{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24,
  "eval_steps": 500,
  "global_step": 450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 474.1432399749756,
      "epoch": 0.0010666666666666667,
      "grad_norm": 0.14331733537223892,
      "kl": 0.0,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0,
      "reward": 0.33333334419876337,
      "reward_std": 0.5078507382422686,
      "rewards/equation_reward_func": 0.08072916837409139,
      "rewards/format_reward_func": 0.25260417629033327,
      "step": 2
    },
    {
      "completion_length": 469.2942810058594,
      "epoch": 0.0021333333333333334,
      "grad_norm": 0.12168694868095008,
      "kl": 0.00039124488830566406,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0,
      "reward": 0.30468750931322575,
      "reward_std": 0.489006832242012,
      "rewards/equation_reward_func": 0.05989583441987634,
      "rewards/format_reward_func": 0.24479167349636555,
      "step": 4
    },
    {
      "completion_length": 475.3724060058594,
      "epoch": 0.0032,
      "grad_norm": 0.11745838648260112,
      "kl": 0.0004004240036010742,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0,
      "reward": 0.3307291741948575,
      "reward_std": 0.48131777672097087,
      "rewards/equation_reward_func": 0.06510416837409139,
      "rewards/format_reward_func": 0.26562500721774995,
      "step": 6
    },
    {
      "completion_length": 457.7838668823242,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.13351596231298812,
      "kl": 0.00042307376861572266,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.3385416753590107,
      "reward_std": 0.44462722539901733,
      "rewards/equation_reward_func": 0.05208333418704569,
      "rewards/format_reward_func": 0.28645834187045693,
      "step": 8
    },
    {
      "completion_length": 498.04428482055664,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.12357075779052926,
      "kl": 0.00043141841888427734,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0,
      "reward": 0.2734375079162419,
      "reward_std": 0.41914427001029253,
      "rewards/equation_reward_func": 0.05208333535119891,
      "rewards/format_reward_func": 0.22135417582467198,
      "step": 10
    },
    {
      "completion_length": 461.9062614440918,
      "epoch": 0.0064,
      "grad_norm": 0.14132096232063104,
      "kl": 0.0004513263702392578,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.36718751583248377,
      "reward_std": 0.5056461896747351,
      "rewards/equation_reward_func": 0.05729166814126074,
      "rewards/format_reward_func": 0.3098958395421505,
      "step": 12
    },
    {
      "completion_length": 452.78126525878906,
      "epoch": 0.007466666666666667,
      "grad_norm": 0.12890382441644657,
      "kl": 0.0007963180541992188,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.4270833432674408,
      "reward_std": 0.5310215894132853,
      "rewards/equation_reward_func": 0.07291666814126074,
      "rewards/format_reward_func": 0.3541666753590107,
      "step": 14
    },
    {
      "completion_length": 441.22657203674316,
      "epoch": 0.008533333333333334,
      "grad_norm": 0.13692556467094522,
      "kl": 0.0010395050048828125,
      "learning_rate": 4.999740409224932e-07,
      "loss": 0.0,
      "reward": 0.47916668467223644,
      "reward_std": 0.5327073819935322,
      "rewards/equation_reward_func": 0.05208333395421505,
      "rewards/format_reward_func": 0.42708334513008595,
      "step": 16
    },
    {
      "completion_length": 445.1119918823242,
      "epoch": 0.0096,
      "grad_norm": 0.11991312926921868,
      "kl": 0.0029892921447753906,
      "learning_rate": 4.998961690809627e-07,
      "loss": 0.0,
      "reward": 0.5859375223517418,
      "reward_std": 0.5514450371265411,
      "rewards/equation_reward_func": 0.049479167675599456,
      "rewards/format_reward_func": 0.5364583488553762,
      "step": 18
    },
    {
      "completion_length": 448.3099060058594,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.09106842614787719,
      "kl": 0.0060882568359375,
      "learning_rate": 4.997664006472578e-07,
      "loss": 0.0,
      "reward": 0.7421875186264515,
      "reward_std": 0.5555396415293217,
      "rewards/equation_reward_func": 0.09375000186264515,
      "rewards/format_reward_func": 0.6484375186264515,
      "step": 20
    },
    {
      "completion_length": 435.3932399749756,
      "epoch": 0.011733333333333333,
      "grad_norm": 0.09786508002778878,
      "kl": 0.0065059661865234375,
      "learning_rate": 4.995847625707292e-07,
      "loss": 0.0,
      "reward": 0.7916666977107525,
      "reward_std": 0.5166244171559811,
      "rewards/equation_reward_func": 0.0807291679084301,
      "rewards/format_reward_func": 0.7109375223517418,
      "step": 22
    },
    {
      "completion_length": 450.3671989440918,
      "epoch": 0.0128,
      "grad_norm": 0.09265470447275229,
      "kl": 0.00585174560546875,
      "learning_rate": 4.993512925726318e-07,
      "loss": 0.0,
      "reward": 0.8411458618938923,
      "reward_std": 0.5050761476159096,
