{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.11428571428571428,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2700.4271850585938,
      "entropy": 0.36767578125,
      "epoch": 0.001142857142857143,
      "grad_norm": 0.11875967681407928,
      "kl": 0.0,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 0.7604166893288493,
      "reward_std": 0.4268697127699852,
      "rewards/accuracy_reward": 0.25000001303851604,
      "rewards/format_reward": 0.5104166669771075,
      "step": 1
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3127.3958435058594,
      "entropy": 0.353515625,
      "epoch": 0.002285714285714286,
      "grad_norm": 0.14844515919685364,
      "kl": 0.0,
      "learning_rate": 1e-07,
      "loss": 0.0,
      "reward": 0.6458333637565374,
      "reward_std": 0.4249730706214905,
      "rewards/accuracy_reward": 0.2812500102445483,
      "rewards/format_reward": 0.3645833386108279,
      "step": 2
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3672.3229370117188,
      "entropy": 0.43701171875,
      "epoch": 0.0034285714285714284,
      "grad_norm": 0.15467818081378937,
      "kl": 4.011392593383789e-05,
      "learning_rate": 2e-07,
      "loss": 0.0,
      "reward": 0.20833333674818277,
      "reward_std": 0.3576926067471504,
      "rewards/accuracy_reward": 0.05208333395421505,
      "rewards/format_reward": 0.15625000838190317,
      "step": 3
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2434.0000610351562,
      "entropy": 0.40673828125,
      "epoch": 0.004571428571428572,
      "grad_norm": 0.1262386441230774,
      "kl": 3.9577484130859375e-05,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.6666666716337204,
      "reward_std": 0.3760714754462242,
      "rewards/accuracy_reward": 0.07291666697710752,
      "rewards/format_reward": 0.5937500074505806,
      "step": 4
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3514.447998046875,
      "entropy": 0.46142578125,
      "epoch": 0.005714285714285714,
      "grad_norm": 0.11896030604839325,
      "kl": 3.7997961044311523e-05,
      "learning_rate": 4e-07,
      "loss": 0.0,
      "reward": 0.4375000149011612,
      "reward_std": 0.45638682320713997,
      "rewards/accuracy_reward": 0.0729166716337204,
      "rewards/format_reward": 0.3645833432674408,
      "step": 5
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3369.8125610351562,
      "entropy": 0.46142578125,
      "epoch": 0.006857142857142857,
      "grad_norm": 0.11138290911912918,
      "kl": 5.46574592590332e-05,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.3333333395421505,
      "reward_std": 0.3872676342725754,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.2708333395421505,
      "step": 6
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3402.6146240234375,
      "entropy": 0.400390625,
      "epoch": 0.008,
      "grad_norm": 0.10288004577159882,
      "kl": 2.6792287826538086e-05,
      "learning_rate": 6e-07,
      "loss": 0.0,
      "reward": 0.833333358168602,
      "reward_std": 0.5691196992993355,
      "rewards/accuracy_reward": 0.2500000074505806,
      "rewards/format_reward": 0.5833333358168602,
      "step": 7
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2891.447998046875,
      "entropy": 0.357421875,
      "epoch": 0.009142857142857144,
      "grad_norm": 0.0705084502696991,
      "kl": 2.35140323638916e-05,
      "learning_rate": 7e-07,
      "loss": 0.0,
      "reward": 0.8541666865348816,
      "reward_std": 0.22278132662177086,
      "rewards/accuracy_reward": 0.3750000102445483,
      "rewards/format_reward": 0.479166679084301,
      "step": 8
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3418.947998046875,
      "entropy": 0.45068359375,
      "epoch": 0.010285714285714285,
      "grad_norm": 0.15328024327754974,
      "kl": 4.3720006942749023e-05,
      "learning_rate": 8e-07,
      "loss": 0.0,
      "reward": 0.5000000223517418,
      "reward_std": 0.6385103613138199,
      "rewards/accuracy_reward": 0.15625000279396772,
      "rewards/format_reward": 0.3437500149011612,
      "step": 9
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2938.40625,
      "entropy": 0.3408203125,
      "epoch": 0.011428571428571429,
      "grad_norm": 0.13008344173431396,
      "kl": 2.7120113372802734e-05,
      "learning_rate": 9e-07,
      "loss": 0.0,
      "reward": 0.5625000149011612,
      "reward_std": 0.4825209602713585,
      "rewards/accuracy_reward": 0.15625000558793545,
      "rewards/format_reward": 0.4062500111758709,
      "step": 10
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3608.0625610351562,
      "entropy": 0.37744140625,
      "epoch": 0.012571428571428572,
      "grad_norm": 0.13492900133132935,
      "kl": 2.568960189819336e-05,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.20833334140479565,
      "reward_std": 0.3463892936706543,
      "rewards/accuracy_reward": 0.06250000093132257,
      "rewards/format_reward": 0.14583333488553762,
      "step": 11
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2577.541748046875,
      "entropy": 0.40869140625,
      "epoch": 0.013714285714285714,
      "grad_norm": 0.17103247344493866,
      "kl": 3.412365913391113e-05,
      "learning_rate": 9.997258721585931e-07,
      "loss": 0.0,
      "reward": 0.7812500298023224,
      "reward_std": 0.4395437017083168,
      "rewards/accuracy_reward": 0.17708333861082792,
      "rewards/format_reward": 0.6041666865348816,
      "step": 12
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3150.9168090820312,
      "entropy": 0.39013671875,
      "epoch": 0.014857142857142857,
      "grad_norm": 0.15444810688495636,
      "kl": 2.7447938919067383e-05,
      "learning_rate": 9.989038226169207e-07,
      "loss": 0.0,
      "reward": 0.6145833507180214,
      "reward_std": 0.32639050483703613,
      "rewards/accuracy_reward": 0.18750000279396772,
