{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.36986301369863,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 463.84375,
      "epoch": 0.0013698630136986301,
      "grad_norm": 3.912295341491699,
      "kl": 0.0006895065307617188,
      "learning_rate": 9.995433789954337e-07,
      "loss": 0.0,
      "reward": 0.734375,
      "reward_std": 0.6482069045305252,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 0.4375,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.53125,
      "epoch": 0.0027397260273972603,
      "grad_norm": 4.425003528594971,
      "kl": 0.0008821487426757812,
      "learning_rate": 9.990867579908674e-07,
      "loss": 0.0,
      "reward": 0.734375,
      "reward_std": 0.5391269624233246,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 0.4375,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 407.46875,
      "epoch": 0.00410958904109589,
      "grad_norm": 4.136044979095459,
      "kl": 0.0013284683227539062,
      "learning_rate": 9.986301369863014e-07,
      "loss": 0.0,
      "reward": 0.909375011920929,
      "reward_std": 0.49507734179496765,
      "rewards/accuracy_reward": 0.25312499701976776,
      "rewards/format_reward": 0.65625,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 463.9375,
      "epoch": 0.005479452054794521,
      "grad_norm": 3.070124864578247,
      "kl": 0.0018978118896484375,
      "learning_rate": 9.98173515981735e-07,
      "loss": 0.0,
      "reward": 0.859375,
      "reward_std": 0.5979855433106422,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 0.59375,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 348.75,
      "epoch": 0.00684931506849315,
      "grad_norm": 3.5212478637695312,
      "kl": 0.00206756591796875,
      "learning_rate": 9.977168949771688e-07,
      "loss": 0.0,
      "reward": 1.3125,
      "reward_std": 0.6811521649360657,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.78125,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 363.0625,
      "epoch": 0.00821917808219178,
      "grad_norm": 3.7381842136383057,
      "kl": 0.002559661865234375,
      "learning_rate": 9.972602739726028e-07,
      "loss": 0.0,
      "reward": 1.109375,
      "reward_std": 0.3039700835943222,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 0.9375,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 332.1875,
      "epoch": 0.009589041095890411,
      "grad_norm": 2.0565128326416016,
      "kl": 0.003376007080078125,
      "learning_rate": 9.968036529680365e-07,
      "loss": 0.0,
      "reward": 1.296875,
      "reward_std": 0.1530819907784462,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 1.0,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 381.5625,
      "epoch": 0.010958904109589041,
      "grad_norm": 3.6933867931365967,
      "kl": 0.004985809326171875,
      "learning_rate": 9.963470319634703e-07,
      "loss": 0.0,
      "reward": 1.390625,
      "reward_std": 0.5591665953397751,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.875,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.625,
      "epoch": 0.012328767123287671,
      "grad_norm": 15.956212043762207,
      "kl": 0.006072998046875,
      "learning_rate": 9.95890410958904e-07,
      "loss": 0.0,
      "reward": 1.1119791865348816,
      "reward_std": 0.5166353359818459,
      "rewards/accuracy_reward": 0.2369791567325592,
      "rewards/format_reward": 0.875,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 557.15625,
      "epoch": 0.0136986301369863,
      "grad_norm": 2.227487802505493,
      "kl": 0.0097808837890625,
      "learning_rate": 9.954337899543377e-07,
      "loss": 0.0,
      "reward": 1.0885416567325592,
      "reward_std": 0.4930662214756012,
      "rewards/accuracy_reward": 0.2760416716337204,
      "rewards/format_reward": 0.8125,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 365.625,
      "epoch": 0.015068493150684932,
      "grad_norm": 4.018799781799316,
      "kl": 0.00763702392578125,
      "learning_rate": 9.949771689497717e-07,
      "loss": 0.0,
      "reward": 1.234375,
      "reward_std": 0.49322642385959625,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 0.90625,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 506.96875,
      "epoch": 0.01643835616438356,
      "grad_norm": 2.298476219177246,
      "kl": 0.01721954345703125,
      "learning_rate": 9.945205479452054e-07,
      "loss": 0.0,
      "reward": 1.15625,
      "reward_std": 0.5379246100783348,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.875,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.78125,
      "epoch": 0.01780821917808219,
      "grad_norm": 5.038638591766357,
      "kl": 0.015594482421875,
      "learning_rate": 9.940639269406391e-07,
      "loss": 0.0,
      "reward": 1.1875,
      "reward_std": 0.48483333736658096,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.875,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 341.40625,
      "epoch": 0.019178082191780823,
      "grad_norm": 3.0773346424102783,
      "kl": 0.0150146484375,
      "learning_rate": 9.93607305936073e-07,
      "loss": 0.0,
      "reward": 1.15625,
      "reward_std": 0.3471629247069359,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.96875,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 564.25,
      "epoch": 0.02054794520547945,
      "grad_norm": 4.9919514656066895,
      "kl": 0.02197265625,
      "learning_rate": 9.931506849315068e-07,
      "loss": 0.0,
      "reward": 1.234375,
      "reward_std": 0.40913281589746475,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 0.96875,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 573.25,
      "epoch": 0.021917808219178082,
      "grad_norm": 4.952914714813232,
      "kl": 0.0227508544921875,
      "learning_rate": 9.926940639269406e-07,
      "loss": 0.0,
      "reward": 1.3359375,
      "reward_std": 0.3955412805080414,
      "rewards/accuracy_reward": 0.4296875,
      "rewards/format_reward": 0.90625,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 447.96875,
      "epoch": 0.023287671232876714,
      "grad_norm": 2.7455577850341797,
      "kl": 0.0212554931640625,
      "learning_rate": 9.922374429223745e-07,
      "loss": 0.0,
      "reward": 1.25,
      "reward_std": 0.4765502139925957,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.96875,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.875,
      "epoch": 0.024657534246575342,
      "grad_norm": 5.91363525390625,
      "kl": 0.021270751953125,
      "learning_rate": 9.917808219178082e-07,
      "loss": 0.0,
      "reward": 1.2265625,
      "reward_std": 0.5312308222055435,
      "rewards/accuracy_reward": 0.3203125,
      "rewards/format_reward": 0.90625,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.09375,
      "epoch": 0.026027397260273973,
      "grad_norm": 4.743810653686523,
      "kl": 0.0315704345703125,
      "learning_rate": 9.91324200913242e-07,
      "loss": 0.0,
      "reward": 1.1875,
      "reward_std": 0.6161536350846291,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.875,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 450.5,
      "epoch": 0.0273972602739726,
      "grad_norm": 1.7055182456970215,
      "kl": 0.027923583984375,
      "learning_rate": 9.908675799086757e-07,
      "loss": 0.0,
      "reward": 1.234375,
      "reward_std": 0.13939543068408966,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 600.6875,
      "epoch": 0.028767123287671233,
      "grad_norm": 2.169266700744629,
      "kl": 0.04107666015625,
      "learning_rate": 9.904109589041094e-07,
      "loss": 0.0,
      "reward": 1.3229166865348816,
      "reward_std": 0.23428862541913986,
      "rewards/accuracy_reward": 0.3541666567325592,
      "rewards/format_reward": 0.96875,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.53125,
      "epoch": 0.030136986301369864,
      "grad_norm": 3.142946720123291,
      "kl": 0.0318145751953125,
      "learning_rate": 9.899543378995434e-07,
      "loss": 0.0,
      "reward": 1.2890625,
      "reward_std": 0.46334072202444077,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 0.9375,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 383.65625,
      "epoch": 0.031506849315068496,
      "grad_norm": 4.148040294647217,
      "kl": 0.021270751953125,
      "learning_rate": 9.894977168949771e-07,
      "loss": 0.0,
      "reward": 1.234375,
      "reward_std": 0.3653144985437393,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.15625,
      "epoch": 0.03287671232876712,
      "grad_norm": 3.037674903869629,
      "kl": 0.03924560546875,
      "learning_rate": 9.89041095890411e-07,
      "loss": 0.0,
      "reward": 1.3385416865348816,
      "reward_std": 0.3099621832370758,
      "rewards/accuracy_reward": 0.3697916567325592,
      "rewards/format_reward": 0.96875,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 554.1875,
      "epoch": 0.03424657534246575,
      "grad_norm": 4.072643280029297,
      "kl": 0.036590576171875,
      "learning_rate": 9.885844748858448e-07,
      "loss": 0.0,
      "reward": 1.453125,
      "reward_std": 0.40609828382730484,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 414.90625,
      "epoch": 0.03561643835616438,
      "grad_norm": 6.202883243560791,
      "kl": 0.03265380859375,
      "learning_rate": 9.881278538812785e-07,
      "loss": 0.0,
      "reward": 1.171875,
      "reward_std": 0.19583626091480255,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.75,
      "epoch": 0.036986301369863014,
      "grad_norm": 2.829223155975342,
      "kl": 0.0391693115234375,
      "learning_rate": 9.876712328767123e-07,
      "loss": 0.0,
      "reward": 1.46875,
      "reward_std": 0.22084104642271996,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 398.96875,
      "epoch": 0.038356164383561646,
      "grad_norm": 3.0730113983154297,
      "kl": 0.043853759765625,
      "learning_rate": 9.87214611872146e-07,
      "loss": 0.0,
      "reward": 1.46875,
      "reward_std": 0.36339621990919113,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.34375,
      "epoch": 0.03972602739726028,
      "grad_norm": 3.6202750205993652,
      "kl": 0.0521240234375,
      "learning_rate": 9.867579908675797e-07,
      "loss": 0.0001,
      "reward": 1.43359375,
      "reward_std": 0.4707936607301235,
      "rewards/accuracy_reward": 0.52734375,
      "rewards/format_reward": 0.90625,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.75,
      "epoch": 0.0410958904109589,
      "grad_norm": 2.320294141769409,
      "kl": 0.040008544921875,
      "learning_rate": 9.863013698630137e-07,
      "loss": 0.0,
      "reward": 1.0,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.03125,
      "rewards/format_reward": 0.96875,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.125,
      "epoch": 0.04246575342465753,
      "grad_norm": 3.044395685195923,
      "kl": 0.05633544921875,
      "learning_rate": 9.858447488584474e-07,
      "loss": 0.0001,
      "reward": 1.46875,
      "reward_std": 0.346555445343256,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 316.65625,
      "epoch": 0.043835616438356165,
      "grad_norm": 2.4935569763183594,
      "kl": 0.04290771484375,
      "learning_rate": 9.853881278538814e-07,
      "loss": 0.0,
      "reward": 1.09375,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.09375,
      "rewards/format_reward": 1.0,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 430.46875,
      "epoch": 0.045205479452054796,
      "grad_norm": 2.3980815410614014,
      "kl": 0.06085205078125,
      "learning_rate": 9.84931506849315e-07,
      "loss": 0.0001,
      "reward": 1.28125,
      "reward_std": 0.30371319502592087,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 0.96875,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.6875,
      "epoch": 0.04657534246575343,
      "grad_norm": 1.6388932466506958,
      "kl": 0.04888916015625,
      "learning_rate": 9.844748858447488e-07,
      "loss": 0.0,
      "reward": 1.421875,
      "reward_std": 0.1711306795477867,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.09375,
      "epoch": 0.04794520547945205,
      "grad_norm": 2.9538469314575195,
      "kl": 0.054229736328125,
      "learning_rate": 9.840182648401826e-07,
      "loss": 0.0001,
      "reward": 1.34375,
      "reward_std": 0.1523548737168312,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 396.28125,
      "epoch": 0.049315068493150684,
      "grad_norm": 1.2823041677474976,
      "kl": 0.032958984375,
      "learning_rate": 9.835616438356163e-07,
      "loss": 0.0,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.96875,
      "epoch": 0.050684931506849315,
      "grad_norm": 10.917831420898438,
      "kl": 0.0721435546875,
      "learning_rate": 9.831050228310502e-07,
      "loss": 0.0001,
      "reward": 1.50390625,
      "reward_std": 0.08883348293602467,
      "rewards/accuracy_reward": 0.50390625,
      "rewards/format_reward": 1.0,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 489.15625,
      "epoch": 0.052054794520547946,
      "grad_norm": 3.121203660964966,
      "kl": 0.04541015625,
      "learning_rate": 9.82648401826484e-07,
      "loss": 0.0,
      "reward": 1.25,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.96875,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 476.96875,
      "epoch": 0.05342465753424658,
      "grad_norm": 2.3028512001037598,
      "kl": 0.0438232421875,
      "learning_rate": 9.821917808219177e-07,
      "loss": 0.0,
      "reward": 1.4739583134651184,
      "reward_std": 0.2700696364045143,
      "rewards/accuracy_reward": 0.4739583432674408,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.96875,
      "epoch": 0.0547945205479452,
      "grad_norm": 1.8145931959152222,
      "kl": 0.056488037109375,
      "learning_rate": 9.817351598173517e-07,
      "loss": 0.0001,
      "reward": 1.2421875,
      "reward_std": 0.3533598557114601,
      "rewards/accuracy_reward": 0.3046875,
      "rewards/format_reward": 0.9375,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.59375,
      "epoch": 0.056164383561643834,
      "grad_norm": 3.3131184577941895,
      "kl": 0.044189453125,
      "learning_rate": 9.812785388127854e-07,
      "loss": 0.0,
      "reward": 1.34375,
      "reward_std": 0.4578060656785965,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.96875,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.9375,
      "epoch": 0.057534246575342465,
      "grad_norm": 8.756609916687012,
      "kl": 0.059814453125,
      "learning_rate": 9.808219178082191e-07,
      "loss": 0.0001,
      "reward": 1.5703125,
      "reward_std": 0.4013843312859535,
      "rewards/accuracy_reward": 0.6015625,
      "rewards/format_reward": 0.96875,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 472.96875,
      "epoch": 0.0589041095890411,
      "grad_norm": 2.866149663925171,
      "kl": 0.0538330078125,
      "learning_rate": 9.803652968036529e-07,
      "loss": 0.0001,
      "reward": 1.3515625,
      "reward_std": 0.34194046072661877,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 667.09375,
      "epoch": 0.06027397260273973,
      "grad_norm": 6.409434795379639,
      "kl": 0.058013916015625,
      "learning_rate": 9.799086757990868e-07,
      "loss": 0.0001,
      "reward": 1.5218749642372131,
      "reward_std": 0.34158414881676435,
      "rewards/accuracy_reward": 0.5843749791383743,
      "rewards/format_reward": 0.9375,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 716.84375,
      "epoch": 0.06164383561643835,
      "grad_norm": 1.960483431816101,
      "kl": 0.0870361328125,
      "learning_rate": 9.794520547945205e-07,
      "loss": 0.0001,
      "reward": 1.3515625,
      "reward_std": 0.26203832402825356,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 0.96875,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.34375,
      "epoch": 0.06301369863013699,
      "grad_norm": 1.6870055198669434,
      "kl": 0.06658935546875,
      "learning_rate": 9.789954337899543e-07,
      "loss": 0.0001,
      "reward": 1.2578125,
      "reward_std": 0.32489965856075287,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 0.875,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.34375,
      "epoch": 0.06438356164383562,
      "grad_norm": 2.6664040088653564,
      "kl": 0.0615234375,
      "learning_rate": 9.78538812785388e-07,
      "loss": 0.0001,
      "reward": 1.3828125,
      "reward_std": 0.1984097883105278,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 758.625,
      "epoch": 0.06575342465753424,
      "grad_norm": 6.174232482910156,
      "kl": 0.07745361328125,
      "learning_rate": 9.78082191780822e-07,
      "loss": 0.0001,
      "reward": 1.2946428656578064,
      "reward_std": 0.6038303673267365,
      "rewards/accuracy_reward": 0.4821428507566452,
      "rewards/format_reward": 0.8125,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 478.3125,
      "epoch": 0.06712328767123288,
      "grad_norm": 4.74515438079834,
      "kl": 0.06597900390625,
      "learning_rate": 9.776255707762557e-07,
      "loss": 0.0001,
      "reward": 1.2734375,
      "reward_std": 0.24417023360729218,
      "rewards/accuracy_reward": 0.2734375,
      "rewards/format_reward": 1.0,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.71875,
      "epoch": 0.0684931506849315,
      "grad_norm": 3.3720693588256836,
      "kl": 0.056671142578125,
      "learning_rate": 9.771689497716894e-07,
      "loss": 0.0001,
      "reward": 1.5,
      "reward_std": 0.23194295540452003,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.125,
      "epoch": 0.06986301369863014,
      "grad_norm": 1.075783133506775,
      "kl": 0.07354736328125,
      "learning_rate": 9.767123287671234e-07,
      "loss": 0.0001,
      "reward": 1.140625,
      "reward_std": 0.1530819907784462,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 0.96875,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 541.5,
      "epoch": 0.07123287671232877,
      "grad_norm": 4.1653876304626465,
      "kl": 0.0670166015625,
      "learning_rate": 9.762557077625571e-07,
      "loss": 0.0001,
      "reward": 1.4375,
      "reward_std": 0.35381053015589714,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.96875,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.1875,
      "epoch": 0.07260273972602739,
      "grad_norm": 2.497183322906494,
      "kl": 0.07391357421875,
      "learning_rate": 9.757990867579908e-07,
      "loss": 0.0001,
      "reward": 1.703125,
      "reward_std": 0.2172447368502617,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.5,
      "epoch": 0.07397260273972603,
      "grad_norm": 1.2262613773345947,
      "kl": 0.06640625,
      "learning_rate": 9.753424657534246e-07,
      "loss": 0.0001,
      "reward": 1.15625,
      "reward_std": 0.1356339044868946,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.96875,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 710.59375,
      "epoch": 0.07534246575342465,
      "grad_norm": 1.7770954370498657,
      "kl": 0.0960693359375,
      "learning_rate": 9.748858447488583e-07,
      "loss": 0.0001,
      "reward": 1.6171875,
      "reward_std": 0.18259718269109726,
      "rewards/accuracy_reward": 0.6171875,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.71875,
      "epoch": 0.07671232876712329,
      "grad_norm": 2.9477083683013916,
      "kl": 0.08294677734375,
      "learning_rate": 9.744292237442923e-07,
      "loss": 0.0001,
      "reward": 1.375,
      "reward_std": 0.2698745857924223,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.40625,
      "epoch": 0.07808219178082192,
      "grad_norm": 2.187389850616455,
      "kl": 0.07904052734375,
      "learning_rate": 9.73972602739726e-07,
      "loss": 0.0001,
      "reward": 1.4620535373687744,
      "reward_std": 0.3073258101940155,
      "rewards/accuracy_reward": 0.493303582072258,
      "rewards/format_reward": 0.96875,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 722.9375,
      "epoch": 0.07945205479452055,
      "grad_norm": 1.8998942375183105,
      "kl": 0.1063232421875,
      "learning_rate": 9.735159817351597e-07,
      "loss": 0.0001,
      "reward": 1.3177083432674408,
      "reward_std": 0.2553338035941124,
      "rewards/accuracy_reward": 0.3489583283662796,
      "rewards/format_reward": 0.96875,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.84375,
      "epoch": 0.08082191780821918,
      "grad_norm": 2.1120872497558594,
      "kl": 0.07696533203125,
      "learning_rate": 9.730593607305937e-07,
      "loss": 0.0001,
      "reward": 1.265625,
      "reward_std": 0.26977966725826263,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 413.28125,
      "epoch": 0.0821917808219178,
      "grad_norm": 3.2552480697631836,
      "kl": 0.06304931640625,
      "learning_rate": 9.726027397260274e-07,
      "loss": 0.0001,
      "reward": 1.125,
      "reward_std": 0.2177756354212761,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 510.9375,
      "epoch": 0.08356164383561644,
      "grad_norm": 2.9126460552215576,
      "kl": 0.0733642578125,
      "learning_rate": 9.721461187214611e-07,
      "loss": 0.0001,
      "reward": 1.328125,
      "reward_std": 0.28930897638201714,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.8125,
      "epoch": 0.08493150684931507,
      "grad_norm": 1.73414146900177,
      "kl": 0.0865478515625,
      "learning_rate": 9.716894977168949e-07,
      "loss": 0.0001,
      "reward": 1.203125,
      "reward_std": 0.33669837564229965,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 0.96875,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 692.09375,
      "epoch": 0.0863013698630137,
      "grad_norm": 1.9965529441833496,
      "kl": 0.1036376953125,
      "learning_rate": 9.712328767123286e-07,
      "loss": 0.0001,
      "reward": 1.7749256193637848,
      "reward_std": 0.1739531122148037,
      "rewards/accuracy_reward": 0.7749256044626236,
      "rewards/format_reward": 1.0,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 695.1875,
      "epoch": 0.08767123287671233,
      "grad_norm": 1.9106062650680542,
      "kl": 0.094482421875,
      "learning_rate": 9.707762557077626e-07,
      "loss": 0.0001,
      "reward": 1.580729216337204,
      "reward_std": 0.07768097147345543,
      "rewards/accuracy_reward": 0.5807291567325592,
      "rewards/format_reward": 1.0,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.75,
      "epoch": 0.08904109589041095,
      "grad_norm": 2.800123929977417,
      "kl": 0.0806884765625,
      "learning_rate": 9.703196347031963e-07,
      "loss": 0.0001,
      "reward": 1.50390625,
      "reward_std": 0.30879483185708523,
      "rewards/accuracy_reward": 0.50390625,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.1875,
      "epoch": 0.09041095890410959,
      "grad_norm": 3.2412710189819336,
      "kl": 0.0830078125,
      "learning_rate": 9.6986301369863e-07,
      "loss": 0.0001,
      "reward": 1.515625,
      "reward_std": 0.5082386285066605,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 457.75,
      "epoch": 0.09178082191780822,
      "grad_norm": 4.994422912597656,
      "kl": 0.0894775390625,
      "learning_rate": 9.69406392694064e-07,
      "loss": 0.0001,
      "reward": 1.1796875,
      "reward_std": 0.12073516845703125,
      "rewards/accuracy_reward": 0.1796875,
      "rewards/format_reward": 1.0,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 532.15625,
      "epoch": 0.09315068493150686,
      "grad_norm": 0.836651623249054,
      "kl": 0.0897216796875,
      "learning_rate": 9.689497716894977e-07,
      "loss": 0.0001,
      "reward": 1.1796875,
      "reward_std": 0.04005437344312668,
      "rewards/accuracy_reward": 0.1796875,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 488.65625,
      "epoch": 0.09452054794520548,
      "grad_norm": 3.8262500762939453,
      "kl": 0.0775146484375,
      "learning_rate": 9.684931506849314e-07,
      "loss": 0.0001,
      "reward": 1.3515625,
      "reward_std": 0.15467960201203823,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 1.0,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.3125,
      "epoch": 0.0958904109589041,
      "grad_norm": 15.291179656982422,
      "kl": 0.069091796875,
      "learning_rate": 9.680365296803652e-07,
      "loss": 0.0001,
      "reward": 1.2265625,
      "reward_std": 0.14807433634996414,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 1.0,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.1875,
      "epoch": 0.09726027397260274,
      "grad_norm": 2.252143383026123,
      "kl": 0.07574462890625,
      "learning_rate": 9.675799086757991e-07,
      "loss": 0.0001,
      "reward": 1.359375,
      "reward_std": 0.2109457477927208,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 575.8125,
      "epoch": 0.09863013698630137,
      "grad_norm": 2.375014305114746,
      "kl": 0.0933837890625,
      "learning_rate": 9.671232876712329e-07,
      "loss": 0.0001,
      "reward": 1.359375,
      "reward_std": 0.2414703369140625,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 424.03125,
      "epoch": 0.1,
      "grad_norm": 3.685784101486206,
      "kl": 0.05877685546875,
      "learning_rate": 9.666666666666666e-07,
      "loss": 0.0001,
      "reward": 1.421875,
      "reward_std": 0.37769732251763344,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 496.0,
      "epoch": 0.10136986301369863,
      "grad_norm": 2.5997228622436523,
      "kl": 0.07012939453125,
      "learning_rate": 9.662100456621003e-07,
      "loss": 0.0001,
      "reward": 1.5,
      "reward_std": 0.31046149134635925,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.75,
      "epoch": 0.10273972602739725,
      "grad_norm": 2.172786235809326,
      "kl": 0.12713623046875,
      "learning_rate": 9.657534246575343e-07,
      "loss": 0.0001,
      "reward": 1.234375,
      "reward_std": 0.17358146235346794,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 653.59375,
      "epoch": 0.10410958904109589,
      "grad_norm": 2.230001449584961,
      "kl": 0.1158447265625,
      "learning_rate": 9.65296803652968e-07,
      "loss": 0.0001,
      "reward": 1.453125,
      "reward_std": 0.23453032225370407,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.0,
      "epoch": 0.10547945205479452,
      "grad_norm": 3.474895477294922,
      "kl": 0.1109619140625,
      "learning_rate": 9.648401826484017e-07,
      "loss": 0.0001,
      "reward": 1.7578125,
      "reward_std": 0.15931576862931252,
      "rewards/accuracy_reward": 0.7578125,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 433.8125,
      "epoch": 0.10684931506849316,
      "grad_norm": 2.6400644779205322,
      "kl": 0.0609130859375,
      "learning_rate": 9.643835616438357e-07,
      "loss": 0.0001,
      "reward": 1.28125,
      "reward_std": 0.2709311693906784,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.4375,
      "epoch": 0.10821917808219178,
      "grad_norm": 3.992541790008545,
      "kl": 0.09515380859375,
      "learning_rate": 9.639269406392694e-07,
      "loss": 0.0001,
      "reward": 1.7578125,
      "reward_std": 0.4703022539615631,
      "rewards/accuracy_reward": 0.7578125,
      "rewards/format_reward": 1.0,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 647.96875,
      "epoch": 0.1095890410958904,
      "grad_norm": 3.139620065689087,
      "kl": 0.106689453125,
      "learning_rate": 9.634703196347032e-07,
      "loss": 0.0001,
      "reward": 1.53515625,
      "reward_std": 0.22270986810326576,
      "rewards/accuracy_reward": 0.53515625,
      "rewards/format_reward": 1.0,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.8125,
      "epoch": 0.11095890410958904,
      "grad_norm": 1.872832179069519,
      "kl": 0.1011962890625,
      "learning_rate": 9.630136986301369e-07,
      "loss": 0.0001,
      "reward": 1.6640625,
      "reward_std": 0.14312389120459557,
      "rewards/accuracy_reward": 0.6640625,
      "rewards/format_reward": 1.0,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.96875,
      "epoch": 0.11232876712328767,
      "grad_norm": 1.6903170347213745,
      "kl": 0.0982666015625,
      "learning_rate": 9.625570776255706e-07,
      "loss": 0.0001,
      "reward": 1.234375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.625,
      "epoch": 0.1136986301369863,
      "grad_norm": 2.276421546936035,
      "kl": 0.095947265625,
      "learning_rate": 9.621004566210046e-07,
      "loss": 0.0001,
      "reward": 1.4453125,
      "reward_std": 0.232578843832016,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 1.0,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.03125,
      "epoch": 0.11506849315068493,
      "grad_norm": 3.133164882659912,
      "kl": 0.109130859375,
      "learning_rate": 9.616438356164383e-07,
      "loss": 0.0001,
      "reward": 1.6015625,
      "reward_std": 0.2483602836728096,
      "rewards/accuracy_reward": 0.6015625,
      "rewards/format_reward": 1.0,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 512.8125,
      "epoch": 0.11643835616438356,
      "grad_norm": 2.0008928775787354,
      "kl": 0.0712890625,
      "learning_rate": 9.61187214611872e-07,
      "loss": 0.0001,
      "reward": 1.328125,
      "reward_std": 0.30617379024624825,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 0.96875,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 523.75,
      "epoch": 0.1178082191780822,
      "grad_norm": 1.2373316287994385,
      "kl": 0.082275390625,
      "learning_rate": 9.60730593607306e-07,
      "loss": 0.0001,
      "reward": 1.390625,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 417.5,
      "epoch": 0.11917808219178082,
      "grad_norm": 22.696931838989258,
      "kl": 0.0906982421875,
      "learning_rate": 9.602739726027397e-07,
      "loss": 0.0001,
      "reward": 1.171875,
      "reward_std": 0.13258251920342445,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 1.0,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 365.84375,
      "epoch": 0.12054794520547946,
      "grad_norm": 4.310155868530273,
      "kl": 0.05987548828125,
      "learning_rate": 9.598173515981735e-07,
      "loss": 0.0001,
      "reward": 1.1875,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.21875,
      "epoch": 0.12191780821917808,
      "grad_norm": 0.8474344611167908,
      "kl": 0.09967041015625,
      "learning_rate": 9.593607305936072e-07,
      "loss": 0.0001,
      "reward": 1.1875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.03125,
      "epoch": 0.1232876712328767,
      "grad_norm": 2.66658091545105,
      "kl": 0.12939453125,
      "learning_rate": 9.58904109589041e-07,
      "loss": 0.0001,
      "reward": 1.4296875,
      "reward_std": 0.2822495624423027,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 0.96875,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.40625,
      "epoch": 0.12465753424657534,
      "grad_norm": 1.2776012420654297,
      "kl": 0.1162109375,
      "learning_rate": 9.584474885844749e-07,
      "loss": 0.0001,
      "reward": 1.34375,
      "reward_std": 0.10888781771063805,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 463.25,
      "epoch": 0.12602739726027398,
      "grad_norm": 1.5968950986862183,
      "kl": 0.09906005859375,
      "learning_rate": 9.579908675799086e-07,
      "loss": 0.0001,
      "reward": 1.3125,
      "reward_std": 0.1356339044868946,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.21875,
      "epoch": 0.1273972602739726,
      "grad_norm": 6.543825149536133,
      "kl": 0.112060546875,
      "learning_rate": 9.575342465753423e-07,
      "loss": 0.0001,
      "reward": 1.5546875,
      "reward_std": 0.29614376835525036,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 1.0,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 710.5625,
      "epoch": 0.12876712328767123,
      "grad_norm": 4.022386074066162,
      "kl": 0.1082763671875,
      "learning_rate": 9.570776255707763e-07,
      "loss": 0.0001,
      "reward": 1.2916666865348816,
      "reward_std": 0.289409551769495,
      "rewards/accuracy_reward": 0.3854166716337204,
      "rewards/format_reward": 0.90625,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 499.5625,
      "epoch": 0.13013698630136986,
      "grad_norm": 1.6726783514022827,
      "kl": 0.103271484375,
      "learning_rate": 9.5662100456621e-07,
      "loss": 0.0001,
      "reward": 1.2265625,
      "reward_std": 0.14807433634996414,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 1.0,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.5,
      "epoch": 0.13150684931506848,
      "grad_norm": 2.4275333881378174,
      "kl": 0.109130859375,
      "learning_rate": 9.561643835616437e-07,
      "loss": 0.0001,
      "reward": 1.4453125,
      "reward_std": 0.35531364381313324,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 0.96875,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.40625,
      "epoch": 0.13287671232876713,
      "grad_norm": 1.4377013444900513,
      "kl": 0.12249755859375,
      "learning_rate": 9.557077625570777e-07,
      "loss": 0.0001,
      "reward": 1.6145833432674408,
      "reward_std": 0.15587851032614708,
      "rewards/accuracy_reward": 0.6145833432674408,
      "rewards/format_reward": 1.0,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 697.5625,
      "epoch": 0.13424657534246576,
      "grad_norm": 2.026230573654175,
      "kl": 0.1431884765625,
      "learning_rate": 9.552511415525114e-07,
      "loss": 0.0001,
      "reward": 1.6875,
      "reward_std": 0.18185461685061455,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.28125,
      "epoch": 0.13561643835616438,
      "grad_norm": 4.95074462890625,
      "kl": 0.153564453125,
      "learning_rate": 9.547945205479452e-07,
      "loss": 0.0002,
      "reward": 2.0625,
      "reward_std": 0.3226073309779167,
      "rewards/accuracy_reward": 1.09375,
      "rewards/format_reward": 0.96875,
      "step": 99
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 694.40625,
      "epoch": 0.136986301369863,
      "grad_norm": 3.2093119621276855,
      "kl": 0.1156005859375,
      "learning_rate": 9.54337899543379e-07,
      "loss": 0.0001,
      "reward": 1.4140625,
      "reward_std": 0.18956539407372475,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 0.96875,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 465.21875,
      "epoch": 0.13835616438356163,
      "grad_norm": 3.3832147121429443,
      "kl": 0.0748291015625,
      "learning_rate": 9.538812785388126e-07,
      "loss": 0.0001,
      "reward": 1.2604166865348816,
      "reward_std": 0.24269168078899384,
      "rewards/accuracy_reward": 0.2604166567325592,
      "rewards/format_reward": 1.0,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 451.53125,
      "epoch": 0.13972602739726028,
      "grad_norm": 0.012525072321295738,
      "kl": 0.08740234375,
      "learning_rate": 9.534246575342465e-07,
      "loss": 0.0001,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 740.4375,
      "epoch": 0.1410958904109589,
      "grad_norm": 1.9927910566329956,
      "kl": 0.1253662109375,
      "learning_rate": 9.529680365296803e-07,
      "loss": 0.0001,
      "reward": 1.59375,
      "reward_std": 0.16675157472491264,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.375,
      "epoch": 0.14246575342465753,
      "grad_norm": 3.8206396102905273,
      "kl": 0.1165771484375,
      "learning_rate": 9.525114155251142e-07,
      "loss": 0.0001,
      "reward": 0.9375,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.9375,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 696.6875,
      "epoch": 0.14383561643835616,
      "grad_norm": 8.162276268005371,
      "kl": 0.134521484375,
      "learning_rate": 9.520547945205479e-07,
      "loss": 0.0001,
      "reward": 1.6171875,
      "reward_std": 0.19097032584249973,
      "rewards/accuracy_reward": 0.6171875,
      "rewards/format_reward": 1.0,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 656.78125,
      "epoch": 0.14520547945205478,
      "grad_norm": 2.644918441772461,
      "kl": 0.1160888671875,
      "learning_rate": 9.515981735159817e-07,
      "loss": 0.0001,
      "reward": 1.86328125,
      "reward_std": 0.41913160867989063,
      "rewards/accuracy_reward": 0.86328125,
      "rewards/format_reward": 1.0,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.90625,
      "epoch": 0.14657534246575343,
      "grad_norm": 3.6577141284942627,
      "kl": 0.1168212890625,
      "learning_rate": 9.511415525114155e-07,
      "loss": 0.0001,
      "reward": 1.359375,
      "reward_std": 0.19939782842993736,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.875,
      "epoch": 0.14794520547945206,
      "grad_norm": 4.23774528503418,
      "kl": 0.1524658203125,
      "learning_rate": 9.506849315068493e-07,
      "loss": 0.0002,
      "reward": 1.7565104365348816,
      "reward_std": 0.21211734786629677,
      "rewards/accuracy_reward": 0.7565104365348816,
      "rewards/format_reward": 1.0,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.34375,
      "epoch": 0.14931506849315068,
      "grad_norm": 0.013699422590434551,
      "kl": 0.1331787109375,
      "learning_rate": 9.50228310502283e-07,
      "loss": 0.0001,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 584.3125,
      "epoch": 0.1506849315068493,
      "grad_norm": 9.379846572875977,
      "kl": 0.1199951171875,
      "learning_rate": 9.497716894977168e-07,
      "loss": 0.0001,
      "reward": 1.640625,
      "reward_std": 0.36036762222647667,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 718.03125,
      "epoch": 0.15205479452054796,
      "grad_norm": 4.443063735961914,
      "kl": 0.126708984375,
      "learning_rate": 9.493150684931507e-07,
      "loss": 0.0001,
      "reward": 1.4427083134651184,
      "reward_std": 0.17040568217635155,
      "rewards/accuracy_reward": 0.4739583283662796,
      "rewards/format_reward": 0.96875,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 735.9375,
      "epoch": 0.15342465753424658,
      "grad_norm": 2.7463929653167725,
      "kl": 0.1220703125,
      "learning_rate": 9.488584474885845e-07,
      "loss": 0.0001,
      "reward": 1.46875,
      "reward_std": 0.47249409183859825,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 0.875,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 444.25,
      "epoch": 0.1547945205479452,
      "grad_norm": 3.2011280059814453,
      "kl": 0.0909423828125,
      "learning_rate": 9.484018264840182e-07,
      "loss": 0.0001,
      "reward": 1.15625,
      "reward_std": 0.3061639815568924,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 504.8125,
      "epoch": 0.15616438356164383,
      "grad_norm": 3.5561389923095703,
      "kl": 0.10040283203125,
      "learning_rate": 9.47945205479452e-07,
      "loss": 0.0001,
      "reward": 1.203125,
      "reward_std": 0.15992168709635735,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 1.0,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 519.15625,
      "epoch": 0.15753424657534246,
      "grad_norm": 1.8714385032653809,
      "kl": 0.100341796875,
      "learning_rate": 9.474885844748858e-07,
      "loss": 0.0001,
      "reward": 1.265625,
      "reward_std": 0.19408093392848969,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 411.625,
      "epoch": 0.1589041095890411,
      "grad_norm": 6.825204849243164,
      "kl": 0.086181640625,
      "learning_rate": 9.470319634703196e-07,
      "loss": 0.0001,
      "reward": 1.21875,
      "reward_std": 0.2756393924355507,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.34375,
      "epoch": 0.16027397260273973,
      "grad_norm": 1.6678544282913208,
      "kl": 0.112060546875,
      "learning_rate": 9.465753424657534e-07,
      "loss": 0.0001,
      "reward": 1.3333333134651184,
      "reward_std": 0.1900147907435894,
      "rewards/accuracy_reward": 0.3645833134651184,
      "rewards/format_reward": 0.96875,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 418.1875,
      "epoch": 0.16164383561643836,
      "grad_norm": 2.2251501083374023,
      "kl": 0.0775146484375,
      "learning_rate": 9.461187214611872e-07,
      "loss": 0.0001,
      "reward": 1.1953125,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.1953125,
      "rewards/format_reward": 1.0,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.8125,
      "epoch": 0.16301369863013698,
      "grad_norm": 1.6605937480926514,
      "kl": 0.1004638671875,
      "learning_rate": 9.45662100456621e-07,
      "loss": 0.0001,
      "reward": 1.6875,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 488.3125,
      "epoch": 0.1643835616438356,
      "grad_norm": 10.110955238342285,
      "kl": 0.1434326171875,
      "learning_rate": 9.452054794520548e-07,
      "loss": 0.0001,
      "reward": 1.171875,
      "reward_std": 0.13258251920342445,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 670.28125,
      "epoch": 0.16575342465753426,
      "grad_norm": 1.275918960571289,
      "kl": 0.1370849609375,
      "learning_rate": 9.447488584474885e-07,
      "loss": 0.0001,
      "reward": 1.1302083432674408,
      "reward_std": 0.014731401577591896,
      "rewards/accuracy_reward": 0.1302083283662796,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.28125,
      "epoch": 0.16712328767123288,
      "grad_norm": 2.001127004623413,
      "kl": 0.1129150390625,
      "learning_rate": 9.442922374429223e-07,
      "loss": 0.0001,
      "reward": 1.3125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 420.59375,
      "epoch": 0.1684931506849315,
      "grad_norm": 4.077696800231934,
      "kl": 0.0904541015625,
      "learning_rate": 9.438356164383561e-07,
      "loss": 0.0001,
      "reward": 1.375,
      "reward_std": 0.2619796171784401,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 491.8125,
      "epoch": 0.16986301369863013,
      "grad_norm": 4.269510269165039,
      "kl": 0.09326171875,
      "learning_rate": 9.4337899543379e-07,
      "loss": 0.0001,
      "reward": 1.390625,
      "reward_std": 0.2665942460298538,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.78125,
      "epoch": 0.17123287671232876,
      "grad_norm": 3.122952938079834,
      "kl": 0.1368408203125,
      "learning_rate": 9.429223744292237e-07,
      "loss": 0.0001,
      "reward": 1.453125,
      "reward_std": 0.16415906324982643,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 522.25,
      "epoch": 0.1726027397260274,
      "grad_norm": 1.53658926486969,
      "kl": 0.129150390625,
      "learning_rate": 9.424657534246575e-07,
      "loss": 0.0001,
      "reward": 1.66015625,
      "reward_std": 0.19974715635180473,
      "rewards/accuracy_reward": 0.66015625,
      "rewards/format_reward": 1.0,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 548.1875,
      "epoch": 0.17397260273972603,
      "grad_norm": 2.161501884460449,
      "kl": 0.117919921875,
      "learning_rate": 9.420091324200913e-07,
      "loss": 0.0001,
      "reward": 1.25,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 301.65625,
      "epoch": 0.17534246575342466,
      "grad_norm": 2.8414783477783203,
      "kl": 0.0875244140625,
      "learning_rate": 9.41552511415525e-07,
      "loss": 0.0001,
      "reward": 1.5625,
      "reward_std": 0.3335031494498253,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.3125,
      "epoch": 0.17671232876712328,
      "grad_norm": 2.295241355895996,
      "kl": 0.1317138671875,
      "learning_rate": 9.410958904109588e-07,
      "loss": 0.0001,
      "reward": 1.6302083134651184,
      "reward_std": 0.18143897131085396,
      "rewards/accuracy_reward": 0.6302083283662796,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.71875,
      "epoch": 0.1780821917808219,
      "grad_norm": 2.436352491378784,
      "kl": 0.13232421875,
      "learning_rate": 9.406392694063926e-07,
      "loss": 0.0001,
      "reward": 1.7109375,
      "reward_std": 0.2758216764777899,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 1.0,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.25,
      "epoch": 0.17945205479452056,
      "grad_norm": 3.0529682636260986,
      "kl": 0.13720703125,
      "learning_rate": 9.401826484018265e-07,
      "loss": 0.0001,
      "reward": 1.50390625,
      "reward_std": 0.07282309234142303,
      "rewards/accuracy_reward": 0.50390625,
      "rewards/format_reward": 1.0,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 483.15625,
      "epoch": 0.18082191780821918,
      "grad_norm": 1.2933306694030762,
      "kl": 0.123291015625,
      "learning_rate": 9.397260273972603e-07,
      "loss": 0.0001,
      "reward": 1.09375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.09375,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.71875,
      "epoch": 0.1821917808219178,
      "grad_norm": 1.558491826057434,
      "kl": 0.11572265625,
      "learning_rate": 9.39269406392694e-07,
      "loss": 0.0001,
      "reward": 1.5677083432674408,
      "reward_std": 0.0725951585918665,
      "rewards/accuracy_reward": 0.5677083432674408,
      "rewards/format_reward": 1.0,
      "step": 133
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 488.78125,
      "epoch": 0.18356164383561643,
