{
  "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": 0.3681640625,
      "epoch": 0.001142857142857143,
      "grad_norm": 0.1187390461564064,
      "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": 0.1484038233757019,
      "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": 3831.6771240234375,
      "entropy": 0.4521484375,
      "epoch": 0.0034285714285714284,
      "grad_norm": 0.1441987007856369,
      "kl": 4.5239925384521484e-05,
      "learning_rate": 2e-07,
      "loss": 0.0,
      "reward": 0.15625000838190317,
      "reward_std": 0.31725750118494034,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.11458333674818277,
      "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": 2465.760498046875,
      "entropy": 0.4189453125,
      "epoch": 0.004571428571428572,
      "grad_norm": 0.1570746898651123,
      "kl": 3.510713577270508e-05,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.7812500298023224,
      "reward_std": 0.4064664840698242,
      "rewards/accuracy_reward": 0.13541667256504297,
      "rewards/format_reward": 0.6458333432674408,
      "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": 3401.7709350585938,
      "entropy": 0.46484375,
      "epoch": 0.005714285714285714,
      "grad_norm": 0.13895763456821442,
      "kl": 4.506111145019531e-05,
      "learning_rate": 4e-07,
      "loss": 0.0,
      "reward": 0.5416666716337204,
      "reward_std": 0.5744388923048973,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.4375000074505806,
      "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": 3493.6146850585938,
      "entropy": 0.466796875,
      "epoch": 0.006857142857142857,
      "grad_norm": 0.12972474098205566,
      "kl": 4.756450653076172e-05,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.3958333507180214,
      "reward_std": 0.4614809453487396,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.2916666716337204,
      "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": 3271.3125610351562,
      "entropy": 0.3828125,
      "epoch": 0.008,
      "grad_norm": 0.11172469705343246,
      "kl": 2.6941299438476562e-05,
      "learning_rate": 6e-07,
      "loss": 0.0,
      "reward": 0.9895833730697632,
      "reward_std": 0.6482726484537125,
      "rewards/accuracy_reward": 0.3229166716337204,
      "rewards/format_reward": 0.6666667014360428,
      "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": 2961.8854370117188,
      "entropy": 0.360107421875,
      "epoch": 0.009142857142857144,
      "grad_norm": 0.10366630554199219,
      "kl": 2.3096799850463867e-05,
      "learning_rate": 7e-07,
      "loss": 0.0,
      "reward": 0.8229166865348816,
      "reward_std": 0.46650417521595955,
      "rewards/accuracy_reward": 0.3645833386108279,
      "rewards/format_reward": 0.45833334885537624,
      "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": 3490.6459350585938,
      "entropy": 0.451171875,
      "epoch": 0.010285714285714285,
      "grad_norm": 0.1490144431591034,
      "kl": 4.25875186920166e-05,
      "learning_rate": 8e-07,
      "loss": 0.0,
      "reward": 0.3750000074505806,
      "reward_std": 0.5181794688105583,
      "rewards/accuracy_reward": 0.08333333488553762,
      "rewards/format_reward": 0.291666679084301,
      "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": 2958.822998046875,
      "entropy": 0.34912109375,
      "epoch": 0.011428571428571429,
      "grad_norm": 0.17549936473369598,
      "kl": 2.4825334548950195e-05,
      "learning_rate": 9e-07,
      "loss": 0.0,
      "reward": 0.5520833358168602,
      "reward_std": 0.48831162601709366,
      "rewards/accuracy_reward": 0.15625000279396772,
      "rewards/format_reward": 0.3958333469927311,
      "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": 3614.1563110351562,
      "entropy": 0.38818359375,
      "epoch": 0.012571428571428572,
      "grad_norm": 0.148716539144516,
      "kl": 3.039836883544922e-05,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.2604166679084301,
      "reward_std": 0.4707208350300789,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.17708333767950535,
      "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": 2686.2709045410156,
      "entropy": 0.39697265625,
      "epoch": 0.013714285714285714,
      "grad_norm": 0.14581860601902008,
      "kl": 2.4810433387756348e-05,
      "learning_rate": 9.997258721585931e-07,
      "loss": 0.0,
      "reward": 0.770833358168602,
      "reward_std": 0.4604961648583412,
      "rewards/accuracy_reward": 0.1562500037252903,
      "rewards/format_reward": 0.614583358168602,
      "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": 3204.9583740234375,
      "entropy": 0.39453125,
      "epoch": 0.014857142857142857,
      "grad_norm": 0.10839570313692093,
      "kl": 2.79843807220459e-05,
      "learning_rate": 9.989038226169207e-07,
      "loss": 0.0,
      "reward": 0.5312500223517418,
      "reward_std": 0.34297536313533783,
      "rewards/accuracy_reward": 0.1666666753590107,
      "rewards/format_reward": 0.3645833432674408,
      "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": 3025.2501220703125,
      "entropy": 0.3623046875,
      "epoch": 0.016,
      "grad_norm": 0.2049231380224228,
      "kl": 2.2601336240768433e-05,
      "learning_rate": 9.975348529157229e-07,
      "loss": 0.0,
      "reward": 0.5937500074505806,
      "reward_std": 0.47151048481464386,
      "rewards/accuracy_reward": 0.16666667070239782,
      "rewards/format_reward": 0.4270833395421505,
      "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": 3013.6979370117188,
      "entropy": 0.36328125,
      "epoch": 0.017142857142857144,