      "rewards/equation_reward_func": 0.07812500232830644,
      "rewards/format_reward_func": 0.763020858168602,
      "step": 24
    },
    {
      "completion_length": 440.473970413208,
      "epoch": 0.013866666666666666,
      "grad_norm": 0.09129193085053657,
      "kl": 0.00798797607421875,
      "learning_rate": 4.990660391382923e-07,
      "loss": 0.0,
      "reward": 0.8072916865348816,
      "reward_std": 0.43093991186469793,
      "rewards/equation_reward_func": 0.039062500931322575,
      "rewards/format_reward_func": 0.7682291828095913,
      "step": 26
    },
    {
      "completion_length": 457.30470085144043,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.08687222126109678,
      "kl": 0.009069442749023438,
      "learning_rate": 4.987290615070384e-07,
      "loss": 0.0,
      "reward": 0.8515625260770321,
      "reward_std": 0.42435176484286785,
      "rewards/equation_reward_func": 0.06250000186264515,
      "rewards/format_reward_func": 0.7890625186264515,
      "step": 28
    },
    {
      "completion_length": 406.19532203674316,
      "epoch": 0.016,
      "grad_norm": 0.1021588765285621,
      "kl": 0.011852264404296875,
      "learning_rate": 4.983404296598978e-07,
      "loss": 0.0,
      "reward": 0.9609375335276127,
      "reward_std": 0.42848627734929323,
      "rewards/equation_reward_func": 0.09375000209547579,
      "rewards/format_reward_func": 0.8671875186264515,
      "step": 30
    },
    {
      "completion_length": 428.82813453674316,
      "epoch": 0.017066666666666667,
      "grad_norm": 0.08826447827353007,
      "kl": 0.0118865966796875,
      "learning_rate": 4.979002243050646e-07,
      "loss": 0.0,
      "reward": 0.9557292014360428,
      "reward_std": 0.42629835614934564,
      "rewards/equation_reward_func": 0.09375000186264515,
      "rewards/format_reward_func": 0.861979179084301,
      "step": 32
    },
    {
      "completion_length": 381.89584159851074,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.08254377419499123,
      "kl": 0.01509857177734375,
      "learning_rate": 4.974085368611381e-07,
      "loss": 0.0,
      "reward": 1.0104167014360428,
      "reward_std": 0.2932055569253862,
      "rewards/equation_reward_func": 0.0781250016298145,
      "rewards/format_reward_func": 0.932291679084301,
      "step": 34
    },
    {
      "completion_length": 402.3255310058594,
      "epoch": 0.0192,
      "grad_norm": 0.08985049726745727,
      "kl": 0.01468658447265625,
      "learning_rate": 4.968654694381379e-07,
      "loss": 0.0,
      "reward": 1.0260416977107525,
      "reward_std": 0.3457729024812579,
      "rewards/equation_reward_func": 0.11197916883975267,
      "rewards/format_reward_func": 0.9140625186264515,
      "step": 36
    },
    {
      "completion_length": 383.4218854904175,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.06937661321251537,
      "kl": 0.016204833984375,
      "learning_rate": 4.962711348162987e-07,
      "loss": 0.0,
      "reward": 1.0312500335276127,
      "reward_std": 0.2866137330420315,
      "rewards/equation_reward_func": 0.10156250325962901,
      "rewards/format_reward_func": 0.9296875186264515,
      "step": 38
    },
    {
      "completion_length": 366.7474060058594,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.08617157373958526,
      "kl": 0.016704559326171875,
      "learning_rate": 4.956256564226487e-07,
      "loss": 0.0,
      "reward": 1.0494791977107525,
      "reward_std": 0.3076633233577013,
      "rewards/equation_reward_func": 0.10677083674818277,
      "rewards/format_reward_func": 0.942708358168602,
      "step": 40
    },
    {
      "completion_length": 374.84636306762695,
      "epoch": 0.0224,
      "grad_norm": 0.07644404910418388,
      "kl": 0.018764495849609375,
      "learning_rate": 4.949291683053768e-07,
      "loss": 0.0,
      "reward": 0.9817708544433117,
      "reward_std": 0.27403224213048816,
      "rewards/equation_reward_func": 0.052083335584029555,
      "rewards/format_reward_func": 0.9296875186264515,
      "step": 42
    },
    {
      "completion_length": 349.61720085144043,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.09147976837649754,
      "kl": 0.0202484130859375,
      "learning_rate": 4.941818151059955e-07,
      "loss": 0.0,
      "reward": 1.1250000409781933,
      "reward_std": 0.3236892116256058,
      "rewards/equation_reward_func": 0.16666667209938169,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 44
    },
    {
      "completion_length": 351.20573806762695,
      "epoch": 0.024533333333333334,
      "grad_norm": 0.080810111519101,
      "kl": 0.02154541015625,
      "learning_rate": 4.933837520293017e-07,
      "loss": 0.0,
      "reward": 1.101562537252903,
      "reward_std": 0.3292490583844483,
      "rewards/equation_reward_func": 0.14843750395812094,
      "rewards/format_reward_func": 0.9531250223517418,
      "step": 46
    },
    {
      "completion_length": 337.18750953674316,
      "epoch": 0.0256,
      "grad_norm": 0.08812991211015588,
      "kl": 0.022308349609375,
      "learning_rate": 4.925351448111454e-07,