      "grad_norm": 2.861250400543213,
      "kl": 0.113525390625,
      "learning_rate": 9.388127853881278e-07,
      "loss": 0.0001,
      "reward": 1.375,
      "reward_std": 0.2966036908328533,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.96875,
      "step": 134
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.28125,
      "epoch": 0.18493150684931506,
      "grad_norm": 2.4764702320098877,
      "kl": 0.123291015625,
      "learning_rate": 9.383561643835616e-07,
      "loss": 0.0001,
      "reward": 1.5825892686843872,
      "reward_std": 0.20715469866991043,
      "rewards/accuracy_reward": 0.5825892686843872,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.53125,
      "epoch": 0.1863013698630137,
      "grad_norm": 2.0686910152435303,
      "kl": 0.1162109375,
      "learning_rate": 9.378995433789953e-07,
      "loss": 0.0001,
      "reward": 1.40234375,
      "reward_std": 0.21405773982405663,
      "rewards/accuracy_reward": 0.40234375,
      "rewards/format_reward": 1.0,
      "step": 136
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.84375,
      "epoch": 0.18767123287671234,
      "grad_norm": 1.7358382940292358,
      "kl": 0.1298828125,
      "learning_rate": 9.374429223744292e-07,
      "loss": 0.0001,
      "reward": 1.6507812142372131,
      "reward_std": 0.2213343046605587,
      "rewards/accuracy_reward": 0.6820312440395355,
      "rewards/format_reward": 0.96875,
      "step": 137
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 395.5,
      "epoch": 0.18904109589041096,
      "grad_norm": 2.3077900409698486,
      "kl": 0.114501953125,
      "learning_rate": 9.36986301369863e-07,
      "loss": 0.0001,
      "reward": 1.2265625,
      "reward_std": 0.22621294669806957,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 1.0,
      "step": 138
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.59375,
      "epoch": 0.19041095890410958,
      "grad_norm": 0.7606338262557983,
      "kl": 0.135498046875,
      "learning_rate": 9.365296803652968e-07,
      "loss": 0.0001,
      "reward": 1.3333333134651184,
      "reward_std": 0.07042950391769409,
      "rewards/accuracy_reward": 0.3333333432674408,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.9375,
      "epoch": 0.1917808219178082,
      "grad_norm": 3.7832796573638916,
      "kl": 0.1358642578125,
      "learning_rate": 9.360730593607306e-07,
      "loss": 0.0001,
      "reward": 1.526041716337204,
      "reward_std": 0.2930229790508747,
      "rewards/accuracy_reward": 0.5885416567325592,
      "rewards/format_reward": 0.9375,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 681.34375,
      "epoch": 0.19315068493150686,
      "grad_norm": 5.2146830558776855,
      "kl": 0.1507568359375,
      "learning_rate": 9.356164383561643e-07,
      "loss": 0.0002,
      "reward": 1.734375,
      "reward_std": 0.16887323930859566,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 141
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.59375,
      "epoch": 0.19452054794520549,
      "grad_norm": 1.6805075407028198,
      "kl": 0.158935546875,
      "learning_rate": 9.351598173515981e-07,
      "loss": 0.0002,
      "reward": 1.3515625,
      "reward_std": 0.07996084354817867,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 453.8125,
      "epoch": 0.1958904109589041,
      "grad_norm": 2.2252542972564697,
      "kl": 0.130615234375,
      "learning_rate": 9.347031963470319e-07,
      "loss": 0.0001,
      "reward": 1.28125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 143
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.96875,
      "epoch": 0.19726027397260273,
      "grad_norm": 2.0735678672790527,
      "kl": 0.164306640625,
      "learning_rate": 9.342465753424658e-07,
      "loss": 0.0002,
      "reward": 1.4791666567325592,
      "reward_std": 0.23902175202965736,
      "rewards/accuracy_reward": 0.4791666567325592,
      "rewards/format_reward": 1.0,
      "step": 144
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.96875,
      "epoch": 0.19863013698630136,
      "grad_norm": 1.4199141263961792,
      "kl": 0.1151123046875,
      "learning_rate": 9.337899543378995e-07,
      "loss": 0.0001,
      "reward": 1.328125,
      "reward_std": 0.220042884349823,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 446.3125,
      "epoch": 0.2,
      "grad_norm": 4.550744533538818,
      "kl": 0.13037109375,
      "learning_rate": 9.333333333333333e-07,
      "loss": 0.0001,
      "reward": 1.53125,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 146
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.625,
      "epoch": 0.20136986301369864,
      "grad_norm": 0.7138487100601196,
      "kl": 0.1424560546875,
      "learning_rate": 9.328767123287671e-07,
      "loss": 0.0001,
      "reward": 1.15625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 147
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.65625,
      "epoch": 0.20273972602739726,
      "grad_norm": 5.086813449859619,
      "kl": 0.1173095703125,
      "learning_rate": 9.324200913242009e-07,
      "loss": 0.0001,
      "reward": 1.4609375,
      "reward_std": 0.05476716160774231,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 1.0,
      "step": 148
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.65625,
      "epoch": 0.20410958904109588,
      "grad_norm": 0.8646766543388367,
      "kl": 0.1317138671875,
      "learning_rate": 9.319634703196346e-07,
      "loss": 0.0001,
      "reward": 1.5104166567325592,
      "reward_std": 0.32082508504390717,
      "rewards/accuracy_reward": 0.5416666865348816,
      "rewards/format_reward": 0.96875,
      "step": 149
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 396.40625,
      "epoch": 0.2054794520547945,
      "grad_norm": 2.680182456970215,
      "kl": 0.12353515625,
      "learning_rate": 9.315068493150684e-07,
      "loss": 0.0001,
      "reward": 1.40625,
      "reward_std": 0.4045617878437042,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.90625,
      "epoch": 0.20684931506849316,
      "grad_norm": 1.8212926387786865,
      "kl": 0.108642578125,
      "learning_rate": 9.310502283105023e-07,
      "loss": 0.0001,
      "reward": 1.3177083432674408,
      "reward_std": 0.13903126679360867,
      "rewards/accuracy_reward": 0.3177083432674408,
      "rewards/format_reward": 1.0,
      "step": 151
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 510.625,
      "epoch": 0.20821917808219179,
      "grad_norm": 2.6356329917907715,
      "kl": 0.1246337890625,
      "learning_rate": 9.30593607305936e-07,
      "loss": 0.0001,
      "reward": 1.4812500178813934,
      "reward_std": 0.45198121294379234,
      "rewards/accuracy_reward": 0.512499988079071,
      "rewards/format_reward": 0.96875,
      "step": 152
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 449.8125,
      "epoch": 0.2095890410958904,
      "grad_norm": 3.026435613632202,
      "kl": 0.1207275390625,
      "learning_rate": 9.301369863013698e-07,
      "loss": 0.0001,
      "reward": 1.3125,
      "reward_std": 0.3230287954211235,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 153
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.15625,
      "epoch": 0.21095890410958903,
      "grad_norm": 0.8813155293464661,
      "kl": 0.1131591796875,
      "learning_rate": 9.296803652968036e-07,
      "loss": 0.0001,
      "reward": 1.3294270634651184,
      "reward_std": 0.18091023340821266,
      "rewards/accuracy_reward": 0.3606770634651184,
      "rewards/format_reward": 0.96875,
      "step": 154
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.59375,
      "epoch": 0.21232876712328766,
      "grad_norm": 2.931427001953125,
      "kl": 0.1202392578125,
      "learning_rate": 9.292237442922374e-07,
      "loss": 0.0001,
      "reward": 1.515625,
      "reward_std": 0.17358146235346794,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 426.25,
      "epoch": 0.2136986301369863,
      "grad_norm": 1.421940803527832,
      "kl": 0.1243896484375,
      "learning_rate": 9.287671232876712e-07,
      "loss": 0.0001,
      "reward": 1.125,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 1.0,
      "step": 156
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.15625,
      "epoch": 0.21506849315068494,
      "grad_norm": 0.9390948414802551,
      "kl": 0.134521484375,
      "learning_rate": 9.28310502283105e-07,
      "loss": 0.0001,
      "reward": 1.35546875,
      "reward_std": 0.06765139661729336,
      "rewards/accuracy_reward": 0.35546875,
      "rewards/format_reward": 1.0,
      "step": 157
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 721.1875,
      "epoch": 0.21643835616438356,
      "grad_norm": 1.8904517889022827,
      "kl": 0.130615234375,
      "learning_rate": 9.278538812785388e-07,
      "loss": 0.0001,
      "reward": 1.3515625,
      "reward_std": 0.10474801808595657,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 1.0,
      "step": 158
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.71875,
      "epoch": 0.21780821917808219,
      "grad_norm": 6.320685386657715,
      "kl": 0.1202392578125,
      "learning_rate": 9.273972602739726e-07,
      "loss": 0.0001,
      "reward": 1.4107142686843872,
      "reward_std": 0.2957366779446602,
      "rewards/accuracy_reward": 0.4107142984867096,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.5625,
      "epoch": 0.2191780821917808,
      "grad_norm": 0.6069585680961609,
      "kl": 0.1357421875,
      "learning_rate": 9.269406392694063e-07,
      "loss": 0.0001,
      "reward": 1.1875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 366.25,
      "epoch": 0.22054794520547946,
      "grad_norm": 3.5432889461517334,
      "kl": 0.18798828125,
      "learning_rate": 9.264840182648401e-07,
      "loss": 0.0002,
      "reward": 1.5,
      "reward_std": 0.2314550280570984,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 161
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.9375,
      "epoch": 0.2219178082191781,
      "grad_norm": 2.56186580657959,
      "kl": 0.149169921875,
      "learning_rate": 9.260273972602739e-07,
      "loss": 0.0001,
      "reward": 1.6875,
      "reward_std": 0.1825428232550621,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 162
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.8125,
      "epoch": 0.2232876712328767,
      "grad_norm": 5.973612308502197,
      "kl": 0.14306640625,
      "learning_rate": 9.255707762557077e-07,
      "loss": 0.0001,
      "reward": 1.359375,
      "reward_std": 0.19939782842993736,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 163
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 401.125,
      "epoch": 0.22465753424657534,
      "grad_norm": 1.2704259157180786,
      "kl": 0.14794921875,
      "learning_rate": 9.251141552511416e-07,
      "loss": 0.0001,
      "reward": 1.1796875,
      "reward_std": 0.04005437344312668,
      "rewards/accuracy_reward": 0.1796875,
      "rewards/format_reward": 1.0,
      "step": 164
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.28125,
      "epoch": 0.22602739726027396,
      "grad_norm": 2.514188289642334,
      "kl": 0.126220703125,
      "learning_rate": 9.246575342465753e-07,
      "loss": 0.0001,
      "reward": 1.515625,
      "reward_std": 0.2561880201101303,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 701.75,
      "epoch": 0.2273972602739726,
      "grad_norm": 1.1485939025878906,
      "kl": 0.116943359375,
      "learning_rate": 9.242009132420091e-07,
      "loss": 0.0001,
      "reward": 1.265625,
      "reward_std": 0.2461063265800476,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 0.9375,
      "step": 166
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.125,
      "epoch": 0.22876712328767124,
      "grad_norm": 16.20563316345215,
      "kl": 0.132568359375,
      "learning_rate": 9.237442922374429e-07,
      "loss": 0.0001,
      "reward": 1.550000011920929,
      "reward_std": 0.15476782992482185,
      "rewards/accuracy_reward": 0.5499999970197678,
      "rewards/format_reward": 1.0,
      "step": 167
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 476.78125,
      "epoch": 0.23013698630136986,
      "grad_norm": 0.7318564057350159,
      "kl": 0.146728515625,
      "learning_rate": 9.232876712328766e-07,
      "loss": 0.0001,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 168
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.28125,
      "epoch": 0.23150684931506849,
      "grad_norm": 3.2940330505371094,
      "kl": 0.136962890625,
      "learning_rate": 9.228310502283104e-07,
      "loss": 0.0001,
      "reward": 1.5013020634651184,
      "reward_std": 0.0952342264354229,
      "rewards/accuracy_reward": 0.5013020783662796,
      "rewards/format_reward": 1.0,
      "step": 169
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 475.6875,
      "epoch": 0.2328767123287671,
      "grad_norm": 6.060558795928955,
      "kl": 0.1484375,
      "learning_rate": 9.223744292237442e-07,
      "loss": 0.0001,
      "reward": 1.4140625,
      "reward_std": 0.2114126794040203,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.96875,
      "epoch": 0.23424657534246576,
      "grad_norm": 2.6800572872161865,
      "kl": 0.1197509765625,
      "learning_rate": 9.219178082191781e-07,
      "loss": 0.0001,
      "reward": 1.4947916567325592,
      "reward_std": 0.1304523590952158,
      "rewards/accuracy_reward": 0.4947916567325592,
      "rewards/format_reward": 1.0,
      "step": 171
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.0625,
      "epoch": 0.2356164383561644,
      "grad_norm": 1.535984992980957,
      "kl": 0.2501220703125,
      "learning_rate": 9.214611872146119e-07,
      "loss": 0.0002,
      "reward": 1.28125,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 172
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 520.125,
      "epoch": 0.236986301369863,
      "grad_norm": 1.8922423124313354,
      "kl": 0.1363525390625,
      "learning_rate": 9.210045662100456e-07,
      "loss": 0.0001,
      "reward": 1.3671875,
      "reward_std": 0.16834918968379498,
      "rewards/accuracy_reward": 0.3671875,
      "rewards/format_reward": 1.0,
      "step": 173
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 489.4375,
      "epoch": 0.23835616438356164,
      "grad_norm": 4.069134712219238,
      "kl": 0.154052734375,
      "learning_rate": 9.205479452054794e-07,
      "loss": 0.0002,
      "reward": 1.5,
      "reward_std": 0.33614395931363106,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 174
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.1875,
      "epoch": 0.23972602739726026,
      "grad_norm": 3.013641119003296,
      "kl": 0.1435546875,
      "learning_rate": 9.200913242009132e-07,
      "loss": 0.0001,
      "reward": 1.21875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.9375,
      "epoch": 0.2410958904109589,
      "grad_norm": 1.713585376739502,
      "kl": 0.1307373046875,
      "learning_rate": 9.196347031963469e-07,
      "loss": 0.0001,
      "reward": 1.6640625,
      "reward_std": 0.202580526471138,
      "rewards/accuracy_reward": 0.6640625,
      "rewards/format_reward": 1.0,
      "step": 176
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 458.90625,
      "epoch": 0.24246575342465754,
      "grad_norm": 1.653497576713562,
      "kl": 0.133544921875,
      "learning_rate": 9.191780821917808e-07,
      "loss": 0.0001,
      "reward": 1.3828125,
      "reward_std": 0.23987272381782532,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 177
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 302.875,
      "epoch": 0.24383561643835616,
      "grad_norm": 1.537979006767273,
      "kl": 0.167236328125,
      "learning_rate": 9.187214611872146e-07,
      "loss": 0.0002,
      "reward": 1.0625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 1.0,
      "step": 178
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.28125,
      "epoch": 0.2452054794520548,
      "grad_norm": 1.1757584810256958,
      "kl": 0.146484375,
      "learning_rate": 9.182648401826484e-07,
      "loss": 0.0001,
      "reward": 1.6953125,
      "reward_std": 0.08516896516084671,
      "rewards/accuracy_reward": 0.6953125,
      "rewards/format_reward": 1.0,
      "step": 179
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.3125,
      "epoch": 0.2465753424657534,
      "grad_norm": 2.4092350006103516,
      "kl": 0.143310546875,
      "learning_rate": 9.178082191780822e-07,
      "loss": 0.0001,
      "reward": 1.4921875,
      "reward_std": 0.22854942083358765,
      "rewards/accuracy_reward": 0.4921875,
      "rewards/format_reward": 1.0,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.53125,
      "epoch": 0.24794520547945206,
      "grad_norm": 2.0598623752593994,
      "kl": 0.1513671875,
      "learning_rate": 9.173515981735159e-07,
      "loss": 0.0002,
      "reward": 1.3125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 181
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.84375,
      "epoch": 0.2493150684931507,
      "grad_norm": 3.4757208824157715,
      "kl": 0.15380859375,
      "learning_rate": 9.168949771689497e-07,
      "loss": 0.0002,
      "reward": 1.4322916567325592,
      "reward_std": 0.11389755457639694,
      "rewards/accuracy_reward": 0.4322916716337204,
      "rewards/format_reward": 1.0,
      "step": 182
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.53125,
      "epoch": 0.25068493150684934,
      "grad_norm": 1.2352172136306763,
      "kl": 0.1416015625,
      "learning_rate": 9.164383561643835e-07,
      "loss": 0.0001,
      "reward": 1.7265625,
      "reward_std": 0.14262642711400986,
      "rewards/accuracy_reward": 0.7265625,
      "rewards/format_reward": 1.0,
      "step": 183
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.46875,
      "epoch": 0.25205479452054796,
      "grad_norm": 2.0595240592956543,
      "kl": 0.143798828125,
      "learning_rate": 9.159817351598174e-07,
      "loss": 0.0001,
      "reward": 1.671875,
      "reward_std": 0.3808670938014984,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 184
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.75,
      "epoch": 0.2534246575342466,
      "grad_norm": 2.4244847297668457,
      "kl": 0.14208984375,
      "learning_rate": 9.155251141552511e-07,
      "loss": 0.0001,
      "reward": 1.3046875,
      "reward_std": 0.04339781776070595,
      "rewards/accuracy_reward": 0.3046875,
      "rewards/format_reward": 1.0,
      "step": 185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 411.1875,
      "epoch": 0.2547945205479452,
      "grad_norm": 2.7618906497955322,
      "kl": 0.138671875,
      "learning_rate": 9.150684931506849e-07,
      "loss": 0.0001,
      "reward": 1.3828125,
      "reward_std": 0.13782460056245327,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 186
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.375,
      "epoch": 0.25616438356164384,
      "grad_norm": 1.3447990417480469,
      "kl": 0.16015625,
      "learning_rate": 9.146118721461187e-07,
      "loss": 0.0002,
      "reward": 1.2942708432674408,
      "reward_std": 0.11083479970693588,
      "rewards/accuracy_reward": 0.2942708283662796,
      "rewards/format_reward": 1.0,
      "step": 187
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 482.59375,
      "epoch": 0.25753424657534246,
      "grad_norm": 1.3052458763122559,
      "kl": 0.17041015625,
      "learning_rate": 9.141552511415525e-07,
      "loss": 0.0002,
      "reward": 1.40625,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 188
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.96875,
      "epoch": 0.2589041095890411,
      "grad_norm": 1.417614221572876,
      "kl": 0.1513671875,
      "learning_rate": 9.136986301369862e-07,
      "loss": 0.0002,
      "reward": 1.2109375,
      "reward_std": 0.17315101623535156,
      "rewards/accuracy_reward": 0.2109375,
      "rewards/format_reward": 1.0,
      "step": 189
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.75,
      "epoch": 0.2602739726027397,
      "grad_norm": 1.8264148235321045,
      "kl": 0.179931640625,
      "learning_rate": 9.1324200913242e-07,
      "loss": 0.0002,
      "reward": 1.359375,
      "reward_std": 0.09143973141908646,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 398.78125,
      "epoch": 0.26164383561643834,
      "grad_norm": 2.271404266357422,
      "kl": 0.276123046875,
      "learning_rate": 9.127853881278539e-07,
      "loss": 0.0003,
      "reward": 1.5,
      "reward_std": 0.1293042004108429,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 191
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 579.90625,
      "epoch": 0.26301369863013696,
      "grad_norm": 1.1621103286743164,
      "kl": 0.146484375,
      "learning_rate": 9.123287671232876e-07,
      "loss": 0.0001,
      "reward": 1.34375,
      "reward_std": 0.10888781771063805,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 192
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 418.71875,
      "epoch": 0.26438356164383564,
      "grad_norm": 2.4608054161071777,
      "kl": 0.1552734375,
      "learning_rate": 9.118721461187214e-07,
      "loss": 0.0002,
      "reward": 1.84375,
      "reward_std": 0.2704022154211998,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 193
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 422.65625,
      "epoch": 0.26575342465753427,
      "grad_norm": 2.2332265377044678,
      "kl": 0.152587890625,
      "learning_rate": 9.114155251141552e-07,
      "loss": 0.0002,
      "reward": 1.5546875,
      "reward_std": 0.27746163308620453,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 1.0,
      "step": 194
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 527.59375,
      "epoch": 0.2671232876712329,
      "grad_norm": 1.5108232498168945,
      "kl": 0.13525390625,
      "learning_rate": 9.10958904109589e-07,
      "loss": 0.0001,
      "reward": 1.375,
      "reward_std": 0.115727499127388,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 195
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 435.625,
      "epoch": 0.2684931506849315,
      "grad_norm": 4.214125633239746,
      "kl": 0.14013671875,
      "learning_rate": 9.105022831050228e-07,
      "loss": 0.0001,
      "reward": 1.3984375,
      "reward_std": 0.18013210594654083,
      "rewards/accuracy_reward": 0.3984375,
      "rewards/format_reward": 1.0,
      "step": 196
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.3125,
      "epoch": 0.26986301369863014,
      "grad_norm": 5.064935207366943,
      "kl": 0.165771484375,
      "learning_rate": 9.100456621004566e-07,
      "loss": 0.0002,
      "reward": 1.7109375,
      "reward_std": 0.27000918984413147,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 1.0,
      "step": 197
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.4375,
      "epoch": 0.27123287671232876,
      "grad_norm": 1.5454697608947754,
      "kl": 0.18994140625,
      "learning_rate": 9.095890410958904e-07,
      "loss": 0.0002,
      "reward": 1.4296875,
      "reward_std": 0.13488983362913132,
      "rewards/accuracy_reward": 0.4296875,
      "rewards/format_reward": 1.0,
      "step": 198
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 391.34375,
      "epoch": 0.2726027397260274,
      "grad_norm": 4.849545478820801,
      "kl": 0.1414794921875,
      "learning_rate": 9.091324200913242e-07,
      "loss": 0.0001,
      "reward": 1.234375,
      "reward_std": 0.432646207511425,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 0.9375,
      "step": 199
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 587.40625,
      "epoch": 0.273972602739726,
      "grad_norm": 1.7864863872528076,
      "kl": 0.173583984375,
      "learning_rate": 9.08675799086758e-07,
      "loss": 0.0002,
      "reward": 1.53125,
      "reward_std": 0.14913516864180565,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.65625,
      "epoch": 0.27534246575342464,
      "grad_norm": 1.9118916988372803,
      "kl": 0.177734375,
      "learning_rate": 9.082191780821917e-07,
      "loss": 0.0002,
      "reward": 1.640625,
      "reward_std": 0.35035815089941025,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 201
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.84375,
      "epoch": 0.27671232876712326,
      "grad_norm": 3.346449613571167,
      "kl": 0.16943359375,
      "learning_rate": 9.077625570776255e-07,
      "loss": 0.0002,
      "reward": 1.875,
      "reward_std": 0.4419417232275009,
      "rewards/accuracy_reward": 0.90625,
      "rewards/format_reward": 0.96875,
      "step": 202
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.90625,
      "epoch": 0.27808219178082194,
      "grad_norm": 2.130422830581665,
      "kl": 0.161865234375,
      "learning_rate": 9.073059360730593e-07,
      "loss": 0.0002,
      "reward": 1.2890625,
      "reward_std": 0.27090023458004,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 0.875,
      "step": 203
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.65625,
      "epoch": 0.27945205479452057,
      "grad_norm": 2.6185011863708496,
      "kl": 0.1533203125,
      "learning_rate": 9.068493150684932e-07,
      "loss": 0.0002,
      "reward": 1.625,
      "reward_std": 0.18702642805874348,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 204
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 430.5625,
      "epoch": 0.2808219178082192,
      "grad_norm": 5.361741065979004,
      "kl": 0.174072265625,
      "learning_rate": 9.063926940639269e-07,
      "loss": 0.0002,
      "reward": 1.34375,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 417.5625,
      "epoch": 0.2821917808219178,
      "grad_norm": 1.3938766717910767,
      "kl": 0.155517578125,
      "learning_rate": 9.059360730593607e-07,
      "loss": 0.0002,
      "reward": 1.453125,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 206
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 437.875,
      "epoch": 0.28356164383561644,
      "grad_norm": 1.8740441799163818,
      "kl": 0.191162109375,
      "learning_rate": 9.054794520547945e-07,
      "loss": 0.0002,
      "reward": 1.3125,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 207
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 438.59375,
      "epoch": 0.28493150684931506,
      "grad_norm": 10.041444778442383,
      "kl": 0.17919921875,
      "learning_rate": 9.050228310502282e-07,
      "loss": 0.0002,
      "reward": 1.4453125,
      "reward_std": 0.26258746162056923,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 1.0,
      "step": 208
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 443.15625,
      "epoch": 0.2863013698630137,
      "grad_norm": 2.726414203643799,
      "kl": 0.166015625,
      "learning_rate": 9.04566210045662e-07,
      "loss": 0.0002,
      "reward": 1.453125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 209
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 337.84375,
      "epoch": 0.2876712328767123,
      "grad_norm": 1.979183554649353,
      "kl": 0.167724609375,
      "learning_rate": 9.041095890410958e-07,
      "loss": 0.0002,
      "reward": 1.28125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 512.84375,
      "epoch": 0.28904109589041094,
      "grad_norm": 1.7484036684036255,
      "kl": 0.17041015625,
      "learning_rate": 9.036529680365297e-07,
      "loss": 0.0002,
      "reward": 1.453125,
      "reward_std": 0.15992168709635735,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 211
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 473.84375,
      "epoch": 0.29041095890410956,
      "grad_norm": 0.7065235376358032,
      "kl": 0.157958984375,
      "learning_rate": 9.031963470319635e-07,
      "loss": 0.0002,
      "reward": 1.1953125,
      "reward_std": 0.09704047441482544,
      "rewards/accuracy_reward": 0.1953125,
      "rewards/format_reward": 1.0,
      "step": 212
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.21875,
      "epoch": 0.29178082191780824,
      "grad_norm": 2.361640453338623,
      "kl": 0.18505859375,
      "learning_rate": 9.027397260273972e-07,
      "loss": 0.0002,
      "reward": 1.6070312559604645,
      "reward_std": 0.23584069684147835,
      "rewards/accuracy_reward": 0.6070312410593033,
      "rewards/format_reward": 1.0,
      "step": 213
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.40625,
      "epoch": 0.29315068493150687,
      "grad_norm": 2.628955125808716,
      "kl": 0.17041015625,
      "learning_rate": 9.02283105022831e-07,
      "loss": 0.0002,
      "reward": 1.42578125,
      "reward_std": 0.028628919273614883,
      "rewards/accuracy_reward": 0.42578125,
      "rewards/format_reward": 1.0,
      "step": 214
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 460.65625,
      "epoch": 0.2945205479452055,
      "grad_norm": 5.236007213592529,
      "kl": 0.173828125,
      "learning_rate": 9.018264840182648e-07,
      "loss": 0.0002,
      "reward": 1.643750011920929,
      "reward_std": 0.30251236632466316,
      "rewards/accuracy_reward": 0.643750011920929,
      "rewards/format_reward": 1.0,
      "step": 215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 440.25,
      "epoch": 0.2958904109589041,
      "grad_norm": 3.1358797550201416,
      "kl": 0.176513671875,
      "learning_rate": 9.013698630136985e-07,
      "loss": 0.0002,
      "reward": 1.4453125,
      "reward_std": 0.22854942083358765,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 1.0,
      "step": 216
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 446.75,
      "epoch": 0.29726027397260274,
      "grad_norm": 1.5145561695098877,
      "kl": 0.185791015625,
      "learning_rate": 9.009132420091324e-07,
      "loss": 0.0002,
      "reward": 1.19140625,
      "reward_std": 0.11935807205736637,
      "rewards/accuracy_reward": 0.19140625,
      "rewards/format_reward": 1.0,
      "step": 217
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 377.40625,
      "epoch": 0.29863013698630136,
      "grad_norm": 1.9805744886398315,
      "kl": 0.181884765625,
      "learning_rate": 9.004566210045662e-07,
      "loss": 0.0002,
      "reward": 1.28125,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 218
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 429.84375,
      "epoch": 0.3,
      "grad_norm": 8.512849807739258,
      "kl": 0.184814453125,
      "learning_rate": 9e-07,
      "loss": 0.0002,
      "reward": 1.53125,
      "reward_std": 0.22301281243562698,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 219
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 415.09375,
      "epoch": 0.3013698630136986,
      "grad_norm": 0.7778679132461548,
      "kl": 0.225341796875,
      "learning_rate": 8.995433789954338e-07,
      "loss": 0.0002,
      "reward": 1.34375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.75,
      "epoch": 0.30273972602739724,
      "grad_norm": 6.034634590148926,
      "kl": 0.19775390625,
      "learning_rate": 8.990867579908675e-07,
      "loss": 0.0002,
      "reward": 1.375,
      "reward_std": 0.16279494389891624,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 221
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.40625,
      "epoch": 0.3041095890410959,
      "grad_norm": 0.8172019720077515,
      "kl": 0.258544921875,
      "learning_rate": 8.986301369863013e-07,
      "loss": 0.0003,
      "reward": 1.578125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 222
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 468.28125,
      "epoch": 0.30547945205479454,
      "grad_norm": 2.73528790473938,
      "kl": 0.19140625,
      "learning_rate": 8.981735159817351e-07,
      "loss": 0.0002,
      "reward": 1.3645833432674408,
      "reward_std": 0.2686460316181183,
      "rewards/accuracy_reward": 0.3645833283662796,
      "rewards/format_reward": 1.0,
      "step": 223
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 475.09375,
      "epoch": 0.30684931506849317,
      "grad_norm": 0.015807894989848137,
      "kl": 0.178466796875,
      "learning_rate": 8.97716894977169e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 224
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.03125,
      "epoch": 0.3082191780821918,
      "grad_norm": 1.2931923866271973,
      "kl": 0.155029296875,
      "learning_rate": 8.972602739726027e-07,
      "loss": 0.0002,
      "reward": 1.3515625,
      "reward_std": 0.2547192648053169,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 0.96875,
      "step": 225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.40625,
      "epoch": 0.3095890410958904,
      "grad_norm": 2.7091639041900635,
      "kl": 0.183349609375,
      "learning_rate": 8.968036529680365e-07,
      "loss": 0.0002,
      "reward": 1.390625,
      "reward_std": 0.2109457477927208,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 226
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 497.9375,
      "epoch": 0.31095890410958904,
      "grad_norm": 1.3506500720977783,
      "kl": 0.1748046875,
      "learning_rate": 8.963470319634703e-07,
      "loss": 0.0002,
      "reward": 1.09375,
      "reward_std": 0.22201896458864212,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.96875,
      "step": 227
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 377.8125,
      "epoch": 0.31232876712328766,
      "grad_norm": 1.4593608379364014,
      "kl": 0.183349609375,
      "learning_rate": 8.958904109589041e-07,
      "loss": 0.0002,
      "reward": 1.34375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 228
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.3125,
      "epoch": 0.3136986301369863,
      "grad_norm": 0.8097538948059082,
      "kl": 0.159423828125,
      "learning_rate": 8.954337899543378e-07,
      "loss": 0.0002,
      "reward": 1.1875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 229
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.09375,
      "epoch": 0.3150684931506849,
      "grad_norm": 2.8190650939941406,
      "kl": 0.18408203125,
      "learning_rate": 8.949771689497716e-07,
      "loss": 0.0002,
      "reward": 1.40625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 519.1875,
      "epoch": 0.31643835616438354,
      "grad_norm": 1.5466049909591675,
      "kl": 0.1650390625,
      "learning_rate": 8.945205479452055e-07,
      "loss": 0.0002,
      "reward": 1.40625,
      "reward_std": 0.27339156717061996,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.96875,
      "step": 231
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.09375,
      "epoch": 0.3178082191780822,
      "grad_norm": 8.43017864227295,
      "kl": 0.15771484375,
      "learning_rate": 8.940639269406392e-07,
      "loss": 0.0002,
      "reward": 1.203125,
      "reward_std": 0.26621313393116,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 0.96875,
      "step": 232
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 458.96875,
      "epoch": 0.31917808219178084,
      "grad_norm": 3.6506810188293457,
      "kl": 0.197509765625,
      "learning_rate": 8.93607305936073e-07,
      "loss": 0.0002,
      "reward": 1.625,
      "reward_std": 0.4045617878437042,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 233
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.0,
      "epoch": 0.32054794520547947,
      "grad_norm": 2.9570024013519287,
      "kl": 0.17041015625,
      "learning_rate": 8.931506849315068e-07,
      "loss": 0.0002,
      "reward": 1.4375,
      "reward_std": 0.3128525000065565,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.9375,
      "step": 234
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.78125,
      "epoch": 0.3219178082191781,
      "grad_norm": 1.4737908840179443,
      "kl": 0.158203125,
      "learning_rate": 8.926940639269406e-07,
      "loss": 0.0002,
      "reward": 1.765625,
      "reward_std": 0.41709377616643906,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 656.59375,
      "epoch": 0.3232876712328767,
      "grad_norm": 3.560518264770508,
      "kl": 0.168701171875,
      "learning_rate": 8.922374429223744e-07,
      "loss": 0.0002,
      "reward": 1.6796875,
      "reward_std": 0.12393621355295181,
      "rewards/accuracy_reward": 0.6796875,
      "rewards/format_reward": 1.0,
      "step": 236
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.59375,
      "epoch": 0.32465753424657534,
      "grad_norm": 1.0155640840530396,
      "kl": 0.16259765625,
      "learning_rate": 8.917808219178081e-07,
      "loss": 0.0002,
      "reward": 1.28125,
      "reward_std": 0.07312605157494545,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 237
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 572.84375,
      "epoch": 0.32602739726027397,
      "grad_norm": 0.018812214955687523,
      "kl": 0.177490234375,
      "learning_rate": 8.91324200913242e-07,
      "loss": 0.0002,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 238
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 714.375,
      "epoch": 0.3273972602739726,
      "grad_norm": 1.2518051862716675,
      "kl": 0.1611328125,
      "learning_rate": 8.908675799086758e-07,
      "loss": 0.0002,
      "reward": 1.8567708134651184,
      "reward_std": 0.23625470884144306,
      "rewards/accuracy_reward": 0.8880208134651184,
      "rewards/format_reward": 0.96875,
      "step": 239
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 471.78125,
      "epoch": 0.3287671232876712,
      "grad_norm": 0.015509501099586487,
      "kl": 0.189453125,
      "learning_rate": 8.904109589041095e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.5625,
      "epoch": 0.33013698630136984,
      "grad_norm": 2.026989459991455,
      "kl": 0.18701171875,
      "learning_rate": 8.899543378995433e-07,
      "loss": 0.0002,
      "reward": 1.6953125,
      "reward_std": 0.229622982442379,
      "rewards/accuracy_reward": 0.6953125,
      "rewards/format_reward": 1.0,
      "step": 241
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.1875,
      "epoch": 0.3315068493150685,
      "grad_norm": 1.3158351182937622,
      "kl": 0.180419921875,
      "learning_rate": 8.894977168949771e-07,
      "loss": 0.0002,
      "reward": 1.65625,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 242
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 721.46875,
      "epoch": 0.33287671232876714,
      "grad_norm": 1.1662763357162476,
      "kl": 0.147216796875,
      "learning_rate": 8.890410958904109e-07,
      "loss": 0.0001,
      "reward": 1.7299107313156128,
      "reward_std": 0.14230500534176826,
      "rewards/accuracy_reward": 0.7299107015132904,
      "rewards/format_reward": 1.0,
      "step": 243
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.09375,
      "epoch": 0.33424657534246577,
      "grad_norm": 2.274336814880371,
      "kl": 0.17529296875,
      "learning_rate": 8.885844748858448e-07,
      "loss": 0.0002,
      "reward": 1.4114583432674408,
      "reward_std": 0.17236988991498947,
      "rewards/accuracy_reward": 0.4114583432674408,
      "rewards/format_reward": 1.0,
      "step": 244
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 450.25,
      "epoch": 0.3356164383561644,
      "grad_norm": 1.401793360710144,
      "kl": 0.201416015625,
      "learning_rate": 8.881278538812785e-07,
      "loss": 0.0002,
      "reward": 1.328125,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 584.71875,
      "epoch": 0.336986301369863,
      "grad_norm": 1.7275149822235107,
      "kl": 0.1708984375,
      "learning_rate": 8.876712328767123e-07,
      "loss": 0.0002,
      "reward": 1.4192708432674408,
      "reward_std": 0.12374339066445827,
      "rewards/accuracy_reward": 0.4192708432674408,
      "rewards/format_reward": 1.0,
      "step": 246
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 548.875,
      "epoch": 0.33835616438356164,
      "grad_norm": 10.723580360412598,
      "kl": 0.16357421875,
      "learning_rate": 8.872146118721461e-07,
      "loss": 0.0002,
      "reward": 1.2890625,
      "reward_std": 0.16597744077444077,
      "rewards/accuracy_reward": 0.2890625,
      "rewards/format_reward": 1.0,
      "step": 247
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 618.375,
      "epoch": 0.33972602739726027,
      "grad_norm": 5.1156816482543945,
      "kl": 0.235107421875,
      "learning_rate": 8.867579908675798e-07,
      "loss": 0.0002,
      "reward": 1.625,
      "reward_std": 0.24456444010138512,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 248
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 449.625,
      "epoch": 0.3410958904109589,
      "grad_norm": 2.348053216934204,
      "kl": 0.185546875,
      "learning_rate": 8.863013698630136e-07,
      "loss": 0.0002,
      "reward": 1.421875,
      "reward_std": 0.29355230554938316,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 249
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 694.375,
      "epoch": 0.3424657534246575,
      "grad_norm": 0.8780984878540039,
      "kl": 0.171630859375,
      "learning_rate": 8.858447488584474e-07,
      "loss": 0.0002,
      "reward": 1.234375,
      "reward_std": 0.1173202246427536,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 0.96875,
      "step": 250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.1875,
      "epoch": 0.34383561643835614,
      "grad_norm": 3.7509207725524902,
      "kl": 0.19140625,
      "learning_rate": 8.853881278538813e-07,
      "loss": 0.0002,
      "reward": 1.4375,
      "reward_std": 0.2734241336584091,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 251
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.5,
      "epoch": 0.3452054794520548,
      "grad_norm": 1.7510371208190918,
      "kl": 0.3330078125,
      "learning_rate": 8.849315068493151e-07,
      "loss": 0.0003,
      "reward": 1.53125,
      "reward_std": 0.26409146934747696,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 252
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 497.0625,
      "epoch": 0.34657534246575344,
      "grad_norm": 0.7126864790916443,
      "kl": 0.1669921875,
      "learning_rate": 8.844748858447488e-07,
      "loss": 0.0002,
      "reward": 1.2890625,
      "reward_std": 0.3738361746072769,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 0.9375,
      "step": 253
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 350.6875,
      "epoch": 0.34794520547945207,
      "grad_norm": 2.1832709312438965,
      "kl": 0.1943359375,
      "learning_rate": 8.840182648401826e-07,
      "loss": 0.0002,
      "reward": 1.3515625,
      "reward_std": 0.1815449744462967,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 1.0,
      "step": 254
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 349.0,
      "epoch": 0.3493150684931507,
      "grad_norm": 2.6141257286071777,
      "kl": 0.217041015625,
      "learning_rate": 8.835616438356164e-07,
      "loss": 0.0002,
      "reward": 1.203125,
      "reward_std": 0.22097086533904076,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 1.0,
      "step": 255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.78125,
      "epoch": 0.3506849315068493,
      "grad_norm": 1.7095686197280884,
      "kl": 0.1640625,
      "learning_rate": 8.831050228310501e-07,
      "loss": 0.0002,
      "reward": 1.3671875,
      "reward_std": 0.21604011207818985,
      "rewards/accuracy_reward": 0.3984375,
      "rewards/format_reward": 0.96875,
      "step": 256
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.25,
      "epoch": 0.35205479452054794,
      "grad_norm": 1.052132487297058,
      "kl": 0.1669921875,
      "learning_rate": 8.826484018264839e-07,
      "loss": 0.0002,
      "reward": 1.609375,
      "reward_std": 0.1315089538693428,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 257
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.0,
      "epoch": 0.35342465753424657,
      "grad_norm": 17.632997512817383,
      "kl": 0.168212890625,
      "learning_rate": 8.821917808219178e-07,
      "loss": 0.0002,
      "reward": 1.671875,
      "reward_std": 0.38394393771886826,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.96875,
      "step": 258
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.46875,
      "epoch": 0.3547945205479452,
      "grad_norm": 1.7030237913131714,
      "kl": 0.17041015625,
      "learning_rate": 8.817351598173516e-07,
      "loss": 0.0002,
      "reward": 1.53125,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 259
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 385.09375,
      "epoch": 0.3561643835616438,
      "grad_norm": 3.2231457233428955,
      "kl": 0.19873046875,
      "learning_rate": 8.812785388127854e-07,
      "loss": 0.0002,
      "reward": 1.15625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 697.28125,
      "epoch": 0.35753424657534244,
      "grad_norm": 1.5386288166046143,
      "kl": 0.18798828125,
      "learning_rate": 8.808219178082191e-07,
      "loss": 0.0002,
      "reward": 1.5078125,
      "reward_std": 0.1916224267333746,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 0.96875,
      "step": 261
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.25,