      "grad_norm": 0.08478636294603348,
      "kl": 2.2374093532562256e-05,
      "learning_rate": 9.956206309337066e-07,
      "loss": 0.0,
      "reward": 0.666666679084301,
      "reward_std": 0.3029968775808811,
      "rewards/accuracy_reward": 0.2604166716337204,
      "rewards/format_reward": 0.4062500074505806,
      "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": 3824.9063720703125,
      "entropy": 0.46728515625,
      "epoch": 0.018285714285714287,
      "grad_norm": 0.14077027142047882,
      "kl": 2.8401613235473633e-05,
      "learning_rate": 9.931634888554935e-07,
      "loss": 0.0,
      "reward": 0.2291666679084301,
      "reward_std": 0.4411991238594055,
      "rewards/accuracy_reward": 0.0833333358168602,
      "rewards/format_reward": 0.1458333395421505,
      "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": 2340.947998046875,
      "entropy": 0.42822265625,
      "epoch": 0.019428571428571427,
      "grad_norm": 0.18051159381866455,
      "kl": 4.7087669372558594e-05,
      "learning_rate": 9.901664203302124e-07,
      "loss": 0.0,
      "reward": 0.9062500149011612,
      "reward_std": 0.4745892286300659,
      "rewards/accuracy_reward": 0.2812500009313226,
      "rewards/format_reward": 0.625,
      "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": 3181.3958740234375,
      "entropy": 0.36865234375,
      "epoch": 0.02057142857142857,
      "grad_norm": 0.1322987973690033,
      "kl": 3.0174851417541504e-05,
      "learning_rate": 9.866330768241983e-07,
      "loss": 0.0,
      "reward": 0.5520833488553762,
      "reward_std": 0.472620353102684,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.38541666977107525,
      "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": 3140.697998046875,
      "entropy": 0.39599609375,
      "epoch": 0.021714285714285714,
      "grad_norm": 0.16092555224895477,
      "kl": 5.456805229187012e-05,
      "learning_rate": 9.825677631722435e-07,
      "loss": 0.0,
      "reward": 0.7083333656191826,
      "reward_std": 0.5550193563103676,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.3958333358168602,
      "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": 2432.9063110351562,
      "entropy": 0.327392578125,
      "epoch": 0.022857142857142857,
      "grad_norm": 0.10998747497797012,
      "kl": 5.906820297241211e-05,
      "learning_rate": 9.779754323328192e-07,
      "loss": 0.0,
      "reward": 0.9895833730697632,
      "reward_std": 0.410064734518528,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.708333358168602,
      "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": 2747.322998046875,
      "entropy": 0.4189453125,
      "epoch": 0.024,
      "grad_norm": 0.3037484884262085,
      "kl": 0.00028693675994873047,
      "learning_rate": 9.728616793536587e-07,
      "loss": 0.0,
      "reward": 0.6562500298023224,
      "reward_std": 0.35315681248903275,
      "rewards/accuracy_reward": 0.1979166716337204,
      "rewards/format_reward": 0.4583333507180214,
      "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": 1896.4583740234375,
      "entropy": 0.404296875,
      "epoch": 0.025142857142857144,
      "grad_norm": 0.15490920841693878,
      "kl": 0.0002675056457519531,
      "learning_rate": 9.672327345550543e-07,
      "loss": 0.0,
      "reward": 1.0833333730697632,
      "reward_std": 0.3764601796865463,
      "rewards/accuracy_reward": 0.27083334140479565,
      "rewards/format_reward": 0.8125000298023224,
      "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": 2586.0521545410156,
      "entropy": 0.3759765625,
      "epoch": 0.026285714285714287,
      "grad_norm": 0.17139679193496704,
      "kl": 0.0001055002212524414,
      "learning_rate": 9.610954559391704e-07,
      "loss": 0.0,
      "reward": 0.6979166865348816,
      "reward_std": 0.5570684000849724,
      "rewards/accuracy_reward": 0.1770833358168602,
      "rewards/format_reward": 0.520833358168602,
      "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": 3055.1458740234375,
      "entropy": 0.37890625,
      "epoch": 0.027428571428571427,
      "grad_norm": 0.22735774517059326,
      "kl": 0.0001550912857055664,
      "learning_rate": 9.54457320834625e-07,
      "loss": 0.0,
      "reward": 0.7083333507180214,
      "reward_std": 0.6170841455459595,
      "rewards/accuracy_reward": 0.229166679084301,
      "rewards/format_reward": 0.4791666865348816,
      "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": 2793.479248046875,
      "entropy": 0.4296875,
      "epoch": 0.02857142857142857,
      "grad_norm": 0.11460871249437332,
      "kl": 0.00021645426750183105,
      "learning_rate": 9.473264167865171e-07,
      "loss": 0.0,
      "reward": 0.6666666669771075,
      "reward_std": 0.37379153445363045,
      "rewards/accuracy_reward": 0.2291666679084301,
      "rewards/format_reward": 0.4375000027939677,
      "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": 3250.635498046875,
      "entropy": 0.42919921875,
      "epoch": 0.029714285714285714,
      "grad_norm": 0.10849029570817947,
      "kl": 4.756450653076172e-05,
      "learning_rate": 9.397114317029974e-07,
      "loss": 0.0,
      "reward": 0.6250000149011612,
      "reward_std": 0.37658975273370743,
      "rewards/accuracy_reward": 0.1875000074505806,
      "rewards/format_reward": 0.4375000074505806,
      "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": 3344.5313110351562,
      "entropy": 0.4423828125,
      "epoch": 0.030857142857142857,
      "grad_norm": 0.11115474253892899,
      "kl": 0.00017783045768737793,
      "learning_rate": 9.316216432703916e-07,
      "loss": 0.0,
      "reward": 0.5833333432674408,
      "reward_std": 0.5297819748520851,