      "loss": 0.0,
      "reward": 1.0703125298023224,
      "reward_std": 0.24853238929063082,
      "rewards/equation_reward_func": 0.0989583374466747,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 48
    },
    {
      "completion_length": 329.7760524749756,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.09791557407500211,
      "kl": 0.02400970458984375,
      "learning_rate": 4.91636169684011e-07,
      "loss": 0.0,
      "reward": 1.109375026077032,
      "reward_std": 0.2923102146014571,
      "rewards/equation_reward_func": 0.145833337912336,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 50
    },
    {
      "completion_length": 335.13802909851074,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.0848949698864576,
      "kl": 0.025177001953125,
      "learning_rate": 4.906870133404186e-07,
      "loss": 0.0,
      "reward": 1.0442708767950535,
      "reward_std": 0.27369245281443,
      "rewards/equation_reward_func": 0.09114583535119891,
      "rewards/format_reward_func": 0.9531250223517418,
      "step": 52
    },
    {
      "completion_length": 339.0677185058594,
      "epoch": 0.0288,
      "grad_norm": 0.08543744935145518,
      "kl": 0.02909088134765625,
      "learning_rate": 4.896878728941531e-07,
      "loss": 0.0,
      "reward": 1.0572917014360428,
      "reward_std": 0.2592460950836539,
      "rewards/equation_reward_func": 0.09895833651535213,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 54
    },
    {
      "completion_length": 306.4062614440918,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.0892429955729079,
      "kl": 0.03144073486328125,
      "learning_rate": 4.886389558393284e-07,
      "loss": 0.0,
      "reward": 1.1380208805203438,
      "reward_std": 0.2522500101476908,
      "rewards/equation_reward_func": 0.14583333698101342,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 56
    },
    {
      "completion_length": 321.0468873977661,
      "epoch": 0.030933333333333334,
      "grad_norm": 0.09832489597643927,
      "kl": 0.033355712890625,
      "learning_rate": 4.875404800072976e-07,
      "loss": 0.0,
      "reward": 1.1223958730697632,
      "reward_std": 0.2980109853670001,
      "rewards/equation_reward_func": 0.15104166977107525,
      "rewards/format_reward_func": 0.9713541902601719,
      "step": 58
    },
    {
      "completion_length": 312.5052194595337,
      "epoch": 0.032,
      "grad_norm": 0.09346855616087738,
      "kl": 0.0359954833984375,
      "learning_rate": 4.86392673521415e-07,
      "loss": 0.0,
      "reward": 1.1041667126119137,
      "reward_std": 0.2926621907390654,
      "rewards/equation_reward_func": 0.1406250037252903,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 60
    },
    {
      "completion_length": 307.67709732055664,
      "epoch": 0.03306666666666667,
      "grad_norm": 0.0928645477348611,
      "kl": 0.0410308837890625,
      "learning_rate": 4.851957747496606e-07,
      "loss": 0.0,
      "reward": 1.0833333656191826,
      "reward_std": 0.26842446345835924,
      "rewards/equation_reward_func": 0.11458333604969084,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 62
    },
    {
      "completion_length": 293.5130252838135,
      "epoch": 0.034133333333333335,
      "grad_norm": 0.10081686377737699,
      "kl": 0.043701171875,
      "learning_rate": 4.839500322551386e-07,
      "loss": 0.0,
      "reward": 1.1328125521540642,
      "reward_std": 0.2438423940911889,
      "rewards/equation_reward_func": 0.15104167046956718,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 64
    },
    {
      "completion_length": 291.0494842529297,
      "epoch": 0.0352,
      "grad_norm": 0.10220556104458557,
      "kl": 0.057830810546875,
      "learning_rate": 4.826557047444563e-07,
      "loss": 0.0001,
      "reward": 1.1510417014360428,
      "reward_std": 0.28418216248974204,
      "rewards/equation_reward_func": 0.169270837912336,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 66
    },
    {
      "completion_length": 274.10417652130127,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.09847033680691569,
      "kl": 0.0567779541015625,
      "learning_rate": 4.813130610139993e-07,
      "loss": 0.0001,
      "reward": 1.171875037252903,
      "reward_std": 0.25645031640306115,
      "rewards/equation_reward_func": 0.1744791711680591,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 68
    },
    {
      "completion_length": 267.9427185058594,
      "epoch": 0.037333333333333336,
      "grad_norm": 0.08978592804347527,
      "kl": 0.0566864013671875,
      "learning_rate": 4.799223798941089e-07,
      "loss": 0.0001,
      "reward": 1.1718750447034836,
      "reward_std": 0.2309985337778926,
      "rewards/equation_reward_func": 0.1770833390764892,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 70
    },
    {
      "completion_length": 273.7968854904175,
      "epoch": 0.0384,
      "grad_norm": 0.11719889448244371,
      "kl": 0.0622711181640625,
      "learning_rate": 4.78483950191177e-07,
      "loss": 0.0001,
      "reward": 1.1354167088866234,