      "epoch": 0.3589041095890411,
      "grad_norm": 17.967370986938477,
      "kl": 0.21240234375,
      "learning_rate": 8.803652968036529e-07,
      "loss": 0.0002,
      "reward": 1.6458333432674408,
      "reward_std": 0.26592448726296425,
      "rewards/accuracy_reward": 0.6458333432674408,
      "rewards/format_reward": 1.0,
      "step": 262
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 553.34375,
      "epoch": 0.36027397260273974,
      "grad_norm": 1.2522979974746704,
      "kl": 0.169921875,
      "learning_rate": 8.799086757990867e-07,
      "loss": 0.0002,
      "reward": 1.4791666269302368,
      "reward_std": 0.12335556373000145,
      "rewards/accuracy_reward": 0.4791666567325592,
      "rewards/format_reward": 1.0,
      "step": 263
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 380.53125,
      "epoch": 0.36164383561643837,
      "grad_norm": 1.5335299968719482,
      "kl": 0.187744140625,
      "learning_rate": 8.794520547945205e-07,
      "loss": 0.0002,
      "reward": 1.28125,
      "reward_std": 0.1962025985121727,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 264
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.9375,
      "epoch": 0.363013698630137,
      "grad_norm": 1.35916006565094,
      "kl": 0.24365234375,
      "learning_rate": 8.789954337899543e-07,
      "loss": 0.0002,
      "reward": 1.390625,
      "reward_std": 0.13258251920342445,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.96875,
      "step": 265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 546.75,
      "epoch": 0.3643835616438356,
      "grad_norm": 0.8804590702056885,
      "kl": 0.2041015625,
      "learning_rate": 8.785388127853881e-07,
      "loss": 0.0002,
      "reward": 1.3515625,
      "reward_std": 0.09704046696424484,
      "rewards/accuracy_reward": 0.3515625,
      "rewards/format_reward": 1.0,
      "step": 266
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.875,
      "epoch": 0.36575342465753424,
      "grad_norm": 1.8236092329025269,
      "kl": 0.188720703125,
      "learning_rate": 8.780821917808219e-07,
      "loss": 0.0002,
      "reward": 2.078125,
      "reward_std": 0.0776018276810646,
      "rewards/accuracy_reward": 1.078125,
      "rewards/format_reward": 1.0,
      "step": 267
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 386.59375,
      "epoch": 0.36712328767123287,
      "grad_norm": 11.162297248840332,
      "kl": 0.190185546875,
      "learning_rate": 8.776255707762557e-07,
      "loss": 0.0002,
      "reward": 1.5859375,
      "reward_std": 0.23224157467484474,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 1.0,
      "step": 268
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.5625,
      "epoch": 0.3684931506849315,
      "grad_norm": 2.6213197708129883,
      "kl": 0.214111328125,
      "learning_rate": 8.771689497716894e-07,
      "loss": 0.0002,
      "reward": 1.58984375,
      "reward_std": 0.33964164927601814,
      "rewards/accuracy_reward": 0.58984375,
      "rewards/format_reward": 1.0,
      "step": 269
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.5,
      "epoch": 0.3698630136986301,
      "grad_norm": 2.206378221511841,
      "kl": 0.1650390625,
      "learning_rate": 8.767123287671232e-07,
      "loss": 0.0002,
      "reward": 1.6437499523162842,
      "reward_std": 0.4507673643529415,
      "rewards/accuracy_reward": 0.6749999970197678,
      "rewards/format_reward": 0.96875,
      "step": 270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.84375,
      "epoch": 0.37123287671232874,
      "grad_norm": 3.658581018447876,
      "kl": 0.167236328125,
      "learning_rate": 8.762557077625571e-07,
      "loss": 0.0002,
      "reward": 1.2125000059604645,
      "reward_std": 0.09996108617633581,
      "rewards/accuracy_reward": 0.2124999761581421,
      "rewards/format_reward": 1.0,
      "step": 271
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 384.375,
      "epoch": 0.3726027397260274,
      "grad_norm": 2.4175188541412354,
      "kl": 0.258056640625,
      "learning_rate": 8.757990867579908e-07,
      "loss": 0.0003,
      "reward": 1.515625,
      "reward_std": 0.30721208080649376,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 272
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.53125,
      "epoch": 0.37397260273972605,
      "grad_norm": 1.4815560579299927,
      "kl": 0.1787109375,
      "learning_rate": 8.753424657534246e-07,
      "loss": 0.0002,
      "reward": 1.359375,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 273
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.0,
      "epoch": 0.37534246575342467,
      "grad_norm": 2.208836078643799,
      "kl": 0.159423828125,
      "learning_rate": 8.748858447488584e-07,
      "loss": 0.0002,
      "reward": 1.5859375,
      "reward_std": 0.19568835757672787,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 1.0,
      "step": 274
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.1875,
      "epoch": 0.3767123287671233,
      "grad_norm": 2.3311572074890137,
      "kl": 0.16748046875,
      "learning_rate": 8.744292237442922e-07,
      "loss": 0.0002,
      "reward": 1.609375,
      "reward_std": 0.269338458776474,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.78125,
      "epoch": 0.3780821917808219,
      "grad_norm": 2.8278110027313232,
      "kl": 0.22265625,
      "learning_rate": 8.73972602739726e-07,
      "loss": 0.0002,
      "reward": 1.296875,
      "reward_std": 0.1583191677927971,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 1.0,
      "step": 276
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 708.03125,
      "epoch": 0.37945205479452054,
      "grad_norm": 1.0638788938522339,
      "kl": 0.224853515625,
      "learning_rate": 8.735159817351597e-07,
      "loss": 0.0002,
      "reward": 1.3984375,
      "reward_std": 0.13941731117665768,
      "rewards/accuracy_reward": 0.4296875,
      "rewards/format_reward": 0.96875,
      "step": 277
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.6875,
      "epoch": 0.38082191780821917,
      "grad_norm": 0.551222026348114,
      "kl": 0.21728515625,
      "learning_rate": 8.730593607305936e-07,
      "loss": 0.0002,
      "reward": 1.078125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.109375,
      "rewards/format_reward": 0.96875,
      "step": 278
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 418.125,
      "epoch": 0.3821917808219178,
      "grad_norm": 2.8419177532196045,
      "kl": 0.177978515625,
      "learning_rate": 8.726027397260274e-07,
      "loss": 0.0002,
      "reward": 1.515625,
      "reward_std": 0.3098084479570389,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 279
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.46875,
      "epoch": 0.3835616438356164,
      "grad_norm": 2.1832010746002197,
      "kl": 0.171875,
      "learning_rate": 8.721461187214611e-07,
      "loss": 0.0002,
      "reward": 1.734375,
      "reward_std": 0.22673699632287025,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 280
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 407.8125,
      "epoch": 0.38493150684931504,
      "grad_norm": 1.6464120149612427,
      "kl": 0.18115234375,
      "learning_rate": 8.716894977168949e-07,
      "loss": 0.0002,
      "reward": 1.203125,
      "reward_std": 0.22097086161375046,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 1.0,
      "step": 281
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.125,
      "epoch": 0.3863013698630137,
      "grad_norm": 1.0540616512298584,
      "kl": 0.1845703125,
      "learning_rate": 8.712328767123287e-07,
      "loss": 0.0002,
      "reward": 1.546875,
      "reward_std": 0.05444390885531902,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 282
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.4375,
      "epoch": 0.38767123287671235,
      "grad_norm": 0.87820965051651,
      "kl": 0.15380859375,
      "learning_rate": 8.707762557077625e-07,
      "loss": 0.0002,
      "reward": 1.3020833134651184,
      "reward_std": 0.029462769627571106,
      "rewards/accuracy_reward": 0.3020833283662796,
      "rewards/format_reward": 1.0,
      "step": 283
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.53125,
      "epoch": 0.38904109589041097,
      "grad_norm": 3.814786434173584,
      "kl": 0.184326171875,
      "learning_rate": 8.703196347031964e-07,
      "loss": 0.0002,
      "reward": 1.5679687559604645,
      "reward_std": 0.125592902302742,
      "rewards/accuracy_reward": 0.5679687559604645,
      "rewards/format_reward": 1.0,
      "step": 284
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.59375,
      "epoch": 0.3904109589041096,
      "grad_norm": 8.2400484085083,
      "kl": 0.18896484375,
      "learning_rate": 8.698630136986301e-07,
      "loss": 0.0002,
      "reward": 1.7677083015441895,
      "reward_std": 0.2733229286968708,
      "rewards/accuracy_reward": 0.7989583611488342,
      "rewards/format_reward": 0.96875,
      "step": 285
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 505.28125,
      "epoch": 0.3917808219178082,
      "grad_norm": 1.994737982749939,
      "kl": 0.18994140625,
      "learning_rate": 8.694063926940639e-07,
      "loss": 0.0002,
      "reward": 1.4140625,
      "reward_std": 0.10613362491130829,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 286
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.8125,
      "epoch": 0.39315068493150684,
      "grad_norm": 5.628194808959961,
      "kl": 0.166748046875,
      "learning_rate": 8.689497716894977e-07,
      "loss": 0.0002,
      "reward": 1.7578125,
      "reward_std": 0.21267853677272797,
      "rewards/accuracy_reward": 0.7578125,
      "rewards/format_reward": 1.0,
      "step": 287
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 445.21875,
      "epoch": 0.39452054794520547,
      "grad_norm": 2.055222749710083,
      "kl": 0.202392578125,
      "learning_rate": 8.684931506849314e-07,
      "loss": 0.0002,
      "reward": 1.7187499701976776,
      "reward_std": 0.23613503947854042,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 288
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 406.75,
      "epoch": 0.3958904109589041,
      "grad_norm": 0.7133424282073975,
      "kl": 0.19189453125,
      "learning_rate": 8.680365296803652e-07,
      "loss": 0.0002,
      "reward": 1.5703125,
      "reward_std": 0.05725783854722977,
      "rewards/accuracy_reward": 0.5703125,
      "rewards/format_reward": 1.0,
      "step": 289
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 573.25,
      "epoch": 0.3972602739726027,
      "grad_norm": 1.4115186929702759,
      "kl": 0.185546875,
      "learning_rate": 8.67579908675799e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.23833239078521729,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 290
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.625,
      "epoch": 0.39863013698630134,
      "grad_norm": 1.865753173828125,
      "kl": 0.2314453125,
      "learning_rate": 8.671232876712329e-07,
      "loss": 0.0002,
      "reward": 1.5546875,
      "reward_std": 0.18630647659301758,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 1.0,
      "step": 291
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 523.46875,
      "epoch": 0.4,
      "grad_norm": 1.9343034029006958,
      "kl": 0.18505859375,
      "learning_rate": 8.666666666666667e-07,
      "loss": 0.0002,
      "reward": 1.640625,
      "reward_std": 0.31587694957852364,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 292
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.9375,
      "epoch": 0.40136986301369865,
      "grad_norm": 1.4373409748077393,
      "kl": 0.17578125,
      "learning_rate": 8.662100456621004e-07,
      "loss": 0.0002,
      "reward": 1.6388888359069824,
      "reward_std": 0.1706457920372486,
      "rewards/accuracy_reward": 0.6388888955116272,
      "rewards/format_reward": 1.0,
      "step": 293
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 509.96875,
      "epoch": 0.40273972602739727,
      "grad_norm": 1.1212681531906128,
      "kl": 0.19921875,
      "learning_rate": 8.657534246575342e-07,
      "loss": 0.0002,
      "reward": 1.203125,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 1.0,
      "step": 294
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.65625,
      "epoch": 0.4041095890410959,
      "grad_norm": 1.6588208675384521,
      "kl": 0.193359375,
      "learning_rate": 8.65296803652968e-07,
      "loss": 0.0002,
      "reward": 1.3971354067325592,
      "reward_std": 0.03875125199556351,
      "rewards/accuracy_reward": 0.3971354067325592,
      "rewards/format_reward": 1.0,
      "step": 295
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.1875,
      "epoch": 0.4054794520547945,
      "grad_norm": 1.8998637199401855,
      "kl": 0.22900390625,
      "learning_rate": 8.648401826484017e-07,
      "loss": 0.0002,
      "reward": 1.96875,
      "reward_std": 0.2519447058439255,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 1.0,
      "step": 296
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.46875,
      "epoch": 0.40684931506849314,
      "grad_norm": 1.1315598487854004,
      "kl": 0.178955078125,
      "learning_rate": 8.643835616438355e-07,
      "loss": 0.0002,
      "reward": 1.3828125,
      "reward_std": 0.09021057933568954,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 297
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 481.6875,
      "epoch": 0.40821917808219177,
      "grad_norm": 1.4266315698623657,
      "kl": 0.209716796875,
      "learning_rate": 8.639269406392694e-07,
      "loss": 0.0002,
      "reward": 1.3203125,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.3203125,
      "rewards/format_reward": 1.0,
      "step": 298
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 432.6875,
      "epoch": 0.4095890410958904,
      "grad_norm": 0.9578768014907837,
      "kl": 0.186279296875,
      "learning_rate": 8.634703196347032e-07,
      "loss": 0.0002,
      "reward": 1.203125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 1.0,
      "step": 299
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.875,
      "epoch": 0.410958904109589,
      "grad_norm": 1.1748281717300415,
      "kl": 0.247802734375,
      "learning_rate": 8.63013698630137e-07,
      "loss": 0.0002,
      "reward": 1.3671875,
      "reward_std": 0.08679073117673397,
      "rewards/accuracy_reward": 0.3671875,
      "rewards/format_reward": 1.0,
      "step": 300
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 459.375,
      "epoch": 0.4123287671232877,
      "grad_norm": 1.2540714740753174,
      "kl": 0.2158203125,
      "learning_rate": 8.625570776255707e-07,
      "loss": 0.0002,
      "reward": 1.9296875,
      "reward_std": 0.1649293377995491,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 1.0,
      "step": 301
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.5625,
      "epoch": 0.4136986301369863,
      "grad_norm": 1.9239349365234375,
      "kl": 0.167724609375,
      "learning_rate": 8.621004566210045e-07,
      "loss": 0.0002,
      "reward": 1.5446428060531616,
      "reward_std": 0.05746740661561489,
      "rewards/accuracy_reward": 0.5446428954601288,
      "rewards/format_reward": 1.0,
      "step": 302
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.46875,
      "epoch": 0.41506849315068495,
      "grad_norm": 2.8443050384521484,
      "kl": 0.185546875,
      "learning_rate": 8.616438356164383e-07,
      "loss": 0.0002,
      "reward": 1.5546875,
      "reward_std": 0.15467960387468338,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 1.0,
      "step": 303
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 369.8125,
      "epoch": 0.41643835616438357,
      "grad_norm": 2.6606931686401367,
      "kl": 0.226318359375,
      "learning_rate": 8.611872146118721e-07,
      "loss": 0.0002,
      "reward": 1.71875,
      "reward_std": 0.38481390848755836,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 1.0,
      "step": 304
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.625,
      "epoch": 0.4178082191780822,
      "grad_norm": 1.6216801404953003,
      "kl": 0.1943359375,
      "learning_rate": 8.607305936073059e-07,
      "loss": 0.0002,
      "reward": 1.21875,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 305
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.3125,
      "epoch": 0.4191780821917808,
      "grad_norm": 2.814319372177124,
      "kl": 0.189453125,
      "learning_rate": 8.602739726027397e-07,
      "loss": 0.0002,
      "reward": 1.515625,
      "reward_std": 0.36036762222647667,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 306
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 473.53125,
      "epoch": 0.42054794520547945,
      "grad_norm": 2.310194730758667,
      "kl": 0.187255859375,
      "learning_rate": 8.598173515981735e-07,
      "loss": 0.0002,
      "reward": 1.578125,
      "reward_std": 0.30617379024624825,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 307
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.25,
      "epoch": 0.42191780821917807,
      "grad_norm": 0.9056942462921143,
      "kl": 0.190673828125,
      "learning_rate": 8.593607305936073e-07,
      "loss": 0.0002,
      "reward": 1.359375,
      "reward_std": 0.19408094882965088,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 308
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 389.1875,
      "epoch": 0.4232876712328767,
      "grad_norm": 0.03478769585490227,
      "kl": 0.239501953125,
      "learning_rate": 8.58904109589041e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 309
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 453.59375,
      "epoch": 0.4246575342465753,
      "grad_norm": 7.321046829223633,
      "kl": 0.204345703125,
      "learning_rate": 8.584474885844748e-07,
      "loss": 0.0002,
      "reward": 1.3046875,
      "reward_std": 0.30941806733608246,
      "rewards/accuracy_reward": 0.3359375,
      "rewards/format_reward": 0.96875,
      "step": 310
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 475.28125,
      "epoch": 0.426027397260274,
      "grad_norm": 3.1986887454986572,
      "kl": 0.221435546875,
      "learning_rate": 8.579908675799087e-07,
      "loss": 0.0002,
      "reward": 1.421875,
      "reward_std": 0.4004939943552017,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.96875,
      "step": 311
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.9375,
      "epoch": 0.4273972602739726,
      "grad_norm": 1.7835962772369385,
      "kl": 0.18798828125,
      "learning_rate": 8.575342465753424e-07,
      "loss": 0.0002,
      "reward": 2.171875,
      "reward_std": 0.23280548676848412,
      "rewards/accuracy_reward": 1.171875,
      "rewards/format_reward": 1.0,
      "step": 312
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 506.5625,
      "epoch": 0.42876712328767125,
      "grad_norm": 2.5982441902160645,
      "kl": 0.175537109375,
      "learning_rate": 8.570776255707762e-07,
      "loss": 0.0002,
      "reward": 1.46875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 313
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 481.71875,
      "epoch": 0.4301369863013699,
      "grad_norm": 0.013101520948112011,
      "kl": 0.212890625,
      "learning_rate": 8.5662100456621e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 314
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 465.71875,
      "epoch": 0.4315068493150685,
      "grad_norm": 5.282835960388184,
      "kl": 0.212890625,
      "learning_rate": 8.561643835616438e-07,
      "loss": 0.0002,
      "reward": 1.1796875,
      "reward_std": 0.12073516845703125,
      "rewards/accuracy_reward": 0.1796875,
      "rewards/format_reward": 1.0,
      "step": 315
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.96875,
      "epoch": 0.4328767123287671,
      "grad_norm": 2.862443447113037,
      "kl": 0.21142578125,
      "learning_rate": 8.557077625570776e-07,
      "loss": 0.0002,
      "reward": 2.234375,
      "reward_std": 0.44417304918169975,
      "rewards/accuracy_reward": 1.234375,
      "rewards/format_reward": 1.0,
      "step": 316
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 707.71875,
      "epoch": 0.43424657534246575,
      "grad_norm": 1.707274317741394,
      "kl": 0.19287109375,
      "learning_rate": 8.552511415525113e-07,
      "loss": 0.0002,
      "reward": 1.7218749523162842,
      "reward_std": 0.0844996627420187,
      "rewards/accuracy_reward": 0.721875011920929,
      "rewards/format_reward": 1.0,
      "step": 317
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 455.0,
      "epoch": 0.43561643835616437,
      "grad_norm": 2.739851713180542,
      "kl": 0.216796875,
      "learning_rate": 8.547945205479452e-07,
      "loss": 0.0002,
      "reward": 1.328125,
      "reward_std": 0.29355230554938316,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 318
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.71875,
      "epoch": 0.436986301369863,
      "grad_norm": 1.7768224477767944,
      "kl": 0.217529296875,
      "learning_rate": 8.54337899543379e-07,
      "loss": 0.0002,
      "reward": 1.25,
      "reward_std": 0.3535533770918846,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.96875,
      "step": 319
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 387.8125,
      "epoch": 0.4383561643835616,
      "grad_norm": 5.192182540893555,
      "kl": 0.357421875,
      "learning_rate": 8.538812785388127e-07,
      "loss": 0.0004,
      "reward": 1.3697916865348816,
      "reward_std": 0.1467284932732582,
      "rewards/accuracy_reward": 0.3697916567325592,
      "rewards/format_reward": 1.0,
      "step": 320
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.8125,
      "epoch": 0.4397260273972603,
      "grad_norm": 1.898450493812561,
      "kl": 0.24365234375,
      "learning_rate": 8.534246575342465e-07,
      "loss": 0.0002,
      "reward": 1.2265625,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 1.0,
      "step": 321
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 412.375,
      "epoch": 0.4410958904109589,
      "grad_norm": 1.9675666093826294,
      "kl": 0.231689453125,
      "learning_rate": 8.529680365296803e-07,
      "loss": 0.0002,
      "reward": 1.375,
      "reward_std": 0.1825428232550621,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 322
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 380.5,
      "epoch": 0.44246575342465755,
      "grad_norm": 1.3606637716293335,
      "kl": 0.257568359375,
      "learning_rate": 8.52511415525114e-07,
      "loss": 0.0003,
      "reward": 1.25,
      "reward_std": 0.0704294964671135,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 323
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.0,
      "epoch": 0.4438356164383562,
      "grad_norm": 1.0777534246444702,
      "kl": 0.18701171875,
      "learning_rate": 8.52054794520548e-07,
      "loss": 0.0002,
      "reward": 1.6614583134651184,
      "reward_std": 0.0961906760931015,
      "rewards/accuracy_reward": 0.6614583432674408,
      "rewards/format_reward": 1.0,
      "step": 324
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.09375,
      "epoch": 0.4452054794520548,
      "grad_norm": 1.1463043689727783,
      "kl": 0.205078125,
      "learning_rate": 8.515981735159817e-07,
      "loss": 0.0002,
      "reward": 1.390625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 325
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 483.25,
      "epoch": 0.4465753424657534,
      "grad_norm": 3.938135862350464,
      "kl": 0.256591796875,
      "learning_rate": 8.511415525114155e-07,
      "loss": 0.0003,
      "reward": 1.60546875,
      "reward_std": 0.2536969259381294,
      "rewards/accuracy_reward": 0.60546875,
      "rewards/format_reward": 1.0,
      "step": 326
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 707.46875,
      "epoch": 0.44794520547945205,
      "grad_norm": 2.0289745330810547,
      "kl": 0.170166015625,
      "learning_rate": 8.506849315068493e-07,
      "loss": 0.0002,
      "reward": 1.75,
      "reward_std": 0.16622394509613514,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 327
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 496.15625,
      "epoch": 0.44931506849315067,
      "grad_norm": 1.461342215538025,
      "kl": 0.2412109375,
      "learning_rate": 8.50228310502283e-07,
      "loss": 0.0002,
      "reward": 1.390625,
      "reward_std": 0.1446593925356865,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 328
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 693.28125,
      "epoch": 0.4506849315068493,
      "grad_norm": 1.320560336112976,
      "kl": 0.18310546875,
      "learning_rate": 8.497716894977168e-07,
      "loss": 0.0002,
      "reward": 1.703125,
      "reward_std": 0.3912949990481138,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 329
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.65625,
      "epoch": 0.4520547945205479,
      "grad_norm": 2.9369583129882812,
      "kl": 0.19384765625,
      "learning_rate": 8.493150684931506e-07,
      "loss": 0.0002,
      "reward": 1.80859375,
      "reward_std": 0.1918780878186226,
      "rewards/accuracy_reward": 0.80859375,
      "rewards/format_reward": 1.0,
      "step": 330
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 461.03125,
      "epoch": 0.4534246575342466,
      "grad_norm": 2.696607828140259,
      "kl": 0.252685546875,
      "learning_rate": 8.488584474885845e-07,
      "loss": 0.0003,
      "reward": 1.45703125,
      "reward_std": 0.16619354858994484,
      "rewards/accuracy_reward": 0.45703125,
      "rewards/format_reward": 1.0,
      "step": 331
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 378.0625,
      "epoch": 0.4547945205479452,
      "grad_norm": 1.493513822555542,
      "kl": 0.244140625,
      "learning_rate": 8.484018264840183e-07,
      "loss": 0.0002,
      "reward": 1.234375,
      "reward_std": 0.1530819907784462,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 332
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.59375,
      "epoch": 0.45616438356164385,
      "grad_norm": 2.641517162322998,
      "kl": 0.220703125,
      "learning_rate": 8.47945205479452e-07,
      "loss": 0.0002,
      "reward": 1.7890625,
      "reward_std": 0.2441160511225462,
      "rewards/accuracy_reward": 0.7890625,
      "rewards/format_reward": 1.0,
      "step": 333
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.84375,
      "epoch": 0.4575342465753425,
      "grad_norm": 1.104002594947815,
      "kl": 0.206298828125,
      "learning_rate": 8.474885844748858e-07,
      "loss": 0.0002,
      "reward": 1.4296875,
      "reward_std": 0.09704047441482544,
      "rewards/accuracy_reward": 0.4296875,
      "rewards/format_reward": 1.0,
      "step": 334
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.6875,
      "epoch": 0.4589041095890411,
      "grad_norm": 2.944154977798462,
      "kl": 0.2412109375,
      "learning_rate": 8.470319634703196e-07,
      "loss": 0.0002,
      "reward": 1.3671875,
      "reward_std": 0.1678851991891861,
      "rewards/accuracy_reward": 0.3671875,
      "rewards/format_reward": 1.0,
      "step": 335
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.9375,
      "epoch": 0.4602739726027397,
      "grad_norm": 1.3954813480377197,
      "kl": 1.03955078125,
      "learning_rate": 8.465753424657533e-07,
      "loss": 0.001,
      "reward": 1.1614583432674408,
      "reward_std": 0.026702914386987686,
      "rewards/accuracy_reward": 0.1614583283662796,
      "rewards/format_reward": 1.0,
      "step": 336
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.25,
      "epoch": 0.46164383561643835,
      "grad_norm": 1.2163310050964355,
      "kl": 0.23876953125,
      "learning_rate": 8.461187214611871e-07,
      "loss": 0.0002,
      "reward": 1.75,
      "reward_std": 0.1733490191400051,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 337
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.625,
      "epoch": 0.46301369863013697,
      "grad_norm": 1.4970253705978394,
      "kl": 0.2275390625,
      "learning_rate": 8.45662100456621e-07,
      "loss": 0.0002,
      "reward": 1.34375,
      "reward_std": 0.1356339044868946,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 338
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.5,
      "epoch": 0.4643835616438356,
      "grad_norm": 2.083859920501709,
      "kl": 0.19677734375,
      "learning_rate": 8.452054794520548e-07,
      "loss": 0.0002,
      "reward": 1.62109375,
      "reward_std": 0.06917708925902843,
      "rewards/accuracy_reward": 0.62109375,
      "rewards/format_reward": 1.0,
      "step": 339
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.34375,
      "epoch": 0.4657534246575342,
      "grad_norm": 0.6715368628501892,
      "kl": 0.2392578125,
      "learning_rate": 8.447488584474886e-07,
      "loss": 0.0002,
      "reward": 1.49609375,
      "reward_std": 0.04555431008338928,
      "rewards/accuracy_reward": 0.49609375,
      "rewards/format_reward": 1.0,
      "step": 340
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.625,
      "epoch": 0.4671232876712329,
      "grad_norm": 1.0103129148483276,
      "kl": 0.19287109375,
      "learning_rate": 8.442922374429223e-07,
      "loss": 0.0002,
      "reward": 1.3606770634651184,
      "reward_std": 0.040511311031877995,
      "rewards/accuracy_reward": 0.3606770783662796,
      "rewards/format_reward": 1.0,
      "step": 341
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 512.5,
      "epoch": 0.4684931506849315,
      "grad_norm": 0.989227831363678,
      "kl": 0.186767578125,
      "learning_rate": 8.438356164383561e-07,
      "loss": 0.0002,
      "reward": 1.84375,
      "reward_std": 0.21564550511538982,
      "rewards/accuracy_reward": 0.84375,
      "rewards/format_reward": 1.0,
      "step": 342
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.0,
      "epoch": 0.46986301369863015,
      "grad_norm": 1.0388092994689941,
      "kl": 0.19189453125,
      "learning_rate": 8.433789954337899e-07,
      "loss": 0.0002,
      "reward": 1.34375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 343
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.3125,
      "epoch": 0.4712328767123288,
      "grad_norm": 2.6138434410095215,
      "kl": 0.216552734375,
      "learning_rate": 8.429223744292237e-07,
      "loss": 0.0002,
      "reward": 1.234375,
      "reward_std": 0.1437886729836464,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 344
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.34375,
      "epoch": 0.4726027397260274,
      "grad_norm": 2.721027135848999,
      "kl": 0.189453125,
      "learning_rate": 8.424657534246576e-07,
      "loss": 0.0002,
      "reward": 1.453125,
      "reward_std": 0.2909114882349968,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.9375,
      "step": 345
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.65625,
      "epoch": 0.473972602739726,
      "grad_norm": 1.8512462377548218,
      "kl": 0.235107421875,
      "learning_rate": 8.420091324200913e-07,
      "loss": 0.0002,
      "reward": 1.2109375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.2109375,
      "rewards/format_reward": 1.0,
      "step": 346
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.125,
      "epoch": 0.47534246575342465,
      "grad_norm": 0.8181569576263428,
      "kl": 0.232177734375,
      "learning_rate": 8.415525114155251e-07,
      "loss": 0.0002,
      "reward": 1.2447916567325592,
      "reward_std": 0.05193428695201874,
      "rewards/accuracy_reward": 0.2447916865348816,
      "rewards/format_reward": 1.0,
      "step": 347
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.875,
      "epoch": 0.4767123287671233,
      "grad_norm": 1.4510343074798584,
      "kl": 0.21044921875,
      "learning_rate": 8.410958904109589e-07,
      "loss": 0.0002,
      "reward": 2.015625,
      "reward_std": 0.3187600150704384,
      "rewards/accuracy_reward": 1.015625,
      "rewards/format_reward": 1.0,
      "step": 348
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.625,
      "epoch": 0.4780821917808219,
      "grad_norm": 0.8320886492729187,
      "kl": 0.21875,
      "learning_rate": 8.406392694063926e-07,
      "loss": 0.0002,
      "reward": 1.328125,
      "reward_std": 0.2824692949652672,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 0.96875,
      "step": 349
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.75,
      "epoch": 0.4794520547945205,
      "grad_norm": 2.6687960624694824,
      "kl": 0.2578125,
      "learning_rate": 8.401826484018264e-07,
      "loss": 0.0003,
      "reward": 1.32421875,
      "reward_std": 0.26008394733071327,
      "rewards/accuracy_reward": 0.32421875,
      "rewards/format_reward": 1.0,
      "step": 350
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.84375,
      "epoch": 0.4808219178082192,
      "grad_norm": 1.2626047134399414,
      "kl": 0.199462890625,
      "learning_rate": 8.397260273972603e-07,
      "loss": 0.0002,
      "reward": 1.453125,
      "reward_std": 0.19960851781070232,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 351
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 760.0,
      "epoch": 0.4821917808219178,
      "grad_norm": 1.4241095781326294,
      "kl": 0.26806640625,
      "learning_rate": 8.39269406392694e-07,
      "loss": 0.0003,
      "reward": 1.5312499701976776,
      "reward_std": 0.19424722902476788,
      "rewards/accuracy_reward": 0.5624999850988388,
      "rewards/format_reward": 0.96875,
      "step": 352
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 389.09375,
      "epoch": 0.48356164383561645,
      "grad_norm": 2.1359047889709473,
      "kl": 0.23779296875,
      "learning_rate": 8.388127853881279e-07,
      "loss": 0.0002,
      "reward": 1.359375,
      "reward_std": 0.27564920112490654,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 353
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 498.59375,
      "epoch": 0.4849315068493151,
      "grad_norm": 2.235769510269165,
      "kl": 0.20361328125,
      "learning_rate": 8.383561643835616e-07,
      "loss": 0.0002,
      "reward": 1.25,
      "reward_std": 0.2314550280570984,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 354
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 754.40625,
      "epoch": 0.4863013698630137,
      "grad_norm": 0.011801384389400482,
      "kl": 0.194091796875,
      "learning_rate": 8.378995433789954e-07,
      "loss": 0.0002,
      "reward": 1.125,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 1.0,
      "step": 355
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.09375,
      "epoch": 0.4876712328767123,
      "grad_norm": 3.116180181503296,
      "kl": 0.214599609375,
      "learning_rate": 8.374429223744292e-07,
      "loss": 0.0002,
      "reward": 1.9453125,
      "reward_std": 0.35993071645498276,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 1.0,
      "step": 356
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.90625,
      "epoch": 0.48904109589041095,
      "grad_norm": 2.013319492340088,
      "kl": 0.21240234375,
      "learning_rate": 8.369863013698629e-07,
      "loss": 0.0002,
      "reward": 1.5546875,
      "reward_std": 0.19226671755313873,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 1.0,
      "step": 357
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.53125,
      "epoch": 0.4904109589041096,
      "grad_norm": 1.1122536659240723,
      "kl": 0.219970703125,
      "learning_rate": 8.365296803652968e-07,
      "loss": 0.0002,
      "reward": 1.6927083134651184,
      "reward_std": 0.13533581793308258,
      "rewards/accuracy_reward": 0.6927083432674408,
      "rewards/format_reward": 1.0,
      "step": 358
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 591.5,
      "epoch": 0.4917808219178082,
      "grad_norm": 3.6900923252105713,
      "kl": 0.253173828125,
      "learning_rate": 8.360730593607306e-07,
      "loss": 0.0003,
      "reward": 1.140625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.140625,
      "rewards/format_reward": 1.0,
      "step": 359
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 402.375,
      "epoch": 0.4931506849315068,
      "grad_norm": 1.4678456783294678,
      "kl": 0.254150390625,
      "learning_rate": 8.356164383561643e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.96875,
      "step": 360
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 464.90625,
      "epoch": 0.4945205479452055,
      "grad_norm": 0.8065510392189026,
      "kl": 0.238037109375,
      "learning_rate": 8.351598173515981e-07,
      "loss": 0.0002,
      "reward": 1.484375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 361
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 437.53125,
      "epoch": 0.4958904109589041,
      "grad_norm": 0.8244228959083557,
      "kl": 0.2685546875,
      "learning_rate": 8.347031963470319e-07,
      "loss": 0.0003,
      "reward": 1.1953125,
      "reward_std": 0.04005437344312668,
      "rewards/accuracy_reward": 0.1953125,
      "rewards/format_reward": 1.0,
      "step": 362
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 360.59375,
      "epoch": 0.49726027397260275,
      "grad_norm": 0.8816384077072144,
      "kl": 0.24267578125,
      "learning_rate": 8.342465753424657e-07,
      "loss": 0.0002,
      "reward": 1.234375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 363
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 694.5,
      "epoch": 0.4986301369863014,
      "grad_norm": 1.5277591943740845,
      "kl": 0.229248046875,
      "learning_rate": 8.337899543378996e-07,
      "loss": 0.0002,
      "reward": 1.9697916805744171,
      "reward_std": 0.2954293917864561,
      "rewards/accuracy_reward": 0.9697916805744171,
      "rewards/format_reward": 1.0,
      "step": 364
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 584.96875,
      "epoch": 0.5,
      "grad_norm": 0.6433670520782471,
      "kl": 0.241455078125,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.0002,
      "reward": 1.453125,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 365
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 321.78125,
      "epoch": 0.5013698630136987,
      "grad_norm": 2.3756937980651855,
      "kl": 0.272216796875,
      "learning_rate": 8.328767123287671e-07,
      "loss": 0.0003,
      "reward": 1.28125,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 366
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 733.40625,
      "epoch": 0.5027397260273972,
      "grad_norm": 2.2720232009887695,
      "kl": 0.206298828125,
      "learning_rate": 8.324200913242009e-07,
      "loss": 0.0002,
      "reward": 1.8359375,
      "reward_std": 0.1747150868177414,
      "rewards/accuracy_reward": 0.8359375,
      "rewards/format_reward": 1.0,
      "step": 367
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 443.53125,
      "epoch": 0.5041095890410959,
      "grad_norm": 2.717717170715332,
      "kl": 0.2490234375,
      "learning_rate": 8.319634703196346e-07,
      "loss": 0.0002,
      "reward": 1.15625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 368
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 319.96875,
      "epoch": 0.5054794520547945,
      "grad_norm": 1.975050449371338,
      "kl": 0.249267578125,
      "learning_rate": 8.315068493150684e-07,
      "loss": 0.0002,
      "reward": 1.21875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 369
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 548.09375,
      "epoch": 0.5068493150684932,
      "grad_norm": 4.310245990753174,
      "kl": 0.254638671875,
      "learning_rate": 8.310502283105022e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.2177756354212761,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 370
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 323.375,
      "epoch": 0.5082191780821917,
      "grad_norm": 487.17987060546875,
      "kl": 0.27783203125,
      "learning_rate": 8.305936073059361e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.2346404492855072,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 371
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 449.71875,
      "epoch": 0.5095890410958904,
      "grad_norm": 6.600228309631348,
      "kl": 0.24365234375,
      "learning_rate": 8.301369863013699e-07,
      "loss": 0.0002,
      "reward": 1.46875,
      "reward_std": 0.17965975776314735,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 372
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 409.6875,
      "epoch": 0.510958904109589,
      "grad_norm": 1.2109966278076172,
      "kl": 0.24609375,
      "learning_rate": 8.296803652968036e-07,
      "loss": 0.0002,
      "reward": 1.33984375,
      "reward_std": 0.08086705580353737,
      "rewards/accuracy_reward": 0.33984375,
      "rewards/format_reward": 1.0,
      "step": 373
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.96875,
      "epoch": 0.5123287671232877,
      "grad_norm": 1.2593879699707031,
      "kl": 0.23974609375,
      "learning_rate": 8.292237442922374e-07,
      "loss": 0.0002,
      "reward": 1.9322916567325592,
      "reward_std": 0.07365694083273411,
      "rewards/accuracy_reward": 0.9322916716337204,
      "rewards/format_reward": 1.0,
      "step": 374
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 654.34375,
      "epoch": 0.5136986301369864,
      "grad_norm": 1.1873910427093506,
      "kl": 0.228515625,
      "learning_rate": 8.287671232876712e-07,
      "loss": 0.0002,
      "reward": 1.5208333730697632,
      "reward_std": 0.06780947372317314,
      "rewards/accuracy_reward": 0.5208333134651184,
      "rewards/format_reward": 1.0,
      "step": 375
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.625,
      "epoch": 0.5150684931506849,
      "grad_norm": 2.3675568103790283,
      "kl": 0.3232421875,
      "learning_rate": 8.283105022831049e-07,
      "loss": 0.0003,
      "reward": 1.74609375,
      "reward_std": 0.18580568581819534,
      "rewards/accuracy_reward": 0.74609375,
      "rewards/format_reward": 1.0,
      "step": 376
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 413.6875,
      "epoch": 0.5164383561643836,
      "grad_norm": 1.0833803415298462,
      "kl": 0.29248046875,
      "learning_rate": 8.278538812785387e-07,
      "loss": 0.0003,
      "reward": 1.2109375,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.2109375,
      "rewards/format_reward": 1.0,
      "step": 377
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.5625,
      "epoch": 0.5178082191780822,
      "grad_norm": 2.686486005783081,
      "kl": 0.234130859375,
      "learning_rate": 8.273972602739726e-07,
      "loss": 0.0002,
      "reward": 1.8828125,
      "reward_std": 0.08731903322041035,
      "rewards/accuracy_reward": 0.8828125,
      "rewards/format_reward": 1.0,
      "step": 378
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.8125,
      "epoch": 0.5191780821917809,
      "grad_norm": 0.7580247521400452,
      "kl": 0.2861328125,
      "learning_rate": 8.269406392694064e-07,
      "loss": 0.0003,
      "reward": 1.2578125,
      "reward_std": 0.09704047441482544,
      "rewards/accuracy_reward": 0.2578125,
      "rewards/format_reward": 1.0,
      "step": 379
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 399.46875,
      "epoch": 0.5205479452054794,
      "grad_norm": 0.9390192627906799,
      "kl": 0.2568359375,
      "learning_rate": 8.264840182648402e-07,
      "loss": 0.0003,
      "reward": 1.359375,
      "reward_std": 0.19408094882965088,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 380
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 440.59375,
      "epoch": 0.5219178082191781,
      "grad_norm": 1.6535121202468872,
      "kl": 0.2880859375,
      "learning_rate": 8.260273972602739e-07,
      "loss": 0.0003,
      "reward": 1.4765625,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 1.0,
      "step": 381
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.625,
      "epoch": 0.5232876712328767,
      "grad_norm": 32.29432678222656,
      "kl": 0.25146484375,
      "learning_rate": 8.255707762557077e-07,
      "loss": 0.0003,
      "reward": 1.578125,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 382
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 654.96875,
      "epoch": 0.5246575342465754,
      "grad_norm": 4.199236869812012,
      "kl": 0.243896484375,
      "learning_rate": 8.251141552511415e-07,
      "loss": 0.0002,
      "reward": 1.4739583134651184,
      "reward_std": 0.10995287448167801,
      "rewards/accuracy_reward": 0.4739583283662796,
      "rewards/format_reward": 1.0,
      "step": 383
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 287.28125,
      "epoch": 0.5260273972602739,
      "grad_norm": 14.366601943969727,
      "kl": 0.306640625,
      "learning_rate": 8.246575342465753e-07,
      "loss": 0.0003,
      "reward": 1.421875,
      "reward_std": 0.2877064570784569,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 384