      "rewards/accuracy_reward": 0.1458333358168602,
      "rewards/format_reward": 0.4375000074505806,
      "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": 2976.8125,
      "entropy": 0.41015625,
      "epoch": 0.032,
      "grad_norm": 0.14693813025951385,
      "kl": 0.00039499998092651367,
      "learning_rate": 9.230669076497687e-07,
      "loss": 0.0,
      "reward": 0.7916666716337204,
      "reward_std": 0.5797645375132561,
      "rewards/accuracy_reward": 0.3333333358168602,
      "rewards/format_reward": 0.4583333432674408,
      "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": 3552.7291870117188,
      "entropy": 0.4306640625,
      "epoch": 0.03314285714285714,
      "grad_norm": 0.17034666240215302,
      "kl": 0.0004671812057495117,
      "learning_rate": 9.140576474687263e-07,
      "loss": 0.0,
      "reward": 0.2708333386108279,
      "reward_std": 0.3285643234848976,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.2500000102445483,
      "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": 2964.6771545410156,
      "entropy": 0.39453125,
      "epoch": 0.03428571428571429,
      "grad_norm": 0.12682349979877472,
      "kl": 0.0011525154113769531,
      "learning_rate": 9.046048391230247e-07,
      "loss": 0.0,
      "reward": 0.6666666939854622,
      "reward_std": 0.5078509747982025,
      "rewards/accuracy_reward": 0.17708334140479565,
      "rewards/format_reward": 0.4895833432674408,
      "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": 3263.3854370117188,
      "entropy": 0.39697265625,
      "epoch": 0.03542857142857143,
      "grad_norm": 0.12711374461650848,
      "kl": 0.0005249977111816406,
      "learning_rate": 8.9471999940354e-07,
      "loss": 0.0,
      "reward": 0.4166666716337204,
      "reward_std": 0.3054000325500965,
      "rewards/accuracy_reward": 0.12500000279396772,
      "rewards/format_reward": 0.2916666716337204,
      "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": 3470.2396240234375,
      "entropy": 0.4599609375,
      "epoch": 0.036571428571428574,
      "grad_norm": 0.18772432208061218,
      "kl": 0.003406047821044922,
      "learning_rate": 8.844151714648274e-07,
      "loss": 0.0001,
      "reward": 0.7187500223517418,
      "reward_std": 0.6219850406050682,
      "rewards/accuracy_reward": 0.2812500111758709,
      "rewards/format_reward": 0.4375000111758709,
      "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": 3453.625,
      "entropy": 0.39208984375,
      "epoch": 0.037714285714285714,
      "grad_norm": 0.11044140160083771,
      "kl": 0.0006644725799560547,
      "learning_rate": 8.737029101523929e-07,
      "loss": 0.0,
      "reward": 0.5937500260770321,
      "reward_std": 0.5192387402057648,
      "rewards/accuracy_reward": 0.21875000558793545,
      "rewards/format_reward": 0.37500000558793545,
      "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": 2518.9583435058594,
      "entropy": 0.45556640625,
      "epoch": 0.038857142857142854,
      "grad_norm": 0.13527297973632812,
      "kl": 0.0011320114135742188,
      "learning_rate": 8.625962667065487e-07,
      "loss": 0.0,
      "reward": 0.9062500353902578,
      "reward_std": 0.40292268246412277,
      "rewards/accuracy_reward": 0.3645833386108279,
      "rewards/format_reward": 0.5416666781529784,
      "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": 3182.2188720703125,
      "entropy": 0.44091796875,
      "epoch": 0.04,
      "grad_norm": 0.14582405984401703,
      "kl": 0.0015320777893066406,
      "learning_rate": 8.511087728614862e-07,
      "loss": 0.0001,
      "reward": 0.572916679084301,
      "reward_std": 0.5398506969213486,
      "rewards/accuracy_reward": 0.19791667442768812,
      "rewards/format_reward": 0.3750000111758709,
      "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": 3468.0521850585938,
      "entropy": 0.5126953125,
      "epoch": 0.04114285714285714,
      "grad_norm": 0.1376534402370453,
      "kl": 0.0011998414993286133,
      "learning_rate": 8.392544243589427e-07,
      "loss": 0.0,
      "reward": 0.26041666977107525,
      "reward_std": 0.3082217201590538,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.2395833395421505,
      "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": 3360.78125,
      "entropy": 0.419921875,
      "epoch": 0.04228571428571429,
      "grad_norm": 0.0872306302189827,
      "kl": 0.0006721019744873047,
      "learning_rate": 8.270476638965461e-07,
      "loss": 0.0,
      "reward": 0.2604166744276881,
      "reward_std": 0.10825317353010178,
      "rewards/accuracy_reward": 0.010416666977107525,
      "rewards/format_reward": 0.2500000027939677,
      "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": 3563.0833740234375,
      "entropy": 0.45947265625,
      "epoch": 0.04342857142857143,
      "grad_norm": 0.07418923079967499,
      "kl": 0.0006561279296875,
      "learning_rate": 8.145033635316128e-07,
      "loss": 0.0,
      "reward": 0.3125000223517418,
      "reward_std": 0.20737335830926895,
      "rewards/accuracy_reward": 0.14583333395421505,
      "rewards/format_reward": 0.1666666716337204,
      "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": 2954.0418090820312,
      "entropy": 0.369140625,
      "epoch": 0.044571428571428574,
      "grad_norm": 0.11493231356143951,
      "kl": 0.0006474852561950684,
      "learning_rate": 8.01636806561836e-07,
      "loss": 0.0,
      "reward": 0.760416679084301,
      "reward_std": 0.34002431482076645,
      "rewards/accuracy_reward": 0.2604166716337204,
      "rewards/format_reward": 0.5000000074505806,
      "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": 2670.6771850585938,