      "rewards/format_reward": 0.4270833432674408,
      "step": 13
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3099.6771240234375,
      "entropy": 0.3623046875,
      "epoch": 0.016,
      "grad_norm": 0.22930943965911865,
      "kl": 2.720952033996582e-05,
      "learning_rate": 9.975348529157229e-07,
      "loss": 0.0,
      "reward": 0.6041666716337204,
      "reward_std": 0.530846893787384,
      "rewards/accuracy_reward": 0.18750000651925802,
      "rewards/format_reward": 0.416666679084301,
      "step": 14
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2916.572998046875,
      "entropy": 0.3740234375,
      "epoch": 0.017142857142857144,
      "grad_norm": 0.09911036491394043,
      "kl": 2.2709369659423828e-05,
      "learning_rate": 9.956206309337066e-07,
      "loss": 0.0,
      "reward": 0.7187500149011612,
      "reward_std": 0.3208988383412361,
      "rewards/accuracy_reward": 0.2604166679084301,
      "rewards/format_reward": 0.4583333507180214,
      "step": 15
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3797.3541870117188,
      "entropy": 0.45654296875,
      "epoch": 0.018285714285714287,
      "grad_norm": 0.12035630643367767,
      "kl": 2.969801425933838e-05,
      "learning_rate": 9.931634888554935e-07,
      "loss": 0.0,
      "reward": 0.18750000651925802,
      "reward_std": 0.3208543583750725,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.12500000465661287,
      "step": 16
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2549.604217529297,
      "entropy": 0.44873046875,
      "epoch": 0.019428571428571427,
      "grad_norm": 0.17906327545642853,
      "kl": 3.573298454284668e-05,
      "learning_rate": 9.901664203302124e-07,
      "loss": 0.0,
      "reward": 0.8437500223517418,
      "reward_std": 0.5051586776971817,
      "rewards/accuracy_reward": 0.27083334885537624,
      "rewards/format_reward": 0.572916679084301,
      "step": 17
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3061.9896240234375,
      "entropy": 0.35693359375,
      "epoch": 0.02057142857142857,
      "grad_norm": 0.10486536473035812,
      "kl": 2.0988285541534424e-05,
      "learning_rate": 9.866330768241983e-07,
      "loss": 0.0,
      "reward": 0.6770833656191826,
      "reward_std": 0.45130809396505356,
      "rewards/accuracy_reward": 0.19791667349636555,
      "rewards/format_reward": 0.4791666828095913,
      "step": 18
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3092.8959350585938,
      "entropy": 0.39306640625,
      "epoch": 0.021714285714285714,
      "grad_norm": 0.1308853030204773,
      "kl": 4.026293754577637e-05,
      "learning_rate": 9.825677631722435e-07,
      "loss": 0.0,
      "reward": 0.864583358168602,
      "reward_std": 0.630811333656311,
      "rewards/accuracy_reward": 0.39583334513008595,
      "rewards/format_reward": 0.4687500149011612,
      "step": 19
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2584.5833740234375,
      "entropy": 0.3388671875,
      "epoch": 0.022857142857142857,
      "grad_norm": 0.16664722561836243,
      "kl": 4.4479966163635254e-05,
      "learning_rate": 9.779754323328192e-07,
      "loss": 0.0,
      "reward": 0.958333358168602,
      "reward_std": 0.539019413292408,
      "rewards/accuracy_reward": 0.2812500037252903,
      "rewards/format_reward": 0.677083358168602,
      "step": 20
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2894.4479370117188,
      "entropy": 0.4404296875,
      "epoch": 0.024,
      "grad_norm": 0.11469059437513351,
      "kl": 0.00022789835929870605,
      "learning_rate": 9.728616793536587e-07,
      "loss": 0.0,
      "reward": 0.5937500055879354,
      "reward_std": 0.3156265318393707,
      "rewards/accuracy_reward": 0.15625000279396772,
      "rewards/format_reward": 0.4375000027939677,
      "step": 21
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 1811.4271697998047,
      "entropy": 0.39404296875,
      "epoch": 0.025142857142857144,
      "grad_norm": 0.1692061573266983,
      "kl": 0.00019675493240356445,
      "learning_rate": 9.672327345550543e-07,
      "loss": 0.0,
      "reward": 1.1250000298023224,
      "reward_std": 0.41911277174949646,
      "rewards/accuracy_reward": 0.30208334419876337,
      "rewards/format_reward": 0.8229166865348816,
      "step": 22
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2823.072998046875,
      "entropy": 0.38671875,
      "epoch": 0.026285714285714287,
      "grad_norm": 0.1204013079404831,
      "kl": 7.50422477722168e-05,
      "learning_rate": 9.610954559391704e-07,
      "loss": 0.0,
      "reward": 0.6562500223517418,
      "reward_std": 0.506660170853138,
      "rewards/accuracy_reward": 0.16666667349636555,
      "rewards/format_reward": 0.4895833432674408,
      "step": 23
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2825.416748046875,
      "entropy": 0.38525390625,
      "epoch": 0.027428571428571427,
      "grad_norm": 0.147517591714859,
      "kl": 0.00011518597602844238,
      "learning_rate": 9.54457320834625e-07,
      "loss": 0.0,
      "reward": 0.8229166865348816,
      "reward_std": 0.6497645676136017,
      "rewards/accuracy_reward": 0.2708333358168602,
      "rewards/format_reward": 0.5520833507180214,
      "step": 24
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2880.7813110351562,
      "entropy": 0.43115234375,
      "epoch": 0.02857142857142857,
      "grad_norm": 0.14937955141067505,
      "kl": 0.0002092123031616211,
      "learning_rate": 9.473264167865171e-07,
      "loss": 0.0,
      "reward": 0.7187500102445483,
      "reward_std": 0.3600482754409313,
      "rewards/accuracy_reward": 0.26041667722165585,
      "rewards/format_reward": 0.4583333386108279,
      "step": 25
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3217.5209350585938,
      "entropy": 0.41455078125,
      "epoch": 0.029714285714285714,
      "grad_norm": 0.10219196230173111,
      "kl": 2.9712915420532227e-05,