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 419.9375,
      "epoch": 0.5273972602739726,
      "grad_norm": 1.405211329460144,
      "kl": 0.2666015625,
      "learning_rate": 8.242009132420092e-07,
      "loss": 0.0003,
      "reward": 1.5494791269302368,
      "reward_std": 0.10189040005207062,
      "rewards/accuracy_reward": 0.5494791567325592,
      "rewards/format_reward": 1.0,
      "step": 385
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.96875,
      "epoch": 0.5287671232876713,
      "grad_norm": 1.3458356857299805,
      "kl": 0.250244140625,
      "learning_rate": 8.237442922374429e-07,
      "loss": 0.0003,
      "reward": 1.4140625,
      "reward_std": 0.15268651396036148,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 386
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.59375,
      "epoch": 0.5301369863013699,
      "grad_norm": 1.6495295763015747,
      "kl": 0.23291015625,
      "learning_rate": 8.232876712328767e-07,
      "loss": 0.0002,
      "reward": 1.7203125059604645,
      "reward_std": 0.15397237055003643,
      "rewards/accuracy_reward": 0.7203124761581421,
      "rewards/format_reward": 1.0,
      "step": 387
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 518.90625,
      "epoch": 0.5315068493150685,
      "grad_norm": 1.8638428449630737,
      "kl": 0.28076171875,
      "learning_rate": 8.228310502283105e-07,
      "loss": 0.0003,
      "reward": 1.421875,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 388
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.375,
      "epoch": 0.5328767123287671,
      "grad_norm": 2.2824759483337402,
      "kl": 0.250732421875,
      "learning_rate": 8.223744292237442e-07,
      "loss": 0.0003,
      "reward": 1.8055555820465088,
      "reward_std": 0.19911032635718584,
      "rewards/accuracy_reward": 0.8055555671453476,
      "rewards/format_reward": 1.0,
      "step": 389
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.59375,
      "epoch": 0.5342465753424658,
      "grad_norm": 2.9996278285980225,
      "kl": 0.259765625,
      "learning_rate": 8.21917808219178e-07,
      "loss": 0.0003,
      "reward": 2.0,
      "reward_std": 0.16151439771056175,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 390
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 446.9375,
      "epoch": 0.5356164383561643,
      "grad_norm": 2.858344316482544,
      "kl": 0.2744140625,
      "learning_rate": 8.214611872146119e-07,
      "loss": 0.0003,
      "reward": 1.3828125,
      "reward_std": 0.2894125059247017,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 391
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.03125,
      "epoch": 0.536986301369863,
      "grad_norm": 5.41072940826416,
      "kl": 0.30810546875,
      "learning_rate": 8.210045662100456e-07,
      "loss": 0.0003,
      "reward": 1.21875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 392
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 717.78125,
      "epoch": 0.5383561643835616,
      "grad_norm": 1.8393654823303223,
      "kl": 0.301025390625,
      "learning_rate": 8.205479452054795e-07,
      "loss": 0.0003,
      "reward": 1.40625,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.96875,
      "step": 393
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.125,
      "epoch": 0.5397260273972603,
      "grad_norm": 1.726664662361145,
      "kl": 0.287109375,
      "learning_rate": 8.200913242009132e-07,
      "loss": 0.0003,
      "reward": 1.265625,
      "reward_std": 0.17730122804641724,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 394
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 593.46875,
      "epoch": 0.541095890410959,
      "grad_norm": 1.4600183963775635,
      "kl": 0.23193359375,
      "learning_rate": 8.19634703196347e-07,
      "loss": 0.0002,
      "reward": 1.578125,
      "reward_std": 0.1983242630958557,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 395
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.34375,
      "epoch": 0.5424657534246575,
      "grad_norm": 1.1658108234405518,
      "kl": 0.25390625,
      "learning_rate": 8.191780821917808e-07,
      "loss": 0.0003,
      "reward": 1.9140625,
      "reward_std": 0.1649293377995491,
      "rewards/accuracy_reward": 0.9140625,
      "rewards/format_reward": 1.0,
      "step": 396
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 634.875,
      "epoch": 0.5438356164383562,
      "grad_norm": 1.2310694456100464,
      "kl": 0.251953125,
      "learning_rate": 8.187214611872145e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 397
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.3125,
      "epoch": 0.5452054794520548,
      "grad_norm": 1.6130096912384033,
      "kl": 0.244140625,
      "learning_rate": 8.182648401826484e-07,
      "loss": 0.0002,
      "reward": 1.6171875,
      "reward_std": 0.0765409953892231,
      "rewards/accuracy_reward": 0.6171875,
      "rewards/format_reward": 1.0,
      "step": 398
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 700.3125,
      "epoch": 0.5465753424657535,
      "grad_norm": 1.6435996294021606,
      "kl": 0.229736328125,
      "learning_rate": 8.178082191780822e-07,
      "loss": 0.0002,
      "reward": 1.5677083134651184,
      "reward_std": 0.3693719878792763,
      "rewards/accuracy_reward": 0.6302083283662796,
      "rewards/format_reward": 0.9375,
      "step": 399
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 347.75,
      "epoch": 0.547945205479452,
      "grad_norm": 0.7573147416114807,
      "kl": 0.2626953125,
      "learning_rate": 8.173515981735159e-07,
      "loss": 0.0003,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 400
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 858.03125,
      "epoch": 0.5493150684931507,
      "grad_norm": 2.761258602142334,
      "kl": 0.218994140625,
      "learning_rate": 8.168949771689498e-07,
      "loss": 0.0002,
      "reward": 1.125,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.875,
      "step": 401
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 362.5625,
      "epoch": 0.5506849315068493,
      "grad_norm": 4.398726463317871,
      "kl": 0.285400390625,
      "learning_rate": 8.164383561643835e-07,
      "loss": 0.0003,
      "reward": 1.40625,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 402
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 824.625,
      "epoch": 0.552054794520548,
      "grad_norm": 1.2996392250061035,
      "kl": 0.2373046875,
      "learning_rate": 8.159817351598172e-07,
      "loss": 0.0002,
      "reward": 1.6875,
      "reward_std": 0.6573142260313034,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.875,
      "step": 403
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 847.0,
      "epoch": 0.5534246575342465,
      "grad_norm": 1.484923243522644,
      "kl": 0.2275390625,
      "learning_rate": 8.155251141552512e-07,
      "loss": 0.0002,
      "reward": 1.640625,
      "reward_std": 0.5366625860333443,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.875,
      "step": 404
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.21875,
      "epoch": 0.5547945205479452,
      "grad_norm": 2.970003843307495,
      "kl": 0.255859375,
      "learning_rate": 8.150684931506849e-07,
      "loss": 0.0003,
      "reward": 1.4921875,
      "reward_std": 0.336070965975523,
      "rewards/accuracy_reward": 0.5234375,
      "rewards/format_reward": 0.96875,
      "step": 405
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 843.25,
      "epoch": 0.5561643835616439,
      "grad_norm": 2.474311113357544,
      "kl": 0.22802734375,
      "learning_rate": 8.146118721461187e-07,
      "loss": 0.0002,
      "reward": 1.90625,
      "reward_std": 0.5572120249271393,
      "rewards/accuracy_reward": 0.96875,
      "rewards/format_reward": 0.9375,
      "step": 406
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 494.5625,
      "epoch": 0.5575342465753425,
      "grad_norm": 3.557459592819214,
      "kl": 0.23876953125,
      "learning_rate": 8.141552511415525e-07,
      "loss": 0.0002,
      "reward": 1.6197916567325592,
      "reward_std": 0.1845790259540081,
      "rewards/accuracy_reward": 0.6197916567325592,
      "rewards/format_reward": 1.0,
      "step": 407
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 509.65625,
      "epoch": 0.5589041095890411,
      "grad_norm": 4.341843605041504,
      "kl": 0.25634765625,
      "learning_rate": 8.136986301369862e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.49022960662841797,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.9375,
      "step": 408
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.375,
      "epoch": 0.5602739726027397,
      "grad_norm": 1.4180370569229126,
      "kl": 0.231689453125,
      "learning_rate": 8.132420091324201e-07,
      "loss": 0.0002,
      "reward": 1.59375,
      "reward_std": 0.30327702313661575,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.96875,
      "step": 409
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 618.65625,
      "epoch": 0.5616438356164384,
      "grad_norm": 1.4287060499191284,
      "kl": 0.273193359375,
      "learning_rate": 8.127853881278538e-07,
      "loss": 0.0003,
      "reward": 1.484375,
      "reward_std": 0.2497537788003683,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.9375,
      "step": 410
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 769.59375,
      "epoch": 0.563013698630137,
      "grad_norm": 1.4561328887939453,
      "kl": 0.249267578125,
      "learning_rate": 8.123287671232877e-07,
      "loss": 0.0002,
      "reward": 1.328125,
      "reward_std": 0.09143973141908646,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 411
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 329.5625,
      "epoch": 0.5643835616438356,
      "grad_norm": 1.4629502296447754,
      "kl": 0.26904296875,
      "learning_rate": 8.118721461187215e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 412
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 458.34375,
      "epoch": 0.5657534246575342,
      "grad_norm": 4.4141669273376465,
      "kl": 0.276611328125,
      "learning_rate": 8.114155251141552e-07,
      "loss": 0.0003,
      "reward": 1.3828125,
      "reward_std": 0.17702843621373177,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 413
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 432.84375,
      "epoch": 0.5671232876712329,
      "grad_norm": 0.015535110607743263,
      "kl": 0.278564453125,
      "learning_rate": 8.10958904109589e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 414
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 547.53125,
      "epoch": 0.5684931506849316,
      "grad_norm": 5.981386661529541,
      "kl": 0.2919921875,
      "learning_rate": 8.105022831050228e-07,
      "loss": 0.0003,
      "reward": 1.640625,
      "reward_std": 0.2777610570192337,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 415
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 462.8125,
      "epoch": 0.5698630136986301,
      "grad_norm": 2.892124652862549,
      "kl": 0.279296875,
      "learning_rate": 8.100456621004565e-07,
      "loss": 0.0003,
      "reward": 1.765625,
      "reward_std": 0.35141606256365776,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 416
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 685.75,
      "epoch": 0.5712328767123288,
      "grad_norm": 1.2724870443344116,
      "kl": 0.252685546875,
      "learning_rate": 8.095890410958903e-07,
      "loss": 0.0003,
      "reward": 1.390625,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 417
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 443.0,
      "epoch": 0.5726027397260274,
      "grad_norm": 2.105574131011963,
      "kl": 0.296875,
      "learning_rate": 8.091324200913242e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 418
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 442.21875,
      "epoch": 0.5739726027397261,
      "grad_norm": 4.214569091796875,
      "kl": 0.3125,
      "learning_rate": 8.08675799086758e-07,
      "loss": 0.0003,
      "reward": 1.640625,
      "reward_std": 0.18648964539170265,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 419
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 436.59375,
      "epoch": 0.5753424657534246,
      "grad_norm": 4.964509963989258,
      "kl": 0.2587890625,
      "learning_rate": 8.082191780821918e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.28664068691432476,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 420
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 669.875,
      "epoch": 0.5767123287671233,
      "grad_norm": 2.2054591178894043,
      "kl": 0.243408203125,
      "learning_rate": 8.077625570776255e-07,
      "loss": 0.0002,
      "reward": 1.71484375,
      "reward_std": 0.11598511412739754,
      "rewards/accuracy_reward": 0.71484375,
      "rewards/format_reward": 1.0,
      "step": 421
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.125,
      "epoch": 0.5780821917808219,
      "grad_norm": 2.5887465476989746,
      "kl": 0.28125,
      "learning_rate": 8.073059360730593e-07,
      "loss": 0.0003,
      "reward": 1.515625,
      "reward_std": 0.23243396915495396,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 422
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.84375,
      "epoch": 0.5794520547945206,
      "grad_norm": 3.1099772453308105,
      "kl": 0.29345703125,
      "learning_rate": 8.068493150684931e-07,
      "loss": 0.0003,
      "reward": 1.4810267686843872,
      "reward_std": 0.10061977338045835,
      "rewards/accuracy_reward": 0.4810267835855484,
      "rewards/format_reward": 1.0,
      "step": 423
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 427.875,
      "epoch": 0.5808219178082191,
      "grad_norm": 2.3216264247894287,
      "kl": 0.57421875,
      "learning_rate": 8.063926940639269e-07,
      "loss": 0.0006,
      "reward": 1.140625,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.140625,
      "rewards/format_reward": 1.0,
      "step": 424
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 314.0,
      "epoch": 0.5821917808219178,
      "grad_norm": 1.1300055980682373,
      "kl": 0.27392578125,
      "learning_rate": 8.059360730593608e-07,
      "loss": 0.0003,
      "reward": 1.140625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.140625,
      "rewards/format_reward": 1.0,
      "step": 425
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.5,
      "epoch": 0.5835616438356165,
      "grad_norm": 1.6136479377746582,
      "kl": 0.2802734375,
      "learning_rate": 8.054794520547945e-07,
      "loss": 0.0003,
      "reward": 1.859375,
      "reward_std": 0.19408093392848969,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 426
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 300.6875,
      "epoch": 0.584931506849315,
      "grad_norm": 3.859463930130005,
      "kl": 0.30029296875,
      "learning_rate": 8.050228310502283e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.4182215705513954,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 427
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 491.25,
      "epoch": 0.5863013698630137,
      "grad_norm": 3.137112617492676,
      "kl": 0.2880859375,
      "learning_rate": 8.045662100456621e-07,
      "loss": 0.0003,
      "reward": 1.4713541865348816,
      "reward_std": 0.18944428488612175,
      "rewards/accuracy_reward": 0.4713541716337204,
      "rewards/format_reward": 1.0,
      "step": 428
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 589.84375,
      "epoch": 0.5876712328767123,
      "grad_norm": 2.1864824295043945,
      "kl": 0.30810546875,
      "learning_rate": 8.041095890410958e-07,
      "loss": 0.0003,
      "reward": 1.6171875,
      "reward_std": 0.14428602531552315,
      "rewards/accuracy_reward": 0.6171875,
      "rewards/format_reward": 1.0,
      "step": 429
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 412.46875,
      "epoch": 0.589041095890411,
      "grad_norm": 4.553158283233643,
      "kl": 0.33642578125,
      "learning_rate": 8.036529680365296e-07,
      "loss": 0.0003,
      "reward": 1.5390625,
      "reward_std": 0.25012245774269104,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 1.0,
      "step": 430
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.96875,
      "epoch": 0.5904109589041096,
      "grad_norm": 2.7253074645996094,
      "kl": 0.31884765625,
      "learning_rate": 8.031963470319635e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 431
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 434.8125,
      "epoch": 0.5917808219178082,
      "grad_norm": 1.4084954261779785,
      "kl": 0.275634765625,
      "learning_rate": 8.027397260273972e-07,
      "loss": 0.0003,
      "reward": 1.19921875,
      "reward_std": 0.06874492764472961,
      "rewards/accuracy_reward": 0.19921875,
      "rewards/format_reward": 1.0,
      "step": 432
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.96875,
      "epoch": 0.5931506849315068,
      "grad_norm": 4.675118446350098,
      "kl": 0.31787109375,
      "learning_rate": 8.022831050228311e-07,
      "loss": 0.0003,
      "reward": 1.640625,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 433
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.9375,
      "epoch": 0.5945205479452055,
      "grad_norm": 2.0734643936157227,
      "kl": 0.2802734375,
      "learning_rate": 8.018264840182648e-07,
      "loss": 0.0003,
      "reward": 1.671875,
      "reward_std": 0.15101328492164612,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 434
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 320.59375,
      "epoch": 0.5958904109589042,
      "grad_norm": 4.225161075592041,
      "kl": 0.31201171875,
      "learning_rate": 8.013698630136985e-07,
      "loss": 0.0003,
      "reward": 1.53125,
      "reward_std": 0.24511480331420898,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 435
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.125,
      "epoch": 0.5972602739726027,
      "grad_norm": 1.829791784286499,
      "kl": 0.298828125,
      "learning_rate": 8.009132420091324e-07,
      "loss": 0.0003,
      "reward": 1.66015625,
      "reward_std": 0.3770730784162879,
      "rewards/accuracy_reward": 0.6914062350988388,
      "rewards/format_reward": 0.96875,
      "step": 436
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 207.25,
      "epoch": 0.5986301369863014,
      "grad_norm": 3.862016439437866,
      "kl": 0.26318359375,
      "learning_rate": 8.004566210045661e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.3198433741927147,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 437
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 546.21875,
      "epoch": 0.6,
      "grad_norm": 3.7712178230285645,
      "kl": 0.283203125,
      "learning_rate": 8e-07,
      "loss": 0.0003,
      "reward": 1.5989583134651184,
      "reward_std": 0.37473164498806,
      "rewards/accuracy_reward": 0.6302083134651184,
      "rewards/format_reward": 0.96875,
      "step": 438
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 540.53125,
      "epoch": 0.6013698630136987,
      "grad_norm": 3.2269580364227295,
      "kl": 0.2978515625,
      "learning_rate": 7.995433789954338e-07,
      "loss": 0.0003,
      "reward": 1.40625,
      "reward_std": 0.18662459589540958,
      "rewards/accuracy_reward": 0.4062499850988388,
      "rewards/format_reward": 1.0,
      "step": 439
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 528.15625,
      "epoch": 0.6027397260273972,
      "grad_norm": 2.537423610687256,
      "kl": 0.30517578125,
      "learning_rate": 7.990867579908675e-07,
      "loss": 0.0003,
      "reward": 1.4791666567325592,
      "reward_std": 0.22078385017812252,
      "rewards/accuracy_reward": 0.4791666567325592,
      "rewards/format_reward": 1.0,
      "step": 440
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 448.1875,
      "epoch": 0.6041095890410959,
      "grad_norm": 2.5535776615142822,
      "kl": 0.28759765625,
      "learning_rate": 7.986301369863014e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.2167138308286667,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 441
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.71875,
      "epoch": 0.6054794520547945,
      "grad_norm": 2.6353836059570312,
      "kl": 0.27392578125,
      "learning_rate": 7.981735159817351e-07,
      "loss": 0.0003,
      "reward": 1.8463541567325592,
      "reward_std": 0.2513718083500862,
      "rewards/accuracy_reward": 0.8776041567325592,
      "rewards/format_reward": 0.96875,
      "step": 442
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.34375,
      "epoch": 0.6068493150684932,
      "grad_norm": 2.2128899097442627,
      "kl": 0.265625,
      "learning_rate": 7.977168949771688e-07,
      "loss": 0.0003,
      "reward": 1.859375,
      "reward_std": 0.17806704714894295,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 443
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 427.8125,
      "epoch": 0.6082191780821918,
      "grad_norm": 2.6061644554138184,
      "kl": 0.287109375,
      "learning_rate": 7.972602739726027e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.09127141162753105,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 444
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 308.46875,
      "epoch": 0.6095890410958904,
      "grad_norm": 13.305716514587402,
      "kl": 0.30712890625,
      "learning_rate": 7.968036529680365e-07,
      "loss": 0.0003,
      "reward": 1.4196428656578064,
      "reward_std": 0.2756448173895478,
      "rewards/accuracy_reward": 0.419642835855484,
      "rewards/format_reward": 1.0,
      "step": 445
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 561.03125,
      "epoch": 0.6109589041095891,
      "grad_norm": 3.1724693775177,
      "kl": 0.27685546875,
      "learning_rate": 7.963470319634703e-07,
      "loss": 0.0003,
      "reward": 1.296875,
      "reward_std": 0.1446593925356865,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 1.0,
      "step": 446
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.15625,
      "epoch": 0.6123287671232877,
      "grad_norm": 3.927133798599243,
      "kl": 0.369140625,
      "learning_rate": 7.958904109589041e-07,
      "loss": 0.0004,
      "reward": 1.564843773841858,
      "reward_std": 0.18459024094045162,
      "rewards/accuracy_reward": 0.5648437440395355,
      "rewards/format_reward": 1.0,
      "step": 447
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.625,
      "epoch": 0.6136986301369863,
      "grad_norm": 1.9001963138580322,
      "kl": 0.26806640625,
      "learning_rate": 7.954337899543378e-07,
      "loss": 0.0003,
      "reward": 1.6171875,
      "reward_std": 0.148872472345829,
      "rewards/accuracy_reward": 0.6171875149011612,
      "rewards/format_reward": 1.0,
      "step": 448
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.96875,
      "epoch": 0.6150684931506849,
      "grad_norm": 0.014812873676419258,
      "kl": 0.29345703125,
      "learning_rate": 7.949771689497717e-07,
      "loss": 0.0003,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 449
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 555.25,
      "epoch": 0.6164383561643836,
      "grad_norm": 4.2387614250183105,
      "kl": 0.345703125,
      "learning_rate": 7.945205479452054e-07,
      "loss": 0.0003,
      "reward": 1.5859375,
      "reward_std": 0.21959786862134933,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 1.0,
      "step": 450
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 483.5,
      "epoch": 0.6178082191780822,
      "grad_norm": 1.1416374444961548,
      "kl": 0.29345703125,
      "learning_rate": 7.940639269406393e-07,
      "loss": 0.0003,
      "reward": 1.4531249701976776,
      "reward_std": 0.11806126311421394,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 451
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.5,
      "epoch": 0.6191780821917808,
      "grad_norm": 2.7325708866119385,
      "kl": 0.296875,
      "learning_rate": 7.936073059360731e-07,
      "loss": 0.0003,
      "reward": 1.8020833432674408,
      "reward_std": 0.1516590639948845,
      "rewards/accuracy_reward": 0.8020833283662796,
      "rewards/format_reward": 1.0,
      "step": 452
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 698.15625,
      "epoch": 0.6205479452054794,
      "grad_norm": 1.962207317352295,
      "kl": 0.2685546875,
      "learning_rate": 7.931506849315068e-07,
      "loss": 0.0003,
      "reward": 1.4609375,
      "reward_std": 0.10126157477498055,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 1.0,
      "step": 453
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.25,
      "epoch": 0.6219178082191781,
      "grad_norm": 2.9467625617980957,
      "kl": 0.254150390625,
      "learning_rate": 7.926940639269406e-07,
      "loss": 0.0003,
      "reward": 1.7825521230697632,
      "reward_std": 0.37998438626527786,
      "rewards/accuracy_reward": 0.8138020634651184,
      "rewards/format_reward": 0.96875,
      "step": 454
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.5625,
      "epoch": 0.6232876712328768,
      "grad_norm": 2.3912293910980225,
      "kl": 0.299072265625,
      "learning_rate": 7.922374429223744e-07,
      "loss": 0.0003,
      "reward": 1.71484375,
      "reward_std": 0.41244056448340416,
      "rewards/accuracy_reward": 0.74609375,
      "rewards/format_reward": 0.96875,
      "step": 455
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 459.84375,
      "epoch": 0.6246575342465753,
      "grad_norm": 2.0481388568878174,
      "kl": 0.271240234375,
      "learning_rate": 7.917808219178081e-07,
      "loss": 0.0003,
      "reward": 1.4296875,
      "reward_std": 0.23987272381782532,
      "rewards/accuracy_reward": 0.4296875,
      "rewards/format_reward": 1.0,
      "step": 456
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.1875,
      "epoch": 0.626027397260274,
      "grad_norm": 1.0337698459625244,
      "kl": 0.2783203125,
      "learning_rate": 7.91324200913242e-07,
      "loss": 0.0003,
      "reward": 1.140625,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.140625,
      "rewards/format_reward": 1.0,
      "step": 457
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.8125,
      "epoch": 0.6273972602739726,
      "grad_norm": 3.085940361022949,
      "kl": 0.267578125,
      "learning_rate": 7.908675799086758e-07,
      "loss": 0.0003,
      "reward": 1.3984375,
      "reward_std": 0.21758441254496574,
      "rewards/accuracy_reward": 0.4296875,
      "rewards/format_reward": 0.96875,
      "step": 458
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 722.125,
      "epoch": 0.6287671232876713,
      "grad_norm": 2.2344439029693604,
      "kl": 0.243408203125,
      "learning_rate": 7.904109589041096e-07,
      "loss": 0.0002,
      "reward": 1.7083333730697632,
      "reward_std": 0.10860283300280571,
      "rewards/accuracy_reward": 0.7083333432674408,
      "rewards/format_reward": 1.0,
      "step": 459
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 448.84375,
      "epoch": 0.6301369863013698,
      "grad_norm": 2.974860429763794,
      "kl": 0.2705078125,
      "learning_rate": 7.899543378995434e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.11100946366786957,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 460
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 468.53125,
      "epoch": 0.6315068493150685,
      "grad_norm": 2.6200695037841797,
      "kl": 0.289794921875,
      "learning_rate": 7.894977168949771e-07,
      "loss": 0.0003,
      "reward": 1.265625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 461
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 731.0,
      "epoch": 0.6328767123287671,
      "grad_norm": 2.422001838684082,
      "kl": 0.307861328125,
      "learning_rate": 7.890410958904109e-07,
      "loss": 0.0003,
      "reward": 1.9296875,
      "reward_std": 0.31823596358299255,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 1.0,
      "step": 462
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 685.46875,
      "epoch": 0.6342465753424658,
      "grad_norm": 2.2778875827789307,
      "kl": 0.2734375,
      "learning_rate": 7.885844748858447e-07,
      "loss": 0.0003,
      "reward": 1.7734375,
      "reward_std": 0.1592222936451435,
      "rewards/accuracy_reward": 0.7734375,
      "rewards/format_reward": 1.0,
      "step": 463
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.90625,
      "epoch": 0.6356164383561644,
      "grad_norm": 1.6627824306488037,
      "kl": 0.251708984375,
      "learning_rate": 7.881278538812784e-07,
      "loss": 0.0003,
      "reward": 1.5598958134651184,
      "reward_std": 0.04366161487996578,
      "rewards/accuracy_reward": 0.5598958283662796,
      "rewards/format_reward": 1.0,
      "step": 464
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 450.0625,
      "epoch": 0.636986301369863,
      "grad_norm": 2.5756311416625977,
      "kl": 0.251220703125,
      "learning_rate": 7.876712328767124e-07,
      "loss": 0.0003,
      "reward": 1.203125,
      "reward_std": 0.2824692949652672,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 0.96875,
      "step": 465
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 682.375,
      "epoch": 0.6383561643835617,
      "grad_norm": 2.5796759128570557,
      "kl": 0.2705078125,
      "learning_rate": 7.872146118721461e-07,
      "loss": 0.0003,
      "reward": 1.4140625,
      "reward_std": 0.03234682232141495,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 466
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 731.21875,
      "epoch": 0.6397260273972603,
      "grad_norm": 2.126073122024536,
      "kl": 0.257568359375,
      "learning_rate": 7.867579908675798e-07,
      "loss": 0.0003,
      "reward": 1.7682291567325592,
      "reward_std": 0.19275827147066593,
      "rewards/accuracy_reward": 0.768229141831398,
      "rewards/format_reward": 1.0,
      "step": 467
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 554.9375,
      "epoch": 0.6410958904109589,
      "grad_norm": 1.904000163078308,
      "kl": 0.2880859375,
      "learning_rate": 7.863013698630137e-07,
      "loss": 0.0003,
      "reward": 1.65625,
      "reward_std": 0.1246790662407875,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 468
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 690.0,
      "epoch": 0.6424657534246575,
      "grad_norm": 1.2763168811798096,
      "kl": 0.28125,
      "learning_rate": 7.858447488584474e-07,
      "loss": 0.0003,
      "reward": 1.421875,
      "reward_std": 0.06646592170000076,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 469
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 689.65625,
      "epoch": 0.6438356164383562,
      "grad_norm": 1.074062705039978,
      "kl": 0.2880859375,
      "learning_rate": 7.853881278538812e-07,
      "loss": 0.0003,
      "reward": 1.578125,
      "reward_std": 0.09300297498703003,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.96875,
      "step": 470
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 320.03125,
      "epoch": 0.6452054794520548,
      "grad_norm": 7.623113632202148,
      "kl": 0.27197265625,
      "learning_rate": 7.849315068493151e-07,
      "loss": 0.0003,
      "reward": 1.28125,
      "reward_std": 0.3471629247069359,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 471
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 330.34375,
      "epoch": 0.6465753424657534,
      "grad_norm": 1.6552479267120361,
      "kl": 0.3212890625,
      "learning_rate": 7.844748858447488e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.16279494389891624,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 472
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 434.875,
      "epoch": 0.647945205479452,
      "grad_norm": 2.9016270637512207,
      "kl": 0.2880859375,
      "learning_rate": 7.840182648401827e-07,
      "loss": 0.0003,
      "reward": 1.421875,
      "reward_std": 0.3365500792860985,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.96875,
      "step": 473
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 436.46875,
      "epoch": 0.6493150684931507,
      "grad_norm": 2.371548652648926,
      "kl": 0.30126953125,
      "learning_rate": 7.835616438356164e-07,
      "loss": 0.0003,
      "reward": 1.6510416567325592,
      "reward_std": 0.15049929916858673,
      "rewards/accuracy_reward": 0.6510416716337204,
      "rewards/format_reward": 1.0,
      "step": 474
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 535.125,
      "epoch": 0.6506849315068494,
      "grad_norm": 1.709067940711975,
      "kl": 0.29541015625,
      "learning_rate": 7.831050228310501e-07,
      "loss": 0.0003,
      "reward": 1.5,
      "reward_std": 0.1825428232550621,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 475
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 317.53125,
      "epoch": 0.6520547945205479,
      "grad_norm": 0.8692818284034729,
      "kl": 0.29931640625,
      "learning_rate": 7.82648401826484e-07,
      "loss": 0.0003,
      "reward": 1.5729166269302368,
      "reward_std": 0.07952611148357391,
      "rewards/accuracy_reward": 0.5729166567325592,
      "rewards/format_reward": 1.0,
      "step": 476
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 441.5,
      "epoch": 0.6534246575342466,
      "grad_norm": 3.1770925521850586,
      "kl": 0.2998046875,
      "learning_rate": 7.821917808219177e-07,
      "loss": 0.0003,
      "reward": 1.875,
      "reward_std": 0.27724190801382065,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 477
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 734.28125,
      "epoch": 0.6547945205479452,
      "grad_norm": 6.152999401092529,
      "kl": 0.271484375,
      "learning_rate": 7.817351598173516e-07,
      "loss": 0.0003,
      "reward": 1.5234375,
      "reward_std": 0.11048543080687523,
      "rewards/accuracy_reward": 0.5234375,
      "rewards/format_reward": 1.0,
      "step": 478
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 575.75,
      "epoch": 0.6561643835616439,
      "grad_norm": 2.7290778160095215,
      "kl": 0.269287109375,
      "learning_rate": 7.812785388127854e-07,
      "loss": 0.0003,
      "reward": 1.5729166865348816,
      "reward_std": 0.18791258335113525,
      "rewards/accuracy_reward": 0.5729166865348816,
      "rewards/format_reward": 1.0,
      "step": 479
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 463.03125,
      "epoch": 0.6575342465753424,
      "grad_norm": 1.3419233560562134,
      "kl": 0.322021484375,
      "learning_rate": 7.808219178082191e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 480
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 624.03125,
      "epoch": 0.6589041095890411,
      "grad_norm": 2.2850310802459717,
      "kl": 0.2744140625,
      "learning_rate": 7.80365296803653e-07,
      "loss": 0.0003,
      "reward": 1.63671875,
      "reward_std": 0.17153325304389,
      "rewards/accuracy_reward": 0.63671875,
      "rewards/format_reward": 1.0,
      "step": 481
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.03125,
      "epoch": 0.6602739726027397,
      "grad_norm": 1.9118156433105469,
      "kl": 0.247802734375,
      "learning_rate": 7.799086757990867e-07,
      "loss": 0.0002,
      "reward": 1.53125,
      "reward_std": 0.102588826790452,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 482
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 325.625,
      "epoch": 0.6616438356164384,
      "grad_norm": 15.575130462646484,
      "kl": 0.304931640625,
      "learning_rate": 7.794520547945204e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.18861131370067596,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 483
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.46875,
      "epoch": 0.663013698630137,
      "grad_norm": 2.0344674587249756,
      "kl": 0.3017578125,
      "learning_rate": 7.789954337899543e-07,
      "loss": 0.0003,
      "reward": 1.5234375,
      "reward_std": 0.29708079993724823,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 0.96875,
      "step": 484
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 773.5625,
      "epoch": 0.6643835616438356,
      "grad_norm": 1.7942973375320435,
      "kl": 0.3359375,
      "learning_rate": 7.785388127853881e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.1514892801642418,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 485
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.84375,
      "epoch": 0.6657534246575343,
      "grad_norm": 4.269023418426514,
      "kl": 0.285400390625,
      "learning_rate": 7.780821917808219e-07,
      "loss": 0.0003,
      "reward": 1.359375,
      "reward_std": 0.43254324793815613,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 0.96875,
      "step": 486
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.53125,
      "epoch": 0.6671232876712329,
      "grad_norm": 1.3577572107315063,
      "kl": 0.267578125,
      "learning_rate": 7.776255707762557e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.13258251920342445,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 487
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 797.6875,
      "epoch": 0.6684931506849315,
      "grad_norm": 0.8058923482894897,
      "kl": 0.247802734375,
      "learning_rate": 7.771689497716894e-07,
      "loss": 0.0002,
      "reward": 1.15625,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 488
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.90625,
      "epoch": 0.6698630136986301,
      "grad_norm": 1.1615171432495117,
      "kl": 0.2861328125,
      "learning_rate": 7.767123287671233e-07,
      "loss": 0.0003,
      "reward": 1.3802083432674408,
      "reward_std": 0.0725951585918665,
      "rewards/accuracy_reward": 0.3802083432674408,
      "rewards/format_reward": 1.0,
      "step": 489
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.8125,
      "epoch": 0.6712328767123288,
      "grad_norm": 2.5159947872161865,
      "kl": 0.248779296875,
      "learning_rate": 7.76255707762557e-07,
      "loss": 0.0002,
      "reward": 1.2421875,
      "reward_std": 0.19887377880513668,
      "rewards/accuracy_reward": 0.2734375,
      "rewards/format_reward": 0.96875,
      "step": 490
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.34375,
      "epoch": 0.6726027397260274,
      "grad_norm": 3.6898646354675293,
      "kl": 0.26513671875,
      "learning_rate": 7.757990867579909e-07,
      "loss": 0.0003,
      "reward": 1.5572916269302368,
      "reward_std": 0.22042623907327652,
      "rewards/accuracy_reward": 0.5572916567325592,
      "rewards/format_reward": 1.0,
      "step": 491
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 443.15625,
      "epoch": 0.673972602739726,
      "grad_norm": 0.8622454404830933,
      "kl": 0.2578125,
      "learning_rate": 7.753424657534247e-07,
      "loss": 0.0003,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 492
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.40625,
      "epoch": 0.6753424657534246,
      "grad_norm": 6.587668418884277,
      "kl": 0.26904296875,
      "learning_rate": 7.748858447488584e-07,
      "loss": 0.0003,
      "reward": 1.7760416269302368,
      "reward_std": 0.11938937567174435,
      "rewards/accuracy_reward": 0.7760416567325592,
      "rewards/format_reward": 1.0,
      "step": 493
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.5625,
      "epoch": 0.6767123287671233,
      "grad_norm": 1.662431001663208,
      "kl": 0.2685546875,
      "learning_rate": 7.744292237442922e-07,
      "loss": 0.0003,
      "reward": 1.5390625,
      "reward_std": 0.15148437581956387,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 1.0,
      "step": 494
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 841.1875,
      "epoch": 0.678082191780822,
      "grad_norm": 2.413971424102783,
      "kl": 0.218994140625,
      "learning_rate": 7.73972602739726e-07,
      "loss": 0.0002,
      "reward": 1.2734375,
      "reward_std": 0.4280551001429558,
      "rewards/accuracy_reward": 0.3671875,
      "rewards/format_reward": 0.90625,
      "step": 495
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 699.34375,
      "epoch": 0.6794520547945205,
      "grad_norm": 1.9714837074279785,
      "kl": 0.2646484375,
      "learning_rate": 7.735159817351597e-07,
      "loss": 0.0003,
      "reward": 1.4921875,
      "reward_std": 0.43332719057798386,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 0.90625,
      "step": 496
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 355.53125,
      "epoch": 0.6808219178082192,
      "grad_norm": 5.755354881286621,
      "kl": 0.29296875,
      "learning_rate": 7.730593607305936e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.3945523276925087,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.96875,
      "step": 497
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 889.03125,
      "epoch": 0.6821917808219178,
      "grad_norm": 1.7405959367752075,
      "kl": 0.22509765625,
      "learning_rate": 7.726027397260274e-07,
      "loss": 0.0002,
      "reward": 1.7942708134651184,
      "reward_std": 0.7046211212873459,
      "rewards/accuracy_reward": 0.9192708432674408,
      "rewards/format_reward": 0.875,
      "step": 498
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 740.78125,
      "epoch": 0.6835616438356165,
      "grad_norm": 2.4895272254943848,
      "kl": 0.259521484375,
      "learning_rate": 7.721461187214611e-07,
      "loss": 0.0003,
      "reward": 1.3125,
      "reward_std": 0.6559129282832146,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.84375,
      "step": 499
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.9375,
      "epoch": 0.684931506849315,
      "grad_norm": 1.2708057165145874,
      "kl": 0.2705078125,
      "learning_rate": 7.71689497716895e-07,
      "loss": 0.0003,
      "reward": 1.078125,
      "reward_std": 0.3463020324707031,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 0.90625,
      "step": 500
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 385.6875,
      "epoch": 0.6863013698630137,
      "grad_norm": 2.6224188804626465,
      "kl": 0.279052734375,
      "learning_rate": 7.712328767123287e-07,
      "loss": 0.0003,
      "reward": 0.96875,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.03125,
      "rewards/format_reward": 0.9375,
      "step": 501
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.8125,
      "epoch": 0.6876712328767123,
      "grad_norm": 3.9175403118133545,
      "kl": 0.35009765625,
      "learning_rate": 7.707762557077625e-07,
      "loss": 0.0004,
      "reward": 1.625,
      "reward_std": 0.29514501616358757,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 502
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.9375,
      "epoch": 0.689041095890411,
      "grad_norm": 6.975749492645264,
      "kl": 0.266357421875,
      "learning_rate": 7.703196347031963e-07,
      "loss": 0.0003,
      "reward": 2.0,
      "reward_std": 0.45176807790994644,
      "rewards/accuracy_reward": 1.03125,
      "rewards/format_reward": 0.96875,
      "step": 503
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 349.9375,
      "epoch": 0.6904109589041096,
      "grad_norm": 5.680784225463867,
      "kl": 0.28369140625,
      "learning_rate": 7.6986301369863e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.4729364886879921,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 504
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 821.1875,
      "epoch": 0.6917808219178082,
      "grad_norm": 1.110270380973816,
      "kl": 0.23046875,
      "learning_rate": 7.69406392694064e-07,
      "loss": 0.0002,
      "reward": 0.9375,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.9375,
      "step": 505
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 656.78125,
      "epoch": 0.6931506849315069,
      "grad_norm": 1.595416784286499,
      "kl": 0.2509765625,
      "learning_rate": 7.689497716894977e-07,
      "loss": 0.0003,
      "reward": 1.8359375,
      "reward_std": 0.26719603314995766,
      "rewards/accuracy_reward": 0.8359375,
      "rewards/format_reward": 1.0,
      "step": 506
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.8125,
      "epoch": 0.6945205479452055,
      "grad_norm": 4.328030586242676,
      "kl": 0.26171875,
      "learning_rate": 7.684931506849314e-07,
      "loss": 0.0003,
      "reward": 1.234375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 507
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.65625,
      "epoch": 0.6958904109589041,
      "grad_norm": 0.007987846620380878,
      "kl": 0.261474609375,
      "learning_rate": 7.680365296803653e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 508
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.03125,
      "epoch": 0.6972602739726027,
      "grad_norm": 3.3370308876037598,
      "kl": 0.268310546875,
      "learning_rate": 7.67579908675799e-07,