      "reward_std": 0.2343482794240117,
      "rewards/equation_reward_func": 0.14843750442378223,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 72
    },
    {
      "completion_length": 293.1041736602783,
      "epoch": 0.039466666666666664,
      "grad_norm": 0.0771085948322575,
      "kl": 0.0579681396484375,
      "learning_rate": 4.769980706276687e-07,
      "loss": 0.0001,
      "reward": 1.0885417014360428,
      "reward_std": 0.18613768089562654,
      "rewards/equation_reward_func": 0.1015625016298145,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 74
    },
    {
      "completion_length": 275.88282012939453,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.09498835999619648,
      "kl": 0.0618438720703125,
      "learning_rate": 4.7546504978008595e-07,
      "loss": 0.0001,
      "reward": 1.1380208693444729,
      "reward_std": 0.23339220695197582,
      "rewards/equation_reward_func": 0.15104167209938169,
      "rewards/format_reward_func": 0.9869791716337204,
      "step": 76
    },
    {
      "completion_length": 260.79428005218506,
      "epoch": 0.0416,
      "grad_norm": 0.09832115718244772,
      "kl": 0.0676727294921875,
      "learning_rate": 4.738852060148848e-07,
      "loss": 0.0001,
      "reward": 1.1666667088866234,
      "reward_std": 0.30688405269756913,
      "rewards/equation_reward_func": 0.1770833395421505,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 78
    },
    {
      "completion_length": 277.85157012939453,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.0817474712194154,
      "kl": 0.0685577392578125,
      "learning_rate": 4.722588674223593e-07,
      "loss": 0.0001,
      "reward": 1.0833333618938923,
      "reward_std": 0.17427472537383437,
      "rewards/equation_reward_func": 0.10156250302679837,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 80
    },
    {
      "completion_length": 272.9739656448364,
      "epoch": 0.04373333333333333,
      "grad_norm": 0.11152056972270338,
      "kl": 0.06781005859375,
      "learning_rate": 4.70586371748506e-07,
      "loss": 0.0001,
      "reward": 1.166666716337204,
      "reward_std": 0.2563547547906637,
      "rewards/equation_reward_func": 0.17447917186655104,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 82
    },
    {
      "completion_length": 266.23438262939453,
      "epoch": 0.0448,
      "grad_norm": 0.08738050204966807,
      "kl": 0.0767974853515625,
      "learning_rate": 4.6886806632488363e-07,
      "loss": 0.0001,
      "reward": 1.1536458767950535,
      "reward_std": 0.252029016148299,
      "rewards/equation_reward_func": 0.17447917186655104,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 84
    },
    {
      "completion_length": 265.9661560058594,
      "epoch": 0.04586666666666667,
      "grad_norm": 0.0921721597746663,
      "kl": 0.075592041015625,
      "learning_rate": 4.6710430799648143e-07,
      "loss": 0.0001,
      "reward": 1.197916716337204,
      "reward_std": 0.273911755066365,
      "rewards/equation_reward_func": 0.21093750628642738,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 86
    },
    {
      "completion_length": 254.08594608306885,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.10824024038176047,
      "kl": 0.0884246826171875,
      "learning_rate": 4.652954630476127e-07,
      "loss": 0.0001,
      "reward": 1.2031250223517418,
      "reward_std": 0.26051403814926744,
      "rewards/equation_reward_func": 0.21875000442378223,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 88
    },
    {
      "completion_length": 273.9270944595337,
      "epoch": 0.048,
      "grad_norm": 0.0759106876406683,
      "kl": 0.076568603515625,
      "learning_rate": 4.6344190712584713e-07,
      "loss": 0.0001,
      "reward": 1.1692708656191826,
      "reward_std": 0.25936984410509467,
      "rewards/equation_reward_func": 0.179687503259629,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 90
    },
    {
      "completion_length": 253.90886211395264,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.11783628557713965,
      "kl": 0.0765533447265625,
      "learning_rate": 4.615440251639995e-07,
      "loss": 0.0001,
      "reward": 1.2265625298023224,
      "reward_std": 0.30153857497498393,
      "rewards/equation_reward_func": 0.23177084000781178,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 92
    },
    {
      "completion_length": 273.47917461395264,
      "epoch": 0.050133333333333335,
      "grad_norm": 0.08532452350344605,
      "kl": 0.0875091552734375,
      "learning_rate": 4.596022113001894e-07,
      "loss": 0.0001,
      "reward": 1.1250000298023224,
      "reward_std": 0.26654769526794553,
      "rewards/equation_reward_func": 0.15364583698101342,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 94
    },
    {
      "completion_length": 254.58594608306885,
      "epoch": 0.0512,
      "grad_norm": 0.08252539758184962,
      "kl": 0.097412109375,
      "learning_rate": 4.576168687959895e-07,
      "loss": 0.0001,
      "reward": 1.203125037252903,
      "reward_std": 0.23845402244478464,
