "step": 980
    },
    {
      "completion_length": 132.1875,
      "epoch": 9.9,
      "grad_norm": 3.171875,
      "kl": 0.1203857421875,
      "learning_rate": 6.703333333333333e-07,
      "loss": 0.0,
      "reward": 0.0844914206303656,
      "reward_std": 0.04573890994070098,
      "rewards/DCR_reward": 0.0844914206303656,
      "step": 990
    },
    {
      "completion_length": 91.7,
      "epoch": 10.0,
      "grad_norm": 0.0,
      "kl": 0.14872217178344727,
      "learning_rate": 6.67e-07,
      "loss": -0.0,
      "reward": 0.22451465255580844,
      "reward_std": 0.03193944031372666,
      "rewards/DCR_reward": 0.22451465255580844,
      "step": 1000
    },
    {
      "epoch": 10.0,
      "eval_completion_length": 88.34125,
      "eval_kl": 0.16021126747131348,
      "eval_loss": -2.1401815786248335e-07,
      "eval_reward": 0.170399680956034,
      "eval_reward_std": 0.052541274107022674,
      "eval_rewards/DCR_reward": 0.170399680956034,
      "eval_runtime": 2514.5153,
      "eval_samples_per_second": 0.04,
      "eval_steps_per_second": 0.005,
      "step": 1000
    },
    {
      "completion_length": 69.625,
      "epoch": 10.1,
      "grad_norm": 0.0,
      "kl": 0.16669921875,
      "learning_rate": 6.636666666666666e-07,
      "loss": -0.0,
      "reward": 0.18533597313798963,
      "reward_std": 0.06167281661574009,
      "rewards/DCR_reward": 0.18533597313798963,
      "step": 1010
    },
    {
      "completion_length": 60.6875,
      "epoch": 10.2,
      "grad_norm": 0.0,
      "kl": 0.1216064453125,
      "learning_rate": 6.603333333333333e-07,
      "loss": -0.0,
      "reward": 0.12192644038586878,
      "reward_std": 0.01787745998954051,
      "rewards/DCR_reward": 0.12192644038586878,
      "step": 1020
    },
    {
      "completion_length": 76.7625,
      "epoch": 10.3,
      "grad_norm": 7.5,
      "kl": 0.17220458984375,
      "learning_rate": 6.57e-07,
      "loss": 0.0,
      "reward": 0.11648766156286001,
      "reward_std": 0.0692232246074127,
      "rewards/DCR_reward": 0.11648766156286001,
      "step": 1030
    },
    {
      "completion_length": 88.8375,
      "epoch": 10.4,
      "grad_norm": 38.5,
      "kl": 0.130328369140625,
      "learning_rate": 6.536666666666666e-07,
      "loss": 0.0,
      "reward": 0.21984734574798495,
      "reward_std": 0.008847105817403644,
      "rewards/DCR_reward": 0.21984734574798495,
      "step": 1040
    },
    {
      "completion_length": 68.9125,
      "epoch": 10.5,
      "grad_norm": 0.0,
      "kl": 0.168310546875,
      "learning_rate": 6.503333333333332e-07,
      "loss": 0.0,
      "reward": 0.0662006882019341,
      "reward_std": 0.089736894213479,
      "rewards/DCR_reward": 0.0662006882019341,
      "step": 1050
    },
    {
      "completion_length": 92.8875,
      "epoch": 10.6,
      "grad_norm": 21.625,
      "kl": 0.1709716796875,
      "learning_rate": 6.47e-07,
      "loss": 0.0,
      "reward": 0.3175446430454031,
      "reward_std": 0.04590535781462677,
      "rewards/DCR_reward": 0.3175446430454031,
      "step": 1060
    },
    {
      "completion_length": 104.0875,
      "epoch": 10.7,
      "grad_norm": 0.0013275146484375,
      "kl": 0.1356201171875,
      "learning_rate": 6.436666666666667e-07,
      "loss": 0.0,
      "reward": 0.17272857704083436,
      "reward_std": 0.04932637963789972,
      "rewards/DCR_reward": 0.17272857704083436,
      "step": 1070
    },
    {
      "completion_length": 73.5875,
      "epoch": 10.8,
      "grad_norm": 0.0,
      "kl": 0.17826080322265625,
      "learning_rate": 6.403333333333332e-07,
      "loss": 0.0,
      "reward": 0.19967932105064393,
      "reward_std": 0.012677951477235183,
      "rewards/DCR_reward": 0.19967932105064393,
      "step": 1080
    },
    {
      "completion_length": 61.4375,
      "epoch": 10.9,
      "grad_norm": 14.9375,
      "kl": 0.2052734375,
      "learning_rate": 6.37e-07,
      "loss": 0.0,
      "reward": 0.21544204794627148,
      "reward_std": 0.016153086804820305,
      "rewards/DCR_reward": 0.21544204794627148,
      "step": 1090
    },
    {
      "completion_length": 89.9125,
      "epoch": 11.0,
      "grad_norm": 18.875,
      "kl": 0.18170166015625,
      "learning_rate": 6.336666666666667e-07,
      "loss": 0.0,
      "reward": 0.09034715148736723,
      "reward_std": 0.07943847334618112,
      "rewards/DCR_reward": 0.09034715148736723,
      "step": 1100
    },
    {
      "completion_length": 64.05,
      "epoch": 11.1,
      "grad_norm": 15.0625,