      "entropy": 0.376953125,
      "epoch": 0.045714285714285714,
      "grad_norm": 0.1400008499622345,
      "kl": 0.002880096435546875,
      "learning_rate": 7.884636689049422e-07,
      "loss": 0.0001,
      "reward": 0.6666666716337204,
      "reward_std": 0.3070584014058113,
      "rewards/accuracy_reward": 0.15625000558793545,
      "rewards/format_reward": 0.5104166865348816,
      "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": 3046.541748046875,
      "entropy": 0.3701171875,
      "epoch": 0.046857142857142854,
      "grad_norm": 0.13457465171813965,
      "kl": 0.0008628368377685547,
      "learning_rate": 7.75e-07,
      "loss": 0.0,
      "reward": 0.5937500149011612,
      "reward_std": 0.5119726955890656,
      "rewards/accuracy_reward": 0.14583333488553762,
      "rewards/format_reward": 0.4479166716337204,
      "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": 2984.7500610351562,
      "entropy": 0.484375,
      "epoch": 0.048,
      "grad_norm": 0.09397904574871063,
      "kl": 0.0007486343383789062,
      "learning_rate": 7.612622032536507e-07,
      "loss": 0.0,
      "reward": 0.3229166716337204,
      "reward_std": 0.13653486967086792,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.3229166716337204,
      "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": 3094.229248046875,
      "entropy": 0.40966796875,
      "epoch": 0.04914285714285714,
      "grad_norm": 0.13279645144939423,
      "kl": 0.0017828941345214844,
      "learning_rate": 7.472670160550848e-07,
      "loss": 0.0001,
      "reward": 0.4895833386108279,
      "reward_std": 0.30805424600839615,
      "rewards/accuracy_reward": 0.13541667256504297,
      "rewards/format_reward": 0.3541666744276881,
      "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": 2804.8438110351562,
      "entropy": 0.39013671875,
      "epoch": 0.05028571428571429,
      "grad_norm": 0.11614151298999786,
      "kl": 0.001705169677734375,
      "learning_rate": 7.330314893841101e-07,
      "loss": 0.0001,
      "reward": 0.7812500325962901,
      "reward_std": 0.34850965440273285,
      "rewards/accuracy_reward": 0.260416679084301,
      "rewards/format_reward": 0.5208333386108279,
      "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": 3612.5209350585938,
      "entropy": 0.43603515625,
      "epoch": 0.05142857142857143,
      "grad_norm": 0.10443254560232162,
      "kl": 0.0018954277038574219,
      "learning_rate": 7.185729670371604e-07,
      "loss": 0.0001,
      "reward": 0.3958333395421505,
      "reward_std": 0.4208246320486069,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 0.2395833358168602,
      "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": 3438.7188720703125,
      "entropy": 0.5146484375,
      "epoch": 0.052571428571428575,
      "grad_norm": 0.09968335181474686,
      "kl": 0.0014276504516601562,
      "learning_rate": 7.039090644965509e-07,
      "loss": 0.0001,
      "reward": 0.18750000186264515,
      "reward_std": 0.15358614921569824,
      "rewards/accuracy_reward": 0.010416666977107525,
      "rewards/format_reward": 0.17708333395421505,
      "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": 3005.3438110351562,
      "entropy": 0.39501953125,
      "epoch": 0.053714285714285714,
      "grad_norm": 0.16180375218391418,
      "kl": 0.001639723777770996,
      "learning_rate": 6.890576474687263e-07,
      "loss": 0.0001,
      "reward": 0.9375000204890966,
      "reward_std": 0.5641577839851379,
      "rewards/accuracy_reward": 0.3750000149011612,
      "rewards/format_reward": 0.5625000055879354,
      "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": 2912.0000915527344,
      "entropy": 0.421875,
      "epoch": 0.054857142857142854,
      "grad_norm": 0.11711370199918747,
      "kl": 0.0034971237182617188,
      "learning_rate": 6.740368101176495e-07,
      "loss": 0.0001,
      "reward": 0.5312500251457095,
      "reward_std": 0.3790072128176689,
      "rewards/accuracy_reward": 0.1145833395421505,
      "rewards/format_reward": 0.4166666818782687,
      "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": 2401.2501220703125,
      "entropy": 0.37939453125,
      "epoch": 0.056,
      "grad_norm": 0.17111879587173462,
      "kl": 0.008028984069824219,
      "learning_rate": 6.588648530198504e-07,
      "loss": 0.0003,
      "reward": 0.9687500149011612,
      "reward_std": 0.4728153273463249,
      "rewards/accuracy_reward": 0.3020833423361182,
      "rewards/format_reward": 0.6666666865348816,
      "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": 3193.947998046875,
      "entropy": 0.349609375,
      "epoch": 0.05714285714285714,
      "grad_norm": 0.10395801067352295,
      "kl": 0.001220703125,
      "learning_rate": 6.435602608679916e-07,
      "loss": 0.0,
      "reward": 0.6979166865348816,
      "reward_std": 0.4201039932668209,
      "rewards/accuracy_reward": 0.3020833432674408,
      "rewards/format_reward": 0.3958333432674408,
      "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": 2429.1563720703125,
      "entropy": 0.453125,
      "epoch": 0.05828571428571429,
      "grad_norm": 0.19003717601299286,
      "kl": 0.0063724517822265625,
      "learning_rate": 6.281416799501187e-07,
      "loss": 0.0003,
      "reward": 0.6145833432674408,
      "reward_std": 0.33619827404618263,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.4895833358168602,
      "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": 3179.7813110351562,
      "entropy": 0.4033203125,
      "epoch": 0.05942857142857143,
      "grad_norm": 0.14212153851985931,
      "kl": 0.0013420581817626953,
      "learning_rate": 6.126278954320294e-07,