      "rewards/equation_reward_func": 0.2109375058207661,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 96
    },
    {
      "completion_length": 275.47136211395264,
      "epoch": 0.05226666666666667,
      "grad_norm": 0.08236324956103444,
      "kl": 0.081390380859375,
      "learning_rate": 4.555884099526793e-07,
      "loss": 0.0001,
      "reward": 1.1901042088866234,
      "reward_std": 0.26423150720074773,
      "rewards/equation_reward_func": 0.19791667512618005,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 98
    },
    {
      "completion_length": 270.8515682220459,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.07926412487583492,
      "kl": 0.074127197265625,
      "learning_rate": 4.5351725602562174e-07,
      "loss": 0.0001,
      "reward": 1.1666667014360428,
      "reward_std": 0.2103340970352292,
      "rewards/equation_reward_func": 0.17187500419095159,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 100
    },
    {
      "completion_length": 289.2786521911621,
      "epoch": 0.0544,
      "grad_norm": 0.09050663026015614,
      "kl": 0.0777587890625,
      "learning_rate": 4.514038371367791e-07,
      "loss": 0.0001,
      "reward": 1.1432292088866234,
      "reward_std": 0.23622727626934648,
      "rewards/equation_reward_func": 0.15364583767950535,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 102
    },
    {
      "completion_length": 281.1744861602783,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.10688958655389184,
      "kl": 0.0772247314453125,
      "learning_rate": 4.4924859218538936e-07,
      "loss": 0.0001,
      "reward": 1.2473958730697632,
      "reward_std": 0.2720282976515591,
      "rewards/equation_reward_func": 0.2604166774544865,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 104
    },
    {
      "completion_length": 278.20313358306885,
      "epoch": 0.05653333333333333,
      "grad_norm": 0.0967601461658064,
      "kl": 0.0747222900390625,
      "learning_rate": 4.470519687568185e-07,
      "loss": 0.0001,
      "reward": 1.2109375521540642,
      "reward_std": 0.2844023099169135,
      "rewards/equation_reward_func": 0.22135417559184134,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 106
    },
    {
      "completion_length": 280.2239685058594,
      "epoch": 0.0576,
      "grad_norm": 0.1073193207656852,
      "kl": 0.083953857421875,
      "learning_rate": 4.4481442302960923e-07,
      "loss": 0.0001,
      "reward": 1.273437537252903,
      "reward_std": 0.34114335058256984,
      "rewards/equation_reward_func": 0.2812500111758709,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 108
    },
    {
      "completion_length": 302.32813262939453,
      "epoch": 0.058666666666666666,
      "grad_norm": 0.07811304956074856,
      "kl": 0.078094482421875,
      "learning_rate": 4.4253641968074505e-07,
      "loss": 0.0001,
      "reward": 1.1875000298023224,
      "reward_std": 0.22219527000561357,
      "rewards/equation_reward_func": 0.21614584070630372,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 110
    },
    {
      "completion_length": 296.5338659286499,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.09384336608062288,
      "kl": 0.07928466796875,
      "learning_rate": 4.402184317891501e-07,
      "loss": 0.0001,
      "reward": 1.1796875447034836,
      "reward_std": 0.2780974409542978,
      "rewards/equation_reward_func": 0.20312500721774995,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 112
    },
    {
      "completion_length": 286.5130319595337,
      "epoch": 0.0608,
      "grad_norm": 0.09470360951060115,
      "kl": 0.086334228515625,
      "learning_rate": 4.37860940737443e-07,
      "loss": 0.0001,
      "reward": 1.1770833693444729,
      "reward_std": 0.2872624071314931,
      "rewards/equation_reward_func": 0.2031250069849193,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 114
    },
    {
      "completion_length": 275.84636306762695,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.10006083180691484,
      "kl": 0.110260009765625,
      "learning_rate": 4.354644361119671e-07,
      "loss": 0.0001,
      "reward": 1.2864583730697632,
      "reward_std": 0.3422618228942156,
      "rewards/equation_reward_func": 0.3046875074505806,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 116
    },
    {
      "completion_length": 279.63282203674316,
      "epoch": 0.06293333333333333,
      "grad_norm": 0.08552614368611813,
      "kl": 0.088287353515625,
      "learning_rate": 4.3302941560111716e-07,
      "loss": 0.0001,
      "reward": 1.2057292088866234,
      "reward_std": 0.2789528788998723,
      "rewards/equation_reward_func": 0.22395833930931985,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 118
    },
    {
      "completion_length": 286.19011306762695,
      "epoch": 0.064,
      "grad_norm": 0.09721854870867883,
      "kl": 0.088653564453125,
      "learning_rate": 4.3055638489198236e-07,
      "loss": 0.0001,
      "reward": 1.226562537252903,
      "reward_std": 0.29836362041532993,
      "rewards/equation_reward_func": 0.24218750768341124,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 120