      "learning_rate": 9.397114317029974e-07,
      "loss": 0.0,
      "reward": 0.6458333507180214,
      "reward_std": 0.3189500719308853,
      "rewards/accuracy_reward": 0.1770833358168602,
      "rewards/format_reward": 0.4687500074505806,
      "step": 26
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3267.0521850585938,
      "entropy": 0.4404296875,
      "epoch": 0.030857142857142857,
      "grad_norm": 0.12435087561607361,
      "kl": 0.0001267567276954651,
      "learning_rate": 9.316216432703916e-07,
      "loss": 0.0,
      "reward": 0.5000000149011612,
      "reward_std": 0.47042689844965935,
      "rewards/accuracy_reward": 0.11458333861082792,
      "rewards/format_reward": 0.3854166716337204,
      "step": 27
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3018.5938110351562,
      "entropy": 0.40185546875,
      "epoch": 0.032,
      "grad_norm": 0.13387200236320496,
      "kl": 0.00012268126010894775,
      "learning_rate": 9.230669076497687e-07,
      "loss": 0.0,
      "reward": 0.7395833432674408,
      "reward_std": 0.40498094633221626,
      "rewards/accuracy_reward": 0.30208334140479565,
      "rewards/format_reward": 0.4375000074505806,
      "step": 28
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3436.9688110351562,
      "entropy": 0.4326171875,
      "epoch": 0.03314285714285714,
      "grad_norm": 0.18476502597332,
      "kl": 0.0002486705780029297,
      "learning_rate": 9.140576474687263e-07,
      "loss": 0.0,
      "reward": 0.3333333460614085,
      "reward_std": 0.3739718273282051,
      "rewards/accuracy_reward": 0.07291666697710752,
      "rewards/format_reward": 0.2604166818782687,
      "step": 29
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3012.510498046875,
      "entropy": 0.38818359375,
      "epoch": 0.03428571428571429,
      "grad_norm": 0.10571455210447311,
      "kl": 0.0007374286651611328,
      "learning_rate": 9.046048391230247e-07,
      "loss": 0.0,
      "reward": 0.7291666716337204,
      "reward_std": 0.5458482205867767,
      "rewards/accuracy_reward": 0.22916667256504297,
      "rewards/format_reward": 0.500000013038516,
      "step": 30
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3214.7083740234375,
      "entropy": 0.38037109375,
      "epoch": 0.03542857142857143,
      "grad_norm": 0.1431698352098465,
      "kl": 0.0004398822784423828,
      "learning_rate": 8.9471999940354e-07,
      "loss": 0.0,
      "reward": 0.5312500298023224,
      "reward_std": 0.5084670260548592,
      "rewards/accuracy_reward": 0.17708334140479565,
      "rewards/format_reward": 0.3541666716337204,
      "step": 31
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3432.5833740234375,
      "entropy": 0.4443359375,
      "epoch": 0.036571428571428574,
      "grad_norm": 0.09518005698919296,
      "kl": 0.0003883838653564453,
      "learning_rate": 8.844151714648274e-07,
      "loss": 0.0,
      "reward": 0.6354166939854622,
      "reward_std": 0.47362349927425385,
      "rewards/accuracy_reward": 0.2708333367481828,
      "rewards/format_reward": 0.36458334140479565,
      "step": 32
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3424.1043090820312,
      "entropy": 0.37109375,
      "epoch": 0.037714285714285714,
      "grad_norm": 0.12133150547742844,
      "kl": 0.0007408857345581055,
      "learning_rate": 8.737029101523929e-07,
      "loss": 0.0,
      "reward": 0.645833358168602,
      "reward_std": 0.5898082852363586,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.4166666753590107,
      "step": 33
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2531.0521240234375,
      "entropy": 0.435302734375,
      "epoch": 0.038857142857142854,
      "grad_norm": 0.16009539365768433,
      "kl": 0.0005445480346679688,
      "learning_rate": 8.625962667065487e-07,
      "loss": 0.0,
      "reward": 0.8958333358168602,
      "reward_std": 0.5157222747802734,
      "rewards/accuracy_reward": 0.35416667722165585,
      "rewards/format_reward": 0.5416666744276881,
      "step": 34
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3209.281280517578,
      "entropy": 0.42431640625,
      "epoch": 0.04,
      "grad_norm": 0.14229460060596466,
      "kl": 0.001219034194946289,
      "learning_rate": 8.511087728614862e-07,
      "loss": 0.0,
      "reward": 0.6562500074505806,
      "reward_std": 0.6340432465076447,
      "rewards/accuracy_reward": 0.25000000931322575,
      "rewards/format_reward": 0.4062500111758709,
      "step": 35
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3487.9063720703125,
      "entropy": 0.48828125,
      "epoch": 0.04114285714285714,
      "grad_norm": 0.1350383758544922,
      "kl": 0.0010302066802978516,
      "learning_rate": 8.392544243589427e-07,
      "loss": 0.0,
      "reward": 0.3020833386108279,
      "reward_std": 0.4083498902618885,
      "rewards/accuracy_reward": 0.05208333395421505,
      "rewards/format_reward": 0.2500000027939677,
      "step": 36
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3341.541748046875,
      "entropy": 0.44482421875,
      "epoch": 0.04228571428571429,
      "grad_norm": 0.09916561841964722,
      "kl": 0.0008041858673095703,
      "learning_rate": 8.270476638965461e-07,
      "loss": 0.0,
      "reward": 0.30208334885537624,
      "reward_std": 0.13339675217866898,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.28125000558793545,
      "step": 37
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3503.510498046875,
      "entropy": 0.4560546875,
      "epoch": 0.04342857142857143,
      "grad_norm": 0.0784042626619339,
      "kl": 0.00047469139099121094,
      "learning_rate": 8.145033635316128e-07,
      "loss": 0.0,
      "reward": 0.322916679084301,
      "reward_std": 0.22953829169273376,
      "rewards/accuracy_reward": 0.13541666697710752,
      "rewards/format_reward": 0.1875000111758709,
      "step": 38