      "loss": 0.0001,
      "reward": 0.677083358168602,
      "reward_std": 0.6127200499176979,
      "rewards/accuracy_reward": 0.3020833386108279,
      "rewards/format_reward": 0.37500000558793545,
      "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": 2806.7918090820312,
      "entropy": 0.41748046875,
      "epoch": 0.060571428571428575,
      "grad_norm": 0.1251596063375473,
      "kl": 0.0011522769927978516,
      "learning_rate": 5.97037808470444e-07,
      "loss": 0.0,
      "reward": 0.8750000149011612,
      "reward_std": 0.5521636754274368,
      "rewards/accuracy_reward": 0.28125001303851604,
      "rewards/format_reward": 0.5937500149011612,
      "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": 2811.312530517578,
      "entropy": 0.37353515625,
      "epoch": 0.061714285714285715,
      "grad_norm": 0.115362748503685,
      "kl": 0.002117633819580078,
      "learning_rate": 5.813904131848564e-07,
      "loss": 0.0001,
      "reward": 1.0208333432674408,
      "reward_std": 0.6431698948144913,
      "rewards/accuracy_reward": 0.4166666828095913,
      "rewards/format_reward": 0.6041666939854622,
      "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": 3230.5938110351562,
      "entropy": 0.41845703125,
      "epoch": 0.06285714285714286,
      "grad_norm": 0.10033451020717621,
      "kl": 0.0014595985412597656,
      "learning_rate": 5.657047735161255e-07,
      "loss": 0.0001,
      "reward": 0.6458333432674408,
      "reward_std": 0.45980407297611237,
      "rewards/accuracy_reward": 0.23958333767950535,
      "rewards/format_reward": 0.4062500149011612,
      "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": 3048.572998046875,
      "entropy": 0.3974609375,
      "epoch": 0.064,
      "grad_norm": 0.11146537959575653,
      "kl": 0.000797271728515625,
      "learning_rate": 5.5e-07,
      "loss": 0.0,
      "reward": 0.7083333618938923,
      "reward_std": 0.3958134725689888,
      "rewards/accuracy_reward": 0.22916666697710752,
      "rewards/format_reward": 0.4791666753590107,
      "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": 3564.4271240234375,
      "entropy": 0.32861328125,
      "epoch": 0.06514285714285714,
      "grad_norm": 0.0911736860871315,
      "kl": 0.0005888938903808594,
      "learning_rate": 5.342952264838747e-07,
      "loss": 0.0,
      "reward": 0.4479166744276881,
      "reward_std": 0.4659876525402069,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.3229166781529784,
      "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": 2263.2709350585938,
      "entropy": 0.31640625,
      "epoch": 0.06628571428571428,
      "grad_norm": 0.35684794187545776,
      "kl": 0.017975807189941406,
      "learning_rate": 5.186095868151436e-07,
      "loss": 0.0007,
      "reward": 1.0833333879709244,
      "reward_std": 0.4603991210460663,
      "rewards/accuracy_reward": 0.3437500074505806,
      "rewards/format_reward": 0.7395833507180214,
      "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": 3056.916748046875,
      "entropy": 0.373046875,
      "epoch": 0.06742857142857143,
      "grad_norm": 0.10368785262107849,
      "kl": 0.0009164810180664062,
      "learning_rate": 5.02962191529556e-07,
      "loss": 0.0,
      "reward": 0.6458333488553762,
      "reward_std": 0.23102688789367676,
      "rewards/accuracy_reward": 0.2604166716337204,
      "rewards/format_reward": 0.38541667722165585,
      "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": 3156.8021850585938,
      "entropy": 0.37548828125,
      "epoch": 0.06857142857142857,
      "grad_norm": 0.1183134987950325,
      "kl": 0.002079010009765625,
      "learning_rate": 4.873721045679706e-07,
      "loss": 0.0001,
      "reward": 0.583333358168602,
      "reward_std": 0.4747445732355118,
      "rewards/accuracy_reward": 0.1562500037252903,
      "rewards/format_reward": 0.4270833469927311,
      "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": 3244.5834350585938,
      "entropy": 0.3857421875,
      "epoch": 0.06971428571428571,
      "grad_norm": 0.14130569994449615,
      "kl": 0.0011582374572753906,
      "learning_rate": 4.7185832004988133e-07,
      "loss": 0.0,
      "reward": 0.6041666865348816,
      "reward_std": 0.5180932655930519,
      "rewards/accuracy_reward": 0.1666666716337204,
      "rewards/format_reward": 0.4375000074505806,
      "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": 2601.604248046875,
      "entropy": 0.349365234375,
      "epoch": 0.07085714285714285,
      "grad_norm": 0.1424204707145691,
      "kl": 0.003665924072265625,
      "learning_rate": 4.5643973913200837e-07,
      "loss": 0.0001,
      "reward": 0.9687500149011612,
      "reward_std": 0.5048108845949173,
      "rewards/accuracy_reward": 0.29166667722165585,
      "rewards/format_reward": 0.677083358168602,
      "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": 2206.0521240234375,
      "entropy": 0.3828125,
      "epoch": 0.072,
      "grad_norm": 0.1572231650352478,
      "kl": 0.002132415771484375,
      "learning_rate": 4.4113514698014953e-07,
      "loss": 0.0001,
      "reward": 1.1145833879709244,
      "reward_std": 0.4276188388466835,
      "rewards/accuracy_reward": 0.37500000558793545,
      "rewards/format_reward": 0.7395833507180214,
      "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": 2939.104248046875,
      "entropy": 0.41796875,
      "epoch": 0.07314285714285715,
      "grad_norm": 0.12188446521759033,
      "kl": 0.0015726089477539062,
      "learning_rate": 4.2596318988235037e-07,
      "loss": 0.0001,
      "reward": 0.697916679084301,
      "reward_std": 0.41498078405857086,