    },
    {
      "completion_length": 264.43750762939453,
      "epoch": 0.06506666666666666,
      "grad_norm": 0.09017140778699437,
      "kl": 0.09814453125,
      "learning_rate": 4.280458575653296e-07,
      "loss": 0.0001,
      "reward": 1.2291666939854622,
      "reward_std": 0.232796979136765,
      "rewards/equation_reward_func": 0.2343750090803951,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 122
    },
    {
      "completion_length": 273.72136402130127,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.09045977124859449,
      "kl": 0.094268798828125,
      "learning_rate": 4.2549835498894665e-07,
      "loss": 0.0001,
      "reward": 1.182291705161333,
      "reward_std": 0.21962608164176345,
      "rewards/equation_reward_func": 0.19270834047347307,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 124
    },
    {
      "completion_length": 267.1041736602783,
      "epoch": 0.0672,
      "grad_norm": 0.10142470255177057,
      "kl": 0.109100341796875,
      "learning_rate": 4.229144062093679e-07,
      "loss": 0.0001,
      "reward": 1.2239583805203438,
      "reward_std": 0.2902214271016419,
      "rewards/equation_reward_func": 0.24739584070630372,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 126
    },
    {
      "completion_length": 259.2317771911621,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.1179722317793916,
      "kl": 0.100433349609375,
      "learning_rate": 4.2029454784200675e-07,
      "loss": 0.0001,
      "reward": 1.2604167088866234,
      "reward_std": 0.2859130958095193,
      "rewards/equation_reward_func": 0.27343750931322575,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 128
    },
    {
      "completion_length": 244.24740505218506,
      "epoch": 0.06933333333333333,
      "grad_norm": 0.11744647822751025,
      "kl": 0.10302734375,
      "learning_rate": 4.1763932395971433e-07,
      "loss": 0.0001,
      "reward": 1.2682292088866234,
      "reward_std": 0.29776600282639265,
      "rewards/equation_reward_func": 0.27604167396202683,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 130
    },
    {
      "completion_length": 262.6770896911621,
      "epoch": 0.0704,
      "grad_norm": 0.0792242116848132,
      "kl": 0.1085205078125,
      "learning_rate": 4.1494928597979117e-07,
      "loss": 0.0001,
      "reward": 1.2526041977107525,
      "reward_std": 0.22840882744640112,
      "rewards/equation_reward_func": 0.2604166704695672,
      "rewards/format_reward_func": 0.9921875037252903,
      "step": 132
    },
    {
      "completion_length": 273.04167556762695,
      "epoch": 0.07146666666666666,
      "grad_norm": 0.09121048756837698,
      "kl": 0.11712646484375,
      "learning_rate": 4.122249925494726e-07,
      "loss": 0.0001,
      "reward": 1.1744792088866234,
      "reward_std": 0.24433040153235197,
      "rewards/equation_reward_func": 0.2031250074505806,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 134
    },
    {
      "completion_length": 257.322922706604,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.10803171417140216,
      "kl": 0.114501953125,
      "learning_rate": 4.094670094299131e-07,
      "loss": 0.0001,
      "reward": 1.3177083618938923,
      "reward_std": 0.2959897918626666,
      "rewards/equation_reward_func": 0.33072917303070426,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 136
    },
    {
      "completion_length": 253.40886116027832,
      "epoch": 0.0736,
      "grad_norm": 0.11919316885402649,
      "kl": 0.11944580078125,
      "learning_rate": 4.066759093786931e-07,
      "loss": 0.0001,
      "reward": 1.2291667088866234,
      "reward_std": 0.29171993816271424,
      "rewards/equation_reward_func": 0.2500000048894435,
      "rewards/format_reward_func": 0.9791666865348816,
      "step": 138
    },
    {
      "completion_length": 254.38542461395264,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.09338771512811811,
      "kl": 0.110992431640625,
      "learning_rate": 4.038522720308732e-07,
      "loss": 0.0001,
      "reward": 1.2942708656191826,
      "reward_std": 0.2597912196069956,
      "rewards/equation_reward_func": 0.31250000977888703,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 140
    },
    {
      "completion_length": 261.5651111602783,
      "epoch": 0.07573333333333333,
      "grad_norm": 0.0978514819443017,
      "kl": 0.113983154296875,
      "learning_rate": 4.009966837786194e-07,
      "loss": 0.0001,
      "reward": 1.2604167014360428,
      "reward_std": 0.23182117100805044,
      "rewards/equation_reward_func": 0.26562500721774995,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 142
    },
    {
      "completion_length": 249.08334159851074,
      "epoch": 0.0768,
      "grad_norm": 0.0996729641628275,
      "kl": 0.123016357421875,
      "learning_rate": 3.981097376494259e-07,
      "loss": 0.0001,
      "reward": 1.3125000447034836,
      "reward_std": 0.22445971937850118,
      "rewards/equation_reward_func": 0.3255208465270698,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 144