      "loss": 0.0003,
      "reward": 1.1875,
      "reward_std": 0.2619796171784401,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 0.96875,
      "step": 509
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.25,
      "epoch": 0.6986301369863014,
      "grad_norm": 3.816755533218384,
      "kl": 0.287109375,
      "learning_rate": 7.671232876712328e-07,
      "loss": 0.0003,
      "reward": 1.09375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.09375,
      "rewards/format_reward": 1.0,
      "step": 510
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 431.0,
      "epoch": 0.7,
      "grad_norm": 0.028308337554335594,
      "kl": 0.325439453125,
      "learning_rate": 7.666666666666667e-07,
      "loss": 0.0003,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 511
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 758.375,
      "epoch": 0.7013698630136986,
      "grad_norm": 2.2149853706359863,
      "kl": 0.2958984375,
      "learning_rate": 7.662100456621004e-07,
      "loss": 0.0003,
      "reward": 1.6875,
      "reward_std": 0.05892554949969053,
      "rewards/accuracy_reward": 0.6874999850988388,
      "rewards/format_reward": 1.0,
      "step": 512
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 593.5625,
      "epoch": 0.7027397260273973,
      "grad_norm": 3.7374677658081055,
      "kl": 0.328857421875,
      "learning_rate": 7.657534246575343e-07,
      "loss": 0.0003,
      "reward": 1.7734375,
      "reward_std": 0.30776159279048443,
      "rewards/accuracy_reward": 0.7734375,
      "rewards/format_reward": 1.0,
      "step": 513
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.09375,
      "epoch": 0.7041095890410959,
      "grad_norm": 1.2948936223983765,
      "kl": 0.260986328125,
      "learning_rate": 7.65296803652968e-07,
      "loss": 0.0003,
      "reward": 1.6953125,
      "reward_std": 0.15467960201203823,
      "rewards/accuracy_reward": 0.6953125,
      "rewards/format_reward": 1.0,
      "step": 514
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 438.53125,
      "epoch": 0.7054794520547946,
      "grad_norm": 6.833772659301758,
      "kl": 0.3134765625,
      "learning_rate": 7.648401826484017e-07,
      "loss": 0.0003,
      "reward": 1.75,
      "reward_std": 0.22778154164552689,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 515
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 288.03125,
      "epoch": 0.7068493150684931,
      "grad_norm": 8.628284454345703,
      "kl": 0.315185546875,
      "learning_rate": 7.643835616438356e-07,
      "loss": 0.0003,
      "reward": 1.5,
      "reward_std": 0.18861131370067596,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 516
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 709.09375,
      "epoch": 0.7082191780821918,
      "grad_norm": 3.9952471256256104,
      "kl": 0.2568359375,
      "learning_rate": 7.639269406392693e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.08548713475465775,
      "rewards/accuracy_reward": 0.4374999850988388,
      "rewards/format_reward": 1.0,
      "step": 517
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.03125,
      "epoch": 0.7095890410958904,
      "grad_norm": 0.01444815844297409,
      "kl": 0.294921875,
      "learning_rate": 7.634703196347032e-07,
      "loss": 0.0003,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 518
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 435.21875,
      "epoch": 0.7109589041095891,
      "grad_norm": 3.509876251220703,
      "kl": 0.31005859375,
      "learning_rate": 7.63013698630137e-07,
      "loss": 0.0003,
      "reward": 1.640625,
      "reward_std": 0.3361537680029869,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 519
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.625,
      "epoch": 0.7123287671232876,
      "grad_norm": 2.812932252883911,
      "kl": 0.3154296875,
      "learning_rate": 7.625570776255707e-07,
      "loss": 0.0003,
      "reward": 1.578125,
      "reward_std": 0.2109457477927208,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 520
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 564.75,
      "epoch": 0.7136986301369863,
      "grad_norm": 1.0527565479278564,
      "kl": 0.28515625,
      "learning_rate": 7.621004566210046e-07,
      "loss": 0.0003,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 521
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.125,
      "epoch": 0.7150684931506849,
      "grad_norm": 1.4227604866027832,
      "kl": 0.26123046875,
      "learning_rate": 7.616438356164383e-07,
      "loss": 0.0003,
      "reward": 1.6328125,
      "reward_std": 0.20593809336423874,
      "rewards/accuracy_reward": 0.6328125,
      "rewards/format_reward": 1.0,
      "step": 522
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 411.34375,
      "epoch": 0.7164383561643836,
      "grad_norm": 1.042665719985962,
      "kl": 0.3056640625,
      "learning_rate": 7.61187214611872e-07,
      "loss": 0.0003,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 523
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 697.65625,
      "epoch": 0.7178082191780822,
      "grad_norm": 4.018407344818115,
      "kl": 0.2666015625,
      "learning_rate": 7.607305936073059e-07,
      "loss": 0.0003,
      "reward": 1.6953125,
      "reward_std": 0.16724733635783195,
      "rewards/accuracy_reward": 0.6953125,
      "rewards/format_reward": 1.0,
      "step": 524
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.78125,
      "epoch": 0.7191780821917808,
      "grad_norm": 1.4768673181533813,
      "kl": 0.28564453125,
      "learning_rate": 7.602739726027397e-07,
      "loss": 0.0003,
      "reward": 1.4114583134651184,
      "reward_std": 0.09781630150973797,
      "rewards/accuracy_reward": 0.4114583283662796,
      "rewards/format_reward": 1.0,
      "step": 525
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.15625,
      "epoch": 0.7205479452054795,
      "grad_norm": 1.8076670169830322,
      "kl": 0.254638671875,
      "learning_rate": 7.598173515981735e-07,
      "loss": 0.0003,
      "reward": 1.875,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 526
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 553.625,
      "epoch": 0.7219178082191781,
      "grad_norm": 2.97518253326416,
      "kl": 0.26708984375,
      "learning_rate": 7.593607305936073e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 527
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 690.125,
      "epoch": 0.7232876712328767,
      "grad_norm": 3.0159289836883545,
      "kl": 0.29248046875,
      "learning_rate": 7.58904109589041e-07,
      "loss": 0.0003,
      "reward": 1.6875,
      "reward_std": 0.3495672009885311,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.96875,
      "step": 528
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 547.53125,
      "epoch": 0.7246575342465753,
      "grad_norm": 0.024200087413191795,
      "kl": 0.3154296875,
      "learning_rate": 7.584474885844749e-07,
      "loss": 0.0003,
      "reward": 1.125,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 1.0,
      "step": 529
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.3125,
      "epoch": 0.726027397260274,
      "grad_norm": 16.51392936706543,
      "kl": 0.2783203125,
      "learning_rate": 7.579908675799086e-07,
      "loss": 0.0003,
      "reward": 1.7708333134651184,
      "reward_std": 0.24923127330839634,
      "rewards/accuracy_reward": 0.7708333283662796,
      "rewards/format_reward": 1.0,
      "step": 530
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 670.75,
      "epoch": 0.7273972602739726,
      "grad_norm": 2.6126623153686523,
      "kl": 0.275390625,
      "learning_rate": 7.575342465753424e-07,
      "loss": 0.0003,
      "reward": 1.7552083432674408,
      "reward_std": 0.14971192181110382,
      "rewards/accuracy_reward": 0.7552083432674408,
      "rewards/format_reward": 1.0,
      "step": 531
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.21875,
      "epoch": 0.7287671232876712,
      "grad_norm": 1.1948318481445312,
      "kl": 0.2802734375,
      "learning_rate": 7.570776255707763e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 532
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 527.375,
      "epoch": 0.7301369863013699,
      "grad_norm": 23.645620346069336,
      "kl": 0.320556640625,
      "learning_rate": 7.5662100456621e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 533
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 403.96875,
      "epoch": 0.7315068493150685,
      "grad_norm": 3.456782579421997,
      "kl": 0.36669921875,
      "learning_rate": 7.561643835616438e-07,
      "loss": 0.0004,
      "reward": 1.6354166269302368,
      "reward_std": 0.20183072239160538,
      "rewards/accuracy_reward": 0.6354166865348816,
      "rewards/format_reward": 1.0,
      "step": 534
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 277.625,
      "epoch": 0.7328767123287672,
      "grad_norm": 2.4733662605285645,
      "kl": 0.30615234375,
      "learning_rate": 7.557077625570776e-07,
      "loss": 0.0003,
      "reward": 1.15625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 535
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.875,
      "epoch": 0.7342465753424657,
      "grad_norm": 3.866555690765381,
      "kl": 0.3017578125,
      "learning_rate": 7.552511415525113e-07,
      "loss": 0.0003,
      "reward": 1.875,
      "reward_std": 0.2177756354212761,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 536
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 422.59375,
      "epoch": 0.7356164383561644,
      "grad_norm": 85.90696716308594,
      "kl": 0.3330078125,
      "learning_rate": 7.547945205479452e-07,
      "loss": 0.0003,
      "reward": 1.75,
      "reward_std": 0.22461533173918724,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 537
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 305.5625,
      "epoch": 0.736986301369863,
      "grad_norm": 5.220126628875732,
      "kl": 0.30078125,
      "learning_rate": 7.54337899543379e-07,
      "loss": 0.0003,
      "reward": 1.3125,
      "reward_std": 0.20044592767953873,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 538
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 261.40625,
      "epoch": 0.7383561643835617,
      "grad_norm": 5.28268575668335,
      "kl": 0.310546875,
      "learning_rate": 7.538812785388127e-07,
      "loss": 0.0003,
      "reward": 1.7890625,
      "reward_std": 0.26380185037851334,
      "rewards/accuracy_reward": 0.7890625,
      "rewards/format_reward": 1.0,
      "step": 539
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.09375,
      "epoch": 0.7397260273972602,
      "grad_norm": 2.411632537841797,
      "kl": 0.29296875,
      "learning_rate": 7.534246575342466e-07,
      "loss": 0.0003,
      "reward": 1.65625,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 540
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 563.8125,
      "epoch": 0.7410958904109589,
      "grad_norm": 1.7370656728744507,
      "kl": 0.31689453125,
      "learning_rate": 7.529680365296803e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 541
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.75,
      "epoch": 0.7424657534246575,
      "grad_norm": 1.8685580492019653,
      "kl": 0.2685546875,
      "learning_rate": 7.525114155251141e-07,
      "loss": 0.0003,
      "reward": 1.5052083432674408,
      "reward_std": 0.1630059964954853,
      "rewards/accuracy_reward": 0.5052083283662796,
      "rewards/format_reward": 1.0,
      "step": 542
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 416.1875,
      "epoch": 0.7438356164383562,
      "grad_norm": 4.826626777648926,
      "kl": 0.28369140625,
      "learning_rate": 7.520547945205479e-07,
      "loss": 0.0003,
      "reward": 1.546875,
      "reward_std": 0.319402813911438,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 543
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.8125,
      "epoch": 0.7452054794520548,
      "grad_norm": 1.5942007303237915,
      "kl": 0.27001953125,
      "learning_rate": 7.515981735159816e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 544
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 548.1875,
      "epoch": 0.7465753424657534,
      "grad_norm": 0.8855934739112854,
      "kl": 0.2890625,
      "learning_rate": 7.511415525114156e-07,
      "loss": 0.0003,
      "reward": 1.4453125,
      "reward_std": 0.0521576851606369,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 1.0,
      "step": 545
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 416.75,
      "epoch": 0.7479452054794521,
      "grad_norm": 9.443655967712402,
      "kl": 0.31640625,
      "learning_rate": 7.506849315068493e-07,
      "loss": 0.0003,
      "reward": 1.53125,
      "reward_std": 0.17358146235346794,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 546
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.25,
      "epoch": 0.7493150684931507,
      "grad_norm": 1.2628391981124878,
      "kl": 0.2822265625,
      "learning_rate": 7.50228310502283e-07,
      "loss": 0.0003,
      "reward": 1.478124976158142,
      "reward_std": 0.08902433887124062,
      "rewards/accuracy_reward": 0.4781249761581421,
      "rewards/format_reward": 1.0,
      "step": 547
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.21875,
      "epoch": 0.7506849315068493,
      "grad_norm": 2.95210599899292,
      "kl": 0.349609375,
      "learning_rate": 7.497716894977169e-07,
      "loss": 0.0004,
      "reward": 1.7135416865348816,
      "reward_std": 0.18521836958825588,
      "rewards/accuracy_reward": 0.7135416716337204,
      "rewards/format_reward": 1.0,
      "step": 548
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 717.96875,
      "epoch": 0.7520547945205479,
      "grad_norm": 17.074256896972656,
      "kl": 0.2978515625,
      "learning_rate": 7.493150684931506e-07,
      "loss": 0.0003,
      "reward": 1.6614583134651184,
      "reward_std": 0.1296813301742077,
      "rewards/accuracy_reward": 0.6614583432674408,
      "rewards/format_reward": 1.0,
      "step": 549
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 460.6875,
      "epoch": 0.7534246575342466,
      "grad_norm": 3.607158660888672,
      "kl": 0.283203125,
      "learning_rate": 7.488584474885844e-07,
      "loss": 0.0003,
      "reward": 1.640625,
      "reward_std": 0.3824852555990219,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.96875,
      "step": 550
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.0,
      "epoch": 0.7547945205479452,
      "grad_norm": 1.7502480745315552,
      "kl": 0.26708984375,
      "learning_rate": 7.484018264840183e-07,
      "loss": 0.0003,
      "reward": 1.6238839328289032,
      "reward_std": 0.12080634757876396,
      "rewards/accuracy_reward": 0.623883917927742,
      "rewards/format_reward": 1.0,
      "step": 551
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 528.1875,
      "epoch": 0.7561643835616438,
      "grad_norm": 6.180083274841309,
      "kl": 0.29345703125,
      "learning_rate": 7.47945205479452e-07,
      "loss": 0.0003,
      "reward": 1.8828125,
      "reward_std": 0.31077960692346096,
      "rewards/accuracy_reward": 0.8828124850988388,
      "rewards/format_reward": 1.0,
      "step": 552
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.1875,
      "epoch": 0.7575342465753425,
      "grad_norm": 0.9904180765151978,
      "kl": 0.2626953125,
      "learning_rate": 7.474885844748859e-07,
      "loss": 0.0003,
      "reward": 1.546875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 553
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 716.78125,
      "epoch": 0.7589041095890411,
      "grad_norm": 2.229074478149414,
      "kl": 0.24365234375,
      "learning_rate": 7.470319634703196e-07,
      "loss": 0.0002,
      "reward": 1.8260416388511658,
      "reward_std": 0.1378844864666462,
      "rewards/accuracy_reward": 0.826041653752327,
      "rewards/format_reward": 1.0,
      "step": 554
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.5625,
      "epoch": 0.7602739726027398,
      "grad_norm": 4.273021697998047,
      "kl": 0.277587890625,
      "learning_rate": 7.465753424657533e-07,
      "loss": 0.0003,
      "reward": 1.8828125,
      "reward_std": 0.15702588856220245,
      "rewards/accuracy_reward": 0.8828125,
      "rewards/format_reward": 1.0,
      "step": 555
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 304.03125,
      "epoch": 0.7616438356164383,
      "grad_norm": 0.02915891632437706,
      "kl": 0.322265625,
      "learning_rate": 7.461187214611872e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 556
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.90625,
      "epoch": 0.763013698630137,
      "grad_norm": 1.1693438291549683,
      "kl": 0.29638671875,
      "learning_rate": 7.456621004566209e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 557
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 465.09375,
      "epoch": 0.7643835616438356,
      "grad_norm": 0.9773749113082886,
      "kl": 0.3115234375,
      "learning_rate": 7.452054794520548e-07,
      "loss": 0.0003,
      "reward": 1.3046875,
      "reward_std": 0.09704047441482544,
      "rewards/accuracy_reward": 0.3046875,
      "rewards/format_reward": 1.0,
      "step": 558
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 331.03125,
      "epoch": 0.7657534246575343,
      "grad_norm": 6.028942108154297,
      "kl": 0.28271484375,
      "learning_rate": 7.447488584474886e-07,
      "loss": 0.0003,
      "reward": 1.765625,
      "reward_std": 0.10698894783854485,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 559
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 469.125,
      "epoch": 0.7671232876712328,
      "grad_norm": 6.394768714904785,
      "kl": 0.3134765625,
      "learning_rate": 7.442922374429223e-07,
      "loss": 0.0003,
      "reward": 1.609375,
      "reward_std": 0.17236988618969917,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 560
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 768.28125,
      "epoch": 0.7684931506849315,
      "grad_norm": 1.2454359531402588,
      "kl": 0.275390625,
      "learning_rate": 7.438356164383562e-07,
      "loss": 0.0003,
      "reward": 1.4140625,
      "reward_std": 0.09704046696424484,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 561
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 476.40625,
      "epoch": 0.7698630136986301,
      "grad_norm": 4.182791233062744,
      "kl": 0.27001953125,
      "learning_rate": 7.433789954337899e-07,
      "loss": 0.0003,
      "reward": 1.484375,
      "reward_std": 0.23144521936774254,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 562
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 430.8125,
      "epoch": 0.7712328767123288,
      "grad_norm": 2.8904168605804443,
      "kl": 0.3212890625,
      "learning_rate": 7.429223744292236e-07,
      "loss": 0.0003,
      "reward": 1.40625,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 563
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.15625,
      "epoch": 0.7726027397260274,
      "grad_norm": 10.556282997131348,
      "kl": 0.28759765625,
      "learning_rate": 7.424657534246575e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.047245558351278305,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 564
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 788.96875,
      "epoch": 0.773972602739726,
      "grad_norm": 2.3608124256134033,
      "kl": 0.26708984375,
      "learning_rate": 7.420091324200913e-07,
      "loss": 0.0003,
      "reward": 1.4427083730697632,
      "reward_std": 0.10979808866977692,
      "rewards/accuracy_reward": 0.4427083432674408,
      "rewards/format_reward": 1.0,
      "step": 565
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 600.4375,
      "epoch": 0.7753424657534247,
      "grad_norm": 1.940722942352295,
      "kl": 0.26904296875,
      "learning_rate": 7.415525114155251e-07,
      "loss": 0.0003,
      "reward": 1.78125,
      "reward_std": 0.16675157472491264,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 566
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.8125,
      "epoch": 0.7767123287671233,
      "grad_norm": 1.715049147605896,
      "kl": 0.27685546875,
      "learning_rate": 7.410958904109589e-07,
      "loss": 0.0003,
      "reward": 1.62109375,
      "reward_std": 0.20764102414250374,
      "rewards/accuracy_reward": 0.65234375,
      "rewards/format_reward": 0.96875,
      "step": 567
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 634.90625,
      "epoch": 0.7780821917808219,
      "grad_norm": 2.0080249309539795,
      "kl": 0.255859375,
      "learning_rate": 7.406392694063926e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 568
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 792.75,
      "epoch": 0.7794520547945205,
      "grad_norm": 1.9278701543807983,
      "kl": 0.27001953125,
      "learning_rate": 7.401826484018265e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.07312605157494545,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 569
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 298.28125,
      "epoch": 0.7808219178082192,
      "grad_norm": 0.017607873305678368,
      "kl": 0.31640625,
      "learning_rate": 7.397260273972602e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 570
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.0625,
      "epoch": 0.7821917808219178,
      "grad_norm": 4.558350086212158,
      "kl": 0.3037109375,
      "learning_rate": 7.39269406392694e-07,
      "loss": 0.0003,
      "reward": 1.25,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.96875,
      "step": 571
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 448.625,
      "epoch": 0.7835616438356164,
      "grad_norm": 5.511167049407959,
      "kl": 0.361328125,
      "learning_rate": 7.388127853881279e-07,
      "loss": 0.0004,
      "reward": 1.6484375,
      "reward_std": 0.2647695615887642,
      "rewards/accuracy_reward": 0.6796875,
      "rewards/format_reward": 0.96875,
      "step": 572
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.5,
      "epoch": 0.7849315068493151,
      "grad_norm": 3.1493077278137207,
      "kl": 0.3115234375,
      "learning_rate": 7.383561643835616e-07,
      "loss": 0.0003,
      "reward": 1.4453125,
      "reward_std": 0.09021057933568954,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 1.0,
      "step": 573
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.9375,
      "epoch": 0.7863013698630137,
      "grad_norm": 2.7337605953216553,
      "kl": 0.266357421875,
      "learning_rate": 7.378995433789954e-07,
      "loss": 0.0003,
      "reward": 1.5859375,
      "reward_std": 0.4949648827314377,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 0.875,
      "step": 574
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 803.21875,
      "epoch": 0.7876712328767124,
      "grad_norm": 1.5574761629104614,
      "kl": 0.269775390625,
      "learning_rate": 7.374429223744292e-07,
      "loss": 0.0003,
      "reward": 1.703125,
      "reward_std": 0.3444985970854759,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 0.96875,
      "step": 575
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.21875,
      "epoch": 0.7890410958904109,
      "grad_norm": 2.1322855949401855,
      "kl": 0.28125,
      "learning_rate": 7.369863013698629e-07,
      "loss": 0.0003,
      "reward": 1.6328125,
      "reward_std": 0.38375694304704666,
      "rewards/accuracy_reward": 0.6640625,
      "rewards/format_reward": 0.96875,
      "step": 576
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 811.1875,
      "epoch": 0.7904109589041096,
      "grad_norm": 2.1284778118133545,
      "kl": 0.27099609375,
      "learning_rate": 7.365296803652968e-07,
      "loss": 0.0003,
      "reward": 1.65625,
      "reward_std": 0.2121911160647869,
      "rewards/accuracy_reward": 0.6875000298023224,
      "rewards/format_reward": 0.96875,
      "step": 577
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 432.84375,
      "epoch": 0.7917808219178082,
      "grad_norm": 2.501584529876709,
      "kl": 0.2880859375,
      "learning_rate": 7.360730593607306e-07,
      "loss": 0.0003,
      "reward": 1.7604166269302368,
      "reward_std": 0.19149437546730042,
      "rewards/accuracy_reward": 0.7604166269302368,
      "rewards/format_reward": 1.0,
      "step": 578
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 709.5625,
      "epoch": 0.7931506849315069,
      "grad_norm": 2.271083354949951,
      "kl": 0.2861328125,
      "learning_rate": 7.356164383561643e-07,
      "loss": 0.0003,
      "reward": 1.7109375,
      "reward_std": 0.29368035681545734,
      "rewards/accuracy_reward": 0.7421875,
      "rewards/format_reward": 0.96875,
      "step": 579
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.21875,
      "epoch": 0.7945205479452054,
      "grad_norm": 0.75428706407547,
      "kl": 0.2734375,
      "learning_rate": 7.351598173515982e-07,
      "loss": 0.0003,
      "reward": 1.3828125,
      "reward_std": 0.26397860050201416,
      "rewards/accuracy_reward": 0.4453125,
      "rewards/format_reward": 0.9375,
      "step": 580
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.28125,
      "epoch": 0.7958904109589041,
      "grad_norm": 6.066575527191162,
      "kl": 0.29150390625,
      "learning_rate": 7.347031963470319e-07,
      "loss": 0.0003,
      "reward": 2.0625,
      "reward_std": 0.21965250372886658,
      "rewards/accuracy_reward": 1.0625,
      "rewards/format_reward": 1.0,
      "step": 581
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.875,
      "epoch": 0.7972602739726027,
      "grad_norm": 2.4375851154327393,
      "kl": 0.28125,
      "learning_rate": 7.342465753424657e-07,
      "loss": 0.0003,
      "reward": 1.7734375,
      "reward_std": 0.14333692379295826,
      "rewards/accuracy_reward": 0.7734375,
      "rewards/format_reward": 1.0,
      "step": 582
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 560.75,
      "epoch": 0.7986301369863014,
      "grad_norm": 4.929551601409912,
      "kl": 0.30712890625,
      "learning_rate": 7.337899543378995e-07,
      "loss": 0.0003,
      "reward": 1.5885416865348816,
      "reward_std": 0.33561520278453827,
      "rewards/accuracy_reward": 0.6197916716337204,
      "rewards/format_reward": 0.96875,
      "step": 583
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.96875,
      "epoch": 0.8,
      "grad_norm": 3.666161298751831,
      "kl": 0.294921875,
      "learning_rate": 7.333333333333332e-07,
      "loss": 0.0003,
      "reward": 1.6875,
      "reward_std": 0.20616560243070126,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 584
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 237.90625,
      "epoch": 0.8013698630136986,
      "grad_norm": 3.3543715476989746,
      "kl": 0.31201171875,
      "learning_rate": 7.328767123287672e-07,
      "loss": 0.0003,
      "reward": 1.4140625,
      "reward_std": 0.22327817976474762,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 585
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 377.5625,
      "epoch": 0.8027397260273973,
      "grad_norm": 4.409224987030029,
      "kl": 0.28857421875,
      "learning_rate": 7.324200913242009e-07,
      "loss": 0.0003,
      "reward": 1.875,
      "reward_std": 0.1934976615011692,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 586
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 545.78125,
      "epoch": 0.8041095890410959,
      "grad_norm": 2.494868278503418,
      "kl": 0.3046875,
      "learning_rate": 7.319634703196346e-07,
      "loss": 0.0003,
      "reward": 1.703125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 587
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 409.3125,
      "epoch": 0.8054794520547945,
      "grad_norm": 4.713694095611572,
      "kl": 0.3271484375,
      "learning_rate": 7.315068493150685e-07,
      "loss": 0.0003,
      "reward": 1.21875,
      "reward_std": 0.29305070638656616,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.9375,
      "step": 588
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.4375,
      "epoch": 0.8068493150684931,
      "grad_norm": 4.22615909576416,
      "kl": 0.288818359375,
      "learning_rate": 7.310502283105022e-07,
      "loss": 0.0003,
      "reward": 1.6979166269302368,
      "reward_std": 0.14746366813778877,
      "rewards/accuracy_reward": 0.6979166567325592,
      "rewards/format_reward": 1.0,
      "step": 589
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.46875,
      "epoch": 0.8082191780821918,
      "grad_norm": 3.3240842819213867,
      "kl": 0.29296875,
      "learning_rate": 7.30593607305936e-07,
      "loss": 0.0003,
      "reward": 1.8515625,
      "reward_std": 0.07912752032279968,
      "rewards/accuracy_reward": 0.8515625,
      "rewards/format_reward": 1.0,
      "step": 590
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 371.59375,
      "epoch": 0.8095890410958904,
      "grad_norm": 2.9559454917907715,
      "kl": 0.45556640625,
      "learning_rate": 7.301369863013699e-07,
      "loss": 0.0005,
      "reward": 1.3984375,
      "reward_std": 0.08679073117673397,
      "rewards/accuracy_reward": 0.3984375,
      "rewards/format_reward": 1.0,
      "step": 591
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 698.65625,
      "epoch": 0.810958904109589,
      "grad_norm": 6.2642693519592285,
      "kl": 0.29150390625,
      "learning_rate": 7.296803652968036e-07,
      "loss": 0.0003,
      "reward": 1.390625,
      "reward_std": 0.1825350895524025,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.96875,
      "step": 592
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.78125,
      "epoch": 0.8123287671232877,
      "grad_norm": 1.58978271484375,
      "kl": 0.2802734375,
      "learning_rate": 7.292237442922375e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.10669417306780815,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 593
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.75,
      "epoch": 0.8136986301369863,
      "grad_norm": 3.9518697261810303,
      "kl": 0.2890625,
      "learning_rate": 7.287671232876712e-07,
      "loss": 0.0003,
      "reward": 1.9453125,
      "reward_std": 0.26032389141619205,
      "rewards/accuracy_reward": 0.9453125,
      "rewards/format_reward": 1.0,
      "step": 594
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 398.9375,
      "epoch": 0.815068493150685,
      "grad_norm": 2.4345860481262207,
      "kl": 0.30078125,
      "learning_rate": 7.283105022831049e-07,
      "loss": 0.0003,
      "reward": 1.390625,
      "reward_std": 0.17358146235346794,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 595
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.90625,
      "epoch": 0.8164383561643835,
      "grad_norm": 1.7585220336914062,
      "kl": 0.260009765625,
      "learning_rate": 7.278538812785388e-07,
      "loss": 0.0003,
      "reward": 1.7447916269302368,
      "reward_std": 0.11100948229432106,
      "rewards/accuracy_reward": 0.7447916567325592,
      "rewards/format_reward": 1.0,
      "step": 596
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 414.03125,
      "epoch": 0.8178082191780822,
      "grad_norm": 10.006606101989746,
      "kl": 0.2744140625,
      "learning_rate": 7.273972602739725e-07,
      "loss": 0.0003,
      "reward": 1.3828125,
      "reward_std": 0.08679073117673397,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 597
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.84375,
      "epoch": 0.8191780821917808,
      "grad_norm": 3.1597371101379395,
      "kl": 0.293701171875,
      "learning_rate": 7.269406392694064e-07,
      "loss": 0.0003,
      "reward": 1.6015625,
      "reward_std": 0.1936504878103733,
      "rewards/accuracy_reward": 0.6015625,
      "rewards/format_reward": 1.0,
      "step": 598
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 715.8125,
      "epoch": 0.8205479452054795,
      "grad_norm": 1.5528578758239746,
      "kl": 0.28076171875,
      "learning_rate": 7.264840182648402e-07,
      "loss": 0.0003,
      "reward": 1.4765625,
      "reward_std": 0.07860555313527584,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 1.0,
      "step": 599
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 751.15625,
      "epoch": 0.821917808219178,
      "grad_norm": 1.6795881986618042,
      "kl": 0.312744140625,
      "learning_rate": 7.260273972602739e-07,
      "loss": 0.0003,
      "reward": 1.875,
      "reward_std": 0.21018434315919876,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 1.0,
      "step": 600
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 279.125,
      "epoch": 0.8232876712328767,
      "grad_norm": 1.0556623935699463,
      "kl": 0.32861328125,
      "learning_rate": 7.255707762557078e-07,
      "loss": 0.0003,
      "reward": 1.171875,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.171875,
      "rewards/format_reward": 1.0,
      "step": 601
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.03125,
      "epoch": 0.8246575342465754,
      "grad_norm": 4.618863105773926,
      "kl": 0.28955078125,
      "learning_rate": 7.251141552511415e-07,
      "loss": 0.0003,
      "reward": 1.609375,
      "reward_std": 0.3098084628582001,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 602
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 724.4375,
      "epoch": 0.826027397260274,
      "grad_norm": 3.4050896167755127,
      "kl": 0.2763671875,
      "learning_rate": 7.246575342465752e-07,
      "loss": 0.0003,
      "reward": 1.6015625,
      "reward_std": 0.1325697861611843,
      "rewards/accuracy_reward": 0.6015625,
      "rewards/format_reward": 1.0,
      "step": 603
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 779.09375,
      "epoch": 0.8273972602739726,
      "grad_norm": 2.157498836517334,
      "kl": 0.27197265625,
      "learning_rate": 7.242009132420091e-07,
      "loss": 0.0003,
      "reward": 2.125,
      "reward_std": 0.4762524124234915,
      "rewards/accuracy_reward": 1.15625,
      "rewards/format_reward": 0.96875,
      "step": 604
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 460.0,
      "epoch": 0.8287671232876712,
      "grad_norm": 2.8244545459747314,
      "kl": 0.29443359375,
      "learning_rate": 7.237442922374429e-07,
      "loss": 0.0003,
      "reward": 1.671875,
      "reward_std": 0.22097086161375046,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 605
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.0,
      "epoch": 0.8301369863013699,
      "grad_norm": 3.9018266201019287,
      "kl": 0.28271484375,
      "learning_rate": 7.232876712328767e-07,
      "loss": 0.0003,
      "reward": 1.5736607313156128,
      "reward_std": 0.1770894043147564,
      "rewards/accuracy_reward": 0.5736607313156128,
      "rewards/format_reward": 1.0,
      "step": 606
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 425.84375,
      "epoch": 0.8315068493150685,
      "grad_norm": 3.512537956237793,
      "kl": 0.3349609375,
      "learning_rate": 7.228310502283105e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.1472245752811432,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 607
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.84375,
      "epoch": 0.8328767123287671,
      "grad_norm": 2.6885643005371094,
      "kl": 0.28564453125,
      "learning_rate": 7.223744292237442e-07,
      "loss": 0.0003,
      "reward": 1.4765625,
      "reward_std": 0.29774628579616547,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 0.9375,
      "step": 608
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 468.8125,
      "epoch": 0.8342465753424657,
      "grad_norm": 2.3910205364227295,
      "kl": 0.345947265625,
      "learning_rate": 7.219178082191781e-07,
      "loss": 0.0003,
      "reward": 1.140625,
      "reward_std": 0.3463020324707031,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 0.9375,
      "step": 609
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 475.6875,
      "epoch": 0.8356164383561644,
      "grad_norm": 1.3065999746322632,
      "kl": 0.326171875,
      "learning_rate": 7.214611872146118e-07,
      "loss": 0.0003,
      "reward": 1.390625,
      "reward_std": 0.08456665836274624,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 610
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.9375,
      "epoch": 0.836986301369863,
      "grad_norm": 1.250556468963623,
      "kl": 0.34521484375,
      "learning_rate": 7.210045662100456e-07,
      "loss": 0.0003,
      "reward": 1.359375,
      "reward_std": 0.27564920112490654,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.9375,
      "step": 611
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 808.1875,
      "epoch": 0.8383561643835616,
      "grad_norm": 1.4332400560379028,
      "kl": 0.263671875,
      "learning_rate": 7.205479452054795e-07,
      "loss": 0.0003,
      "reward": 1.8890624940395355,
      "reward_std": 0.47034546732902527,
      "rewards/accuracy_reward": 0.9828124940395355,
      "rewards/format_reward": 0.90625,
      "step": 612
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 819.65625,
      "epoch": 0.8397260273972603,
      "grad_norm": 1.6501152515411377,
      "kl": 0.258544921875,
      "learning_rate": 7.200913242009132e-07,
      "loss": 0.0003,
      "reward": 1.75,
      "reward_std": 0.4389287531375885,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 0.9375,
      "step": 613
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.34375,
      "epoch": 0.8410958904109589,
      "grad_norm": 0.8255899548530579,
      "kl": 0.30078125,
      "learning_rate": 7.19634703196347e-07,
      "loss": 0.0003,
      "reward": 1.765625,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 614
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 467.28125,
      "epoch": 0.8424657534246576,
      "grad_norm": 0.9710884094238281,
      "kl": 0.30859375,
      "learning_rate": 7.191780821917808e-07,
      "loss": 0.0003,
      "reward": 1.484375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 615
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 790.28125,
      "epoch": 0.8438356164383561,
      "grad_norm": 1.5235772132873535,
      "kl": 0.28076171875,
      "learning_rate": 7.187214611872145e-07,
      "loss": 0.0003,
      "reward": 1.75,
      "reward_std": 0.15580293536186218,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 616
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 753.9375,
      "epoch": 0.8452054794520548,
      "grad_norm": 2.7367019653320312,
      "kl": 0.279296875,
      "learning_rate": 7.182648401826484e-07,
      "loss": 0.0003,
      "reward": 2.5677083134651184,
      "reward_std": 0.48400574177503586,
      "rewards/accuracy_reward": 1.5989583134651184,
      "rewards/format_reward": 0.96875,
      "step": 617
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 550.375,
      "epoch": 0.8465753424657534,
      "grad_norm": 3.4342260360717773,
      "kl": 0.29248046875,
      "learning_rate": 7.178082191780822e-07,
      "loss": 0.0003,
      "reward": 1.97265625,
      "reward_std": 0.23257054761052132,
      "rewards/accuracy_reward": 0.97265625,
      "rewards/format_reward": 1.0,
      "step": 618
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 558.65625,
      "epoch": 0.8479452054794521,
      "grad_norm": 5.605570316314697,
      "kl": 0.28662109375,
      "learning_rate": 7.173515981735159e-07,
      "loss": 0.0003,
      "reward": 1.8854166269302368,
      "reward_std": 0.23936405219137669,
      "rewards/accuracy_reward": 0.885416641831398,
      "rewards/format_reward": 1.0,
      "step": 619
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 275.84375,
      "epoch": 0.8493150684931506,
      "grad_norm": 2.4277820587158203,
      "kl": 0.318359375,
      "learning_rate": 7.168949771689498e-07,
      "loss": 0.0003,
      "reward": 1.6328125,
      "reward_std": 0.13782460056245327,
      "rewards/accuracy_reward": 0.6328125,
      "rewards/format_reward": 1.0,
      "step": 620
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 668.28125,
      "epoch": 0.8506849315068493,
      "grad_norm": 1.7949193716049194,
      "kl": 0.28271484375,
      "learning_rate": 7.164383561643835e-07,
      "loss": 0.0003,
      "reward": 1.6276041567325592,
      "reward_std": 0.09027346037328243,
      "rewards/accuracy_reward": 0.6276041567325592,
      "rewards/format_reward": 1.0,
      "step": 621
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 706.1875,
      "epoch": 0.852054794520548,
      "grad_norm": 3.527575969696045,
      "kl": 0.27978515625,
      "learning_rate": 7.159817351598173e-07,
      "loss": 0.0003,
      "reward": 1.421875,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 622
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 700.46875,
      "epoch": 0.8534246575342466,
      "grad_norm": 1.6975399255752563,
      "kl": 0.28515625,
      "learning_rate": 7.155251141552511e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.0936255231499672,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 623
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.59375,
      "epoch": 0.8547945205479452,
      "grad_norm": 4.774352550506592,
      "kl": 0.3291015625,
      "learning_rate": 7.150684931506848e-07,
      "loss": 0.0003,
      "reward": 1.6901041865348816,
      "reward_std": 0.15677691251039505,
      "rewards/accuracy_reward": 0.690104141831398,
      "rewards/format_reward": 1.0,
      "step": 624
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.5625,
      "epoch": 0.8561643835616438,
      "grad_norm": 3.3761487007141113,
      "kl": 0.3056640625,
      "learning_rate": 7.146118721461188e-07,
      "loss": 0.0003,
      "reward": 1.68359375,
      "reward_std": 0.26938531920313835,
      "rewards/accuracy_reward": 0.68359375,
      "rewards/format_reward": 1.0,
      "step": 625
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 267.375,
      "epoch": 0.8575342465753425,
      "grad_norm": 3.5964181423187256,
      "kl": 0.38427734375,
      "learning_rate": 7.141552511415525e-07,
      "loss": 0.0004,
      "reward": 1.515625,
      "reward_std": 0.17782479152083397,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 626
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 200.78125,
      "epoch": 0.8589041095890411,
      "grad_norm": 1.3047560453414917,
      "kl": 0.3671875,
      "learning_rate": 7.136986301369862e-07,
      "loss": 0.0004,
      "reward": 1.453125,
      "reward_std": 0.02670290134847164,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 627
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 500.15625,
      "epoch": 0.8602739726027397,
      "grad_norm": 3.4180963039398193,
      "kl": 0.37451171875,
      "learning_rate": 7.132420091324201e-07,
      "loss": 0.0004,
      "reward": 1.75,
      "reward_std": 0.11659536883234978,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 628
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.59375,
      "epoch": 0.8616438356164383,
      "grad_norm": 3.0317859649658203,
      "kl": 0.31884765625,
      "learning_rate": 7.127853881278538e-07,
      "loss": 0.0003,
      "reward": 1.9609375,
      "reward_std": 0.40251583606004715,
      "rewards/accuracy_reward": 0.9921875,
      "rewards/format_reward": 0.96875,
      "step": 629
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 349.46875,
      "epoch": 0.863013698630137,
      "grad_norm": 5.503389358520508,
      "kl": 0.404296875,
      "learning_rate": 7.123287671232876e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.3335031494498253,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 630
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 446.84375,
      "epoch": 0.8643835616438356,
      "grad_norm": 1.256745457649231,
      "kl": 0.33642578125,
      "learning_rate": 7.118721461187215e-07,
      "loss": 0.0003,
      "reward": 1.6197916269302368,
      "reward_std": 0.051934316754341125,
      "rewards/accuracy_reward": 0.6197916269302368,
      "rewards/format_reward": 1.0,
      "step": 631
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 481.125,
      "epoch": 0.8657534246575342,
      "grad_norm": 2.3345208168029785,
      "kl": 0.35302734375,
      "learning_rate": 7.114155251141552e-07,
      "loss": 0.0004,
      "reward": 1.40625,
      "reward_std": 0.10888781771063805,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 632
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 465.625,
      "epoch": 0.8671232876712329,
      "grad_norm": 2.5264086723327637,
      "kl": 0.33544921875,
      "learning_rate": 7.109589041095891e-07,
      "loss": 0.0003,
      "reward": 1.609375,
      "reward_std": 0.2382849156856537,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 1.0,
      "step": 633
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.78125,
      "epoch": 0.8684931506849315,
      "grad_norm": 2.600130081176758,
      "kl": 0.34765625,