      "rewards/accuracy_reward": 0.2083333358168602,
      "rewards/format_reward": 0.4895833432674408,
      "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": 2967.729248046875,
      "entropy": 0.388671875,
      "epoch": 0.07428571428571429,
      "grad_norm": 0.10674432665109634,
      "kl": 0.0023851394653320312,
      "learning_rate": 4.1094235253127374e-07,
      "loss": 0.0001,
      "reward": 0.572916679084301,
      "reward_std": 0.33290695026516914,
      "rewards/accuracy_reward": 0.14583333395421505,
      "rewards/format_reward": 0.4270833432674408,
      "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": 2160.760543823242,
      "entropy": 0.3271484375,
      "epoch": 0.07542857142857143,
      "grad_norm": 0.19801469147205353,
      "kl": 0.0012044906616210938,
      "learning_rate": 3.9609093550344907e-07,
      "loss": 0.0,
      "reward": 0.8958333637565374,
      "reward_std": 0.2828211933374405,
      "rewards/accuracy_reward": 0.3645833432674408,
      "rewards/format_reward": 0.5312500055879354,
      "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": 3693.2188110351562,
      "entropy": 0.3583984375,
      "epoch": 0.07657142857142857,
      "grad_norm": 0.05972021073102951,
      "kl": 0.0014994144439697266,
      "learning_rate": 3.8142703296283953e-07,
      "loss": 0.0001,
      "reward": 0.22916667722165585,
      "reward_std": 0.2201046198606491,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.18750000279396772,
      "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": 2197.3750610351562,
      "entropy": 0.400390625,
      "epoch": 0.07771428571428571,
      "grad_norm": 0.19413244724273682,
      "kl": 0.004848480224609375,
      "learning_rate": 3.6696851061588994e-07,
      "loss": 0.0002,
      "reward": 0.854166679084301,
      "reward_std": 0.3819599561393261,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.625,
      "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": 2442.1251220703125,
      "entropy": 0.4423828125,
      "epoch": 0.07885714285714286,
      "grad_norm": 0.16839611530303955,
      "kl": 0.00460052490234375,
      "learning_rate": 3.5273298394491515e-07,
      "loss": 0.0002,
      "reward": 0.5833333432674408,
      "reward_std": 0.2996686138212681,
      "rewards/accuracy_reward": 0.052083334885537624,
      "rewards/format_reward": 0.5312500149011612,
      "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": 3122.604217529297,
      "entropy": 0.37255859375,
      "epoch": 0.08,
      "grad_norm": 0.10607440769672394,
      "kl": 0.0026311874389648438,
      "learning_rate": 3.387377967463493e-07,
      "loss": 0.0001,
      "reward": 0.5312500176951289,
      "reward_std": 0.39171791821718216,
      "rewards/accuracy_reward": 0.1250000037252903,
      "rewards/format_reward": 0.4062500027939677,
      "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": 2814.3541870117188,
      "entropy": 0.40185546875,
      "epoch": 0.08114285714285714,
      "grad_norm": 0.14608389139175415,
      "kl": 0.0019731521606445312,
      "learning_rate": 3.250000000000001e-07,
      "loss": 0.0001,
      "reward": 0.5833333656191826,
      "reward_std": 0.2592732161283493,
      "rewards/accuracy_reward": 0.1770833395421505,
      "rewards/format_reward": 0.4062500074505806,
      "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": 3092.6146240234375,
      "entropy": 0.4951171875,
      "epoch": 0.08228571428571428,
      "grad_norm": 0.24482354521751404,
      "kl": 0.00508880615234375,
      "learning_rate": 3.115363310950578e-07,
      "loss": 0.0002,
      "reward": 0.4791666716337204,
      "reward_std": 0.5286267623305321,
      "rewards/accuracy_reward": 0.06250000279396772,
      "rewards/format_reward": 0.4166666716337204,
      "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": 3823.2083740234375,
      "entropy": 0.47607421875,
      "epoch": 0.08342857142857144,
      "grad_norm": 0.0865115225315094,
      "kl": 0.0011157989501953125,
      "learning_rate": 2.9836319343816397e-07,
      "loss": 0.0,
      "reward": 0.2708333358168602,
      "reward_std": 0.38550007343292236,
      "rewards/accuracy_reward": 0.09375000186264515,
      "rewards/format_reward": 0.1770833432674408,
      "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": 3367.6251220703125,
      "entropy": 0.3984375,
      "epoch": 0.08457142857142858,
      "grad_norm": 0.13148467242717743,
      "kl": 0.002348661422729492,
      "learning_rate": 2.854966364683872e-07,
      "loss": 0.0001,
      "reward": 0.5312500074505806,
      "reward_std": 0.4689357727766037,
      "rewards/accuracy_reward": 0.2395833395421505,
      "rewards/format_reward": 0.2916666753590107,
      "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": 3184.9688110351562,
      "entropy": 0.384765625,
      "epoch": 0.08571428571428572,
      "grad_norm": 0.11933936178684235,
      "kl": 0.0034360885620117188,
      "learning_rate": 2.729523361034538e-07,
      "loss": 0.0001,
      "reward": 0.7604166865348816,
      "reward_std": 0.4268460050225258,
      "rewards/accuracy_reward": 0.2395833358168602,
      "rewards/format_reward": 0.5208333432674408,
      "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": 2654.729217529297,
      "entropy": 0.4208984375,
      "epoch": 0.08685714285714285,
      "grad_norm": 0.12776491045951843,
      "kl": 0.0011835098266601562,
      "learning_rate": 2.6074557564105724e-07,
      "loss": 0.0,
      "reward": 0.6145833507180214,
      "reward_std": 0.31801582872867584,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.5520833507180214,
      "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": 3385.1771850585938,