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3080.7293090820312,
      "entropy": 0.36767578125,
      "epoch": 0.044571428571428574,
      "grad_norm": 0.12402305006980896,
      "kl": 0.0004640817642211914,
      "learning_rate": 8.01636806561836e-07,
      "loss": 0.0,
      "reward": 0.7083333507180214,
      "reward_std": 0.32364466041326523,
      "rewards/accuracy_reward": 0.23958333861082792,
      "rewards/format_reward": 0.4687500074505806,
      "step": 39
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2654.135498046875,
      "entropy": 0.36572265625,
      "epoch": 0.045714285714285714,
      "grad_norm": 0.1607232689857483,
      "kl": 0.0020923614501953125,
      "learning_rate": 7.884636689049422e-07,
      "loss": 0.0001,
      "reward": 0.718750037252903,
      "reward_std": 0.39055147767066956,
      "rewards/accuracy_reward": 0.19791666697710752,
      "rewards/format_reward": 0.520833358168602,
      "step": 40
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3220.4688720703125,
      "entropy": 0.38232421875,
      "epoch": 0.046857142857142854,
      "grad_norm": 0.11851893365383148,
      "kl": 0.0005588531494140625,
      "learning_rate": 7.75e-07,
      "loss": 0.0,
      "reward": 0.5729166865348816,
      "reward_std": 0.5243071913719177,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.4479166865348816,
      "step": 41
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3080.8959350585938,
      "entropy": 0.48876953125,
      "epoch": 0.048,
      "grad_norm": 0.057032641023397446,
      "kl": 0.0005335807800292969,
      "learning_rate": 7.612622032536507e-07,
      "loss": 0.0,
      "reward": 0.32291666977107525,
      "reward_std": 0.13653486967086792,
      "rewards/accuracy_reward": 0.010416666977107525,
      "rewards/format_reward": 0.3125000027939677,
      "step": 42
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3166.4584350585938,
      "entropy": 0.40380859375,
      "epoch": 0.04914285714285714,
      "grad_norm": 0.1308462768793106,
      "kl": 0.001255035400390625,
      "learning_rate": 7.472670160550848e-07,
      "loss": 0.0001,
      "reward": 0.3958333432674408,
      "reward_std": 0.26092809066176414,
      "rewards/accuracy_reward": 0.12500000093132257,
      "rewards/format_reward": 0.2708333358168602,
      "step": 43
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2762.635467529297,
      "entropy": 0.3974609375,
      "epoch": 0.05028571428571429,
      "grad_norm": 0.15369124710559845,
      "kl": 0.0009186267852783203,
      "learning_rate": 7.330314893841101e-07,
      "loss": 0.0,
      "reward": 0.854166692122817,
      "reward_std": 0.41533562541007996,
      "rewards/accuracy_reward": 0.3020833507180214,
      "rewards/format_reward": 0.5520833414047956,
      "step": 44
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3691.5521240234375,
      "entropy": 0.4072265625,
      "epoch": 0.05142857142857143,
      "grad_norm": 0.08415117114782333,
      "kl": 0.0013604164123535156,
      "learning_rate": 7.185729670371604e-07,
      "loss": 0.0001,
      "reward": 0.5520833488553762,
      "reward_std": 0.4458491802215576,
      "rewards/accuracy_reward": 0.23958334606140852,
      "rewards/format_reward": 0.3125000102445483,
      "step": 45
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3427.3438720703125,
      "entropy": 0.49755859375,
      "epoch": 0.052571428571428575,
      "grad_norm": 0.13379938900470734,
      "kl": 0.0010638236999511719,
      "learning_rate": 7.039090644965509e-07,
      "loss": 0.0,
      "reward": 0.23958334606140852,
      "reward_std": 0.25305089354515076,
      "rewards/accuracy_reward": 0.031250000931322575,
      "rewards/format_reward": 0.20833333488553762,
      "step": 46
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3061.1250610351562,
      "entropy": 0.4091796875,
      "epoch": 0.053714285714285714,
      "grad_norm": 0.15724660456180573,
      "kl": 0.0008676052093505859,
      "learning_rate": 6.890576474687263e-07,
      "loss": 0.0,
      "reward": 0.6979166939854622,
      "reward_std": 0.6069704815745354,
      "rewards/accuracy_reward": 0.2395833395421505,
      "rewards/format_reward": 0.4583333432674408,
      "step": 47
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2838.947998046875,
      "entropy": 0.38330078125,
      "epoch": 0.054857142857142854,
      "grad_norm": 0.13557472825050354,
      "kl": 0.0023670196533203125,
      "learning_rate": 6.740368101176495e-07,
      "loss": 0.0001,
      "reward": 0.7083333656191826,
      "reward_std": 0.4954182878136635,
      "rewards/accuracy_reward": 0.20833334047347307,
      "rewards/format_reward": 0.5000000111758709,
      "step": 48
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2608.3438110351562,
      "entropy": 0.39111328125,
      "epoch": 0.056,
      "grad_norm": 0.27688586711883545,
      "kl": 0.01327657699584961,
      "learning_rate": 6.588648530198504e-07,
      "loss": 0.0005,
      "reward": 0.7291666865348816,
      "reward_std": 0.3509138487279415,
      "rewards/accuracy_reward": 0.14583333488553762,
      "rewards/format_reward": 0.5833333432674408,
      "step": 49
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3195.7396240234375,
      "entropy": 0.3486328125,
      "epoch": 0.05714285714285714,
      "grad_norm": 0.11028607189655304,
      "kl": 0.0009064674377441406,
      "learning_rate": 6.435602608679916e-07,
      "loss": 0.0,
      "reward": 0.6458333637565374,
      "reward_std": 0.3640809394419193,
      "rewards/accuracy_reward": 0.2812500074505806,
      "rewards/format_reward": 0.36458333395421505,
      "step": 50
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2502.166748046875,
      "entropy": 0.453125,
      "epoch": 0.05828571428571429,
      "grad_norm": 0.13727469742298126,
      "kl": 0.004588127136230469,
      "learning_rate": 6.281416799501187e-07,