      "kl": 0.154913330078125,
      "learning_rate": 6.303333333333332e-07,
      "loss": 0.0,
      "reward": 0.24353665355592966,
      "reward_std": 0.08834277796122478,
      "rewards/DCR_reward": 0.24353665355592966,
      "step": 1110
    },
    {
      "completion_length": 61.9,
      "epoch": 11.2,
      "grad_norm": 0.1220703125,
      "kl": 0.15966796875,
      "learning_rate": 6.27e-07,
      "loss": -0.0,
      "reward": 0.08767885738052428,
      "reward_std": 0.05409979920323167,
      "rewards/DCR_reward": 0.08767885738052428,
      "step": 1120
    },
    {
      "completion_length": 54.05,
      "epoch": 11.3,
      "grad_norm": 0.0,
      "kl": 0.251708984375,
      "learning_rate": 6.236666666666667e-07,
      "loss": -0.0,
      "reward": 0.1877214941661805,
      "reward_std": 0.014956248462726762,
      "rewards/DCR_reward": 0.1877214941661805,
      "step": 1130
    },
    {
      "completion_length": 79.475,
      "epoch": 11.4,
      "grad_norm": 16.5,
      "kl": 0.163287353515625,
      "learning_rate": 6.203333333333333e-07,
      "loss": 0.0,
      "reward": 0.18192651600111276,
      "reward_std": 0.04434406632919945,
      "rewards/DCR_reward": 0.18192651600111276,
      "step": 1140
    },
    {
      "completion_length": 139.775,
      "epoch": 11.5,
      "grad_norm": 2.921875,
      "kl": 0.231005859375,
      "learning_rate": 6.17e-07,
      "loss": -0.0,
      "reward": 0.22696539172902702,
      "reward_std": 0.03848955475841649,
      "rewards/DCR_reward": 0.22696539172902702,
      "step": 1150
    },
    {
      "completion_length": 95.55,
      "epoch": 11.6,
      "grad_norm": 23.25,
      "kl": 0.170263671875,
      "learning_rate": 6.136666666666666e-07,
      "loss": -0.0,
      "reward": 0.24207095536403359,
      "reward_std": 0.07178211783611914,
      "rewards/DCR_reward": 0.24207095536403359,
      "step": 1160
    },
    {
      "completion_length": 89.1,
      "epoch": 11.7,
      "grad_norm": 15.5,
      "kl": 0.126556396484375,
      "learning_rate": 6.103333333333333e-07,
      "loss": 0.0,
      "reward": 0.11935207918286324,
      "reward_std": 0.07003376996144653,
      "rewards/DCR_reward": 0.11935207918286324,
      "step": 1170
    },
    {
      "completion_length": 63.4375,
      "epoch": 11.8,
      "grad_norm": 28.625,
      "kl": 0.18448925018310547,
      "learning_rate": 6.07e-07,
      "loss": -0.0,
      "reward": 0.11633564964868129,
      "reward_std": 0.026135455832263687,
      "rewards/DCR_reward": 0.11633564964868129,
      "step": 1180
    },
    {
      "completion_length": 89.6125,
      "epoch": 11.9,
      "grad_norm": 0.0299072265625,
      "kl": 0.12894287109375,
      "learning_rate": 6.036666666666666e-07,
      "loss": 0.0,
      "reward": 0.16700884115416556,
      "reward_std": 0.036743837507117405,
      "rewards/DCR_reward": 0.16700884115416556,
      "step": 1190
    },
    {
      "completion_length": 89.675,
      "epoch": 12.0,
      "grad_norm": 0.0009307861328125,
      "kl": 0.133929443359375,
      "learning_rate": 6.003333333333334e-07,
      "loss": -0.0,
      "reward": 0.2129200980300084,
      "reward_std": 0.017186377505953487,
      "rewards/DCR_reward": 0.2129200980300084,
      "step": 1200
    },
    {
      "completion_length": 65.125,
      "epoch": 12.1,
      "grad_norm": 0.0,
      "kl": 0.1867431640625,
      "learning_rate": 5.97e-07,
      "loss": 0.0,
      "reward": 0.2869118741014972,
      "reward_std": 0.05356231460464187,
      "rewards/DCR_reward": 0.2869118741014972,
      "step": 1210
    },
    {
      "completion_length": 61.1875,
      "epoch": 12.2,
      "grad_norm": 19.125,
      "kl": 0.209033203125,
      "learning_rate": 5.936666666666666e-07,
      "loss": -0.0,
      "reward": 0.2106538300169632,
      "reward_std": 0.01354630084197197,
      "rewards/DCR_reward": 0.2106538300169632,
      "step": 1220
    },
    {
      "completion_length": 122.4125,
      "epoch": 12.3,
      "grad_norm": 3.703125,
      "kl": 0.15078125,
      "learning_rate": 5.903333333333334e-07,
      "loss": 0.0,
      "reward": 0.19904457703232764,
      "reward_std": 0.015228879620144653,
      "rewards/DCR_reward": 0.19904457703232764,
      "step": 1230
    },
    {
      "completion_length": 60.0625,
      "epoch": 12.4,
      "grad_norm": 0.263671875,
      "kl": 0.17626953125,
      "learning_rate": 5.87e-07,
      "loss": -0.0,
      "reward": 0.06604900020174682,
      "reward_std": 0.015421837849709163,
      "rewards/DCR_reward": 0.06604900020174682,