    },
    {
      "completion_length": 250.35938167572021,
      "epoch": 0.07786666666666667,
      "grad_norm": 0.08419615699604895,
      "kl": 0.132476806640625,
      "learning_rate": 3.951920331829592e-07,
      "loss": 0.0001,
      "reward": 1.2838542088866234,
      "reward_std": 0.2362030209042132,
      "rewards/equation_reward_func": 0.2942708428017795,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 146
    },
    {
      "completion_length": 249.89323616027832,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.1283208171287915,
      "kl": 0.132720947265625,
      "learning_rate": 3.922441763065506e-07,
      "loss": 0.0001,
      "reward": 1.3151041977107525,
      "reward_std": 0.2713711801916361,
      "rewards/equation_reward_func": 0.33072917768731713,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 148
    },
    {
      "completion_length": 241.3046932220459,
      "epoch": 0.08,
      "grad_norm": 0.09375259938311675,
      "kl": 0.19842529296875,
      "learning_rate": 3.8926677920936093e-07,
      "loss": 0.0002,
      "reward": 1.317708358168602,
      "reward_std": 0.2631534468382597,
      "rewards/equation_reward_func": 0.3255208395421505,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 150
    },
    {
      "completion_length": 233.75521755218506,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.1141556527946053,
      "kl": 0.137664794921875,
      "learning_rate": 3.862604602152464e-07,
      "loss": 0.0001,
      "reward": 1.320312537252903,
      "reward_std": 0.22954290360212326,
      "rewards/equation_reward_func": 0.3307291741948575,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 152
    },
    {
      "completion_length": 250.05990409851074,
      "epoch": 0.08213333333333334,
      "grad_norm": 0.11170521900211575,
      "kl": 0.134246826171875,
      "learning_rate": 3.8322584365434934e-07,
      "loss": 0.0001,
      "reward": 1.302083358168602,
      "reward_std": 0.25723502691835165,
      "rewards/equation_reward_func": 0.3072916744276881,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 154
    },
    {
      "completion_length": 262.64844512939453,
      "epoch": 0.0832,
      "grad_norm": 0.11715928802149779,
      "kl": 0.13623046875,
      "learning_rate": 3.8016355973344173e-07,
      "loss": 0.0001,
      "reward": 1.260416716337204,
      "reward_std": 0.2844986612908542,
      "rewards/equation_reward_func": 0.2734375086147338,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 156
    },
    {
      "completion_length": 240.10417556762695,
      "epoch": 0.08426666666666667,
      "grad_norm": 0.10437969988785332,
      "kl": 0.14324951171875,
      "learning_rate": 3.7707424440504863e-07,
      "loss": 0.0001,
      "reward": 1.3385417088866234,
      "reward_std": 0.2518824371509254,
      "rewards/equation_reward_func": 0.3515625111758709,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 158
    },
    {
      "completion_length": 254.15886116027832,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.10197264180728077,
      "kl": 0.140045166015625,
      "learning_rate": 3.739585392353787e-07,
      "loss": 0.0001,
      "reward": 1.2864583656191826,
      "reward_std": 0.17625536350533366,
      "rewards/equation_reward_func": 0.299479179084301,
      "rewards/format_reward_func": 0.9869791716337204,
      "step": 160
    },
    {
      "completion_length": 240.49740314483643,
      "epoch": 0.0864,
      "grad_norm": 0.12368241208002842,
      "kl": 0.1563720703125,
      "learning_rate": 3.7081709127108767e-07,
      "loss": 0.0002,
      "reward": 1.2552083730697632,
      "reward_std": 0.30754117481410503,
      "rewards/equation_reward_func": 0.26562500977888703,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 162
    },
    {
      "completion_length": 246.41146659851074,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.11117407054253553,
      "kl": 0.154571533203125,
      "learning_rate": 3.6765055290490513e-07,
      "loss": 0.0002,
      "reward": 1.3098958730697632,
      "reward_std": 0.23842021962627769,
      "rewards/equation_reward_func": 0.31510417629033327,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 164
    },
    {
      "completion_length": 241.39323616027832,
      "epoch": 0.08853333333333334,
      "grad_norm": 0.14524308584181014,
      "kl": 0.165283203125,
      "learning_rate": 3.644595817401501e-07,
      "loss": 0.0002,
      "reward": 1.3463542014360428,
      "reward_std": 0.3047106293961406,
      "rewards/equation_reward_func": 0.35156250768341124,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 166
    },
    {
      "completion_length": 243.39063358306885,
      "epoch": 0.0896,
      "grad_norm": 0.09625292068136962,
      "kl": 0.14849853515625,
      "learning_rate": 3.6124484045416483e-07,
      "loss": 0.0001,
      "reward": 1.3307292237877846,
      "reward_std": 0.23321589175611734,
      "rewards/equation_reward_func": 0.3385416797827929,
      "rewards/format_reward_func": 0.9921875037252903,
      "step": 168
    },
    {