      "loss": 0.0002,
      "reward": 0.6354166716337204,
      "reward_std": 0.30697082355618477,
      "rewards/accuracy_reward": 0.1354166679084301,
      "rewards/format_reward": 0.5,
      "step": 51
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3103.4688110351562,
      "entropy": 0.404296875,
      "epoch": 0.05942857142857143,
      "grad_norm": 0.11003529280424118,
      "kl": 0.001992940902709961,
      "learning_rate": 6.126278954320294e-07,
      "loss": 0.0001,
      "reward": 0.7500000102445483,
      "reward_std": 0.47105513513088226,
      "rewards/accuracy_reward": 0.3020833358168602,
      "rewards/format_reward": 0.4479166669771075,
      "step": 52
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2895.6146850585938,
      "entropy": 0.41748046875,
      "epoch": 0.060571428571428575,
      "grad_norm": 0.11098165810108185,
      "kl": 0.0010890960693359375,
      "learning_rate": 5.97037808470444e-07,
      "loss": 0.0,
      "reward": 0.895833358168602,
      "reward_std": 0.6053251773118973,
      "rewards/accuracy_reward": 0.3020833469927311,
      "rewards/format_reward": 0.5937500074505806,
      "step": 53
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2898.0938720703125,
      "entropy": 0.36865234375,
      "epoch": 0.061714285714285715,
      "grad_norm": 0.13266167044639587,
      "kl": 0.0010752677917480469,
      "learning_rate": 5.813904131848564e-07,
      "loss": 0.0,
      "reward": 0.9791666939854622,
      "reward_std": 0.6714620888233185,
      "rewards/accuracy_reward": 0.4062500037252903,
      "rewards/format_reward": 0.5729166828095913,
      "step": 54
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3296.2084350585938,
      "entropy": 0.42138671875,
      "epoch": 0.06285714285714286,
      "grad_norm": 0.11708013713359833,
      "kl": 0.0011484622955322266,
      "learning_rate": 5.657047735161255e-07,
      "loss": 0.0,
      "reward": 0.5625000149011612,
      "reward_std": 0.468790166079998,
      "rewards/accuracy_reward": 0.18750000465661287,
      "rewards/format_reward": 0.3750000037252903,
      "step": 55
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3217.760498046875,
      "entropy": 0.40234375,
      "epoch": 0.064,
      "grad_norm": 0.10522608458995819,
      "kl": 0.0005931854248046875,
      "learning_rate": 5.5e-07,
      "loss": 0.0,
      "reward": 0.5833333414047956,
      "reward_std": 0.3264950141310692,
      "rewards/accuracy_reward": 0.1770833432674408,
      "rewards/format_reward": 0.40625001303851604,
      "step": 56
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3551.9063110351562,
      "entropy": 0.32470703125,
      "epoch": 0.06514285714285714,
      "grad_norm": 0.08617226034402847,
      "kl": 0.00047588348388671875,
      "learning_rate": 5.342952264838747e-07,
      "loss": 0.0,
      "reward": 0.5312500260770321,
      "reward_std": 0.5744922757148743,
      "rewards/accuracy_reward": 0.14583334140479565,
      "rewards/format_reward": 0.38541668467223644,
      "step": 57
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2254.4375610351562,
      "entropy": 0.337646484375,
      "epoch": 0.06628571428571428,
      "grad_norm": 0.13133276998996735,
      "kl": 0.0045833587646484375,
      "learning_rate": 5.186095868151436e-07,
      "loss": 0.0002,
      "reward": 1.0208333879709244,
      "reward_std": 0.42615631967782974,
      "rewards/accuracy_reward": 0.291666679084301,
      "rewards/format_reward": 0.7291667014360428,
      "step": 58
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3097.7083740234375,
      "entropy": 0.3427734375,
      "epoch": 0.06742857142857143,
      "grad_norm": 0.12775066494941711,
      "kl": 0.0004324913024902344,
      "learning_rate": 5.02962191529556e-07,
      "loss": 0.0,
      "reward": 0.5312500009313226,
      "reward_std": 0.3815914988517761,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.36458334419876337,
      "step": 59
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3122.7396240234375,
      "entropy": 0.37744140625,
      "epoch": 0.06857142857142857,
      "grad_norm": 0.09946239739656448,
      "kl": 0.0018048286437988281,
      "learning_rate": 4.873721045679706e-07,
      "loss": 0.0001,
      "reward": 0.5520833432674408,
      "reward_std": 0.2942040413618088,
      "rewards/accuracy_reward": 0.13541666977107525,
      "rewards/format_reward": 0.41666668467223644,
      "step": 60
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3085.6771240234375,
      "entropy": 0.37646484375,
      "epoch": 0.06971428571428571,
      "grad_norm": 0.12444304674863815,
      "kl": 0.0009784698486328125,
      "learning_rate": 4.7185832004988133e-07,
      "loss": 0.0,
      "reward": 0.8541667088866234,
      "reward_std": 0.39620716124773026,
      "rewards/accuracy_reward": 0.2708333432674408,
      "rewards/format_reward": 0.5833333507180214,
      "step": 61
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2567.8646240234375,
      "entropy": 0.326171875,
      "epoch": 0.07085714285714285,
      "grad_norm": 0.11932501196861267,
      "kl": 0.001983642578125,
      "learning_rate": 4.5643973913200837e-07,
      "loss": 0.0001,
      "reward": 0.9791666939854622,
      "reward_std": 0.4785246327519417,
      "rewards/accuracy_reward": 0.32291666977107525,
      "rewards/format_reward": 0.6562500149011612,
      "step": 62
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2210.6459045410156,
      "entropy": 0.3583984375,
      "epoch": 0.072,
      "grad_norm": 0.13491779565811157,
      "kl": 0.0017650127410888672,
      "learning_rate": 4.4113514698014953e-07,
      "loss": 0.0001,
      "reward": 1.0937500447034836,
      "reward_std": 0.5017538368701935,
      "rewards/accuracy_reward": 0.3229166716337204,
      "rewards/format_reward": 0.7708333432674408,
      "step": 63
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2907.500030517578,