      "step": 1240
    },
    {
      "completion_length": 77.025,
      "epoch": 12.5,
      "grad_norm": 0.0,
      "kl": 0.1548828125,
      "learning_rate": 5.836666666666666e-07,
      "loss": 0.0,
      "reward": 0.24018247241619975,
      "reward_std": 0.11031100240943488,
      "rewards/DCR_reward": 0.24018247241619975,
      "step": 1250
    },
    {
      "epoch": 12.5,
      "eval_completion_length": 79.80125,
      "eval_kl": 0.1788983154296875,
      "eval_loss": -2.466062483108544e-07,
      "eval_reward": 0.18051586833898908,
      "eval_reward_std": 0.04664378996535504,
      "eval_rewards/DCR_reward": 0.18051586833898908,
      "eval_runtime": 2637.5023,
      "eval_samples_per_second": 0.038,
      "eval_steps_per_second": 0.005,
      "step": 1250
    },
    {
      "completion_length": 109.5375,
      "epoch": 12.6,
      "grad_norm": 0.0,
      "kl": 0.1450439453125,
      "learning_rate": 5.803333333333334e-07,
      "loss": 0.0,
      "reward": 0.18088525887578727,
      "reward_std": 0.07833153888532252,
      "rewards/DCR_reward": 0.18088525887578727,
      "step": 1260
    },
    {
      "completion_length": 93.2375,
      "epoch": 12.7,
      "grad_norm": 24.625,
      "kl": 0.15348119735717775,
      "learning_rate": 5.769999999999999e-07,
      "loss": -0.0,
      "reward": 0.22176313707605005,
      "reward_std": 0.08856261784967501,
      "rewards/DCR_reward": 0.22176313707605005,
      "step": 1270
    },
    {
      "completion_length": 82.95,
      "epoch": 12.8,
      "grad_norm": 0.0,
      "kl": 0.184228515625,
      "learning_rate": 5.736666666666666e-07,
      "loss": 0.0,
      "reward": 0.1458639702643268,
      "reward_std": 0.0540214991623742,
      "rewards/DCR_reward": 0.1458639702643268,
      "step": 1280
    },
    {
      "completion_length": 108.1875,
      "epoch": 12.9,
      "grad_norm": 25.125,
      "kl": 0.19346923828125,
      "learning_rate": 5.703333333333334e-07,
      "loss": -0.0,
      "reward": 0.09487569569610059,
      "reward_std": 0.09456775116559583,
      "rewards/DCR_reward": 0.09487569569610059,
      "step": 1290
    },
    {
      "completion_length": 60.0375,
      "epoch": 13.0,
      "grad_norm": 12.1875,
      "kl": 0.14573974609375,
      "learning_rate": 5.669999999999999e-07,
      "loss": 0.0,
      "reward": 0.09252699612407014,
      "reward_std": 0.0609970541823742,
      "rewards/DCR_reward": 0.09252699612407014,
      "step": 1300
    },
    {
      "completion_length": 87.975,
      "epoch": 13.1,
      "grad_norm": 8.375,
      "kl": 0.24951171875,
      "learning_rate": 5.636666666666666e-07,
      "loss": -0.0,
      "reward": 0.18639756126794965,
      "reward_std": 0.059623961660099666,
      "rewards/DCR_reward": 0.18639756126794965,
      "step": 1310
    },
    {
      "completion_length": 46.525,
      "epoch": 13.2,
      "grad_norm": 0.0,
      "kl": 0.183843994140625,
      "learning_rate": 5.603333333333334e-07,
      "loss": 0.0,
      "reward": 0.27727809102507306,
      "reward_std": 0.012234579344567464,
      "rewards/DCR_reward": 0.27727809102507306,
      "step": 1320
    },
    {
      "completion_length": 81.5375,
      "epoch": 13.3,
      "grad_norm": 15.25,
      "kl": 0.1978515625,
      "learning_rate": 5.57e-07,
      "loss": -0.0,
      "reward": 0.34632972672116014,
      "reward_std": 0.02853981898369966,
      "rewards/DCR_reward": 0.34632972672116014,
      "step": 1330
    },
    {
      "completion_length": 116.1875,
      "epoch": 13.4,
      "grad_norm": 7.46875,
      "kl": 0.14244384765625,
      "learning_rate": 5.536666666666666e-07,
      "loss": -0.0,
      "reward": 0.1364185765822185,
      "reward_std": 0.03053054096526466,
      "rewards/DCR_reward": 0.1364185765822185,
      "step": 1340
    },
    {
      "completion_length": 96.25,
      "epoch": 13.5,
      "grad_norm": 16.875,
      "kl": 0.1332763671875,
      "learning_rate": 5.503333333333334e-07,
      "loss": -0.0,
      "reward": 0.174199710926041,
      "reward_std": 0.06987538231981034,
      "rewards/DCR_reward": 0.174199710926041,
      "step": 1350
    },
    {
      "completion_length": 73.625,
      "epoch": 13.6,
      "grad_norm": 0.01190185546875,
      "kl": 0.145501708984375,
      "learning_rate": 5.47e-07,
      "loss": 0.0,
      "reward": 0.1467902946518734,
      "reward_std": 0.021008528914126145,
      "rewards/DCR_reward": 0.1467902946518734,
      "step": 1360
    },
    {
      "completion_length": 102.9,