      "entropy": 0.42529296875,
      "epoch": 0.088,
      "grad_norm": 0.12118097394704819,
      "kl": 0.0011730194091796875,
      "learning_rate": 2.488912271385139e-07,
      "loss": 0.0,
      "reward": 0.45833333395421505,
      "reward_std": 0.4228808395564556,
      "rewards/accuracy_reward": 0.1458333432674408,
      "rewards/format_reward": 0.31250000558793545,
      "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": 3198.354248046875,
      "entropy": 0.3857421875,
      "epoch": 0.08914285714285715,
      "grad_norm": 0.12059911340475082,
      "kl": 0.0008554458618164062,
      "learning_rate": 2.374037332934512e-07,
      "loss": 0.0,
      "reward": 0.7395833488553762,
      "reward_std": 0.579352393746376,
      "rewards/accuracy_reward": 0.322916679084301,
      "rewards/format_reward": 0.41666667722165585,
      "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": 2450.9376068115234,
      "entropy": 0.35791015625,
      "epoch": 0.09028571428571429,
      "grad_norm": 0.11260655522346497,
      "kl": 0.0017566680908203125,
      "learning_rate": 2.2629708984760706e-07,
      "loss": 0.0001,
      "reward": 0.8645833432674408,
      "reward_std": 0.30639345943927765,
      "rewards/accuracy_reward": 0.2395833358168602,
      "rewards/format_reward": 0.625,
      "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": 3539.8333740234375,
      "entropy": 0.50439453125,
      "epoch": 0.09142857142857143,
      "grad_norm": 0.10055667161941528,
      "kl": 0.0021123886108398438,
      "learning_rate": 2.1558482853517253e-07,
      "loss": 0.0001,
      "reward": 0.6666666828095913,
      "reward_std": 0.41628430783748627,
      "rewards/accuracy_reward": 0.26041666977107525,
      "rewards/format_reward": 0.40625001303851604,
      "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": 3229.6459045410156,
      "entropy": 0.56689453125,
      "epoch": 0.09257142857142857,
      "grad_norm": 0.1098402738571167,
      "kl": 0.0049724578857421875,
      "learning_rate": 2.0528000059645995e-07,
      "loss": 0.0002,
      "reward": 0.5000000149011612,
      "reward_std": 0.33728423714637756,
      "rewards/accuracy_reward": 0.13541667070239782,
      "rewards/format_reward": 0.3645833358168602,
      "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": 2786.8438110351562,
      "entropy": 0.40576171875,
      "epoch": 0.09371428571428571,
      "grad_norm": 0.15689286589622498,
      "kl": 0.0017375946044921875,
      "learning_rate": 1.9539516087697517e-07,
      "loss": 0.0001,
      "reward": 0.7916666865348816,
      "reward_std": 0.39974234253168106,
      "rewards/accuracy_reward": 0.322916679084301,
      "rewards/format_reward": 0.4687500149011612,
      "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": 3007.010498046875,
      "entropy": 0.46875,
      "epoch": 0.09485714285714286,
      "grad_norm": 0.1386316418647766,
      "kl": 0.0024261474609375,
      "learning_rate": 1.8594235253127372e-07,
      "loss": 0.0001,
      "reward": 0.572916679084301,
      "reward_std": 0.47454773262143135,
      "rewards/accuracy_reward": 0.1979166679084301,
      "rewards/format_reward": 0.3750000111758709,
      "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": 3275.729248046875,
      "entropy": 0.4599609375,
      "epoch": 0.096,
      "grad_norm": 0.15495306253433228,
      "kl": 0.0011186599731445312,
      "learning_rate": 1.7693309235023127e-07,
      "loss": 0.0,
      "reward": 0.6562500204890966,
      "reward_std": 0.4965464398264885,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.42708333395421505,
      "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": 3036.8334350585938,
      "entropy": 0.3447265625,
      "epoch": 0.09714285714285714,
      "grad_norm": 0.12624689936637878,
      "kl": 0.0011987686157226562,
      "learning_rate": 1.6837835672960831e-07,
      "loss": 0.0,
      "reward": 0.7500000149011612,
      "reward_std": 0.6647925078868866,
      "rewards/accuracy_reward": 0.20833334140479565,
      "rewards/format_reward": 0.541666679084301,
      "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": 2982.281280517578,
      "entropy": 0.4560546875,
      "epoch": 0.09828571428571428,
      "grad_norm": 0.09444669634103775,
      "kl": 0.0027751922607421875,
      "learning_rate": 1.6028856829700258e-07,
      "loss": 0.0001,
      "reward": 0.65625,
      "reward_std": 0.16639107465744019,
      "rewards/accuracy_reward": 0.18750000279396772,
      "rewards/format_reward": 0.4687500149011612,
      "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": 2746.73974609375,
      "entropy": 0.494140625,
      "epoch": 0.09942857142857142,
      "grad_norm": 0.23193266987800598,
      "kl": 0.00341033935546875,
      "learning_rate": 1.5267358321348285e-07,
      "loss": 0.0001,
      "reward": 0.7604166865348816,
      "reward_std": 0.5403826013207436,
      "rewards/accuracy_reward": 0.20833333861082792,
      "rewards/format_reward": 0.5520833358168602,
      "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": 2746.791717529297,
      "entropy": 0.45751953125,
      "epoch": 0.10057142857142858,
      "grad_norm": 0.17896543443202972,
      "kl": 0.002384185791015625,
      "learning_rate": 1.4554267916537495e-07,
      "loss": 0.0001,
      "reward": 0.9583333730697632,
      "reward_std": 0.6643739938735962,
      "rewards/accuracy_reward": 0.3437500074505806,
      "rewards/format_reward": 0.6145833507180214,
      "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": 3201.4271240234375,
      "entropy": 0.41064453125,
      "epoch": 0.10171428571428572,
      "grad_norm": 0.18984779715538025,