      "entropy": 0.388671875,
      "epoch": 0.07314285714285715,
      "grad_norm": 0.11815346777439117,
      "kl": 0.0008115768432617188,
      "learning_rate": 4.2596318988235037e-07,
      "loss": 0.0,
      "reward": 0.739583358168602,
      "reward_std": 0.5059329643845558,
      "rewards/accuracy_reward": 0.2395833358168602,
      "rewards/format_reward": 0.5000000149011612,
      "step": 64
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2948.166748046875,
      "entropy": 0.35888671875,
      "epoch": 0.07428571428571429,
      "grad_norm": 0.10616273432970047,
      "kl": 0.0011377334594726562,
      "learning_rate": 4.1094235253127374e-07,
      "loss": 0.0,
      "reward": 0.6458333637565374,
      "reward_std": 0.3507690578699112,
      "rewards/accuracy_reward": 0.2083333432674408,
      "rewards/format_reward": 0.43750001303851604,
      "step": 65
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2217.9792098999023,
      "entropy": 0.32080078125,
      "epoch": 0.07542857142857143,
      "grad_norm": 0.12008016556501389,
      "kl": 0.0014886856079101562,
      "learning_rate": 3.9609093550344907e-07,
      "loss": 0.0001,
      "reward": 0.8437500204890966,
      "reward_std": 0.2087036520242691,
      "rewards/accuracy_reward": 0.33333334885537624,
      "rewards/format_reward": 0.5104166669771075,
      "step": 66
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3613.9375610351562,
      "entropy": 0.3525390625,
      "epoch": 0.07657142857142857,
      "grad_norm": 0.06719610840082169,
      "kl": 0.001219034194946289,
      "learning_rate": 3.8142703296283953e-07,
      "loss": 0.0,
      "reward": 0.23958334606140852,
      "reward_std": 0.245176762342453,
      "rewards/accuracy_reward": 0.03125,
      "rewards/format_reward": 0.20833333861082792,
      "step": 67
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2248.8021697998047,
      "entropy": 0.38134765625,
      "epoch": 0.07771428571428571,
      "grad_norm": 0.183990478515625,
      "kl": 0.00318145751953125,
      "learning_rate": 3.6696851061588994e-07,
      "loss": 0.0001,
      "reward": 0.8437500298023224,
      "reward_std": 0.49483707919716835,
      "rewards/accuracy_reward": 0.2500000102445483,
      "rewards/format_reward": 0.5937500074505806,
      "step": 68
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2372.416748046875,
      "entropy": 0.40625,
      "epoch": 0.07885714285714286,
      "grad_norm": 0.16380108892917633,
      "kl": 0.0032024383544921875,
      "learning_rate": 3.5273298394491515e-07,
      "loss": 0.0001,
      "reward": 0.7187500298023224,
      "reward_std": 0.420441672205925,
      "rewards/accuracy_reward": 0.12500000279396772,
      "rewards/format_reward": 0.5937500223517418,
      "step": 69
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3206.5521850585938,
      "entropy": 0.3671875,
      "epoch": 0.08,
      "grad_norm": 0.10725324600934982,
      "kl": 0.0023908615112304688,
      "learning_rate": 3.387377967463493e-07,
      "loss": 0.0001,
      "reward": 0.4895833535119891,
      "reward_std": 0.3960477001965046,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.4062500102445483,
      "step": 70
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2797.8958435058594,
      "entropy": 0.40966796875,
      "epoch": 0.08114285714285714,
      "grad_norm": 0.15477971732616425,
      "kl": 0.00348663330078125,
      "learning_rate": 3.250000000000001e-07,
      "loss": 0.0001,
      "reward": 0.6458333535119891,
      "reward_std": 0.31787312403321266,
      "rewards/accuracy_reward": 0.197916679084301,
      "rewards/format_reward": 0.4479166669771075,
      "step": 71
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2927.3958740234375,
      "entropy": 0.46826171875,
      "epoch": 0.08228571428571428,
      "grad_norm": 0.22557654976844788,
      "kl": 0.0026330947875976562,
      "learning_rate": 3.115363310950578e-07,
      "loss": 0.0001,
      "reward": 0.510416679084301,
      "reward_std": 0.47425539791584015,
      "rewards/accuracy_reward": 0.052083334885537624,
      "rewards/format_reward": 0.4583333432674408,
      "step": 72
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3653.78125,
      "entropy": 0.47802734375,
      "epoch": 0.08342857142857144,
      "grad_norm": 0.08231060206890106,
      "kl": 0.0008916854858398438,
      "learning_rate": 2.9836319343816397e-07,
      "loss": 0.0,
      "reward": 0.385416679084301,
      "reward_std": 0.3866991028189659,
      "rewards/accuracy_reward": 0.13541666977107525,
      "rewards/format_reward": 0.25,
      "step": 73
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3289.0834350585938,
      "entropy": 0.38671875,
      "epoch": 0.08457142857142858,
      "grad_norm": 0.11783669143915176,
      "kl": 0.001605987548828125,
      "learning_rate": 2.854966364683872e-07,
      "loss": 0.0001,
      "reward": 0.6145833432674408,
      "reward_std": 0.4752888083457947,
      "rewards/accuracy_reward": 0.2604166716337204,
      "rewards/format_reward": 0.3541666679084301,
      "step": 74
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3183.7916870117188,
      "entropy": 0.39501953125,
      "epoch": 0.08571428571428572,
      "grad_norm": 0.10990124195814133,
      "kl": 0.0020608901977539062,
      "learning_rate": 2.729523361034538e-07,
      "loss": 0.0001,
      "reward": 0.5833333432674408,
      "reward_std": 0.29638050496578217,
      "rewards/accuracy_reward": 0.1770833358168602,
      "rewards/format_reward": 0.4062500149011612,
      "step": 75
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2836.3750610351562,
      "entropy": 0.40185546875,
      "epoch": 0.08685714285714285,
      "grad_norm": 0.13064220547676086,
      "kl": 0.000682830810546875,
      "learning_rate": 2.6074557564105724e-07,
      "loss": 0.0,
      "reward": 0.5729166716337204,