      "epoch": 13.7,
      "grad_norm": 0.0,
      "kl": 0.150146484375,
      "learning_rate": 5.436666666666666e-07,
      "loss": -0.0,
      "reward": 0.14278438028413803,
      "reward_std": 0.06553829507544151,
      "rewards/DCR_reward": 0.14278438028413803,
      "step": 1370
    },
    {
      "completion_length": 67.125,
      "epoch": 13.8,
      "grad_norm": 34.75,
      "kl": 0.18953857421875,
      "learning_rate": 5.403333333333333e-07,
      "loss": -0.0,
      "reward": 0.04754302315413952,
      "reward_std": 0.0033664418617263435,
      "rewards/DCR_reward": 0.04754302315413952,
      "step": 1380
    },
    {
      "completion_length": 80.1875,
      "epoch": 13.9,
      "grad_norm": 27.875,
      "kl": 0.1549560546875,
      "learning_rate": 5.37e-07,
      "loss": 0.0,
      "reward": 0.06470744522521273,
      "reward_std": 0.06195358677759941,
      "rewards/DCR_reward": 0.06470744522521273,
      "step": 1390
    },
    {
      "completion_length": 76.525,
      "epoch": 14.0,
      "grad_norm": 20.5,
      "kl": 0.18204002380371093,
      "learning_rate": 5.336666666666666e-07,
      "loss": -0.0,
      "reward": 0.25884356582537293,
      "reward_std": 0.07922300189136422,
      "rewards/DCR_reward": 0.25884356582537293,
      "step": 1400
    },
    {
      "completion_length": 61.475,
      "epoch": 14.1,
      "grad_norm": 0.0,
      "kl": 0.1508544921875,
      "learning_rate": 5.303333333333333e-07,
      "loss": 0.0,
      "reward": 0.20719661605544387,
      "reward_std": 0.0746672638963446,
      "rewards/DCR_reward": 0.20719661605544387,
      "step": 1410
    },
    {
      "completion_length": 79.1875,
      "epoch": 14.2,
      "grad_norm": 0.0,
      "kl": 0.1884521484375,
      "learning_rate": 5.27e-07,
      "loss": 0.0,
      "reward": 0.1095911561860703,
      "reward_std": 0.02852085893282492,
      "rewards/DCR_reward": 0.1095911561860703,
      "step": 1420
    },
    {
      "completion_length": 99.4875,
      "epoch": 14.3,
      "grad_norm": 8.5625,
      "kl": 0.183203125,
      "learning_rate": 5.236666666666666e-07,
      "loss": 0.0,
      "reward": 0.26417584040900693,
      "reward_std": 0.03152382288160993,
      "rewards/DCR_reward": 0.26417584040900693,
      "step": 1430
    },
    {
      "completion_length": 81.4375,
      "epoch": 14.4,
      "grad_norm": 11.8125,
      "kl": 0.147418212890625,
      "learning_rate": 5.203333333333333e-07,
      "loss": 0.0,
      "reward": 0.27334287738776764,
      "reward_std": 0.059623092689435,
      "rewards/DCR_reward": 0.27334287738776764,
      "step": 1440
    },
    {
      "completion_length": 64.7375,
      "epoch": 14.5,
      "grad_norm": 22.0,
      "kl": 0.223583984375,
      "learning_rate": 5.17e-07,
      "loss": -0.0,
      "reward": 0.10534097602358088,
      "reward_std": 0.01799371653714843,
      "rewards/DCR_reward": 0.10534097602358088,
      "step": 1450
    },
    {
      "completion_length": 69.275,
      "epoch": 14.6,
      "grad_norm": 0.0,
      "kl": 0.224200439453125,
      "learning_rate": 5.136666666666666e-07,
      "loss": -0.0,
      "reward": 0.2516032636165619,
      "reward_std": 0.07302998011000454,
      "rewards/DCR_reward": 0.2516032636165619,
      "step": 1460
    },
    {
      "completion_length": 99.375,
      "epoch": 14.7,
      "grad_norm": 19.5,
      "kl": 0.16717529296875,
      "learning_rate": 5.103333333333333e-07,
      "loss": -0.0,
      "reward": 0.10548254007007926,
      "reward_std": 0.02692217687581433,
      "rewards/DCR_reward": 0.10548254007007926,
      "step": 1470
    },
    {
      "completion_length": 79.325,
      "epoch": 14.8,
      "grad_norm": 26.375,
      "kl": 0.181884765625,
      "learning_rate": 5.07e-07,
      "loss": 0.0,
      "reward": 0.07880783905275165,
      "reward_std": 0.05647614029903707,
      "rewards/DCR_reward": 0.07880783905275165,
      "step": 1480
    },
    {
      "completion_length": 99.9125,
      "epoch": 14.9,
      "grad_norm": 18.25,
      "kl": 0.09996337890625,
      "learning_rate": 5.036666666666666e-07,
      "loss": -0.0,
      "reward": 0.15730613842606544,
      "reward_std": 0.0505793450953206,
      "rewards/DCR_reward": 0.15730613842606544,
      "step": 1490
    },
    {
      "completion_length": 74.3375,
      "epoch": 15.0,
      "grad_norm": 0.0,
      "kl": 0.167626953125,
      "learning_rate": 5.003333333333333e-07,
      "loss": -0.0,
      "reward": 0.20553053587209433,
      "reward_std": 0.06437594342569355,