      "learning_rate": 7.105022831050228e-07,
      "loss": 0.0003,
      "reward": 1.9562499523162842,
      "reward_std": 0.1930173598229885,
      "rewards/accuracy_reward": 0.956250011920929,
      "rewards/format_reward": 1.0,
      "step": 634
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 490.625,
      "epoch": 0.8698630136986302,
      "grad_norm": 3.292919874191284,
      "kl": 0.35791015625,
      "learning_rate": 7.100456621004565e-07,
      "loss": 0.0004,
      "reward": 1.8541666567325592,
      "reward_std": 0.193493926897645,
      "rewards/accuracy_reward": 0.8541666716337204,
      "rewards/format_reward": 1.0,
      "step": 635
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.59375,
      "epoch": 0.8712328767123287,
      "grad_norm": 14.443897247314453,
      "kl": 0.42919921875,
      "learning_rate": 7.095890410958904e-07,
      "loss": 0.0004,
      "reward": 1.1875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 636
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 234.125,
      "epoch": 0.8726027397260274,
      "grad_norm": 1.4970674514770508,
      "kl": 0.3974609375,
      "learning_rate": 7.091324200913241e-07,
      "loss": 0.0004,
      "reward": 1.2265625,
      "reward_std": 0.03234682232141495,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 1.0,
      "step": 637
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 227.375,
      "epoch": 0.873972602739726,
      "grad_norm": 4.312900543212891,
      "kl": 0.39453125,
      "learning_rate": 7.08675799086758e-07,
      "loss": 0.0004,
      "reward": 1.484375,
      "reward_std": 0.27092268504202366,
      "rewards/accuracy_reward": 0.4843750298023224,
      "rewards/format_reward": 1.0,
      "step": 638
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 266.6875,
      "epoch": 0.8753424657534247,
      "grad_norm": 1.0829259157180786,
      "kl": 0.49462890625,
      "learning_rate": 7.082191780821918e-07,
      "loss": 0.0005,
      "reward": 1.3125,
      "reward_std": 0.10022296756505966,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 639
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 379.5,
      "epoch": 0.8767123287671232,
      "grad_norm": 5.12158203125,
      "kl": 0.37353515625,
      "learning_rate": 7.077625570776255e-07,
      "loss": 0.0004,
      "reward": 1.375,
      "reward_std": 0.25693800300359726,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.96875,
      "step": 640
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 413.96875,
      "epoch": 0.8780821917808219,
      "grad_norm": 3.457838773727417,
      "kl": 0.376953125,
      "learning_rate": 7.073059360730594e-07,
      "loss": 0.0004,
      "reward": 1.578125,
      "reward_std": 0.30377669632434845,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 641
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.09375,
      "epoch": 0.8794520547945206,
      "grad_norm": 3.6919641494750977,
      "kl": 0.31689453125,
      "learning_rate": 7.068493150684931e-07,
      "loss": 0.0003,
      "reward": 1.859375,
      "reward_std": 0.34486518055200577,
      "rewards/accuracy_reward": 0.890625,
      "rewards/format_reward": 0.96875,
      "step": 642
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.125,
      "epoch": 0.8808219178082192,
      "grad_norm": 1.8475896120071411,
      "kl": 0.34619140625,
      "learning_rate": 7.063926940639268e-07,
      "loss": 0.0003,
      "reward": 1.78125,
      "reward_std": 0.4355708882212639,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 0.84375,
      "step": 643
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 680.34375,
      "epoch": 0.8821917808219178,
      "grad_norm": 2.7510480880737305,
      "kl": 0.29150390625,
      "learning_rate": 7.059360730593607e-07,
      "loss": 0.0003,
      "reward": 1.578125,
      "reward_std": 0.2629348188638687,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.9375,
      "step": 644
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 421.53125,
      "epoch": 0.8835616438356164,
      "grad_norm": 4.39385986328125,
      "kl": 0.3525390625,
      "learning_rate": 7.054794520547945e-07,
      "loss": 0.0004,
      "reward": 1.125,
      "reward_std": 0.3104073107242584,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.9375,
      "step": 645
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 332.90625,
      "epoch": 0.8849315068493151,
      "grad_norm": 3.439211368560791,
      "kl": 0.32080078125,
      "learning_rate": 7.050228310502283e-07,
      "loss": 0.0003,
      "reward": 1.15625,
      "reward_std": 0.3438149690628052,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 0.9375,
      "step": 646
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 331.5625,
      "epoch": 0.8863013698630137,
      "grad_norm": 2.4772706031799316,
      "kl": 0.3466796875,
      "learning_rate": 7.045662100456621e-07,
      "loss": 0.0003,
      "reward": 1.15625,
      "reward_std": 0.22201896458864212,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 647
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 590.21875,
      "epoch": 0.8876712328767123,
      "grad_norm": 1.3561179637908936,
      "kl": 0.341796875,
      "learning_rate": 7.041095890410958e-07,
      "loss": 0.0003,
      "reward": 1.4765625,
      "reward_std": 0.10474801808595657,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 1.0,
      "step": 648
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 464.78125,
      "epoch": 0.8890410958904109,
      "grad_norm": 45.053157806396484,
      "kl": 0.33154296875,
      "learning_rate": 7.036529680365297e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.4626970961689949,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.9375,
      "step": 649
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 336.5,
      "epoch": 0.8904109589041096,
      "grad_norm": 3.172072649002075,
      "kl": 0.3642578125,
      "learning_rate": 7.031963470319634e-07,
      "loss": 0.0004,
      "reward": 1.3515625,
      "reward_std": 0.3067816346883774,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 0.96875,
      "step": 650
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 448.53125,
      "epoch": 0.8917808219178082,
      "grad_norm": 1.1374866962432861,
      "kl": 0.3310546875,
      "learning_rate": 7.027397260273972e-07,
      "loss": 0.0003,
      "reward": 1.184374988079071,
      "reward_std": 0.022903122007846832,
      "rewards/accuracy_reward": 0.18437501788139343,
      "rewards/format_reward": 1.0,
      "step": 651
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.3125,
      "epoch": 0.8931506849315068,
      "grad_norm": 2.0539586544036865,
      "kl": 0.29931640625,
      "learning_rate": 7.022831050228311e-07,
      "loss": 0.0003,
      "reward": 1.6953125,
      "reward_std": 0.24830512702465057,
      "rewards/accuracy_reward": 0.7265625,
      "rewards/format_reward": 0.96875,
      "step": 652
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 467.65625,
      "epoch": 0.8945205479452055,
      "grad_norm": 3.450404405593872,
      "kl": 0.3037109375,
      "learning_rate": 7.018264840182648e-07,
      "loss": 0.0003,
      "reward": 1.65625,
      "reward_std": 0.49012404680252075,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.96875,
      "step": 653
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 453.96875,
      "epoch": 0.8958904109589041,
      "grad_norm": 5.490792274475098,
      "kl": 0.3271484375,
      "learning_rate": 7.013698630136986e-07,
      "loss": 0.0003,
      "reward": 1.2750000059604645,
      "reward_std": 0.13220207020640373,
      "rewards/accuracy_reward": 0.2750000059604645,
      "rewards/format_reward": 1.0,
      "step": 654
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 474.78125,
      "epoch": 0.8972602739726028,
      "grad_norm": 4.403593063354492,
      "kl": 0.3408203125,
      "learning_rate": 7.009132420091324e-07,
      "loss": 0.0003,
      "reward": 1.203125,
      "reward_std": 0.30935921147465706,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 0.9375,
      "step": 655
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.5,
      "epoch": 0.8986301369863013,
      "grad_norm": 1.4611150026321411,
      "kl": 0.3193359375,
      "learning_rate": 7.004566210045661e-07,
      "loss": 0.0003,
      "reward": 1.359375,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 0.96875,
      "step": 656
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 347.375,
      "epoch": 0.9,
      "grad_norm": 3.657111644744873,
      "kl": 0.330078125,
      "learning_rate": 7e-07,
      "loss": 0.0003,
      "reward": 1.640625,
      "reward_std": 0.28412990644574165,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 657
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 547.5625,
      "epoch": 0.9013698630136986,
      "grad_norm": 2.3929004669189453,
      "kl": 0.3193359375,
      "learning_rate": 6.995433789954338e-07,
      "loss": 0.0003,
      "reward": 1.859375,
      "reward_std": 0.30617378652095795,
      "rewards/accuracy_reward": 0.859375,
      "rewards/format_reward": 1.0,
      "step": 658
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.4375,
      "epoch": 0.9027397260273973,
      "grad_norm": 2.0510473251342773,
      "kl": 0.3330078125,
      "learning_rate": 6.990867579908675e-07,
      "loss": 0.0003,
      "reward": 1.6953125,
      "reward_std": 0.2706219367682934,
      "rewards/accuracy_reward": 0.6953125,
      "rewards/format_reward": 1.0,
      "step": 659
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.28125,
      "epoch": 0.9041095890410958,
      "grad_norm": 9.588709831237793,
      "kl": 0.32958984375,
      "learning_rate": 6.986301369863014e-07,
      "loss": 0.0003,
      "reward": 1.9114583730697632,
      "reward_std": 0.13045240752398968,
      "rewards/accuracy_reward": 0.9114583432674408,
      "rewards/format_reward": 1.0,
      "step": 660
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 266.28125,
      "epoch": 0.9054794520547945,
      "grad_norm": 3.2910804748535156,
      "kl": 0.373046875,
      "learning_rate": 6.981735159817351e-07,
      "loss": 0.0004,
      "reward": 1.53125,
      "reward_std": 0.3061639815568924,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.96875,
      "step": 661
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.09375,
      "epoch": 0.9068493150684932,
      "grad_norm": 1.6819573640823364,
      "kl": 0.32080078125,
      "learning_rate": 6.977168949771689e-07,
      "loss": 0.0003,
      "reward": 1.5052083134651184,
      "reward_std": 0.11100949719548225,
      "rewards/accuracy_reward": 0.5052083134651184,
      "rewards/format_reward": 1.0,
      "step": 662
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 657.21875,
      "epoch": 0.9082191780821918,
      "grad_norm": 1.5459468364715576,
      "kl": 0.3544921875,
      "learning_rate": 6.972602739726027e-07,
      "loss": 0.0004,
      "reward": 1.453125,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 663
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.4375,
      "epoch": 0.9095890410958904,
      "grad_norm": 2.212812900543213,
      "kl": 0.3076171875,
      "learning_rate": 6.968036529680364e-07,
      "loss": 0.0003,
      "reward": 1.7638888359069824,
      "reward_std": 0.22503389045596123,
      "rewards/accuracy_reward": 0.7951388657093048,
      "rewards/format_reward": 0.96875,
      "step": 664
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 762.3125,
      "epoch": 0.910958904109589,
      "grad_norm": 2.2994203567504883,
      "kl": 0.29345703125,
      "learning_rate": 6.963470319634704e-07,
      "loss": 0.0003,
      "reward": 1.9127604067325592,
      "reward_std": 0.33594274893403053,
      "rewards/accuracy_reward": 0.9440104067325592,
      "rewards/format_reward": 0.96875,
      "step": 665
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.3125,
      "epoch": 0.9123287671232877,
      "grad_norm": 6.928981304168701,
      "kl": 0.2998046875,
      "learning_rate": 6.958904109589041e-07,
      "loss": 0.0003,
      "reward": 1.2291666567325592,
      "reward_std": 0.022271782159805298,
      "rewards/accuracy_reward": 0.2291666567325592,
      "rewards/format_reward": 1.0,
      "step": 666
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 253.15625,
      "epoch": 0.9136986301369863,
      "grad_norm": 7.005854606628418,
      "kl": 0.3505859375,
      "learning_rate": 6.954337899543378e-07,
      "loss": 0.0004,
      "reward": 1.3125,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 667
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 367.78125,
      "epoch": 0.915068493150685,
      "grad_norm": 4.68707275390625,
      "kl": 0.37255859375,
      "learning_rate": 6.949771689497717e-07,
      "loss": 0.0004,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.96875,
      "step": 668
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 447.375,
      "epoch": 0.9164383561643835,
      "grad_norm": 3.7123045921325684,
      "kl": 0.380859375,
      "learning_rate": 6.945205479452054e-07,
      "loss": 0.0004,
      "reward": 1.671875,
      "reward_std": 0.41054617613554,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 669
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.65625,
      "epoch": 0.9178082191780822,
      "grad_norm": 1.5095117092132568,
      "kl": 0.30859375,
      "learning_rate": 6.940639269406392e-07,
      "loss": 0.0003,
      "reward": 1.44921875,
      "reward_std": 0.055597566068172455,
      "rewards/accuracy_reward": 0.44921875,
      "rewards/format_reward": 1.0,
      "step": 670
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.25,
      "epoch": 0.9191780821917809,
      "grad_norm": 1.7842953205108643,
      "kl": 0.40478515625,
      "learning_rate": 6.93607305936073e-07,
      "loss": 0.0004,
      "reward": 1.1796875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.2109375,
      "rewards/format_reward": 0.96875,
      "step": 671
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 496.84375,
      "epoch": 0.9205479452054794,
      "grad_norm": 3.0466763973236084,
      "kl": 0.32470703125,
      "learning_rate": 6.931506849315068e-07,
      "loss": 0.0003,
      "reward": 1.62109375,
      "reward_std": 0.17449257895350456,
      "rewards/accuracy_reward": 0.62109375,
      "rewards/format_reward": 1.0,
      "step": 672
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.59375,
      "epoch": 0.9219178082191781,
      "grad_norm": 6.0069193840026855,
      "kl": 0.3232421875,
      "learning_rate": 6.926940639269407e-07,
      "loss": 0.0003,
      "reward": 1.4479166567325592,
      "reward_std": 0.17747542820870876,
      "rewards/accuracy_reward": 0.4479166567325592,
      "rewards/format_reward": 1.0,
      "step": 673
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.59375,
      "epoch": 0.9232876712328767,
      "grad_norm": 2.055826425552368,
      "kl": 0.3134765625,
      "learning_rate": 6.922374429223744e-07,
      "loss": 0.0003,
      "reward": 1.75,
      "reward_std": 0.24671732261776924,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 674
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.25,
      "epoch": 0.9246575342465754,
      "grad_norm": 0.03342582285404205,
      "kl": 0.3740234375,
      "learning_rate": 6.917808219178081e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 675
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 472.21875,
      "epoch": 0.9260273972602739,
      "grad_norm": 2.0904788970947266,
      "kl": 0.34375,
      "learning_rate": 6.91324200913242e-07,
      "loss": 0.0003,
      "reward": 1.3541666865348816,
      "reward_std": 0.10767630115151405,
      "rewards/accuracy_reward": 0.3541666865348816,
      "rewards/format_reward": 1.0,
      "step": 676
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 589.59375,
      "epoch": 0.9273972602739726,
      "grad_norm": 2.1896920204162598,
      "kl": 0.32666015625,
      "learning_rate": 6.908675799086757e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.20648781582713127,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 0.96875,
      "step": 677
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.15625,
      "epoch": 0.9287671232876712,
      "grad_norm": 1.9878590106964111,
      "kl": 0.32666015625,
      "learning_rate": 6.904109589041097e-07,
      "loss": 0.0003,
      "reward": 1.546875,
      "reward_std": 0.1446593925356865,
      "rewards/accuracy_reward": 0.609375,
      "rewards/format_reward": 0.9375,
      "step": 678
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.78125,
      "epoch": 0.9301369863013699,
      "grad_norm": 4.627434253692627,
      "kl": 0.3154296875,
      "learning_rate": 6.899543378995434e-07,
      "loss": 0.0003,
      "reward": 1.811718761920929,
      "reward_std": 0.12913026381283998,
      "rewards/accuracy_reward": 0.811718761920929,
      "rewards/format_reward": 1.0,
      "step": 679
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 703.84375,
      "epoch": 0.9315068493150684,
      "grad_norm": 2.76521897315979,
      "kl": 0.3232421875,
      "learning_rate": 6.894977168949771e-07,
      "loss": 0.0003,
      "reward": 2.1796875,
      "reward_std": 0.18400542438030243,
      "rewards/accuracy_reward": 1.1796875,
      "rewards/format_reward": 1.0,
      "step": 680
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 573.75,
      "epoch": 0.9328767123287671,
      "grad_norm": 3.8087267875671387,
      "kl": 0.43505859375,
      "learning_rate": 6.89041095890411e-07,
      "loss": 0.0004,
      "reward": 1.359375,
      "reward_std": 0.17782479152083397,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 681
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 346.0625,
      "epoch": 0.9342465753424658,
      "grad_norm": 1.986623764038086,
      "kl": 0.35986328125,
      "learning_rate": 6.885844748858447e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 682
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.40625,
      "epoch": 0.9356164383561644,
      "grad_norm": 2.0661747455596924,
      "kl": 0.3330078125,
      "learning_rate": 6.881278538812784e-07,
      "loss": 0.0003,
      "reward": 1.515625,
      "reward_std": 0.18997547030448914,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 683
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 703.46875,
      "epoch": 0.936986301369863,
      "grad_norm": 13.223034858703613,
      "kl": 0.33251953125,
      "learning_rate": 6.876712328767123e-07,
      "loss": 0.0003,
      "reward": 2.14453125,
      "reward_std": 0.1712184101343155,
      "rewards/accuracy_reward": 1.14453125,
      "rewards/format_reward": 1.0,
      "step": 684
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 377.0625,
      "epoch": 0.9383561643835616,
      "grad_norm": 3.540472984313965,
      "kl": 0.341796875,
      "learning_rate": 6.872146118721461e-07,
      "loss": 0.0003,
      "reward": 1.37109375,
      "reward_std": 0.14941447600722313,
      "rewards/accuracy_reward": 0.37109375,
      "rewards/format_reward": 1.0,
      "step": 685
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.71875,
      "epoch": 0.9397260273972603,
      "grad_norm": 4.10709285736084,
      "kl": 0.36865234375,
      "learning_rate": 6.867579908675799e-07,
      "loss": 0.0004,
      "reward": 1.09375,
      "reward_std": 0.22201896458864212,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.96875,
      "step": 686
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 766.8125,
      "epoch": 0.9410958904109589,
      "grad_norm": 2.0390625,
      "kl": 0.3642578125,
      "learning_rate": 6.863013698630137e-07,
      "loss": 0.0004,
      "reward": 1.7942708134651184,
      "reward_std": 0.28445861861109734,
      "rewards/accuracy_reward": 0.8255208432674408,
      "rewards/format_reward": 0.96875,
      "step": 687
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 573.40625,
      "epoch": 0.9424657534246575,
      "grad_norm": 2.7066261768341064,
      "kl": 0.3212890625,
      "learning_rate": 6.858447488584474e-07,
      "loss": 0.0003,
      "reward": 1.390625,
      "reward_std": 0.10205793008208275,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 688
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 447.59375,
      "epoch": 0.9438356164383561,
      "grad_norm": 1.330697774887085,
      "kl": 0.36767578125,
      "learning_rate": 6.853881278538813e-07,
      "loss": 0.0004,
      "reward": 1.40625,
      "reward_std": 0.22301282733678818,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 689
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.25,
      "epoch": 0.9452054794520548,
      "grad_norm": 1.8517435789108276,
      "kl": 0.3125,
      "learning_rate": 6.84931506849315e-07,
      "loss": 0.0003,
      "reward": 1.4921875,
      "reward_std": 0.23285578191280365,
      "rewards/accuracy_reward": 0.5234375,
      "rewards/format_reward": 0.96875,
      "step": 690
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.96875,
      "epoch": 0.9465753424657535,
      "grad_norm": 7.800582408905029,
      "kl": 0.33154296875,
      "learning_rate": 6.844748858447487e-07,
      "loss": 0.0003,
      "reward": 1.21875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 691
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 326.5,
      "epoch": 0.947945205479452,
      "grad_norm": 2.3165299892425537,
      "kl": 0.3642578125,
      "learning_rate": 6.840182648401827e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 692
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 488.1875,
      "epoch": 0.9493150684931507,
      "grad_norm": 1.7786669731140137,
      "kl": 0.31591796875,
      "learning_rate": 6.835616438356164e-07,
      "loss": 0.0003,
      "reward": 1.3828125,
      "reward_std": 0.16173411160707474,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 693
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 457.96875,
      "epoch": 0.9506849315068493,
      "grad_norm": 0.027479765936732292,
      "kl": 0.33447265625,
      "learning_rate": 6.831050228310502e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 694
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 417.96875,
      "epoch": 0.952054794520548,
      "grad_norm": 2.6095175743103027,
      "kl": 0.36181640625,
      "learning_rate": 6.82648401826484e-07,
      "loss": 0.0004,
      "reward": 1.4609375,
      "reward_std": 0.20912351459264755,
      "rewards/accuracy_reward": 0.4921875,
      "rewards/format_reward": 0.96875,
      "step": 695
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 600.96875,
      "epoch": 0.9534246575342465,
      "grad_norm": 1.3289512395858765,
      "kl": 0.328125,
      "learning_rate": 6.821917808219177e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 696
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.1875,
      "epoch": 0.9547945205479452,
      "grad_norm": 0.9503890872001648,
      "kl": 0.33642578125,
      "learning_rate": 6.817351598173516e-07,
      "loss": 0.0003,
      "reward": 1.243749976158142,
      "reward_std": 0.04580627381801605,
      "rewards/accuracy_reward": 0.24375002086162567,
      "rewards/format_reward": 1.0,
      "step": 697
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 600.0625,
      "epoch": 0.9561643835616438,
      "grad_norm": 6.267343044281006,
      "kl": 0.34130859375,
      "learning_rate": 6.812785388127854e-07,
      "loss": 0.0003,
      "reward": 1.2890625,
      "reward_std": 0.14807433634996414,
      "rewards/accuracy_reward": 0.2890625,
      "rewards/format_reward": 1.0,
      "step": 698
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.59375,
      "epoch": 0.9575342465753425,
      "grad_norm": 3.089637517929077,
      "kl": 0.33935546875,
      "learning_rate": 6.808219178082191e-07,
      "loss": 0.0003,
      "reward": 1.4895833134651184,
      "reward_std": 0.2777084931731224,
      "rewards/accuracy_reward": 0.4895833134651184,
      "rewards/format_reward": 1.0,
      "step": 699
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.03125,
      "epoch": 0.958904109589041,
      "grad_norm": 1.276449203491211,
      "kl": 1.47900390625,
      "learning_rate": 6.80365296803653e-07,
      "loss": 0.0015,
      "reward": 1.7890625,
      "reward_std": 0.06629125773906708,
      "rewards/accuracy_reward": 0.7890625,
      "rewards/format_reward": 1.0,
      "step": 700
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 490.3125,
      "epoch": 0.9602739726027397,
      "grad_norm": 0.03158112242817879,
      "kl": 0.37646484375,
      "learning_rate": 6.799086757990867e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 701
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 446.4375,
      "epoch": 0.9616438356164384,
      "grad_norm": 2.683041572570801,
      "kl": 0.31982421875,
      "learning_rate": 6.794520547945205e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.1751839816570282,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 702
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 744.75,
      "epoch": 0.963013698630137,
      "grad_norm": 3.1106810569763184,
      "kl": 0.32958984375,
      "learning_rate": 6.789954337899543e-07,
      "loss": 0.0003,
      "reward": 1.9375,
      "reward_std": 0.328794926404953,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 703
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 721.6875,
      "epoch": 0.9643835616438357,
      "grad_norm": 2.060368776321411,
      "kl": 0.31201171875,
      "learning_rate": 6.78538812785388e-07,
      "loss": 0.0003,
      "reward": 1.53125,
      "reward_std": 0.1422954723238945,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 704
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.1875,
      "epoch": 0.9657534246575342,
      "grad_norm": 1.8324370384216309,
      "kl": 0.30078125,
      "learning_rate": 6.78082191780822e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 705
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 412.0,
      "epoch": 0.9671232876712329,
      "grad_norm": 6.562459468841553,
      "kl": 0.341796875,
      "learning_rate": 6.776255707762557e-07,
      "loss": 0.0003,
      "reward": 1.5546875,
      "reward_std": 0.4100441411137581,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 1.0,
      "step": 706
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 461.4375,
      "epoch": 0.9684931506849315,
      "grad_norm": 2.6876327991485596,
      "kl": 0.36376953125,
      "learning_rate": 6.771689497716894e-07,
      "loss": 0.0004,
      "reward": 1.453125,
      "reward_std": 0.18807452730834484,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 707
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 315.4375,
      "epoch": 0.9698630136986301,
      "grad_norm": 1.8336397409439087,
      "kl": 0.35595703125,
      "learning_rate": 6.767123287671233e-07,
      "loss": 0.0004,
      "reward": 1.703125,
      "reward_std": 0.13258251920342445,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 708
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 563.90625,
      "epoch": 0.9712328767123287,
      "grad_norm": 4.356468200683594,
      "kl": 0.32275390625,
      "learning_rate": 6.76255707762557e-07,
      "loss": 0.0003,
      "reward": 1.3828125,
      "reward_std": 0.09069566056132317,
      "rewards/accuracy_reward": 0.3828125,
      "rewards/format_reward": 1.0,
      "step": 709
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 452.8125,
      "epoch": 0.9726027397260274,
      "grad_norm": 1.405023217201233,
      "kl": 0.328125,
      "learning_rate": 6.757990867579907e-07,
      "loss": 0.0003,
      "reward": 1.484375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 710
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 324.1875,
      "epoch": 0.9739726027397261,
      "grad_norm": 3.4619688987731934,
      "kl": 0.361328125,
      "learning_rate": 6.753424657534246e-07,
      "loss": 0.0004,
      "reward": 1.8125,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.8125,
      "rewards/format_reward": 1.0,
      "step": 711
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 290.59375,
      "epoch": 0.9753424657534246,
      "grad_norm": 3.3762919902801514,
      "kl": 0.33642578125,
      "learning_rate": 6.748858447488584e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.3072218894958496,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 712
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 757.53125,
      "epoch": 0.9767123287671233,
      "grad_norm": 1.4996120929718018,
      "kl": 0.29296875,
      "learning_rate": 6.744292237442923e-07,
      "loss": 0.0003,
      "reward": 1.518750011920929,
      "reward_std": 0.2324894331395626,
      "rewards/accuracy_reward": 0.5812499821186066,
      "rewards/format_reward": 0.9375,
      "step": 713
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.5,
      "epoch": 0.9780821917808219,
      "grad_norm": 1.4486969709396362,
      "kl": 0.3935546875,
      "learning_rate": 6.73972602739726e-07,
      "loss": 0.0004,
      "reward": 1.421875,
      "reward_std": 0.11100948229432106,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 714
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 668.375,
      "epoch": 0.9794520547945206,
      "grad_norm": 2.0050597190856934,
      "kl": 0.310546875,
      "learning_rate": 6.735159817351597e-07,
      "loss": 0.0003,
      "reward": 1.625,
      "reward_std": 0.11230766586959362,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 715
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 544.59375,
      "epoch": 0.9808219178082191,
      "grad_norm": 1.8048930168151855,
      "kl": 0.31787109375,
      "learning_rate": 6.730593607305936e-07,
      "loss": 0.0003,
      "reward": 1.59375,
      "reward_std": 0.08065321296453476,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 716
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.78125,
      "epoch": 0.9821917808219178,
      "grad_norm": 0.010665824636816978,
      "kl": 0.32568359375,
      "learning_rate": 6.726027397260273e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 717
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.6875,
      "epoch": 0.9835616438356164,
      "grad_norm": 1.4615073204040527,
      "kl": 0.3134765625,
      "learning_rate": 6.721461187214613e-07,
      "loss": 0.0003,
      "reward": 1.5078125,
      "reward_std": 0.12534979730844498,
      "rewards/accuracy_reward": 0.5078125,
      "rewards/format_reward": 1.0,
      "step": 718
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 437.4375,
      "epoch": 0.9849315068493151,
      "grad_norm": 1.7689210176467896,
      "kl": 0.341796875,
      "learning_rate": 6.71689497716895e-07,
      "loss": 0.0003,
      "reward": 1.509374976158142,
      "reward_std": 0.10276375338435173,
      "rewards/accuracy_reward": 0.5093749761581421,
      "rewards/format_reward": 1.0,
      "step": 719
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.53125,
      "epoch": 0.9863013698630136,
      "grad_norm": 7.66063117980957,
      "kl": 0.33544921875,
      "learning_rate": 6.712328767123287e-07,
      "loss": 0.0003,
      "reward": 1.63671875,
      "reward_std": 0.25478895008563995,
      "rewards/accuracy_reward": 0.63671875,
      "rewards/format_reward": 1.0,
      "step": 720
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 672.34375,
      "epoch": 0.9876712328767123,
      "grad_norm": 3.410891532897949,
      "kl": 0.3125,
      "learning_rate": 6.707762557077626e-07,
      "loss": 0.0003,
      "reward": 2.002604156732559,
      "reward_std": 0.156676534563303,
      "rewards/accuracy_reward": 1.0026041567325592,
      "rewards/format_reward": 1.0,
      "step": 721
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 421.59375,
      "epoch": 0.989041095890411,
      "grad_norm": 2.7770752906799316,
      "kl": 0.359375,
      "learning_rate": 6.703196347031963e-07,
      "loss": 0.0004,
      "reward": 1.390625,
      "reward_std": 0.2824692949652672,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 722
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 403.40625,
      "epoch": 0.9904109589041096,
      "grad_norm": 4.406149387359619,
      "kl": 0.3603515625,
      "learning_rate": 6.6986301369863e-07,
      "loss": 0.0004,
      "reward": 1.7109375,
      "reward_std": 0.3282610699534416,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 1.0,
      "step": 723
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 426.5625,
      "epoch": 0.9917808219178083,
      "grad_norm": 3.31536602973938,
      "kl": 0.390625,
      "learning_rate": 6.694063926940639e-07,
      "loss": 0.0004,
      "reward": 1.453125,
      "reward_std": 0.24556982144713402,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 724
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.75,
      "epoch": 0.9931506849315068,
      "grad_norm": 4.0502424240112305,
      "kl": 0.39697265625,
      "learning_rate": 6.689497716894977e-07,
      "loss": 0.0004,
      "reward": 1.8671875,
      "reward_std": 0.16954397037625313,
      "rewards/accuracy_reward": 0.8671875,
      "rewards/format_reward": 1.0,
      "step": 725
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.25,
      "epoch": 0.9945205479452055,
      "grad_norm": 4.198473930358887,
      "kl": 0.31005859375,
      "learning_rate": 6.684931506849316e-07,
      "loss": 0.0003,
      "reward": 1.7578125,
      "reward_std": 0.17160805128514767,
      "rewards/accuracy_reward": 0.7578125,
      "rewards/format_reward": 1.0,
      "step": 726
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 530.1875,
      "epoch": 0.9958904109589041,
      "grad_norm": 2.3717732429504395,
      "kl": 0.33544921875,
      "learning_rate": 6.680365296803653e-07,
      "loss": 0.0003,
      "reward": 2.010416656732559,
      "reward_std": 0.2956680431962013,
      "rewards/accuracy_reward": 1.0104166567325592,
      "rewards/format_reward": 1.0,
      "step": 727
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 523.90625,
      "epoch": 0.9972602739726028,
      "grad_norm": 1.4983484745025635,
      "kl": 0.3330078125,
      "learning_rate": 6.67579908675799e-07,
      "loss": 0.0003,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 728
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 431.34375,
      "epoch": 0.9986301369863013,
      "grad_norm": 2.0858328342437744,
      "kl": 0.42578125,
      "learning_rate": 6.671232876712329e-07,
      "loss": 0.0004,
      "reward": 1.22265625,
      "reward_std": 0.11446718871593475,
      "rewards/accuracy_reward": 0.22265625,
      "rewards/format_reward": 1.0,
      "step": 729
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 510.15625,
      "epoch": 1.0,
      "grad_norm": 3.0337514877319336,
      "kl": 0.3486328125,
      "learning_rate": 6.666666666666666e-07,
      "loss": 0.0003,
      "reward": 1.5234375,
      "reward_std": 0.1662898138165474,
      "rewards/accuracy_reward": 0.5234375,
      "rewards/format_reward": 1.0,
      "step": 730
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 391.5625,
      "epoch": 1.0013698630136987,
      "grad_norm": 1.1392573118209839,
      "kl": 0.33984375,
      "learning_rate": 6.662100456621003e-07,
      "loss": 0.0003,
      "reward": 1.2265625,
      "reward_std": 0.03234682232141495,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 1.0,
      "step": 731
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 482.75,
      "epoch": 1.0027397260273974,
      "grad_norm": 3.7406272888183594,
      "kl": 0.3173828125,
      "learning_rate": 6.657534246575343e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.9375,
      "step": 732
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 407.84375,
      "epoch": 1.0041095890410958,
      "grad_norm": 1.7799715995788574,
      "kl": 0.34912109375,
      "learning_rate": 6.65296803652968e-07,
      "loss": 0.0003,
      "reward": 1.484375,
      "reward_std": 0.22097086161375046,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 733
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 498.59375,
      "epoch": 1.0054794520547945,
      "grad_norm": 6.9609904289245605,
      "kl": 0.34375,
      "learning_rate": 6.648401826484019e-07,
      "loss": 0.0003,
      "reward": 1.8333333134651184,
      "reward_std": 0.17943118885159492,
      "rewards/accuracy_reward": 0.8333333134651184,
      "rewards/format_reward": 1.0,
      "step": 734
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 586.25,
      "epoch": 1.0068493150684932,
      "grad_norm": 2.033268928527832,
      "kl": 0.3408203125,
      "learning_rate": 6.643835616438356e-07,
      "loss": 0.0003,
      "reward": 1.7291666269302368,
      "reward_std": 0.1323188804090023,
      "rewards/accuracy_reward": 0.7291666567325592,
      "rewards/format_reward": 1.0,
      "step": 735
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 522.78125,
      "epoch": 1.0082191780821919,
      "grad_norm": 1.8167824745178223,
      "kl": 0.35107421875,
      "learning_rate": 6.639269406392693e-07,
      "loss": 0.0004,
      "reward": 1.81640625,
      "reward_std": 0.2947664186358452,
      "rewards/accuracy_reward": 0.81640625,
      "rewards/format_reward": 1.0,
      "step": 736
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 311.4375,
      "epoch": 1.0095890410958903,
      "grad_norm": 8.44912338256836,
      "kl": 0.37841796875,
      "learning_rate": 6.634703196347032e-07,
      "loss": 0.0004,
      "reward": 1.4609375,
      "reward_std": 0.3800046369433403,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 1.0,
      "step": 737
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 499.15625,
      "epoch": 1.010958904109589,
      "grad_norm": 2.3125061988830566,
      "kl": 0.34716796875,
      "learning_rate": 6.63013698630137e-07,
      "loss": 0.0003,
      "reward": 1.7330728769302368,
      "reward_std": 0.23545794188976288,
      "rewards/accuracy_reward": 0.7643229067325592,
      "rewards/format_reward": 0.96875,
      "step": 738
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 435.9375,
      "epoch": 1.0123287671232877,
      "grad_norm": 3.3824410438537598,
      "kl": 0.34521484375,
      "learning_rate": 6.625570776255707e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.1173202246427536,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 739
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 402.25,
      "epoch": 1.0136986301369864,
      "grad_norm": 2.3408203125,
      "kl": 0.35205078125,
      "learning_rate": 6.621004566210046e-07,
      "loss": 0.0004,
      "reward": 1.4609375,
      "reward_std": 0.19531989470124245,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 1.0,
      "step": 740
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.28125,
      "epoch": 1.015068493150685,
      "grad_norm": 2.0831456184387207,
      "kl": 0.349609375,
      "learning_rate": 6.616438356164383e-07,
      "loss": 0.0003,
      "reward": 1.7421875,
      "reward_std": 0.09021057933568954,
      "rewards/accuracy_reward": 0.7421875,
      "rewards/format_reward": 1.0,
      "step": 741
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 312.03125,
      "epoch": 1.0164383561643835,
      "grad_norm": 2.4400479793548584,
      "kl": 0.333984375,
      "learning_rate": 6.61187214611872e-07,
      "loss": 0.0003,
      "reward": 1.5234375,
      "reward_std": 0.22642776370048523,
      "rewards/accuracy_reward": 0.5234375,
      "rewards/format_reward": 1.0,
      "step": 742
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.15625,
      "epoch": 1.0178082191780822,
      "grad_norm": 1.4530420303344727,
      "kl": 0.35107421875,
      "learning_rate": 6.607305936073059e-07,
      "loss": 0.0004,
      "reward": 1.4765625,
      "reward_std": 0.1708463504910469,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 1.0,
      "step": 743
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 352.125,
      "epoch": 1.0191780821917809,
      "grad_norm": 3.7937586307525635,
      "kl": 0.419921875,
      "learning_rate": 6.602739726027396e-07,
      "loss": 0.0004,
      "reward": 1.375,
      "reward_std": 0.1629730723798275,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 744
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 467.0,
      "epoch": 1.0205479452054795,
      "grad_norm": 1.1267149448394775,
      "kl": 0.40283203125,
      "learning_rate": 6.598173515981736e-07,
      "loss": 0.0004,
      "reward": 0.96875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.96875,
      "step": 745
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 455.9375,
      "epoch": 1.021917808219178,
      "grad_norm": 1.195306658744812,
      "kl": 0.35009765625,
      "learning_rate": 6.593607305936073e-07,
      "loss": 0.0003,
      "reward": 1.34375,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 1.0,
      "step": 746
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.15625,
      "epoch": 1.0232876712328767,
      "grad_norm": 1.5478026866912842,
      "kl": 0.3720703125,
      "learning_rate": 6.58904109589041e-07,
      "loss": 0.0004,
      "reward": 1.703125,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 747
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 431.1875,
      "epoch": 1.0246575342465754,
      "grad_norm": 2.914033889770508,
      "kl": 0.33740234375,
      "learning_rate": 6.584474885844749e-07,
      "loss": 0.0003,
      "reward": 1.86328125,
      "reward_std": 0.5137732066214085,
      "rewards/accuracy_reward": 0.89453125,
      "rewards/format_reward": 0.96875,
      "step": 748
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 556.71875,
      "epoch": 1.026027397260274,
      "grad_norm": 1.969694972038269,
      "kl": 0.3681640625,
      "learning_rate": 6.579908675799086e-07,
      "loss": 0.0004,
      "reward": 1.6796875,
      "reward_std": 0.11048541404306889,
      "rewards/accuracy_reward": 0.6796875298023224,
      "rewards/format_reward": 1.0,
      "step": 749
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 338.75,
      "epoch": 1.0273972602739727,
      "grad_norm": 5.528670787811279,
      "kl": 0.39990234375,
      "learning_rate": 6.575342465753423e-07,
      "loss": 0.0004,
      "reward": 1.6015625,
      "reward_std": 0.17859892547130585,
      "rewards/accuracy_reward": 0.6015625,
      "rewards/format_reward": 1.0,
      "step": 750
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.46875,
      "epoch": 1.0287671232876712,
      "grad_norm": 4.909811973571777,
      "kl": 0.39013671875,
      "learning_rate": 6.570776255707762e-07,
      "loss": 0.0004,
      "reward": 1.640625,
      "reward_std": 0.0776018276810646,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 751
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 455.4375,
      "epoch": 1.0301369863013699,
      "grad_norm": 5.930173397064209,
      "kl": 0.35498046875,
      "learning_rate": 6.5662100456621e-07,
      "loss": 0.0004,
      "reward": 1.7734375,
      "reward_std": 0.2153516486287117,
      "rewards/accuracy_reward": 0.7734375,
      "rewards/format_reward": 1.0,
      "step": 752
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.5,
      "epoch": 1.0315068493150685,
      "grad_norm": 2.4682910442352295,
      "kl": 0.3642578125,
      "learning_rate": 6.561643835616439e-07,
      "loss": 0.0004,
      "reward": 1.7369791865348816,
      "reward_std": 0.2828553803265095,
      "rewards/accuracy_reward": 0.7682291865348816,
      "rewards/format_reward": 0.96875,
      "step": 753
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 442.28125,
      "epoch": 1.0328767123287672,
      "grad_norm": 3.169569969177246,
      "kl": 0.3759765625,
      "learning_rate": 6.557077625570776e-07,
      "loss": 0.0004,
      "reward": 1.625,
      "reward_std": 0.21937815099954605,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 754
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 558.6875,
      "epoch": 1.0342465753424657,
      "grad_norm": 3.002521514892578,
      "kl": 0.35986328125,
      "learning_rate": 6.552511415525113e-07,
      "loss": 0.0004,
      "reward": 1.671875,
      "reward_std": 0.22208165377378464,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 755
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 546.5625,
      "epoch": 1.0356164383561643,
      "grad_norm": 2.0301320552825928,
      "kl": 0.36865234375,
      "learning_rate": 6.547945205479452e-07,
      "loss": 0.0004,
      "reward": 1.8772321939468384,
      "reward_std": 0.2395726516842842,
      "rewards/accuracy_reward": 0.908482164144516,
      "rewards/format_reward": 0.96875,
      "step": 756
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 335.46875,
      "epoch": 1.036986301369863,
      "grad_norm": 2.746330738067627,
      "kl": 0.4189453125,
      "learning_rate": 6.543378995433789e-07,
      "loss": 0.0004,
      "reward": 1.75,