      "reward_std": 0.33378810063004494,
      "rewards/accuracy_reward": 0.07291666697710752,
      "rewards/format_reward": 0.5,
      "step": 76
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3366.4688720703125,
      "entropy": 0.38916015625,
      "epoch": 0.088,
      "grad_norm": 0.08818019181489944,
      "kl": 0.0007886886596679688,
      "learning_rate": 2.488912271385139e-07,
      "loss": 0.0,
      "reward": 0.5000000009313226,
      "reward_std": 0.370398610830307,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.3750000009313226,
      "step": 77
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3164.4688720703125,
      "entropy": 0.38525390625,
      "epoch": 0.08914285714285715,
      "grad_norm": 0.1075669676065445,
      "kl": 0.00061798095703125,
      "learning_rate": 2.374037332934512e-07,
      "loss": 0.0,
      "reward": 0.6354166865348816,
      "reward_std": 0.5042389929294586,
      "rewards/accuracy_reward": 0.2187500111758709,
      "rewards/format_reward": 0.416666679084301,
      "step": 78
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2405.572982788086,
      "entropy": 0.33349609375,
      "epoch": 0.09028571428571429,
      "grad_norm": 0.09348474442958832,
      "kl": 0.0013494491577148438,
      "learning_rate": 2.2629708984760706e-07,
      "loss": 0.0001,
      "reward": 0.9270833730697632,
      "reward_std": 0.3268127888441086,
      "rewards/accuracy_reward": 0.2604166744276881,
      "rewards/format_reward": 0.6666666865348816,
      "step": 79
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3390.2604370117188,
      "entropy": 0.48388671875,
      "epoch": 0.09142857142857143,
      "grad_norm": 0.1434425562620163,
      "kl": 0.0012698173522949219,
      "learning_rate": 2.1558482853517253e-07,
      "loss": 0.0001,
      "reward": 0.5208333507180214,
      "reward_std": 0.44375673681497574,
      "rewards/accuracy_reward": 0.1562500074505806,
      "rewards/format_reward": 0.3645833432674408,
      "step": 80
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3077.854248046875,
      "entropy": 0.533203125,
      "epoch": 0.09257142857142857,
      "grad_norm": 0.08657144755125046,
      "kl": 0.00328826904296875,
      "learning_rate": 2.0528000059645995e-07,
      "loss": 0.0001,
      "reward": 0.5208333432674408,
      "reward_std": 0.27500855177640915,
      "rewards/accuracy_reward": 0.13541666697710752,
      "rewards/format_reward": 0.3854166716337204,
      "step": 81
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3044.9480590820312,
      "entropy": 0.4287109375,
      "epoch": 0.09371428571428571,
      "grad_norm": 0.1634312868118286,
      "kl": 0.001422882080078125,
      "learning_rate": 1.9539516087697517e-07,
      "loss": 0.0001,
      "reward": 0.6666666865348816,
      "reward_std": 0.4855338782072067,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.4375000149011612,
      "step": 82
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2948.4896850585938,
      "entropy": 0.43408203125,
      "epoch": 0.09485714285714286,
      "grad_norm": 0.13163359463214874,
      "kl": 0.0017070770263671875,
      "learning_rate": 1.8594235253127372e-07,
      "loss": 0.0001,
      "reward": 0.5208333544433117,
      "reward_std": 0.3695184215903282,
      "rewards/accuracy_reward": 0.16666667256504297,
      "rewards/format_reward": 0.35416667722165585,
      "step": 83
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3189.625030517578,
      "entropy": 0.43603515625,
      "epoch": 0.096,
      "grad_norm": 0.1409156173467636,
      "kl": 0.0008630752563476562,
      "learning_rate": 1.7693309235023127e-07,
      "loss": 0.0,
      "reward": 0.677083358168602,
      "reward_std": 0.49923326447606087,
      "rewards/accuracy_reward": 0.2395833358168602,
      "rewards/format_reward": 0.4375000074505806,
      "step": 84
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3121.0938720703125,
      "entropy": 0.32177734375,
      "epoch": 0.09714285714285714,
      "grad_norm": 0.10903389751911163,
      "kl": 0.0006976127624511719,
      "learning_rate": 1.6837835672960831e-07,
      "loss": 0.0,
      "reward": 0.7187500149011612,
      "reward_std": 0.6352673023939133,
      "rewards/accuracy_reward": 0.1875000074505806,
      "rewards/format_reward": 0.5312500223517418,
      "step": 85
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2938.385498046875,
      "entropy": 0.4287109375,
      "epoch": 0.09828571428571428,
      "grad_norm": 0.14300768077373505,
      "kl": 0.0020380020141601562,
      "learning_rate": 1.6028856829700258e-07,
      "loss": 0.0001,
      "reward": 0.6041666818782687,
      "reward_std": 0.26244670152664185,
      "rewards/accuracy_reward": 0.1562500037252903,
      "rewards/format_reward": 0.4479166744276881,
      "step": 86
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3037.6563110351562,
      "entropy": 0.50390625,
      "epoch": 0.09942857142857142,
      "grad_norm": 0.16961807012557983,
      "kl": 0.0024051666259765625,
      "learning_rate": 1.5267358321348285e-07,
      "loss": 0.0001,
      "reward": 0.7395833432674408,
      "reward_std": 0.5077016204595566,
      "rewards/accuracy_reward": 0.1979166716337204,
      "rewards/format_reward": 0.541666679084301,
      "step": 87
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2826.947998046875,
      "entropy": 0.439453125,
      "epoch": 0.10057142857142858,
      "grad_norm": 0.14141874015331268,
      "kl": 0.0017194747924804688,
      "learning_rate": 1.4554267916537495e-07,
      "loss": 0.0001,
      "reward": 0.9062500298023224,
      "reward_std": 0.5221241116523743,
      "rewards/accuracy_reward": 0.3333333358168602,
      "rewards/format_reward": 0.572916679084301,
      "step": 88
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2963.260498046875,
      "entropy": 0.36962890625,
      "epoch": 0.10171428571428572,