      "rewards/DCR_reward": 0.20553053587209433,
      "step": 1500
    },
    {
      "epoch": 15.0,
      "eval_completion_length": 83.67125,
      "eval_kl": 0.21793760299682619,
      "eval_loss": -1.1107649697805755e-06,
      "eval_reward": 0.1789407900039805,
      "eval_reward_std": 0.04753444435028598,
      "eval_rewards/DCR_reward": 0.1789407900039805,
      "eval_runtime": 2622.8316,
      "eval_samples_per_second": 0.038,
      "eval_steps_per_second": 0.005,
      "step": 1500
    },
    {
      "completion_length": 77.8125,
      "epoch": 15.1,
      "grad_norm": 8.75,
      "kl": 0.12496776580810547,
      "learning_rate": 4.97e-07,
      "loss": 0.0,
      "reward": 0.2007469806820154,
      "reward_std": 0.11496121380478144,
      "rewards/DCR_reward": 0.2007469806820154,
      "step": 1510
    },
    {
      "completion_length": 112.4375,
      "epoch": 15.2,
      "grad_norm": 17.375,
      "kl": 0.1972900390625,
      "learning_rate": 4.936666666666666e-07,
      "loss": -0.0,
      "reward": 0.1523078629281372,
      "reward_std": 0.0487046109745279,
      "rewards/DCR_reward": 0.1523078629281372,
      "step": 1520
    },
    {
      "completion_length": 106.325,
      "epoch": 15.3,
      "grad_norm": 33.75,
      "kl": 0.1178955078125,
      "learning_rate": 4.903333333333333e-07,
      "loss": -0.0,
      "reward": 0.14267266684328206,
      "reward_std": 0.13161113108944847,
      "rewards/DCR_reward": 0.14267266684328206,
      "step": 1530
    },
    {
      "completion_length": 70.2875,
      "epoch": 15.4,
      "grad_norm": 14.8125,
      "kl": 0.2324951171875,
      "learning_rate": 4.87e-07,
      "loss": 0.0,
      "reward": 0.2017348323017359,
      "reward_std": 0.04265181252852699,
      "rewards/DCR_reward": 0.2017348323017359,
      "step": 1540
    },
    {
      "completion_length": 45.4625,
      "epoch": 15.5,
      "grad_norm": 0.0,
      "kl": 0.15888671875,
      "learning_rate": 4.836666666666666e-07,
      "loss": -0.0,
      "reward": 0.2259118565125391,
      "reward_std": 0.026894852996608164,
      "rewards/DCR_reward": 0.2259118565125391,
      "step": 1550
    },
    {
      "completion_length": 88.6625,
      "epoch": 15.6,
      "grad_norm": 23.5,
      "kl": 0.164312744140625,
      "learning_rate": 4.803333333333333e-07,
      "loss": 0.0,
      "reward": 0.15437748426338657,
      "reward_std": 0.032754498023996347,
      "rewards/DCR_reward": 0.15437748426338657,
      "step": 1560
    },
    {
      "completion_length": 78.4,
      "epoch": 15.7,
      "grad_norm": 0.0,
      "kl": 0.164697265625,
      "learning_rate": 4.769999999999999e-07,
      "loss": -0.0,
      "reward": 0.09430048232898117,
      "reward_std": 0.03383164100494014,
      "rewards/DCR_reward": 0.09430048232898117,
      "step": 1570
    },
    {
      "completion_length": 79.2875,
      "epoch": 15.8,
      "grad_norm": 0.0,
      "kl": 0.176513671875,
      "learning_rate": 4.7366666666666666e-07,
      "loss": -0.0,
      "reward": 0.23642000226536766,
      "reward_std": 0.04199942027457837,
      "rewards/DCR_reward": 0.23642000226536766,
      "step": 1580
    },
    {
      "completion_length": 86.6875,
      "epoch": 15.9,
      "grad_norm": 28.875,
      "kl": 0.234124755859375,
      "learning_rate": 4.703333333333333e-07,
      "loss": 0.0,
      "reward": 0.20127527262666262,
      "reward_std": 0.026311779970637873,
      "rewards/DCR_reward": 0.20127527262666262,
      "step": 1590
    },
    {
      "completion_length": 70.875,
      "epoch": 16.0,
      "grad_norm": 0.0,
      "kl": 0.175665283203125,
      "learning_rate": 4.67e-07,
      "loss": -0.0,
      "reward": 0.14691403629258276,
      "reward_std": 0.04396011229930537,
      "rewards/DCR_reward": 0.14691403629258276,
      "step": 1600
    },
    {
      "completion_length": 57.625,
      "epoch": 16.1,
      "grad_norm": 17.0,
      "kl": 0.20440673828125,
      "learning_rate": 4.6366666666666665e-07,
      "loss": -0.0,
      "reward": 0.13289597362745553,
      "reward_std": 0.0672353014729822,
      "rewards/DCR_reward": 0.13289597362745553,
      "step": 1610
    },
    {
      "completion_length": 99.6375,
      "epoch": 16.2,
      "grad_norm": 0.0,
      "kl": 0.117718505859375,
      "learning_rate": 4.603333333333333e-07,
      "loss": -0.0,
      "reward": 0.26933112973347306,
      "reward_std": 0.06138738352313169,
      "rewards/DCR_reward": 0.26933112973347306,
      "step": 1620
    },
    {
      "completion_length": 85.85,