      "reward_std": 0.249358132481575,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 757
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 454.34375,
      "epoch": 1.0383561643835617,
      "grad_norm": 2.1871116161346436,
      "kl": 0.40625,
      "learning_rate": 6.538812785388129e-07,
      "loss": 0.0004,
      "reward": 1.6692708134651184,
      "reward_std": 0.11145787499845028,
      "rewards/accuracy_reward": 0.6692708134651184,
      "rewards/format_reward": 1.0,
      "step": 758
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 458.96875,
      "epoch": 1.0397260273972604,
      "grad_norm": 15.84890365600586,
      "kl": 0.40673828125,
      "learning_rate": 6.534246575342466e-07,
      "loss": 0.0004,
      "reward": 1.7109375,
      "reward_std": 0.21829968504607677,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 1.0,
      "step": 759
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 354.09375,
      "epoch": 1.0410958904109588,
      "grad_norm": 2.4201431274414062,
      "kl": 0.37158203125,
      "learning_rate": 6.529680365296803e-07,
      "loss": 0.0004,
      "reward": 1.5390625,
      "reward_std": 0.18201877176761627,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 1.0,
      "step": 760
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 325.375,
      "epoch": 1.0424657534246575,
      "grad_norm": 1.7726272344589233,
      "kl": 0.4150390625,
      "learning_rate": 6.525114155251142e-07,
      "loss": 0.0004,
      "reward": 1.1640625,
      "reward_std": 0.03234682232141495,
      "rewards/accuracy_reward": 0.1640625,
      "rewards/format_reward": 1.0,
      "step": 761
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 406.59375,
      "epoch": 1.0438356164383562,
      "grad_norm": 3.5865135192871094,
      "kl": 0.35986328125,
      "learning_rate": 6.520547945205479e-07,
      "loss": 0.0004,
      "reward": 1.703125,
      "reward_std": 0.36083250865340233,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 1.0,
      "step": 762
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.4375,
      "epoch": 1.0452054794520549,
      "grad_norm": 4.15688943862915,
      "kl": 0.37939453125,
      "learning_rate": 6.515981735159816e-07,
      "loss": 0.0004,
      "reward": 1.78125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 1.0,
      "step": 763
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.5625,
      "epoch": 1.0465753424657533,
      "grad_norm": 1.4329558610916138,
      "kl": 0.361328125,
      "learning_rate": 6.511415525114155e-07,
      "loss": 0.0004,
      "reward": 1.5026041865348816,
      "reward_std": 0.19893107935786247,
      "rewards/accuracy_reward": 0.5338541716337204,
      "rewards/format_reward": 0.96875,
      "step": 764
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 389.625,
      "epoch": 1.047945205479452,
      "grad_norm": 2.296790599822998,
      "kl": 0.3896484375,
      "learning_rate": 6.506849315068493e-07,
      "loss": 0.0004,
      "reward": 1.8984375,
      "reward_std": 0.13098490238189697,
      "rewards/accuracy_reward": 0.8984375,
      "rewards/format_reward": 1.0,
      "step": 765
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 280.0625,
      "epoch": 1.0493150684931507,
      "grad_norm": 4.319180011749268,
      "kl": 0.47119140625,
      "learning_rate": 6.502283105022832e-07,
      "loss": 0.0005,
      "reward": 1.44921875,
      "reward_std": 0.29051198065280914,
      "rewards/accuracy_reward": 0.44921875,
      "rewards/format_reward": 1.0,
      "step": 766
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 392.8125,
      "epoch": 1.0506849315068494,
      "grad_norm": 1.033851146697998,
      "kl": 0.45166015625,
      "learning_rate": 6.497716894977169e-07,
      "loss": 0.0005,
      "reward": 1.375,
      "reward_std": 0.1602174937725067,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 0.96875,
      "step": 767
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.625,
      "epoch": 1.0520547945205478,
      "grad_norm": 3.8175528049468994,
      "kl": 0.408203125,
      "learning_rate": 6.493150684931506e-07,
      "loss": 0.0004,
      "reward": 1.3125,
      "reward_std": 0.1872510462999344,
      "rewards/accuracy_reward": 0.3125,
      "rewards/format_reward": 1.0,
      "step": 768
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.90625,
      "epoch": 1.0534246575342465,
      "grad_norm": 3.4146058559417725,
      "kl": 0.37744140625,
      "learning_rate": 6.488584474885845e-07,
      "loss": 0.0004,
      "reward": 1.46875,
      "reward_std": 0.3107999712228775,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.9375,
      "step": 769
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 364.4375,
      "epoch": 1.0547945205479452,
      "grad_norm": 2.580096960067749,
      "kl": 0.390625,
      "learning_rate": 6.484018264840182e-07,
      "loss": 0.0004,
      "reward": 1.5703125,
      "reward_std": 0.11353681609034538,
      "rewards/accuracy_reward": 0.5703125,
      "rewards/format_reward": 1.0,
      "step": 770
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 299.6875,
      "epoch": 1.0561643835616439,
      "grad_norm": 3.621427297592163,
      "kl": 0.4423828125,
      "learning_rate": 6.479452054794519e-07,
      "loss": 0.0004,
      "reward": 1.09375,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.09375,
      "rewards/format_reward": 1.0,
      "step": 771
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 444.125,
      "epoch": 1.0575342465753426,
      "grad_norm": 6.84398889541626,
      "kl": 0.5302734375,
      "learning_rate": 6.474885844748859e-07,
      "loss": 0.0005,
      "reward": 1.328125,
      "reward_std": 0.30537302792072296,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 0.96875,
      "step": 772
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 696.125,
      "epoch": 1.058904109589041,
      "grad_norm": 2.4137730598449707,
      "kl": 0.3642578125,
      "learning_rate": 6.470319634703196e-07,
      "loss": 0.0004,
      "reward": 1.65625,
      "reward_std": 0.5298669151961803,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.90625,
      "step": 773
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.1875,
      "epoch": 1.0602739726027397,
      "grad_norm": 2.266726493835449,
      "kl": 0.36181640625,
      "learning_rate": 6.465753424657535e-07,
      "loss": 0.0004,
      "reward": 1.390625,
      "reward_std": 0.2643740847706795,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 0.9375,
      "step": 774
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.53125,
      "epoch": 1.0616438356164384,
      "grad_norm": 2.0069427490234375,
      "kl": 0.39306640625,
      "learning_rate": 6.461187214611872e-07,
      "loss": 0.0004,
      "reward": 1.6328125,
      "reward_std": 0.20269311219453812,
      "rewards/accuracy_reward": 0.6640625,
      "rewards/format_reward": 0.96875,
      "step": 775
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 539.0625,
      "epoch": 1.063013698630137,
      "grad_norm": 2.6422464847564697,
      "kl": 0.35205078125,
      "learning_rate": 6.456621004566209e-07,
      "loss": 0.0004,
      "reward": 1.3671875,
      "reward_std": 0.08891239576041698,
      "rewards/accuracy_reward": 0.3671875,
      "rewards/format_reward": 1.0,
      "step": 776
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 396.0,
      "epoch": 1.0643835616438355,
      "grad_norm": 2.95044207572937,
      "kl": 0.4296875,
      "learning_rate": 6.452054794520548e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.2651650384068489,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.96875,
      "step": 777
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 618.21875,
      "epoch": 1.0657534246575342,
      "grad_norm": 3.2683823108673096,
      "kl": 0.33740234375,
      "learning_rate": 6.447488584474886e-07,
      "loss": 0.0003,
      "reward": 1.296875,
      "reward_std": 0.7756812795996666,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.71875,
      "step": 778
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 453.40625,
      "epoch": 1.0671232876712329,
      "grad_norm": 3.376239061355591,
      "kl": 0.38623046875,
      "learning_rate": 6.442922374429223e-07,
      "loss": 0.0004,
      "reward": 1.4140625,
      "reward_std": 0.5326974391937256,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 0.9375,
      "step": 779
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 429.84375,
      "epoch": 1.0684931506849316,
      "grad_norm": 0.967551052570343,
      "kl": 0.42138671875,
      "learning_rate": 6.438356164383562e-07,
      "loss": 0.0004,
      "reward": 1.40625,
      "reward_std": 0.1735912710428238,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.96875,
      "step": 780
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 412.125,
      "epoch": 1.0698630136986302,
      "grad_norm": 7.696181297302246,
      "kl": 0.38427734375,
      "learning_rate": 6.433789954337899e-07,
      "loss": 0.0004,
      "reward": 1.265625,
      "reward_std": 0.1446593925356865,
      "rewards/accuracy_reward": 0.265625,
      "rewards/format_reward": 1.0,
      "step": 781
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 382.5625,
      "epoch": 1.0712328767123287,
      "grad_norm": 3.0563812255859375,
      "kl": 0.4501953125,
      "learning_rate": 6.429223744292238e-07,
      "loss": 0.0005,
      "reward": 1.640625,
      "reward_std": 0.17358146235346794,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 782
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 391.96875,
      "epoch": 1.0726027397260274,
      "grad_norm": 0.024443048983812332,
      "kl": 0.388671875,
      "learning_rate": 6.424657534246575e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 783
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 412.40625,
      "epoch": 1.073972602739726,
      "grad_norm": 1.1630728244781494,
      "kl": 0.365234375,
      "learning_rate": 6.420091324200912e-07,
      "loss": 0.0004,
      "reward": 1.484375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 784
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.1875,
      "epoch": 1.0753424657534247,
      "grad_norm": 3.8575427532196045,
      "kl": 0.38916015625,
      "learning_rate": 6.415525114155252e-07,
      "loss": 0.0004,
      "reward": 1.375,
      "reward_std": 0.4482543617486954,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 0.9375,
      "step": 785
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 401.6875,
      "epoch": 1.0767123287671232,
      "grad_norm": 3.8265440464019775,
      "kl": 0.37060546875,
      "learning_rate": 6.410958904109589e-07,
      "loss": 0.0004,
      "reward": 1.3984375,
      "reward_std": 0.2753252908587456,
      "rewards/accuracy_reward": 0.3984375,
      "rewards/format_reward": 1.0,
      "step": 786
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 389.5625,
      "epoch": 1.0780821917808219,
      "grad_norm": 10.559552192687988,
      "kl": 0.38232421875,
      "learning_rate": 6.406392694063926e-07,
      "loss": 0.0004,
      "reward": 1.4765625,
      "reward_std": 0.1541428230702877,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 1.0,
      "step": 787
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 516.8125,
      "epoch": 1.0794520547945206,
      "grad_norm": 2.3642754554748535,
      "kl": 0.35791015625,
      "learning_rate": 6.401826484018265e-07,
      "loss": 0.0004,
      "reward": 1.40625,
      "reward_std": 0.10888781771063805,
      "rewards/accuracy_reward": 0.40625,
      "rewards/format_reward": 1.0,
      "step": 788
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.8125,
      "epoch": 1.0808219178082192,
      "grad_norm": 2.5423686504364014,
      "kl": 0.3876953125,
      "learning_rate": 6.397260273972602e-07,
      "loss": 0.0004,
      "reward": 1.8723958432674408,
      "reward_std": 0.20485981926321983,
      "rewards/accuracy_reward": 0.8723958283662796,
      "rewards/format_reward": 1.0,
      "step": 789
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.28125,
      "epoch": 1.0821917808219177,
      "grad_norm": 0.012947800569236279,
      "kl": 0.38232421875,
      "learning_rate": 6.39269406392694e-07,
      "loss": 0.0004,
      "reward": 1.75,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 1.0,
      "step": 790
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 514.875,
      "epoch": 1.0835616438356164,
      "grad_norm": 2.922663688659668,
      "kl": 0.38818359375,
      "learning_rate": 6.388127853881278e-07,
      "loss": 0.0004,
      "reward": 2.0546875,
      "reward_std": 0.2646155208349228,
      "rewards/accuracy_reward": 1.0546875,
      "rewards/format_reward": 1.0,
      "step": 791
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.09375,
      "epoch": 1.084931506849315,
      "grad_norm": 1.6584970951080322,
      "kl": 0.4111328125,
      "learning_rate": 6.383561643835616e-07,
      "loss": 0.0004,
      "reward": 1.84375,
      "reward_std": 0.19999710842967033,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 792
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 349.8125,
      "epoch": 1.0863013698630137,
      "grad_norm": 1.3865472078323364,
      "kl": 0.39599609375,
      "learning_rate": 6.378995433789955e-07,
      "loss": 0.0004,
      "reward": 1.2135416567325592,
      "reward_std": 0.026702916249632835,
      "rewards/accuracy_reward": 0.2135416567325592,
      "rewards/format_reward": 1.0,
      "step": 793
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.65625,
      "epoch": 1.0876712328767124,
      "grad_norm": 9.972989082336426,
      "kl": 0.33984375,
      "learning_rate": 6.374429223744292e-07,
      "loss": 0.0003,
      "reward": 1.453125,
      "reward_std": 0.3874029070138931,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.90625,
      "step": 794
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 418.5,
      "epoch": 1.0890410958904109,
      "grad_norm": 3.65055513381958,
      "kl": 0.39306640625,
      "learning_rate": 6.369863013698629e-07,
      "loss": 0.0004,
      "reward": 1.734375,
      "reward_std": 0.33669837564229965,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 0.96875,
      "step": 795
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 385.75,
      "epoch": 1.0904109589041096,
      "grad_norm": 7.232997417449951,
      "kl": 0.38671875,
      "learning_rate": 6.365296803652968e-07,
      "loss": 0.0004,
      "reward": 1.4375,
      "reward_std": 0.23827510699629784,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 796
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 293.34375,
      "epoch": 1.0917808219178082,
      "grad_norm": 4.632352828979492,
      "kl": 0.39208984375,
      "learning_rate": 6.360730593607305e-07,
      "loss": 0.0004,
      "reward": 1.578125,
      "reward_std": 0.24831003323197365,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 1.0,
      "step": 797
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.90625,
      "epoch": 1.093150684931507,
      "grad_norm": 2.5987162590026855,
      "kl": 0.3681640625,
      "learning_rate": 6.356164383561645e-07,
      "loss": 0.0004,
      "reward": 1.640625,
      "reward_std": 0.3057369217276573,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 1.0,
      "step": 798
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.25,
      "epoch": 1.0945205479452054,
      "grad_norm": 2.42681884765625,
      "kl": 0.380859375,
      "learning_rate": 6.351598173515982e-07,
      "loss": 0.0004,
      "reward": 1.953125,
      "reward_std": 0.2366182692348957,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 799
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.6875,
      "epoch": 1.095890410958904,
      "grad_norm": 2.0286333560943604,
      "kl": 0.3779296875,
      "learning_rate": 6.347031963470319e-07,
      "loss": 0.0004,
      "reward": 1.6484375,
      "reward_std": 0.13488983362913132,
      "rewards/accuracy_reward": 0.6484375,
      "rewards/format_reward": 1.0,
      "step": 800
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 397.625,
      "epoch": 1.0972602739726027,
      "grad_norm": 3.4189932346343994,
      "kl": 0.36962890625,
      "learning_rate": 6.342465753424658e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 801
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 670.96875,
      "epoch": 1.0986301369863014,
      "grad_norm": 2.2040603160858154,
      "kl": 0.39013671875,
      "learning_rate": 6.337899543378995e-07,
      "loss": 0.0004,
      "reward": 1.7276785373687744,
      "reward_std": 0.21270384266972542,
      "rewards/accuracy_reward": 0.7589285373687744,
      "rewards/format_reward": 0.96875,
      "step": 802
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 318.96875,
      "epoch": 1.1,
      "grad_norm": 2.4739291667938232,
      "kl": 0.35107421875,
      "learning_rate": 6.333333333333332e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 803
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.8125,
      "epoch": 1.1013698630136985,
      "grad_norm": 3.376389980316162,
      "kl": 0.39453125,
      "learning_rate": 6.328767123287671e-07,
      "loss": 0.0004,
      "reward": 2.2265625,
      "reward_std": 0.40516645461320877,
      "rewards/accuracy_reward": 1.2265625,
      "rewards/format_reward": 1.0,
      "step": 804
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 271.3125,
      "epoch": 1.1027397260273972,
      "grad_norm": 3.8003089427948,
      "kl": 0.36767578125,
      "learning_rate": 6.324200913242009e-07,
      "loss": 0.0004,
      "reward": 1.6328125,
      "reward_std": 0.23646268248558044,
      "rewards/accuracy_reward": 0.6328125,
      "rewards/format_reward": 1.0,
      "step": 805
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 488.375,
      "epoch": 1.104109589041096,
      "grad_norm": 0.9171125888824463,
      "kl": 0.39794921875,
      "learning_rate": 6.319634703196348e-07,
      "loss": 0.0004,
      "reward": 1.234375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 0.234375,
      "rewards/format_reward": 1.0,
      "step": 806
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.375,
      "epoch": 1.1054794520547946,
      "grad_norm": 2.2937474250793457,
      "kl": 0.46240234375,
      "learning_rate": 6.315068493150685e-07,
      "loss": 0.0005,
      "reward": 1.953125,
      "reward_std": 0.22367356345057487,
      "rewards/accuracy_reward": 0.953125,
      "rewards/format_reward": 1.0,
      "step": 807
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 499.25,
      "epoch": 1.106849315068493,
      "grad_norm": 1.2261008024215698,
      "kl": 0.39794921875,
      "learning_rate": 6.310502283105022e-07,
      "loss": 0.0004,
      "reward": 1.6875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 808
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.65625,
      "epoch": 1.1082191780821917,
      "grad_norm": 10.46350383758545,
      "kl": 0.41455078125,
      "learning_rate": 6.305936073059361e-07,
      "loss": 0.0004,
      "reward": 1.5546875,
      "reward_std": 0.19728106819093227,
      "rewards/accuracy_reward": 0.5859375,
      "rewards/format_reward": 0.96875,
      "step": 809
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 393.0625,
      "epoch": 1.1095890410958904,
      "grad_norm": 0.021392615512013435,
      "kl": 0.39013671875,
      "learning_rate": 6.301369863013698e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 810
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.0,
      "epoch": 1.110958904109589,
      "grad_norm": 7.004191875457764,
      "kl": 0.33984375,
      "learning_rate": 6.296803652968035e-07,
      "loss": 0.0003,
      "reward": 1.03125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.03125,
      "rewards/format_reward": 1.0,
      "step": 811
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 397.21875,
      "epoch": 1.1123287671232878,
      "grad_norm": 1.1587871313095093,
      "kl": 0.38330078125,
      "learning_rate": 6.292237442922375e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 812
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 497.53125,
      "epoch": 1.1136986301369862,
      "grad_norm": 1.5848588943481445,
      "kl": 0.42041015625,
      "learning_rate": 6.287671232876712e-07,
      "loss": 0.0004,
      "reward": 1.9352678656578064,
      "reward_std": 0.08208167925477028,
      "rewards/accuracy_reward": 0.9352678507566452,
      "rewards/format_reward": 1.0,
      "step": 813
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.90625,
      "epoch": 1.115068493150685,
      "grad_norm": 2.042673349380493,
      "kl": 0.3955078125,
      "learning_rate": 6.283105022831051e-07,
      "loss": 0.0004,
      "reward": 1.6822916865348816,
      "reward_std": 0.031000984832644463,
      "rewards/accuracy_reward": 0.6822916716337204,
      "rewards/format_reward": 1.0,
      "step": 814
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 400.625,
      "epoch": 1.1164383561643836,
      "grad_norm": 3.0837056636810303,
      "kl": 0.3740234375,
      "learning_rate": 6.278538812785388e-07,
      "loss": 0.0004,
      "reward": 1.296875,
      "reward_std": 0.15992168709635735,
      "rewards/accuracy_reward": 0.296875,
      "rewards/format_reward": 1.0,
      "step": 815
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.84375,
      "epoch": 1.1178082191780823,
      "grad_norm": 2.6489899158477783,
      "kl": 0.4033203125,
      "learning_rate": 6.273972602739725e-07,
      "loss": 0.0004,
      "reward": 1.66015625,
      "reward_std": 0.12530867382884026,
      "rewards/accuracy_reward": 0.66015625,
      "rewards/format_reward": 1.0,
      "step": 816
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 481.25,
      "epoch": 1.1191780821917807,
      "grad_norm": 2.778820514678955,
      "kl": 0.412109375,
      "learning_rate": 6.269406392694064e-07,
      "loss": 0.0004,
      "reward": 1.6093749403953552,
      "reward_std": 0.2615504954010248,
      "rewards/accuracy_reward": 0.6406249850988388,
      "rewards/format_reward": 0.96875,
      "step": 817
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.5,
      "epoch": 1.1205479452054794,
      "grad_norm": 2.434713363647461,
      "kl": 0.39013671875,
      "learning_rate": 6.264840182648402e-07,
      "loss": 0.0004,
      "reward": 1.421875,
      "reward_std": 0.15992168709635735,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 818
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 287.0625,
      "epoch": 1.121917808219178,
      "grad_norm": 5.153371334075928,
      "kl": 0.4091796875,
      "learning_rate": 6.260273972602739e-07,
      "loss": 0.0004,
      "reward": 1.796875,
      "reward_std": 0.2057085707783699,
      "rewards/accuracy_reward": 0.796875,
      "rewards/format_reward": 1.0,
      "step": 819
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 396.0625,
      "epoch": 1.1232876712328768,
      "grad_norm": 1.432228922843933,
      "kl": 0.37451171875,
      "learning_rate": 6.255707762557078e-07,
      "loss": 0.0004,
      "reward": 1.2421875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 0.2734375,
      "rewards/format_reward": 0.96875,
      "step": 820
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.5625,
      "epoch": 1.1246575342465754,
      "grad_norm": 0.9751577377319336,
      "kl": 0.396484375,
      "learning_rate": 6.251141552511415e-07,
      "loss": 0.0004,
      "reward": 1.484375,
      "reward_std": 0.1966201364994049,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 0.96875,
      "step": 821
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.125,
      "epoch": 1.126027397260274,
      "grad_norm": 1.529309868812561,
      "kl": 0.3896484375,
      "learning_rate": 6.246575342465754e-07,
      "loss": 0.0004,
      "reward": 1.4140625,
      "reward_std": 0.0810895636677742,
      "rewards/accuracy_reward": 0.4140625,
      "rewards/format_reward": 1.0,
      "step": 822
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 498.65625,
      "epoch": 1.1273972602739726,
      "grad_norm": 2.0530521869659424,
      "kl": 0.404296875,
      "learning_rate": 6.242009132420091e-07,
      "loss": 0.0004,
      "reward": 1.796875,
      "reward_std": 0.24911179021000862,
      "rewards/accuracy_reward": 0.828125,
      "rewards/format_reward": 0.96875,
      "step": 823
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.65625,
      "epoch": 1.1287671232876713,
      "grad_norm": 1.90122389793396,
      "kl": 0.376953125,
      "learning_rate": 6.237442922374428e-07,
      "loss": 0.0004,
      "reward": 1.5924479365348816,
      "reward_std": 0.15607357770204544,
      "rewards/accuracy_reward": 0.5924479216337204,
      "rewards/format_reward": 1.0,
      "step": 824
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.5,
      "epoch": 1.13013698630137,
      "grad_norm": 2.433654546737671,
      "kl": 0.3994140625,
      "learning_rate": 6.232876712328768e-07,
      "loss": 0.0004,
      "reward": 1.9285714030265808,
      "reward_std": 0.07229206152260303,
      "rewards/accuracy_reward": 0.9285714030265808,
      "rewards/format_reward": 1.0,
      "step": 825
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 371.78125,
      "epoch": 1.1315068493150684,
      "grad_norm": 1.0846009254455566,
      "kl": 0.40673828125,
      "learning_rate": 6.228310502283105e-07,
      "loss": 0.0004,
      "reward": 1.390625,
      "reward_std": 0.08010874688625336,
      "rewards/accuracy_reward": 0.390625,
      "rewards/format_reward": 1.0,
      "step": 826
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 353.375,
      "epoch": 1.132876712328767,
      "grad_norm": 3.237276554107666,
      "kl": 0.42041015625,
      "learning_rate": 6.223744292237442e-07,
      "loss": 0.0004,
      "reward": 1.609375,
      "reward_std": 0.24567490443587303,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 827
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 261.875,
      "epoch": 1.1342465753424658,
      "grad_norm": 1.206854224205017,
      "kl": 0.37841796875,
      "learning_rate": 6.219178082191781e-07,
      "loss": 0.0004,
      "reward": 1.1640625,
      "reward_std": 0.046501487493515015,
      "rewards/accuracy_reward": 0.1640625,
      "rewards/format_reward": 1.0,
      "step": 828
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 397.8125,
      "epoch": 1.1356164383561644,
      "grad_norm": 0.02847045287489891,
      "kl": 0.41259765625,
      "learning_rate": 6.214611872146118e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 829
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 459.15625,
      "epoch": 1.1369863013698631,
      "grad_norm": 9.929925918579102,
      "kl": 0.390625,
      "learning_rate": 6.210045662100457e-07,
      "loss": 0.0004,
      "reward": 1.8880208730697632,
      "reward_std": 0.16971025243401527,
      "rewards/accuracy_reward": 0.8880208432674408,
      "rewards/format_reward": 1.0,
      "step": 830
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 378.3125,
      "epoch": 1.1383561643835616,
      "grad_norm": 1.0134655237197876,
      "kl": 0.404296875,
      "learning_rate": 6.205479452054794e-07,
      "loss": 0.0004,
      "reward": 1.4375,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 831
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 472.71875,
      "epoch": 1.1397260273972603,
      "grad_norm": 1.1163495779037476,
      "kl": 0.39990234375,
      "learning_rate": 6.200913242009132e-07,
      "loss": 0.0004,
      "reward": 1.4375,
      "reward_std": 0.09449111670255661,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 832
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 301.1875,
      "epoch": 1.141095890410959,
      "grad_norm": 2.4487106800079346,
      "kl": 0.40185546875,
      "learning_rate": 6.196347031963471e-07,
      "loss": 0.0004,
      "reward": 1.46875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 833
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 516.15625,
      "epoch": 1.1424657534246576,
      "grad_norm": 8.19469165802002,
      "kl": 0.376953125,
      "learning_rate": 6.191780821917808e-07,
      "loss": 0.0004,
      "reward": 1.8093750476837158,
      "reward_std": 0.04783010669052601,
      "rewards/accuracy_reward": 0.809374988079071,
      "rewards/format_reward": 1.0,
      "step": 834
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 454.21875,
      "epoch": 1.143835616438356,
      "grad_norm": 0.01866794563829899,
      "kl": 0.4033203125,
      "learning_rate": 6.187214611872145e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 835
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.21875,
      "epoch": 1.1452054794520548,
      "grad_norm": 0.9545954465866089,
      "kl": 0.41162109375,
      "learning_rate": 6.182648401826484e-07,
      "loss": 0.0004,
      "reward": 2.1468749940395355,
      "reward_std": 0.048065248876810074,
      "rewards/accuracy_reward": 1.1468749940395355,
      "rewards/format_reward": 1.0,
      "step": 836
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 488.8125,
      "epoch": 1.1465753424657534,
      "grad_norm": 2.1958019733428955,
      "kl": 0.40283203125,
      "learning_rate": 6.178082191780821e-07,
      "loss": 0.0004,
      "reward": 1.8671875,
      "reward_std": 0.20912351086735725,
      "rewards/accuracy_reward": 0.8671875,
      "rewards/format_reward": 1.0,
      "step": 837
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 416.125,
      "epoch": 1.1479452054794521,
      "grad_norm": 1.0978032350540161,
      "kl": 0.3486328125,
      "learning_rate": 6.173515981735161e-07,
      "loss": 0.0003,
      "reward": 1.44921875,
      "reward_std": 0.0817250907421112,
      "rewards/accuracy_reward": 0.44921875,
      "rewards/format_reward": 1.0,
      "step": 838
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.03125,
      "epoch": 1.1493150684931508,
      "grad_norm": 0.019236719235777855,
      "kl": 0.4033203125,
      "learning_rate": 6.168949771689498e-07,
      "loss": 0.0004,
      "reward": 1.625,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 839
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 448.65625,
      "epoch": 1.1506849315068493,
      "grad_norm": 0.01787232607603073,
      "kl": 0.431640625,
      "learning_rate": 6.164383561643835e-07,
      "loss": 0.0004,
      "reward": 1.4375,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 840
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 390.53125,
      "epoch": 1.152054794520548,
      "grad_norm": 5.282984733581543,
      "kl": 0.40185546875,
      "learning_rate": 6.159817351598174e-07,
      "loss": 0.0004,
      "reward": 1.734375,
      "reward_std": 0.22097086533904076,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 841
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 537.6875,
      "epoch": 1.1534246575342466,
      "grad_norm": 0.8666876554489136,
      "kl": 0.39599609375,
      "learning_rate": 6.155251141552511e-07,
      "loss": 0.0004,
      "reward": 1.65625,
      "reward_std": 0.11410887539386749,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 842
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 270.90625,
      "epoch": 1.1547945205479453,
      "grad_norm": 2.4815289974212646,
      "kl": 0.43505859375,
      "learning_rate": 6.150684931506848e-07,
      "loss": 0.0004,
      "reward": 1.3984375,
      "reward_std": 0.2441160511225462,
      "rewards/accuracy_reward": 0.3984375,
      "rewards/format_reward": 1.0,
      "step": 843
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 360.84375,
      "epoch": 1.1561643835616437,
      "grad_norm": 8.32512378692627,
      "kl": 0.37744140625,
      "learning_rate": 6.146118721461187e-07,
      "loss": 0.0004,
      "reward": 1.6875,
      "reward_std": 0.1356339044868946,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 844
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 637.96875,
      "epoch": 1.1575342465753424,
      "grad_norm": 5.9347052574157715,
      "kl": 0.40380859375,
      "learning_rate": 6.141552511415525e-07,
      "loss": 0.0004,
      "reward": 1.9375,
      "reward_std": 0.19219962693750858,
      "rewards/accuracy_reward": 0.9375,
      "rewards/format_reward": 1.0,
      "step": 845
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 155.78125,
      "epoch": 1.158904109589041,
      "grad_norm": 2.557178020477295,
      "kl": 0.35595703125,
      "learning_rate": 6.136986301369864e-07,
      "loss": 0.0004,
      "reward": 1.28125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 846
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 364.09375,
      "epoch": 1.1602739726027398,
      "grad_norm": 0.9839144945144653,
      "kl": 0.369140625,
      "learning_rate": 6.132420091324201e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 847
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 416.1875,
      "epoch": 1.1616438356164385,
      "grad_norm": 0.01052401214838028,
      "kl": 0.3525390625,
      "learning_rate": 6.127853881278538e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 848
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 586.25,
      "epoch": 1.163013698630137,
      "grad_norm": 2.7697222232818604,
      "kl": 0.35546875,
      "learning_rate": 6.123287671232877e-07,
      "loss": 0.0004,
      "reward": 1.774999976158142,
      "reward_std": 0.13462574779987335,
      "rewards/accuracy_reward": 0.7749999463558197,
      "rewards/format_reward": 1.0,
      "step": 849
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.90625,
      "epoch": 1.1643835616438356,
      "grad_norm": 3.054478406906128,
      "kl": 0.35205078125,
      "learning_rate": 6.118721461187214e-07,
      "loss": 0.0004,
      "reward": 1.5390625,
      "reward_std": 0.2896413579583168,
      "rewards/accuracy_reward": 0.5390625,
      "rewards/format_reward": 1.0,
      "step": 850
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 540.125,
      "epoch": 1.1657534246575343,
      "grad_norm": 1.4414507150650024,
      "kl": 0.3310546875,
      "learning_rate": 6.114155251141551e-07,
      "loss": 0.0003,
      "reward": 1.4375,
      "reward_std": 0.07312605157494545,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 851
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.875,
      "epoch": 1.167123287671233,
      "grad_norm": 0.9515066146850586,
      "kl": 0.35302734375,
      "learning_rate": 6.109589041095891e-07,
      "loss": 0.0004,
      "reward": 1.421875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 852
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 430.6875,
      "epoch": 1.1684931506849314,
      "grad_norm": 2.1378560066223145,
      "kl": 0.36083984375,
      "learning_rate": 6.105022831050228e-07,
      "loss": 0.0004,
      "reward": 2.0,
      "reward_std": 0.12179600074887276,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 853
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.9375,
      "epoch": 1.16986301369863,
      "grad_norm": 1.0797525644302368,
      "kl": 0.3515625,
      "learning_rate": 6.100456621004567e-07,
      "loss": 0.0004,
      "reward": 1.421875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 854
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 308.71875,
      "epoch": 1.1712328767123288,
      "grad_norm": 2.498821258544922,
      "kl": 0.341796875,
      "learning_rate": 6.095890410958904e-07,
      "loss": 0.0003,
      "reward": 1.375,
      "reward_std": 0.13363061845302582,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 855
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 443.53125,
      "epoch": 1.1726027397260275,
      "grad_norm": 1.0982065200805664,
      "kl": 0.36279296875,
      "learning_rate": 6.091324200913241e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.033407654613256454,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 856
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 404.75,
      "epoch": 1.1739726027397261,
      "grad_norm": 6.744996547698975,
      "kl": 0.3828125,
      "learning_rate": 6.08675799086758e-07,
      "loss": 0.0004,
      "reward": 1.8333333730697632,
      "reward_std": 0.13888481445610523,
      "rewards/accuracy_reward": 0.8333333432674408,
      "rewards/format_reward": 1.0,
      "step": 857
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 579.25,
      "epoch": 1.1753424657534246,
      "grad_norm": 3.830198049545288,
      "kl": 0.36181640625,
      "learning_rate": 6.082191780821918e-07,
      "loss": 0.0004,
      "reward": 1.4375,
      "reward_std": 0.3230287954211235,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 0.96875,
      "step": 858
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 463.59375,
      "epoch": 1.1767123287671233,
      "grad_norm": 2.6295642852783203,
      "kl": 0.35009765625,
      "learning_rate": 6.077625570776255e-07,
      "loss": 0.0003,
      "reward": 1.1875,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 859
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 428.25,
      "epoch": 1.178082191780822,
      "grad_norm": 3.0436506271362305,
      "kl": 0.35498046875,
      "learning_rate": 6.073059360730594e-07,
      "loss": 0.0004,
      "reward": 1.921875,
      "reward_std": 0.1446593925356865,
      "rewards/accuracy_reward": 0.921875,
      "rewards/format_reward": 1.0,
      "step": 860
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 590.28125,
      "epoch": 1.1794520547945206,
      "grad_norm": 3.9150311946868896,
      "kl": 0.39208984375,
      "learning_rate": 6.068493150684931e-07,
      "loss": 0.0004,
      "reward": 1.484375,
      "reward_std": 0.1366081517189741,
      "rewards/accuracy_reward": 0.4843749701976776,
      "rewards/format_reward": 1.0,
      "step": 861
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.53125,
      "epoch": 1.180821917808219,
      "grad_norm": 1.7237287759780884,
      "kl": 0.33984375,
      "learning_rate": 6.06392694063927e-07,
      "loss": 0.0003,
      "reward": 1.609375,
      "reward_std": 0.39774755015969276,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 0.9375,
      "step": 862
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 558.21875,
      "epoch": 1.1821917808219178,
      "grad_norm": 4.21384334564209,
      "kl": 0.33203125,
      "learning_rate": 6.059360730593607e-07,
      "loss": 0.0003,
      "reward": 1.7738094925880432,
      "reward_std": 0.27128167636692524,
      "rewards/accuracy_reward": 0.8050595223903656,
      "rewards/format_reward": 0.96875,
      "step": 863
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 703.3125,
      "epoch": 1.1835616438356165,
      "grad_norm": 1.8811252117156982,
      "kl": 0.36279296875,
      "learning_rate": 6.054794520547944e-07,
      "loss": 0.0004,
      "reward": 1.9895833134651184,
      "reward_std": 0.1106601133942604,
      "rewards/accuracy_reward": 0.9895833283662796,
      "rewards/format_reward": 1.0,
      "step": 864
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.3125,
      "epoch": 1.1849315068493151,
      "grad_norm": 1.4166743755340576,
      "kl": 0.357421875,
      "learning_rate": 6.050228310502284e-07,
      "loss": 0.0004,
      "reward": 1.8359375,
      "reward_std": 0.061278700828552246,
      "rewards/accuracy_reward": 0.8359375,
      "rewards/format_reward": 1.0,
      "step": 865
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 352.34375,
      "epoch": 1.1863013698630138,
      "grad_norm": 3.5078675746917725,
      "kl": 0.37109375,
      "learning_rate": 6.045662100456621e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.2644027303904295,
      "rewards/accuracy_reward": 0.4999999850988388,
      "rewards/format_reward": 1.0,
      "step": 866
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 494.53125,
      "epoch": 1.1876712328767123,
      "grad_norm": 2.0775046348571777,
      "kl": 0.44140625,
      "learning_rate": 6.041095890410958e-07,
      "loss": 0.0004,
      "reward": 1.4375,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.4375,
      "rewards/format_reward": 1.0,
      "step": 867
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.40625,
      "epoch": 1.189041095890411,
      "grad_norm": 3.112804651260376,
      "kl": 0.349609375,
      "learning_rate": 6.036529680365297e-07,
      "loss": 0.0003,
      "reward": 1.7890625,
      "reward_std": 0.3950253389775753,
      "rewards/accuracy_reward": 0.8203125,
      "rewards/format_reward": 0.96875,
      "step": 868
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 738.625,
      "epoch": 1.1904109589041096,
      "grad_norm": 0.8393311500549316,
      "kl": 0.34765625,
      "learning_rate": 6.031963470319634e-07,
      "loss": 0.0003,
      "reward": 1.46875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 869
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 455.5,
      "epoch": 1.191780821917808,
      "grad_norm": 1.9282116889953613,
      "kl": 0.3583984375,
      "learning_rate": 6.027397260273972e-07,
      "loss": 0.0004,
      "reward": 1.1875,
      "reward_std": 0.06681530922651291,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 1.0,
      "step": 870
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.90625,
      "epoch": 1.1931506849315068,
      "grad_norm": 1.3332931995391846,
      "kl": 0.37451171875,
      "learning_rate": 6.02283105022831e-07,
      "loss": 0.0004,
      "reward": 1.7109375,
      "reward_std": 0.0765409953892231,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 1.0,
      "step": 871
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 515.5625,
      "epoch": 1.1945205479452055,
      "grad_norm": 2.1982920169830322,
      "kl": 0.36279296875,
      "learning_rate": 6.018264840182648e-07,
      "loss": 0.0004,
      "reward": 1.2083333134651184,
      "reward_std": 0.031497031450271606,
      "rewards/accuracy_reward": 0.2083333432674408,
      "rewards/format_reward": 1.0,
      "step": 872
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.4375,
      "epoch": 1.1958904109589041,
      "grad_norm": 1.4797850847244263,
      "kl": 0.37353515625,
      "learning_rate": 6.013698630136987e-07,
      "loss": 0.0004,
      "reward": 1.3776041567325592,
      "reward_std": 0.10277634114027023,
      "rewards/accuracy_reward": 0.3776041567325592,
      "rewards/format_reward": 1.0,
      "step": 873
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.28125,
      "epoch": 1.1972602739726028,
      "grad_norm": 1.9213581085205078,
      "kl": 0.37158203125,
      "learning_rate": 6.009132420091324e-07,
      "loss": 0.0004,
      "reward": 1.5833333432674408,
      "reward_std": 0.09757299907505512,
      "rewards/accuracy_reward": 0.5833333283662796,
      "rewards/format_reward": 1.0,
      "step": 874
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 425.0625,
      "epoch": 1.1986301369863013,
      "grad_norm": 3.607898712158203,
      "kl": 0.4130859375,
      "learning_rate": 6.004566210045661e-07,
      "loss": 0.0004,
      "reward": 1.7734375,
      "reward_std": 0.2514927387237549,
      "rewards/accuracy_reward": 0.7734375,
      "rewards/format_reward": 1.0,
      "step": 875
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 402.34375,
      "epoch": 1.2,
      "grad_norm": 5.052751541137695,
      "kl": 0.46923828125,
      "learning_rate": 6e-07,
      "loss": 0.0005,
      "reward": 1.546875,
      "reward_std": 0.15992168709635735,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 1.0,
      "step": 876
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.96875,
      "epoch": 1.2013698630136986,
      "grad_norm": 4.230776309967041,
      "kl": 0.46435546875,
      "learning_rate": 5.995433789954337e-07,
      "loss": 0.0005,
      "reward": 1.8444940447807312,
      "reward_std": 0.1678207330405712,
      "rewards/accuracy_reward": 0.8444940149784088,
      "rewards/format_reward": 1.0,
      "step": 877
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 316.78125,
      "epoch": 1.2027397260273973,
      "grad_norm": 8.833257675170898,
      "kl": 0.41357421875,
      "learning_rate": 5.990867579908675e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.1462521031498909,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 878
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 437.96875,
      "epoch": 1.2041095890410958,
      "grad_norm": 6.765456199645996,
      "kl": 0.45166015625,
      "learning_rate": 5.986301369863014e-07,
      "loss": 0.0005,
      "reward": 1.8531250357627869,
      "reward_std": 0.34787509217858315,
      "rewards/accuracy_reward": 0.8843750059604645,
      "rewards/format_reward": 0.96875,
      "step": 879
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.6875,
      "epoch": 1.2054794520547945,
      "grad_norm": 1.5083391666412354,
      "kl": 0.39013671875,
      "learning_rate": 5.981735159817351e-07,
      "loss": 0.0004,
      "reward": 1.5703125,
      "reward_std": 0.0657544769346714,