      "grad_norm": 0.16021747887134552,
      "kl": 0.0021514892578125,
      "learning_rate": 1.3890454406082956e-07,
      "loss": 0.0001,
      "reward": 0.6250000149011612,
      "reward_std": 0.4761141389608383,
      "rewards/accuracy_reward": 0.1562500037252903,
      "rewards/format_reward": 0.4687500149011612,
      "step": 89
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2510.8958740234375,
      "entropy": 0.5478515625,
      "epoch": 0.10285714285714286,
      "grad_norm": 0.08637464791536331,
      "kl": 0.0036640167236328125,
      "learning_rate": 1.3276726544494571e-07,
      "loss": 0.0001,
      "reward": 0.5520833386108279,
      "reward_std": 0.18445607274770737,
      "rewards/accuracy_reward": 0.031250000931322575,
      "rewards/format_reward": 0.5208333432674408,
      "step": 90
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3139.1771850585938,
      "entropy": 0.42236328125,
      "epoch": 0.104,
      "grad_norm": 0.09102143347263336,
      "kl": 0.0018377304077148438,
      "learning_rate": 1.2713832064634125e-07,
      "loss": 0.0001,
      "reward": 0.6250000018626451,
      "reward_std": 0.40364614874124527,
      "rewards/accuracy_reward": 0.19791667256504297,
      "rewards/format_reward": 0.42708334419876337,
      "step": 91
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2716.0938415527344,
      "entropy": 0.354736328125,
      "epoch": 0.10514285714285715,
      "grad_norm": 0.12420715391635895,
      "kl": 0.002556324005126953,
      "learning_rate": 1.220245676671809e-07,
      "loss": 0.0001,
      "reward": 0.6354166939854622,
      "reward_std": 0.4150289446115494,
      "rewards/accuracy_reward": 0.1041666716337204,
      "rewards/format_reward": 0.5312500223517418,
      "step": 92
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3754.8229370117188,
      "entropy": 0.5556640625,
      "epoch": 0.10628571428571429,
      "grad_norm": 0.1010187566280365,
      "kl": 0.0030183792114257812,
      "learning_rate": 1.1743223682775649e-07,
      "loss": 0.0001,
      "reward": 0.10416667256504297,
      "reward_std": 0.14910665899515152,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.10416667256504297,
      "step": 93
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3241.3333740234375,
      "entropy": 0.51806640625,
      "epoch": 0.10742857142857143,
      "grad_norm": 0.11734016239643097,
      "kl": 0.0025129318237304688,
      "learning_rate": 1.1336692317580158e-07,
      "loss": 0.0001,
      "reward": 0.5208333432674408,
      "reward_std": 0.3755134493112564,
      "rewards/accuracy_reward": 0.16666666697710752,
      "rewards/format_reward": 0.354166679084301,
      "step": 94
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3691.875,
      "entropy": 0.4384765625,
      "epoch": 0.10857142857142857,
      "grad_norm": 0.10122071951627731,
      "kl": 0.0010461807250976562,
      "learning_rate": 1.0983357966978745e-07,
      "loss": 0.0,
      "reward": 0.2500000037252903,
      "reward_std": 0.39341451972723007,
      "rewards/accuracy_reward": 0.07291666883975267,
      "rewards/format_reward": 0.17708333767950535,
      "step": 95
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2813.8751220703125,
      "entropy": 0.3935546875,
      "epoch": 0.10971428571428571,
      "grad_norm": 0.1897701621055603,
      "kl": 0.0020542144775390625,
      "learning_rate": 1.068365111445064e-07,
      "loss": 0.0001,
      "reward": 0.7291666939854622,
      "reward_std": 0.500264085829258,
      "rewards/accuracy_reward": 0.2708333432674408,
      "rewards/format_reward": 0.4583333507180214,
      "step": 96
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3200.7188110351562,
      "entropy": 0.43798828125,
      "epoch": 0.11085714285714286,
      "grad_norm": 0.1710478961467743,
      "kl": 0.0015239715576171875,
      "learning_rate": 1.0437936906629334e-07,
      "loss": 0.0001,
      "reward": 0.6562500204890966,
      "reward_std": 0.44586458802223206,
      "rewards/accuracy_reward": 0.22916667722165585,
      "rewards/format_reward": 0.4270833386108279,
      "step": 97
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2654.9688110351562,
      "entropy": 0.374267578125,
      "epoch": 0.112,
      "grad_norm": 0.16486391425132751,
      "kl": 0.0010390281677246094,
      "learning_rate": 1.0246514708427701e-07,
      "loss": 0.0,
      "reward": 0.6979166939854622,
      "reward_std": 0.26594046503305435,
      "rewards/accuracy_reward": 0.14583333395421505,
      "rewards/format_reward": 0.5520833507180214,
      "step": 98
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 3058.5833435058594,
      "entropy": 0.36572265625,
      "epoch": 0.11314285714285714,
      "grad_norm": 0.09798528254032135,
      "kl": 0.0014705657958984375,
      "learning_rate": 1.0109617738307911e-07,
      "loss": 0.0001,
      "reward": 0.5312500149011612,
      "reward_std": 0.20392338931560516,
      "rewards/accuracy_reward": 0.197916679084301,
      "rewards/format_reward": 0.3333333358168602,
      "step": 99
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 2583.0521240234375,
      "entropy": 0.34912109375,
      "epoch": 0.11428571428571428,
      "grad_norm": 0.1591912806034088,
      "kl": 0.0018825531005859375,
      "learning_rate": 1.002741278414069e-07,
      "loss": 0.0001,
      "reward": 0.9375,
      "reward_std": 0.5781587325036526,
      "rewards/accuracy_reward": 0.3333333432674408,
      "rewards/format_reward": 0.6041666865348816,
      "step": 100
    },
    {
      "epoch": 0.11428571428571428,
      "step": 100,
      "total_flos": 0.0,
      "train_loss": 4.974420491237197e-05,
      "train_runtime": 8330.7391,
      "train_samples_per_second": 1.152,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "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": 6,
  "trial_name": null,
  "trial_params": null
}