      "epoch": 16.3,
      "grad_norm": 0.0,
      "kl": 0.206988525390625,
      "learning_rate": 4.57e-07,
      "loss": -0.0,
      "reward": 0.1808759123814525,
      "reward_std": 0.023974006343632937,
      "rewards/DCR_reward": 0.1808759123814525,
      "step": 1630
    },
    {
      "completion_length": 74.9625,
      "epoch": 16.4,
      "grad_norm": 19.125,
      "kl": 0.205908203125,
      "learning_rate": 4.5366666666666664e-07,
      "loss": -0.0,
      "reward": 0.2561442313250154,
      "reward_std": 0.05999695781356422,
      "rewards/DCR_reward": 0.2561442313250154,
      "step": 1640
    },
    {
      "completion_length": 51.425,
      "epoch": 16.5,
      "grad_norm": 0.0,
      "kl": 0.19951171875,
      "learning_rate": 4.503333333333333e-07,
      "loss": 0.0,
      "reward": 0.23342568413354456,
      "reward_std": 0.030879499143338762,
      "rewards/DCR_reward": 0.23342568413354456,
      "step": 1650
    },
    {
      "completion_length": 83.3625,
      "epoch": 16.6,
      "grad_norm": 0.0,
      "kl": 0.13411979675292968,
      "learning_rate": 4.4699999999999997e-07,
      "loss": 0.0,
      "reward": 0.10717827337794006,
      "reward_std": 0.055167136660065806,
      "rewards/DCR_reward": 0.10717827337794006,
      "step": 1660
    },
    {
      "completion_length": 80.0875,
      "epoch": 16.7,
      "grad_norm": 14.75,
      "kl": 0.2007080078125,
      "learning_rate": 4.4366666666666663e-07,
      "loss": 0.0,
      "reward": 0.21203166521154343,
      "reward_std": 0.05453803092241287,
      "rewards/DCR_reward": 0.21203166521154343,
      "step": 1670
    },
    {
      "completion_length": 106.975,
      "epoch": 16.8,
      "grad_norm": 16.625,
      "kl": 0.1846435546875,
      "learning_rate": 4.4033333333333335e-07,
      "loss": 0.0,
      "reward": 0.08938784400233998,
      "reward_std": 0.05009205757160089,
      "rewards/DCR_reward": 0.08938784400233998,
      "step": 1680
    },
    {
      "completion_length": 105.3875,
      "epoch": 16.9,
      "grad_norm": 0.0,
      "kl": 0.1232177734375,
      "learning_rate": 4.3699999999999996e-07,
      "loss": -0.0,
      "reward": 0.18415257817832753,
      "reward_std": 0.007105620090851516,
      "rewards/DCR_reward": 0.18415257817832753,
      "step": 1690
    },
    {
      "completion_length": 89.2375,
      "epoch": 17.0,
      "grad_norm": 17.625,
      "kl": 0.1628662109375,
      "learning_rate": 4.336666666666666e-07,
      "loss": 0.0,
      "reward": 0.10958491688361391,
      "reward_std": 0.041172702021503936,
      "rewards/DCR_reward": 0.10958491688361391,
      "step": 1700
    },
    {
      "completion_length": 90.1875,
      "epoch": 17.1,
      "grad_norm": 0.0,
      "kl": 0.15859375,
      "learning_rate": 4.3033333333333334e-07,
      "loss": 0.0,
      "reward": 0.09857021539355629,
      "reward_std": 0.02386656640956062,
      "rewards/DCR_reward": 0.09857021539355629,
      "step": 1710
    },
    {
      "completion_length": 63.25,
      "epoch": 17.2,
      "grad_norm": 0.0,
      "kl": 0.16219091415405273,
      "learning_rate": 4.2699999999999995e-07,
      "loss": -0.0,
      "reward": 0.054912311234511436,
      "reward_std": 0.021657621535587167,
      "rewards/DCR_reward": 0.054912311234511436,
      "step": 1720
    },
    {
      "completion_length": 75.025,
      "epoch": 17.3,
      "grad_norm": 23.875,
      "kl": 0.208575439453125,
      "learning_rate": 4.2366666666666666e-07,
      "loss": 0.0,
      "reward": 0.3105363720096648,
      "reward_std": 0.03136685772915371,
      "rewards/DCR_reward": 0.3105363720096648,
      "step": 1730
    },
    {
      "completion_length": 76.5875,
      "epoch": 17.4,
      "grad_norm": 11.875,
      "kl": 0.22567138671875,
      "learning_rate": 4.203333333333333e-07,
      "loss": 0.0,
      "reward": 0.1657306909793988,
      "reward_std": 0.06016184531727049,
      "rewards/DCR_reward": 0.1657306909793988,
      "step": 1740
    },
    {
      "completion_length": 76.5625,
      "epoch": 17.5,
      "grad_norm": 0.10302734375,
      "kl": 0.21397705078125,
      "learning_rate": 4.17e-07,
      "loss": 0.0,
      "reward": 0.3156662947498262,
      "reward_std": 0.017444778925437276,
      "rewards/DCR_reward": 0.3156662947498262,
      "step": 1750
    },
    {
      "epoch": 17.5,
      "eval_completion_length": 83.015,
      "eval_kl": 0.1820037841796875,
      "eval_loss": 1.8720327261689818e-06,
      "eval_reward": 0.1803470617614221,