      "rewards/accuracy_reward": 0.5703125,
      "rewards/format_reward": 1.0,
      "step": 880
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 284.84375,
      "epoch": 1.2068493150684931,
      "grad_norm": 0.8882075548171997,
      "kl": 0.4072265625,
      "learning_rate": 5.97716894977169e-07,
      "loss": 0.0004,
      "reward": 1.453125,
      "reward_std": 0.0867956355214119,
      "rewards/accuracy_reward": 0.453125,
      "rewards/format_reward": 1.0,
      "step": 881
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 305.40625,
      "epoch": 1.2082191780821918,
      "grad_norm": 1.7317863702774048,
      "kl": 0.427734375,
      "learning_rate": 5.972602739726027e-07,
      "loss": 0.0004,
      "reward": 1.65625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 882
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 560.3125,
      "epoch": 1.2095890410958905,
      "grad_norm": 11.200748443603516,
      "kl": 0.46435546875,
      "learning_rate": 5.968036529680364e-07,
      "loss": 0.0005,
      "reward": 1.5625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 883
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 425.90625,
      "epoch": 1.210958904109589,
      "grad_norm": 5.06436014175415,
      "kl": 0.44775390625,
      "learning_rate": 5.963470319634703e-07,
      "loss": 0.0004,
      "reward": 1.625,
      "reward_std": 0.2314550280570984,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 1.0,
      "step": 884
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 322.125,
      "epoch": 1.2123287671232876,
      "grad_norm": 3.117278575897217,
      "kl": 0.39013671875,
      "learning_rate": 5.958904109589041e-07,
      "loss": 0.0004,
      "reward": 1.15625,
      "reward_std": 0.3061639815568924,
      "rewards/accuracy_reward": 0.15625,
      "rewards/format_reward": 1.0,
      "step": 885
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.0625,
      "epoch": 1.2136986301369863,
      "grad_norm": 3.001260757446289,
      "kl": 0.390625,
      "learning_rate": 5.95433789954338e-07,
      "loss": 0.0004,
      "reward": 2.359375,
      "reward_std": 0.0289318785071373,
      "rewards/accuracy_reward": 1.359375,
      "rewards/format_reward": 1.0,
      "step": 886
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 422.78125,
      "epoch": 1.215068493150685,
      "grad_norm": 4.375697135925293,
      "kl": 0.39306640625,
      "learning_rate": 5.949771689497717e-07,
      "loss": 0.0004,
      "reward": 1.5625,
      "reward_std": 0.2177756354212761,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 887
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.71875,
      "epoch": 1.2164383561643834,
      "grad_norm": 1.2586559057235718,
      "kl": 0.3837890625,
      "learning_rate": 5.945205479452054e-07,
      "loss": 0.0004,
      "reward": 1.421875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 888
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.21875,
      "epoch": 1.2178082191780821,
      "grad_norm": 0.010667498223483562,
      "kl": 0.3828125,
      "learning_rate": 5.940639269406393e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 889
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.0625,
      "epoch": 1.2191780821917808,
      "grad_norm": 4.928837299346924,
      "kl": 0.37646484375,
      "learning_rate": 5.93607305936073e-07,
      "loss": 0.0004,
      "reward": 2.0390625,
      "reward_std": 0.061278700828552246,
      "rewards/accuracy_reward": 1.0390625,
      "rewards/format_reward": 1.0,
      "step": 890
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.6875,
      "epoch": 1.2205479452054795,
      "grad_norm": 3.472158193588257,
      "kl": 0.39794921875,
      "learning_rate": 5.931506849315067e-07,
      "loss": 0.0004,
      "reward": 1.6875,
      "reward_std": 0.2177756354212761,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 1.0,
      "step": 891
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.375,
      "epoch": 1.2219178082191782,
      "grad_norm": 1.6116597652435303,
      "kl": 0.41357421875,
      "learning_rate": 5.926940639269407e-07,
      "loss": 0.0004,
      "reward": 1.109375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.109375,
      "rewards/format_reward": 1.0,
      "step": 892
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 548.46875,
      "epoch": 1.2232876712328766,
      "grad_norm": 2.2251503467559814,
      "kl": 0.39697265625,
      "learning_rate": 5.922374429223744e-07,
      "loss": 0.0004,
      "reward": 1.8541666865348816,
      "reward_std": 0.32778636924922466,
      "rewards/accuracy_reward": 0.8854166567325592,
      "rewards/format_reward": 0.96875,
      "step": 893
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.15625,
      "epoch": 1.2246575342465753,
      "grad_norm": 2.8448946475982666,
      "kl": 0.3955078125,
      "learning_rate": 5.917808219178083e-07,
      "loss": 0.0004,
      "reward": 1.375,
      "reward_std": 0.1552036553621292,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 894
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.8125,
      "epoch": 1.226027397260274,
      "grad_norm": 2.6070163249969482,
      "kl": 0.400390625,
      "learning_rate": 5.91324200913242e-07,
      "loss": 0.0004,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 895
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.21875,
      "epoch": 1.2273972602739727,
      "grad_norm": 1.7733893394470215,
      "kl": 0.38232421875,
      "learning_rate": 5.908675799086757e-07,
      "loss": 0.0004,
      "reward": 1.55859375,
      "reward_std": 0.07789094373583794,
      "rewards/accuracy_reward": 0.55859375,
      "rewards/format_reward": 1.0,
      "step": 896
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 431.40625,
      "epoch": 1.2287671232876711,
      "grad_norm": 4.120666027069092,
      "kl": 0.40087890625,
      "learning_rate": 5.904109589041096e-07,
      "loss": 0.0004,
      "reward": 1.6651785373687744,
      "reward_std": 0.17251565493643284,
      "rewards/accuracy_reward": 0.6651785373687744,
      "rewards/format_reward": 1.0,
      "step": 897
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 547.0625,
      "epoch": 1.2301369863013698,
      "grad_norm": 2.522732973098755,
      "kl": 0.38720703125,
      "learning_rate": 5.899543378995433e-07,
      "loss": 0.0004,
      "reward": 1.6119791567325592,
      "reward_std": 0.25090846233069897,
      "rewards/accuracy_reward": 0.6119791716337204,
      "rewards/format_reward": 1.0,
      "step": 898
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 572.9375,
      "epoch": 1.2315068493150685,
      "grad_norm": 2.1330246925354004,
      "kl": 0.390625,
      "learning_rate": 5.894977168949771e-07,
      "loss": 0.0004,
      "reward": 1.25,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 0.96875,
      "step": 899
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 482.0625,
      "epoch": 1.2328767123287672,
      "grad_norm": 6.462682247161865,
      "kl": 0.38671875,
      "learning_rate": 5.89041095890411e-07,
      "loss": 0.0004,
      "reward": 1.4609375,
      "reward_std": 0.1662898138165474,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 1.0,
      "step": 900
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 550.21875,
      "epoch": 1.2342465753424658,
      "grad_norm": 2.9799962043762207,
      "kl": 0.37939453125,
      "learning_rate": 5.885844748858447e-07,
      "loss": 0.0004,
      "reward": 2.119791656732559,
      "reward_std": 0.014731401577591896,
      "rewards/accuracy_reward": 1.1197916567325592,
      "rewards/format_reward": 1.0,
      "step": 901
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 422.90625,
      "epoch": 1.2356164383561643,
      "grad_norm": 3.0818071365356445,
      "kl": 0.4287109375,
      "learning_rate": 5.881278538812785e-07,
      "loss": 0.0004,
      "reward": 1.546875,
      "reward_std": 0.22733328863978386,
      "rewards/accuracy_reward": 0.5468750149011612,
      "rewards/format_reward": 1.0,
      "step": 902
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 457.28125,
      "epoch": 1.236986301369863,
      "grad_norm": 0.7816328406333923,
      "kl": 0.41259765625,
      "learning_rate": 5.876712328767123e-07,
      "loss": 0.0004,
      "reward": 1.609375,
      "reward_std": 0.17278572916984558,
      "rewards/accuracy_reward": 0.640625,
      "rewards/format_reward": 0.96875,
      "step": 903
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.96875,
      "epoch": 1.2383561643835617,
      "grad_norm": 2.238757848739624,
      "kl": 0.37060546875,
      "learning_rate": 5.87214611872146e-07,
      "loss": 0.0004,
      "reward": 1.8671875,
      "reward_std": 0.27381716668605804,
      "rewards/accuracy_reward": 0.8984375,
      "rewards/format_reward": 0.96875,
      "step": 904
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.5625,
      "epoch": 1.2397260273972603,
      "grad_norm": 2.3652467727661133,
      "kl": 0.38232421875,
      "learning_rate": 5.8675799086758e-07,
      "loss": 0.0004,
      "reward": 1.5364583134651184,
      "reward_std": 0.13729207031428814,
      "rewards/accuracy_reward": 0.5364583283662796,
      "rewards/format_reward": 1.0,
      "step": 905
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.53125,
      "epoch": 1.2410958904109588,
      "grad_norm": 0.013698413036763668,
      "kl": 0.3779296875,
      "learning_rate": 5.863013698630137e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 906
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 468.875,
      "epoch": 1.2424657534246575,
      "grad_norm": 2.4582483768463135,
      "kl": 0.35595703125,
      "learning_rate": 5.858447488584474e-07,
      "loss": 0.0004,
      "reward": 1.484375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 907
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 414.53125,
      "epoch": 1.2438356164383562,
      "grad_norm": 1.652496576309204,
      "kl": 0.4560546875,
      "learning_rate": 5.853881278538813e-07,
      "loss": 0.0005,
      "reward": 1.03125,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.03125,
      "rewards/format_reward": 1.0,
      "step": 908
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 557.53125,
      "epoch": 1.2452054794520548,
      "grad_norm": 1.0350927114486694,
      "kl": 0.3720703125,
      "learning_rate": 5.84931506849315e-07,
      "loss": 0.0004,
      "reward": 1.671875,
      "reward_std": 0.0646936446428299,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 909
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 528.375,
      "epoch": 1.2465753424657535,
      "grad_norm": 2.5157017707824707,
      "kl": 0.40625,
      "learning_rate": 5.844748858447488e-07,
      "loss": 0.0004,
      "reward": 1.4947916567325592,
      "reward_std": 0.1395920068025589,
      "rewards/accuracy_reward": 0.4947916567325592,
      "rewards/format_reward": 1.0,
      "step": 910
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 358.375,
      "epoch": 1.247945205479452,
      "grad_norm": 1.965340256690979,
      "kl": 0.37060546875,
      "learning_rate": 5.840182648401826e-07,
      "loss": 0.0004,
      "reward": 1.203125,
      "reward_std": 0.1530819907784462,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 1.0,
      "step": 911
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 463.15625,
      "epoch": 1.2493150684931507,
      "grad_norm": 1.2656036615371704,
      "kl": 0.39697265625,
      "learning_rate": 5.835616438356164e-07,
      "loss": 0.0004,
      "reward": 1.359375,
      "reward_std": 0.12255740165710449,
      "rewards/accuracy_reward": 0.359375,
      "rewards/format_reward": 1.0,
      "step": 912
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 378.375,
      "epoch": 1.2506849315068493,
      "grad_norm": 2.7013890743255615,
      "kl": 0.37451171875,
      "learning_rate": 5.831050228310503e-07,
      "loss": 0.0004,
      "reward": 1.734375,
      "reward_std": 0.22097086533904076,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 913
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.34375,
      "epoch": 1.252054794520548,
      "grad_norm": 2.896122932434082,
      "kl": 0.3740234375,
      "learning_rate": 5.82648401826484e-07,
      "loss": 0.0004,
      "reward": 2.0625,
      "reward_std": 0.26197961531579494,
      "rewards/accuracy_reward": 1.0625,
      "rewards/format_reward": 1.0,
      "step": 914
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 382.09375,
      "epoch": 1.2534246575342465,
      "grad_norm": 3.789921998977661,
      "kl": 0.3935546875,
      "learning_rate": 5.821917808219177e-07,
      "loss": 0.0004,
      "reward": 1.328125,
      "reward_std": 0.28930897638201714,
      "rewards/accuracy_reward": 0.328125,
      "rewards/format_reward": 1.0,
      "step": 915
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.21875,
      "epoch": 1.2547945205479452,
      "grad_norm": 2.50876522064209,
      "kl": 0.408203125,
      "learning_rate": 5.817351598173516e-07,
      "loss": 0.0004,
      "reward": 1.671875,
      "reward_std": 0.1583191677927971,
      "rewards/accuracy_reward": 0.671875,
      "rewards/format_reward": 1.0,
      "step": 916
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.125,
      "epoch": 1.2561643835616438,
      "grad_norm": 2.9093079566955566,
      "kl": 0.380859375,
      "learning_rate": 5.812785388127853e-07,
      "loss": 0.0004,
      "reward": 2.0078125,
      "reward_std": 0.163336630910635,
      "rewards/accuracy_reward": 1.0078125,
      "rewards/format_reward": 1.0,
      "step": 917
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.65625,
      "epoch": 1.2575342465753425,
      "grad_norm": 1.01085364818573,
      "kl": 0.37353515625,
      "learning_rate": 5.808219178082191e-07,
      "loss": 0.0004,
      "reward": 1.484375,
      "reward_std": 0.021564556285738945,
      "rewards/accuracy_reward": 0.484375,
      "rewards/format_reward": 1.0,
      "step": 918
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 572.0625,
      "epoch": 1.2589041095890412,
      "grad_norm": 1.2397353649139404,
      "kl": 0.36962890625,
      "learning_rate": 5.80365296803653e-07,
      "loss": 0.0004,
      "reward": 1.1875,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 0.96875,
      "step": 919
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 478.5,
      "epoch": 1.2602739726027397,
      "grad_norm": 1.3165117502212524,
      "kl": 0.3916015625,
      "learning_rate": 5.799086757990867e-07,
      "loss": 0.0004,
      "reward": 1.5625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 920
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 682.6875,
      "epoch": 1.2616438356164383,
      "grad_norm": 2.271702289581299,
      "kl": 0.3701171875,
      "learning_rate": 5.794520547945206e-07,
      "loss": 0.0004,
      "reward": 1.77734375,
      "reward_std": 0.32882310450077057,
      "rewards/accuracy_reward": 0.80859375,
      "rewards/format_reward": 0.96875,
      "step": 921
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 375.09375,
      "epoch": 1.263013698630137,
      "grad_norm": 0.017480703070759773,
      "kl": 0.3984375,
      "learning_rate": 5.789954337899543e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 1.0,
      "step": 922
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 504.65625,
      "epoch": 1.2643835616438357,
      "grad_norm": 1.695114016532898,
      "kl": 0.4033203125,
      "learning_rate": 5.78538812785388e-07,
      "loss": 0.0004,
      "reward": 1.5234375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.5234375,
      "rewards/format_reward": 1.0,
      "step": 923
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.90625,
      "epoch": 1.2657534246575342,
      "grad_norm": 1.5401252508163452,
      "kl": 0.353515625,
      "learning_rate": 5.780821917808219e-07,
      "loss": 0.0004,
      "reward": 1.7708333134651184,
      "reward_std": 0.05891544930636883,
      "rewards/accuracy_reward": 0.7708333283662796,
      "rewards/format_reward": 1.0,
      "step": 924
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 469.78125,
      "epoch": 1.2671232876712328,
      "grad_norm": 3.2033495903015137,
      "kl": 0.39013671875,
      "learning_rate": 5.776255707762557e-07,
      "loss": 0.0004,
      "reward": 1.55859375,
      "reward_std": 0.27621358167380095,
      "rewards/accuracy_reward": 0.58984375,
      "rewards/format_reward": 0.96875,
      "step": 925
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.28125,
      "epoch": 1.2684931506849315,
      "grad_norm": 3.77433705329895,
      "kl": 0.4111328125,
      "learning_rate": 5.771689497716896e-07,
      "loss": 0.0004,
      "reward": 1.3177083134651184,
      "reward_std": 0.4321344643831253,
      "rewards/accuracy_reward": 0.3802083283662796,
      "rewards/format_reward": 0.9375,
      "step": 926
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 591.78125,
      "epoch": 1.2698630136986302,
      "grad_norm": 4.000980854034424,
      "kl": 0.3642578125,
      "learning_rate": 5.767123287671233e-07,
      "loss": 0.0004,
      "reward": 1.515625,
      "reward_std": 0.0828370526432991,
      "rewards/accuracy_reward": 0.5156249701976776,
      "rewards/format_reward": 1.0,
      "step": 927
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 379.28125,
      "epoch": 1.2712328767123289,
      "grad_norm": 2.266773223876953,
      "kl": 0.37060546875,
      "learning_rate": 5.76255707762557e-07,
      "loss": 0.0004,
      "reward": 1.53125,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 1.0,
      "step": 928
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 498.125,
      "epoch": 1.2726027397260273,
      "grad_norm": 5.215886116027832,
      "kl": 0.4228515625,
      "learning_rate": 5.757990867579909e-07,
      "loss": 0.0004,
      "reward": 1.6822916269302368,
      "reward_std": 0.24831003323197365,
      "rewards/accuracy_reward": 0.7135416269302368,
      "rewards/format_reward": 0.96875,
      "step": 929
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.78125,
      "epoch": 1.273972602739726,
      "grad_norm": 1.8586878776550293,
      "kl": 0.44384765625,
      "learning_rate": 5.753424657534246e-07,
      "loss": 0.0004,
      "reward": 1.390625,
      "reward_std": 0.2302001230418682,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 0.96875,
      "step": 930
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 773.1875,
      "epoch": 1.2753424657534247,
      "grad_norm": 1.8589192628860474,
      "kl": 0.36376953125,
      "learning_rate": 5.748858447488583e-07,
      "loss": 0.0004,
      "reward": 2.052604168653488,
      "reward_std": 0.25580091029405594,
      "rewards/accuracy_reward": 1.0838541686534882,
      "rewards/format_reward": 0.96875,
      "step": 931
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.84375,
      "epoch": 1.2767123287671232,
      "grad_norm": 1.5999672412872314,
      "kl": 1.4580078125,
      "learning_rate": 5.744292237442923e-07,
      "loss": 0.0015,
      "reward": 1.46875,
      "reward_std": 0.033407654613256454,
      "rewards/accuracy_reward": 0.46875,
      "rewards/format_reward": 1.0,
      "step": 932
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 668.84375,
      "epoch": 1.2780821917808218,
      "grad_norm": 1.7803735733032227,
      "kl": 0.39013671875,
      "learning_rate": 5.73972602739726e-07,
      "loss": 0.0004,
      "reward": 2.435156285762787,
      "reward_std": 0.16129423771053553,
      "rewards/accuracy_reward": 1.4351562857627869,
      "rewards/format_reward": 1.0,
      "step": 933
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.46875,
      "epoch": 1.2794520547945205,
      "grad_norm": 3.8296427726745605,
      "kl": 0.39404296875,
      "learning_rate": 5.735159817351598e-07,
      "loss": 0.0004,
      "reward": 1.65625,
      "reward_std": 0.22201896272599697,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 934
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 509.03125,
      "epoch": 1.2808219178082192,
      "grad_norm": 0.021630197763442993,
      "kl": 0.4111328125,
      "learning_rate": 5.730593607305936e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 1.0,
      "step": 935
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 511.90625,
      "epoch": 1.2821917808219179,
      "grad_norm": 1.9150142669677734,
      "kl": 0.42236328125,
      "learning_rate": 5.726027397260273e-07,
      "loss": 0.0004,
      "reward": 1.4791666269302368,
      "reward_std": 0.3344755917787552,
      "rewards/accuracy_reward": 0.5104166716337204,
      "rewards/format_reward": 0.96875,
      "step": 936
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 778.40625,
      "epoch": 1.2835616438356166,
      "grad_norm": 1.2064579725265503,
      "kl": 0.3486328125,
      "learning_rate": 5.721461187214612e-07,
      "loss": 0.0003,
      "reward": 1.671875,
      "reward_std": 0.22097086533904076,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.96875,
      "step": 937
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 283.5625,
      "epoch": 1.284931506849315,
      "grad_norm": 2.410706043243408,
      "kl": 0.44140625,
      "learning_rate": 5.716894977168949e-07,
      "loss": 0.0004,
      "reward": 1.0625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 1.0,
      "step": 938
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.0,
      "epoch": 1.2863013698630137,
      "grad_norm": 2.2251358032226562,
      "kl": 0.3798828125,
      "learning_rate": 5.712328767123287e-07,
      "loss": 0.0004,
      "reward": 1.734375,
      "reward_std": 0.24511976540088654,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 939
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 461.78125,
      "epoch": 1.2876712328767124,
      "grad_norm": 2.1631131172180176,
      "kl": 0.4345703125,
      "learning_rate": 5.707762557077626e-07,
      "loss": 0.0004,
      "reward": 1.3177083134651184,
      "reward_std": 0.3885781615972519,
      "rewards/accuracy_reward": 0.3489583134651184,
      "rewards/format_reward": 0.96875,
      "step": 940
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.125,
      "epoch": 1.2890410958904108,
      "grad_norm": 2.293469190597534,
      "kl": 0.51708984375,
      "learning_rate": 5.703196347031963e-07,
      "loss": 0.0005,
      "reward": 1.421875,
      "reward_std": 0.1446593925356865,
      "rewards/accuracy_reward": 0.421875,
      "rewards/format_reward": 1.0,
      "step": 941
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.53125,
      "epoch": 1.2904109589041095,
      "grad_norm": 0.01647823676466942,
      "kl": 0.3896484375,
      "learning_rate": 5.698630136986301e-07,
      "loss": 0.0004,
      "reward": 2.5,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.5,
      "rewards/format_reward": 1.0,
      "step": 942
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.71875,
      "epoch": 1.2917808219178082,
      "grad_norm": 2.266577959060669,
      "kl": 0.80126953125,
      "learning_rate": 5.694063926940639e-07,
      "loss": 0.0008,
      "reward": 1.5963541865348816,
      "reward_std": 0.2827804908156395,
      "rewards/accuracy_reward": 0.6276041865348816,
      "rewards/format_reward": 0.96875,
      "step": 943
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.25,
      "epoch": 1.2931506849315069,
      "grad_norm": 3.0220417976379395,
      "kl": 0.40234375,
      "learning_rate": 5.689497716894976e-07,
      "loss": 0.0004,
      "reward": 1.515625,
      "reward_std": 0.16828217171132565,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 944
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 693.875,
      "epoch": 1.2945205479452055,
      "grad_norm": 0.9308492541313171,
      "kl": 0.3837890625,
      "learning_rate": 5.684931506849316e-07,
      "loss": 0.0004,
      "reward": 1.6796875,
      "reward_std": 0.09704047441482544,
      "rewards/accuracy_reward": 0.6796875,
      "rewards/format_reward": 1.0,
      "step": 945
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.5,
      "epoch": 1.2958904109589042,
      "grad_norm": 1.6099046468734741,
      "kl": 0.5966796875,
      "learning_rate": 5.680365296803653e-07,
      "loss": 0.0006,
      "reward": 1.3203125,
      "reward_std": 0.10436524450778961,
      "rewards/accuracy_reward": 0.3203125,
      "rewards/format_reward": 1.0,
      "step": 946
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 635.0625,
      "epoch": 1.2972602739726027,
      "grad_norm": 2.74033784866333,
      "kl": 0.3798828125,
      "learning_rate": 5.67579908675799e-07,
      "loss": 0.0004,
      "reward": 1.8953125476837158,
      "reward_std": 0.3537486009299755,
      "rewards/accuracy_reward": 0.9265625178813934,
      "rewards/format_reward": 0.96875,
      "step": 947
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 654.90625,
      "epoch": 1.2986301369863014,
      "grad_norm": 1.12486732006073,
      "kl": 0.41796875,
      "learning_rate": 5.671232876712329e-07,
      "loss": 0.0004,
      "reward": 1.5078125,
      "reward_std": 0.306659497320652,
      "rewards/accuracy_reward": 0.5703125,
      "rewards/format_reward": 0.9375,
      "step": 948
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.1875,
      "epoch": 1.3,
      "grad_norm": 0.8883933424949646,
      "kl": 0.41259765625,
      "learning_rate": 5.666666666666666e-07,
      "loss": 0.0004,
      "reward": 1.734375,
      "reward_std": 0.04419417306780815,
      "rewards/accuracy_reward": 0.734375,
      "rewards/format_reward": 1.0,
      "step": 949
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.03125,
      "epoch": 1.3013698630136985,
      "grad_norm": 3.347505569458008,
      "kl": 0.39111328125,
      "learning_rate": 5.662100456621004e-07,
      "loss": 0.0004,
      "reward": 1.515625,
      "reward_std": 0.1304589118808508,
      "rewards/accuracy_reward": 0.515625,
      "rewards/format_reward": 1.0,
      "step": 950
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 493.15625,
      "epoch": 1.3027397260273972,
      "grad_norm": 2.3713202476501465,
      "kl": 0.46923828125,
      "learning_rate": 5.657534246575342e-07,
      "loss": 0.0005,
      "reward": 1.2864583730697632,
      "reward_std": 0.13258253410458565,
      "rewards/accuracy_reward": 0.3177083432674408,
      "rewards/format_reward": 0.96875,
      "step": 951
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 650.3125,
      "epoch": 1.3041095890410959,
      "grad_norm": 3.497941017150879,
      "kl": 0.37890625,
      "learning_rate": 5.65296803652968e-07,
      "loss": 0.0004,
      "reward": 1.0625,
      "reward_std": 0.1157275140285492,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 1.0,
      "step": 952
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 620.25,
      "epoch": 1.3054794520547945,
      "grad_norm": 2.711280107498169,
      "kl": 0.38818359375,
      "learning_rate": 5.648401826484019e-07,
      "loss": 0.0004,
      "reward": 1.70703125,
      "reward_std": 0.19521116837859154,
      "rewards/accuracy_reward": 0.70703125,
      "rewards/format_reward": 1.0,
      "step": 953
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 506.21875,
      "epoch": 1.3068493150684932,
      "grad_norm": 2.053891181945801,
      "kl": 0.4052734375,
      "learning_rate": 5.643835616438356e-07,
      "loss": 0.0004,
      "reward": 1.5625,
      "reward_std": 0.249358132481575,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 954
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 473.5625,
      "epoch": 1.308219178082192,
      "grad_norm": 6.742806911468506,
      "kl": 0.4111328125,
      "learning_rate": 5.639269406392693e-07,
      "loss": 0.0004,
      "reward": 1.3177083134651184,
      "reward_std": 0.16407955065369606,
      "rewards/accuracy_reward": 0.3489583283662796,
      "rewards/format_reward": 0.96875,
      "step": 955
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 397.0,
      "epoch": 1.3095890410958904,
      "grad_norm": 3.2806613445281982,
      "kl": 0.6689453125,
      "learning_rate": 5.634703196347032e-07,
      "loss": 0.0007,
      "reward": 1.34375,
      "reward_std": 0.482940673828125,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 0.96875,
      "step": 956
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 721.625,
      "epoch": 1.310958904109589,
      "grad_norm": 2.4216856956481934,
      "kl": 0.380859375,
      "learning_rate": 5.630136986301369e-07,
      "loss": 0.0004,
      "reward": 1.8515625,
      "reward_std": 0.1936504878103733,
      "rewards/accuracy_reward": 0.8515625,
      "rewards/format_reward": 1.0,
      "step": 957
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.96875,
      "epoch": 1.3123287671232877,
      "grad_norm": 2.3214023113250732,
      "kl": 0.375,
      "learning_rate": 5.625570776255707e-07,
      "loss": 0.0004,
      "reward": 1.46875,
      "reward_std": 0.2884480655193329,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 958
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.1875,
      "epoch": 1.3136986301369862,
      "grad_norm": 0.8679585456848145,
      "kl": 0.41845703125,
      "learning_rate": 5.621004566210046e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 1.0,
      "step": 959
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 591.90625,
      "epoch": 1.3150684931506849,
      "grad_norm": 1.6525453329086304,
      "kl": 0.37451171875,
      "learning_rate": 5.616438356164383e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 960
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.59375,
      "epoch": 1.3164383561643835,
      "grad_norm": 1.9350600242614746,
      "kl": 0.37255859375,
      "learning_rate": 5.611872146118722e-07,
      "loss": 0.0004,
      "reward": 1.8984375,
      "reward_std": 0.11048543266952038,
      "rewards/accuracy_reward": 0.9296875,
      "rewards/format_reward": 0.96875,
      "step": 961
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 417.8125,
      "epoch": 1.3178082191780822,
      "grad_norm": 2.944469928741455,
      "kl": 0.416015625,
      "learning_rate": 5.607305936073059e-07,
      "loss": 0.0004,
      "reward": 1.5234375,
      "reward_std": 0.4395580645650625,
      "rewards/accuracy_reward": 0.5546875,
      "rewards/format_reward": 0.96875,
      "step": 962
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.84375,
      "epoch": 1.319178082191781,
      "grad_norm": 3.110598564147949,
      "kl": 0.39990234375,
      "learning_rate": 5.602739726027396e-07,
      "loss": 0.0004,
      "reward": 1.4921875,
      "reward_std": 0.1297733597457409,
      "rewards/accuracy_reward": 0.4921874701976776,
      "rewards/format_reward": 1.0,
      "step": 963
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 516.3125,
      "epoch": 1.3205479452054796,
      "grad_norm": 3.190995216369629,
      "kl": 0.39306640625,
      "learning_rate": 5.598173515981735e-07,
      "loss": 0.0004,
      "reward": 1.6953125,
      "reward_std": 0.2972095049917698,
      "rewards/accuracy_reward": 0.6953125,
      "rewards/format_reward": 1.0,
      "step": 964
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 624.34375,
      "epoch": 1.321917808219178,
      "grad_norm": 1.7255425453186035,
      "kl": 0.41748046875,
      "learning_rate": 5.593607305936073e-07,
      "loss": 0.0004,
      "reward": 1.515625,
      "reward_std": 0.24831003323197365,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.96875,
      "step": 965
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.375,
      "epoch": 1.3232876712328767,
      "grad_norm": 2.058529853820801,
      "kl": 0.36181640625,
      "learning_rate": 5.589041095890411e-07,
      "loss": 0.0004,
      "reward": 1.84375,
      "reward_std": 0.30173346400260925,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 966
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.5625,
      "epoch": 1.3246575342465754,
      "grad_norm": 0.08733844757080078,
      "kl": 0.435546875,
      "learning_rate": 5.584474885844749e-07,
      "loss": 0.0004,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 967
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.5,
      "epoch": 1.3260273972602739,
      "grad_norm": 4.318344593048096,
      "kl": 0.40234375,
      "learning_rate": 5.579908675799086e-07,
      "loss": 0.0004,
      "reward": 1.09375,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.09375,
      "rewards/format_reward": 1.0,
      "step": 968
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.4375,
      "epoch": 1.3273972602739725,
      "grad_norm": 7.782508373260498,
      "kl": 0.4287109375,
      "learning_rate": 5.575342465753425e-07,
      "loss": 0.0004,
      "reward": 1.28125,
      "reward_std": 0.4218914955854416,
      "rewards/accuracy_reward": 0.28125,
      "rewards/format_reward": 1.0,
      "step": 969
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 784.46875,
      "epoch": 1.3287671232876712,
      "grad_norm": 3.4223484992980957,
      "kl": 0.35205078125,
      "learning_rate": 5.570776255707762e-07,
      "loss": 0.0004,
      "reward": 1.15625,
      "reward_std": 0.2651650384068489,
      "rewards/accuracy_reward": 0.21875,
      "rewards/format_reward": 0.9375,
      "step": 970
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 512.125,
      "epoch": 1.33013698630137,
      "grad_norm": 1.055464267730713,
      "kl": 0.39599609375,
      "learning_rate": 5.566210045662099e-07,
      "loss": 0.0004,
      "reward": 1.45703125,
      "reward_std": 0.028628919273614883,
      "rewards/accuracy_reward": 0.45703125,
      "rewards/format_reward": 1.0,
      "step": 971
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 455.53125,
      "epoch": 1.3315068493150686,
      "grad_norm": 3.628035068511963,
      "kl": 0.41015625,
      "learning_rate": 5.561643835616439e-07,
      "loss": 0.0004,
      "reward": 1.46875,
      "reward_std": 0.4397946000099182,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.96875,
      "step": 972
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 442.3125,
      "epoch": 1.3328767123287673,
      "grad_norm": 1.331992745399475,
      "kl": 0.4189453125,
      "learning_rate": 5.557077625570776e-07,
      "loss": 0.0004,
      "reward": 1.59375,
      "reward_std": 0.1293872892856598,
      "rewards/accuracy_reward": 0.59375,
      "rewards/format_reward": 1.0,
      "step": 973
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.0625,
      "epoch": 1.3342465753424657,
      "grad_norm": 1.930756688117981,
      "kl": 0.373046875,
      "learning_rate": 5.552511415525114e-07,
      "loss": 0.0004,
      "reward": 1.625,
      "reward_std": 0.2925042062997818,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 0.96875,
      "step": 974
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.09375,
      "epoch": 1.3356164383561644,
      "grad_norm": 9.222930908203125,
      "kl": 0.39697265625,
      "learning_rate": 5.547945205479452e-07,
      "loss": 0.0004,
      "reward": 1.5,
      "reward_std": 0.3777071312069893,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.96875,
      "step": 975
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.3125,
      "epoch": 1.336986301369863,
      "grad_norm": 2.1612720489501953,
      "kl": 0.38427734375,
      "learning_rate": 5.543378995433789e-07,
      "loss": 0.0004,
      "reward": 1.7994791567325592,
      "reward_std": 0.37991958670318127,
      "rewards/accuracy_reward": 0.8307291567325592,
      "rewards/format_reward": 0.96875,
      "step": 976
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 667.65625,
      "epoch": 1.3383561643835615,
      "grad_norm": 3.2220661640167236,
      "kl": 0.3798828125,
      "learning_rate": 5.538812785388128e-07,
      "loss": 0.0004,
      "reward": 1.1953125,
      "reward_std": 0.12073516845703125,
      "rewards/accuracy_reward": 0.2265625,
      "rewards/format_reward": 0.96875,
      "step": 977
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 709.03125,
      "epoch": 1.3397260273972602,
      "grad_norm": 7.830074310302734,
      "kl": 0.39794921875,
      "learning_rate": 5.534246575342465e-07,
      "loss": 0.0004,
      "reward": 1.765625,
      "reward_std": 0.05444390885531902,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 978
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.5625,
      "epoch": 1.341095890410959,
      "grad_norm": 2.3540923595428467,
      "kl": 0.3916015625,
      "learning_rate": 5.529680365296803e-07,
      "loss": 0.0004,
      "reward": 1.03125,
      "reward_std": 0.2041158601641655,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.96875,
      "step": 979
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.375,
      "epoch": 1.3424657534246576,
      "grad_norm": 12.305758476257324,
      "kl": 0.44287109375,
      "learning_rate": 5.525114155251142e-07,
      "loss": 0.0004,
      "reward": 1.9375,
      "reward_std": 0.2884863168001175,
      "rewards/accuracy_reward": 0.9375000596046448,
      "rewards/format_reward": 1.0,
      "step": 980
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 665.3125,
      "epoch": 1.3438356164383563,
      "grad_norm": 5.535678863525391,
      "kl": 0.3837890625,
      "learning_rate": 5.520547945205479e-07,
      "loss": 0.0004,
      "reward": 1.576562523841858,
      "reward_std": 0.16898519545793533,
      "rewards/accuracy_reward": 0.5765625238418579,
      "rewards/format_reward": 1.0,
      "step": 981
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 740.0,
      "epoch": 1.345205479452055,
      "grad_norm": 0.8980807065963745,
      "kl": 0.4013671875,
      "learning_rate": 5.515981735159817e-07,
      "loss": 0.0004,
      "reward": 2.253125011920929,
      "reward_std": 0.13258251547813416,
      "rewards/accuracy_reward": 1.253125011920929,
      "rewards/format_reward": 1.0,
      "step": 982
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 435.0,
      "epoch": 1.3465753424657534,
      "grad_norm": 1.950405240058899,
      "kl": 0.40380859375,
      "learning_rate": 5.511415525114155e-07,
      "loss": 0.0004,
      "reward": 1.5625,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 1.0,
      "step": 983
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 785.625,
      "epoch": 1.347945205479452,
      "grad_norm": 0.8518653512001038,
      "kl": 0.357421875,
      "learning_rate": 5.506849315068492e-07,
      "loss": 0.0004,
      "reward": 1.21875,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 0.96875,
      "step": 984
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 726.53125,
      "epoch": 1.3493150684931507,
      "grad_norm": 1.3810690641403198,
      "kl": 0.40283203125,
      "learning_rate": 5.502283105022832e-07,
      "loss": 0.0004,
      "reward": 1.7838541269302368,
      "reward_std": 0.05391141213476658,
      "rewards/accuracy_reward": 0.7838541567325592,
      "rewards/format_reward": 1.0,
      "step": 985
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 675.5625,
      "epoch": 1.3506849315068492,
      "grad_norm": 3.373875141143799,
      "kl": 0.36474609375,
      "learning_rate": 5.497716894977169e-07,
      "loss": 0.0004,
      "reward": 1.84375,
      "reward_std": 0.28276579082012177,
      "rewards/accuracy_reward": 0.875,
      "rewards/format_reward": 0.96875,
      "step": 986
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.59375,
      "epoch": 1.3520547945205479,
      "grad_norm": 1.4311509132385254,
      "kl": 0.4384765625,
      "learning_rate": 5.493150684931506e-07,
      "loss": 0.0004,
      "reward": 1.5364583134651184,
      "reward_std": 0.36767221987247467,
      "rewards/accuracy_reward": 0.5677083283662796,
      "rewards/format_reward": 0.96875,
      "step": 987
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.25,
      "epoch": 1.3534246575342466,
      "grad_norm": 4.090206146240234,
      "kl": 0.40185546875,
      "learning_rate": 5.488584474885845e-07,
      "loss": 0.0004,
      "reward": 2.08203125,
      "reward_std": 0.19203272461891174,
      "rewards/accuracy_reward": 1.08203125,
      "rewards/format_reward": 1.0,
      "step": 988
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 418.34375,
      "epoch": 1.3547945205479452,
      "grad_norm": 0.023737359791994095,
      "kl": 0.44921875,
      "learning_rate": 5.484018264840182e-07,
      "loss": 0.0005,
      "reward": 1.25,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 989
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.9375,
      "epoch": 1.356164383561644,
      "grad_norm": 3.110661029815674,
      "kl": 0.40283203125,
      "learning_rate": 5.47945205479452e-07,
      "loss": 0.0004,
      "reward": 1.5104166567325592,
      "reward_std": 0.18662460334599018,
      "rewards/accuracy_reward": 0.5104166567325592,
      "rewards/format_reward": 1.0,
      "step": 990
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.4375,
      "epoch": 1.3575342465753424,
      "grad_norm": 2.7677109241485596,
      "kl": 0.3935546875,
      "learning_rate": 5.474885844748858e-07,
      "loss": 0.0004,
      "reward": 1.7395833134651184,
      "reward_std": 0.09139656275510788,
      "rewards/accuracy_reward": 0.7395833432674408,
      "rewards/format_reward": 1.0,
      "step": 991
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.59375,
      "epoch": 1.358904109589041,
      "grad_norm": 1.7430964708328247,
      "kl": 0.37451171875,
      "learning_rate": 5.470319634703196e-07,
      "loss": 0.0004,
      "reward": 1.765625,
      "reward_std": 0.1804211586713791,
      "rewards/accuracy_reward": 0.765625,
      "rewards/format_reward": 1.0,
      "step": 992
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 731.75,
      "epoch": 1.3602739726027397,
      "grad_norm": 3.005925178527832,
      "kl": 0.38525390625,
      "learning_rate": 5.465753424657535e-07,
      "loss": 0.0004,
      "reward": 2.6171875,
      "reward_std": 0.022097086533904076,
      "rewards/accuracy_reward": 1.6171875,
      "rewards/format_reward": 1.0,
      "step": 993
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.21875,
      "epoch": 1.3616438356164384,
      "grad_norm": 1.7805390357971191,
      "kl": 0.39306640625,
      "learning_rate": 5.461187214611872e-07,
      "loss": 0.0004,
      "reward": 1.8854166269302368,
      "reward_std": 0.10773438680917025,
      "rewards/accuracy_reward": 0.8854166269302368,
      "rewards/format_reward": 1.0,
      "step": 994
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 307.5,
      "epoch": 1.3630136986301369,
      "grad_norm": 3.8102054595947266,
      "kl": 0.46337890625,
      "learning_rate": 5.456621004566209e-07,
      "loss": 0.0005,
      "reward": 1.65625,
      "reward_std": 0.4218914955854416,
      "rewards/accuracy_reward": 0.65625,
      "rewards/format_reward": 1.0,
      "step": 995
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 712.1875,
      "epoch": 1.3643835616438356,
      "grad_norm": 1.8652178049087524,
      "kl": 0.3828125,
      "learning_rate": 5.452054794520548e-07,
      "loss": 0.0004,
      "reward": 1.7265625,
      "reward_std": 0.03234682232141495,
      "rewards/accuracy_reward": 0.7265625,
      "rewards/format_reward": 1.0,
      "step": 996
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 396.96875,
      "epoch": 1.3657534246575342,
      "grad_norm": 3.4569380283355713,
      "kl": 0.4541015625,
      "learning_rate": 5.447488584474885e-07,
      "loss": 0.0005,
      "reward": 1.25,
      "reward_std": 0.2314550280570984,
      "rewards/accuracy_reward": 0.25,
      "rewards/format_reward": 1.0,
      "step": 997
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 550.40625,
      "epoch": 1.367123287671233,
      "grad_norm": 0.03730163723230362,
      "kl": 0.39794921875,
      "learning_rate": 5.442922374429223e-07,
      "loss": 0.0004,
      "reward": 1.375,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.375,
      "rewards/format_reward": 1.0,
      "step": 998
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 727.78125,
      "epoch": 1.3684931506849316,
      "grad_norm": 1.656997799873352,
      "kl": 0.39501953125,
      "learning_rate": 5.438356164383562e-07,
      "loss": 0.0004,
      "reward": 1.676041603088379,
      "reward_std": 0.11559091322124004,
      "rewards/accuracy_reward": 0.6760416626930237,
      "rewards/format_reward": 1.0,
      "step": 999
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 724.75,
      "epoch": 1.36986301369863,
      "grad_norm": 1.8627666234970093,
      "kl": 0.45703125,
      "learning_rate": 5.433789954337899e-07,
      "loss": 0.0005,
      "reward": 1.546875,
      "reward_std": 0.2690962068736553,
      "rewards/accuracy_reward": 0.578125,
      "rewards/format_reward": 0.96875,
      "step": 1000
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 2190,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}