      "kl": 0.002773284912109375,
      "learning_rate": 1.3890454406082956e-07,
      "loss": 0.0001,
      "reward": 0.6250000149011612,
      "reward_std": 0.5604990497231483,
      "rewards/accuracy_reward": 0.2083333358168602,
      "rewards/format_reward": 0.4166666716337204,
      "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": 2363.229248046875,
      "entropy": 0.57080078125,
      "epoch": 0.10285714285714286,
      "grad_norm": 0.12273731827735901,
      "kl": 0.006702423095703125,
      "learning_rate": 1.3276726544494571e-07,
      "loss": 0.0003,
      "reward": 0.6250000102445483,
      "reward_std": 0.17295432463288307,
      "rewards/accuracy_reward": 0.031250000931322575,
      "rewards/format_reward": 0.5937500102445483,
      "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": 3230.1875610351562,
      "entropy": 0.4580078125,
      "epoch": 0.104,
      "grad_norm": 0.11221471428871155,
      "kl": 0.0031261444091796875,
      "learning_rate": 1.2713832064634125e-07,
      "loss": 0.0001,
      "reward": 0.6041666995733976,
      "reward_std": 0.4464225322008133,
      "rewards/accuracy_reward": 0.18750000838190317,
      "rewards/format_reward": 0.4166666818782687,
      "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": 2720.2709350585938,
      "entropy": 0.388671875,
      "epoch": 0.10514285714285715,
      "grad_norm": 0.12194187194108963,
      "kl": 0.002674102783203125,
      "learning_rate": 1.220245676671809e-07,
      "loss": 0.0001,
      "reward": 0.6770833693444729,
      "reward_std": 0.4861874431371689,
      "rewards/accuracy_reward": 0.15625000558793545,
      "rewards/format_reward": 0.5208333544433117,
      "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": 3754.322998046875,
      "entropy": 0.5771484375,
      "epoch": 0.10628571428571429,
      "grad_norm": 0.09679470956325531,
      "kl": 0.00360107421875,
      "learning_rate": 1.1743223682775649e-07,
      "loss": 0.0001,
      "reward": 0.11458333767950535,
      "reward_std": 0.2338472381234169,
      "rewards/accuracy_reward": 0.010416666977107525,
      "rewards/format_reward": 0.10416667070239782,
      "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": 3023.7084350585938,
      "entropy": 0.52783203125,
      "epoch": 0.10742857142857143,
      "grad_norm": 0.12803490459918976,
      "kl": 0.00276947021484375,
      "learning_rate": 1.1336692317580158e-07,
      "loss": 0.0001,
      "reward": 0.5937500074505806,
      "reward_std": 0.3088596239686012,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.4062500074505806,
      "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": 3628.541748046875,
      "entropy": 0.443359375,
      "epoch": 0.10857142857142857,
      "grad_norm": 0.10912292450666428,
      "kl": 0.0014452934265136719,
      "learning_rate": 1.0983357966978745e-07,
      "loss": 0.0001,
      "reward": 0.3645833432674408,
      "reward_std": 0.4797569513320923,
      "rewards/accuracy_reward": 0.0937500037252903,
      "rewards/format_reward": 0.2708333358168602,
      "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": 2806.2708740234375,
      "entropy": 0.390625,
      "epoch": 0.10971428571428571,
      "grad_norm": 0.12556815147399902,
      "kl": 0.0026493072509765625,
      "learning_rate": 1.068365111445064e-07,
      "loss": 0.0001,
      "reward": 0.6979166716337204,
      "reward_std": 0.3458334431052208,
      "rewards/accuracy_reward": 0.2500000111758709,
      "rewards/format_reward": 0.447916679084301,
      "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": 3243.8334350585938,
      "entropy": 0.423828125,
      "epoch": 0.11085714285714286,
      "grad_norm": 0.14200259745121002,
      "kl": 0.001922607421875,
      "learning_rate": 1.0437936906629334e-07,
      "loss": 0.0001,
      "reward": 0.6250000027939677,
      "reward_std": 0.48585430532693863,
      "rewards/accuracy_reward": 0.2395833395421505,
      "rewards/format_reward": 0.3854166818782687,
      "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": 2731.697998046875,
      "entropy": 0.3837890625,
      "epoch": 0.112,
      "grad_norm": 0.19670821726322174,
      "kl": 0.0012807846069335938,
      "learning_rate": 1.0246514708427701e-07,
      "loss": 0.0001,
      "reward": 0.7291667088866234,
      "reward_std": 0.4080168679356575,
      "rewards/accuracy_reward": 0.16666667256504297,
      "rewards/format_reward": 0.5625000149011612,
      "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": 3088.635482788086,
      "entropy": 0.391845703125,
      "epoch": 0.11314285714285714,
      "grad_norm": 0.09702961891889572,
      "kl": 0.0017948150634765625,
      "learning_rate": 1.0109617738307911e-07,
      "loss": 0.0001,
      "reward": 0.5104166772216558,
      "reward_std": 0.24148838222026825,
      "rewards/accuracy_reward": 0.19791667815297842,
      "rewards/format_reward": 0.3125000027939677,
      "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": 2679.916748046875,
      "entropy": 0.361083984375,
      "epoch": 0.11428571428571428,
      "grad_norm": 0.15485654771327972,
      "kl": 0.0039215087890625,
      "learning_rate": 1.002741278414069e-07,
      "loss": 0.0002,
      "reward": 0.9479166865348816,
      "reward_std": 0.6187009885907173,
      "rewards/accuracy_reward": 0.3437500102445483,
      "rewards/format_reward": 0.6041666865348816,
      "step": 100
    },
    {
      "epoch": 0.11428571428571428,
      "step": 100,
      "total_flos": 0.0,
      "train_loss": 7.051572029013187e-05,
      "train_runtime": 8438.3056,
      "train_samples_per_second": 1.138,
      "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
}