      "eval_reward_std": 0.047306431781344714,
      "eval_rewards/DCR_reward": 0.1803470617614221,
      "eval_runtime": 2666.1674,
      "eval_samples_per_second": 0.038,
      "eval_steps_per_second": 0.005,
      "step": 1750
    },
    {
      "completion_length": 71.3125,
      "epoch": 17.6,
      "grad_norm": 18.25,
      "kl": 0.187255859375,
      "learning_rate": 4.1366666666666665e-07,
      "loss": 0.0,
      "reward": 0.20512793064117432,
      "reward_std": 0.047461129271408706,
      "rewards/DCR_reward": 0.20512793064117432,
      "step": 1760
    },
    {
      "completion_length": 59.45,
      "epoch": 17.7,
      "grad_norm": 17.0,
      "kl": 0.1356689453125,
      "learning_rate": 4.103333333333333e-07,
      "loss": 0.0,
      "reward": 0.14152073024306447,
      "reward_std": 0.09260614971240103,
      "rewards/DCR_reward": 0.14152073024306447,
      "step": 1770
    },
    {
      "completion_length": 66.0625,
      "epoch": 17.8,
      "grad_norm": 0.0,
      "kl": 0.1925048828125,
      "learning_rate": 4.07e-07,
      "loss": -0.0,
      "reward": 0.25652417142409834,
      "reward_std": 0.04226240784919355,
      "rewards/DCR_reward": 0.25652417142409834,
      "step": 1780
    },
    {
      "completion_length": 109.575,
      "epoch": 17.9,
      "grad_norm": 0.0,
      "kl": 0.165313720703125,
      "learning_rate": 4.0366666666666664e-07,
      "loss": 0.0,
      "reward": 0.21949415714479983,
      "reward_std": 0.08642580564569471,
      "rewards/DCR_reward": 0.21949415714479983,
      "step": 1790
    },
    {
      "completion_length": 64.35,
      "epoch": 18.0,
      "grad_norm": 20.875,
      "kl": 0.1558349609375,
      "learning_rate": 4.003333333333333e-07,
      "loss": -0.0,
      "reward": 0.16233385398518294,
      "reward_std": 0.04793143280548975,
      "rewards/DCR_reward": 0.16233385398518294,
      "step": 1800
    },
    {
      "completion_length": 47.7375,
      "epoch": 18.1,
      "grad_norm": 13.125,
      "kl": 0.280615234375,
      "learning_rate": 3.97e-07,
      "loss": -0.0,
      "reward": 0.1926513019599952,
      "reward_std": 0.022014413893526808,
      "rewards/DCR_reward": 0.1926513019599952,
      "step": 1810
    },
    {
      "completion_length": 86.2875,
      "epoch": 18.2,
      "grad_norm": 0.0,
      "kl": 0.19091796875,
      "learning_rate": 3.9366666666666663e-07,
      "loss": 0.0,
      "reward": 0.18852764302864672,
      "reward_std": 0.033248070168701814,
      "rewards/DCR_reward": 0.18852764302864672,
      "step": 1820
    },
    {
      "completion_length": 80.175,
      "epoch": 18.3,
      "grad_norm": 13.0,
      "kl": 0.12606201171875,
      "learning_rate": 3.903333333333333e-07,
      "loss": 0.0,
      "reward": 0.17205136871198193,
      "reward_std": 0.019140477599285076,
      "rewards/DCR_reward": 0.17205136871198193,
      "step": 1830
    },
    {
      "completion_length": 79.5125,
      "epoch": 18.4,
      "grad_norm": 0.96484375,
      "kl": 0.17064599990844725,
      "learning_rate": 3.87e-07,
      "loss": 0.0,
      "reward": 0.07664455490885302,
      "reward_std": 0.04274343762583612,
      "rewards/DCR_reward": 0.07664455490885302,
      "step": 1840
    },
    {
      "completion_length": 81.4375,
      "epoch": 18.5,
      "grad_norm": 0.0,
      "kl": 0.1732177734375,
      "learning_rate": 3.836666666666666e-07,
      "loss": 0.0,
      "reward": 0.23127210177481175,
      "reward_std": 0.04357012182008475,
      "rewards/DCR_reward": 0.23127210177481175,
      "step": 1850
    },
    {
      "completion_length": 75.625,
      "epoch": 18.6,
      "grad_norm": 3.03125,
      "kl": 0.159375,
      "learning_rate": 3.8033333333333334e-07,
      "loss": 0.0,
      "reward": 0.27262264720629903,
      "reward_std": 0.05521087486195313,
      "rewards/DCR_reward": 0.27262264720629903,
      "step": 1860
    },
    {
      "completion_length": 134.3125,
      "epoch": 18.7,
      "grad_norm": 0.36328125,
      "kl": 0.1446624755859375,
      "learning_rate": 3.77e-07,
      "loss": 0.0,
      "reward": 0.11476055827224627,
      "reward_std": 0.02939585350050038,
      "rewards/DCR_reward": 0.11476055827224627,
      "step": 1870
    },
    {
      "completion_length": 72.4125,
      "epoch": 18.8,
      "grad_norm": 23.5,
      "kl": 0.1951904296875,
      "learning_rate": 3.736666666666666e-07,
      "loss": -0.0,
      "reward": 0.14638857576064765,
      "reward_std": 0.04362176135448124,
      "rewards/DCR_reward": 0.14638857576064765,
      "step": 1880
    },
    {
      "completion_length": 70.75,
      "epoch": 18.9,
      "grad_norm": 0.0,
      "kl": 0.186572265625,
      "learning_rate": 3.7033333333333333e-07,
      "loss": 0.0,
      "reward": 0.1708667165134102,
      "reward_std": 0.07184823253192008,
      "rewards/DCR_reward": 0.1708667165134102,
      "step": 1890
    },
    {
      "completion_length": 52.675,
      "epoch": 19.0,
      "grad_norm": 19.625,
      "kl": 0.183404541015625,
      "learning_rate": 3.67e-07,
      "loss": -0.0,
      "reward": 0.2897725820541382,
      "reward_std": 0.0576688679928111,
      "rewards/DCR_reward": 0.2897725820541382,
      "step": 1900
    },
    {
      "completion_length": 81.7625,
      "epoch": 19.1,
      "grad_norm": 11.5,
      "kl": 0.217041015625,
      "learning_rate": 3.6366666666666665e-07,
      "loss": 0.0,
      "reward": 0.156359511311166,
      "reward_std": 0.051926983702384175,
      "rewards/DCR_reward": 0.156359511311166,
      "step": 1910
    },
    {
      "completion_length": 130.8125,
      "epoch": 19.2,
      "grad_norm": 9.8125,
      "kl": 0.10965576171875,
      "learning_rate": 3.603333333333333e-07,
      "loss": 0.0,
      "reward": 0.10657796601299196,
      "reward_std": 0.0012149818532634527,
      "rewards/DCR_reward": 0.10657796601299196,
      "step": 1920
    },
    {
      "completion_length": 63.2375,
      "epoch": 19.3,
      "grad_norm": 19.0,
      "kl": 0.26494140625,
      "learning_rate": 3.57e-07,
      "loss": 0.0,
      "reward": 0.25602573398500683,
      "reward_std": 0.04811685611639405,
      "rewards/DCR_reward": 0.25602573398500683,
      "step": 1930
    },
    {
      "completion_length": 71.2125,
      "epoch": 19.4,
      "grad_norm": 0.039794921875,
      "kl": 0.2076171875,
      "learning_rate": 3.5366666666666664e-07,
      "loss": 0.0,
      "reward": 0.2313489816733636,
      "reward_std": 0.037926042393394255,
      "rewards/DCR_reward": 0.2313489816733636,
      "step": 1940
    },
    {
      "completion_length": 80.075,
      "epoch": 19.5,
      "grad_norm": 4.40625,
      "kl": 0.194873046875,
      "learning_rate": 3.503333333333333e-07,
      "loss": 0.0,
      "reward": 0.2062260712031275,
      "reward_std": 0.07047836606834608,
      "rewards/DCR_reward": 0.2062260712031275,
      "step": 1950
    },
    {
      "completion_length": 47.3875,
      "epoch": 19.6,
      "grad_norm": 9.125,
      "kl": 0.159039306640625,
      "learning_rate": 3.4699999999999997e-07,
      "loss": 0.0,
      "reward": 0.3275539556518197,
      "reward_std": 0.056122380661634,
      "rewards/DCR_reward": 0.3275539556518197,
      "step": 1960
    },
    {
      "completion_length": 81.525,
      "epoch": 19.7,
      "grad_norm": 17.875,
      "kl": 0.14603710174560547,
      "learning_rate": 3.436666666666667e-07,
      "loss": -0.0,
      "reward": 0.08158219500910491,
      "reward_std": 0.0475987725701998,
      "rewards/DCR_reward": 0.08158219500910491,
      "step": 1970
    },
    {
      "completion_length": 90.35,
      "epoch": 19.8,
      "grad_norm": 0.0,
      "kl": 0.16241455078125,
      "learning_rate": 3.403333333333333e-07,
      "loss": 0.0,
      "reward": 0.14252315481426195,
      "reward_std": 0.0510443922455579,
      "rewards/DCR_reward": 0.14252315481426195,
      "step": 1980
    },
    {
      "completion_length": 77.0875,
      "epoch": 19.9,
      "grad_norm": 0.8359375,
      "kl": 0.1235107421875,
      "learning_rate": 3.37e-07,
      "loss": -0.0,
      "reward": 0.11888001729967072,
      "reward_std": 0.041191360527292886,
      "rewards/DCR_reward": 0.11888001729967072,
      "step": 1990
    },
    {
      "completion_length": 100.05,
      "epoch": 20.0,
      "grad_norm": 8.125,
      "kl": 0.209521484375,
      "learning_rate": 3.336666666666667e-07,
      "loss": 0.0,
      "reward": 0.16669638943858445,
      "reward_std": 0.028858381987083702,
      "rewards/DCR_reward": 0.16669638943858445,
      "step": 2000
    },
    {
      "epoch": 20.0,
      "eval_completion_length": 84.2325,
      "eval_kl": 0.17778059005737304,
      "eval_loss": 6.65434640723106e-07,
      "eval_reward": 0.18110041963984258,
      "eval_reward_std": 0.0566114658890848,
      "eval_rewards/DCR_reward": 0.18110041963984258,
      "eval_runtime": 2663.7287,
      "eval_samples_per_second": 0.038,
      "eval_steps_per_second": 0.005,
      "step": 2000
    }
  ],
  "logging_steps": 10,
  "max_steps": 3000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}