      "completion_length": 247.43490028381348,
      "epoch": 0.09066666666666667,
      "grad_norm": 0.10289615592492905,
      "kl": 0.15081787109375,
      "learning_rate": 3.580069966606949e-07,
      "loss": 0.0002,
      "reward": 1.304687537252903,
      "reward_std": 0.26835791766643524,
      "rewards/equation_reward_func": 0.31770834350027144,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 170
    },
    {
      "completion_length": 256.49219703674316,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.09079412799818862,
      "kl": 0.155517578125,
      "learning_rate": 3.547467227712444e-07,
      "loss": 0.0002,
      "reward": 1.2942708656191826,
      "reward_std": 0.19367968942970037,
      "rewards/equation_reward_func": 0.315104179084301,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 172
    },
    {
      "completion_length": 263.32032012939453,
      "epoch": 0.0928,
      "grad_norm": 0.09149503314135422,
      "kl": 0.161041259765625,
      "learning_rate": 3.5146469585543386e-07,
      "loss": 0.0002,
      "reward": 1.2473958656191826,
      "reward_std": 0.21927247568964958,
      "rewards/equation_reward_func": 0.2656250074505806,
      "rewards/format_reward_func": 0.9817708395421505,
      "step": 174
    },
    {
      "completion_length": 258.46094512939453,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.12225866423305806,
      "kl": 0.16229248046875,
      "learning_rate": 3.481615975003922e-07,
      "loss": 0.0002,
      "reward": 1.3203125521540642,
      "reward_std": 0.27134765265509486,
      "rewards/equation_reward_func": 0.3437500160653144,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 176
    },
    {
      "completion_length": 263.02605056762695,
      "epoch": 0.09493333333333333,
      "grad_norm": 0.10889388889974946,
      "kl": 0.15594482421875,
      "learning_rate": 3.448381136692089e-07,
      "loss": 0.0002,
      "reward": 1.3828125298023224,
      "reward_std": 0.3108144081197679,
      "rewards/equation_reward_func": 0.39843750884756446,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 178
    },
    {
      "completion_length": 239.73177909851074,
      "epoch": 0.096,
      "grad_norm": 0.12028128132666989,
      "kl": 0.18389892578125,
      "learning_rate": 3.4149493455847897e-07,
      "loss": 0.0002,
      "reward": 1.4036458656191826,
      "reward_std": 0.2842292613349855,
      "rewards/equation_reward_func": 0.4218750086147338,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 180
    },
    {
      "completion_length": 255.11719608306885,
      "epoch": 0.09706666666666666,
      "grad_norm": 0.11721944064081667,
      "kl": 0.1566162109375,
      "learning_rate": 3.3813275445496766e-07,
      "loss": 0.0002,
      "reward": 1.3958333656191826,
      "reward_std": 0.2701294063590467,
      "rewards/equation_reward_func": 0.41666667885147035,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 182
    },
    {
      "completion_length": 253.9192762374878,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.10053395121193144,
      "kl": 0.1617431640625,
      "learning_rate": 3.347522715914262e-07,
      "loss": 0.0002,
      "reward": 1.2760417014360428,
      "reward_std": 0.2605799976736307,
      "rewards/equation_reward_func": 0.2942708367481828,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 184
    },
    {
      "completion_length": 265.5234441757202,
      "epoch": 0.0992,
      "grad_norm": 0.07307015368809015,
      "kl": 0.20306396484375,
      "learning_rate": 3.313541880015877e-07,
      "loss": 0.0002,
      "reward": 1.3203125223517418,
      "reward_std": 0.1439679628238082,
      "rewards/equation_reward_func": 0.33072917140088975,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 186
    },
    {
      "completion_length": 276.56251335144043,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.08556221607477679,
      "kl": 0.1552734375,
      "learning_rate": 3.279392093743747e-07,
      "loss": 0.0002,
      "reward": 1.226562537252903,
      "reward_std": 0.19421058846637607,
      "rewards/equation_reward_func": 0.23437500931322575,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 188
    },
    {
      "completion_length": 264.04948806762695,
      "epoch": 0.10133333333333333,
      "grad_norm": 0.1313876809674113,
      "kl": 0.16461181640625,
      "learning_rate": 3.245080449073459e-07,
      "loss": 0.0002,
      "reward": 1.338541705161333,
      "reward_std": 0.2719450327567756,
      "rewards/equation_reward_func": 0.36197917396202683,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 190
    },
    {
      "completion_length": 256.4974021911621,
      "epoch": 0.1024,
      "grad_norm": 0.08299025727419519,
      "kl": 0.16888427734375,
      "learning_rate": 3.210614071594162e-07,
      "loss": 0.0002,
      "reward": 1.3333333730697632,
      "reward_std": 0.22525584790855646,
      "rewards/equation_reward_func": 0.3411458428017795,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 192
    },
    {
      "completion_length": 250.9817771911621,
      "epoch": 0.10346666666666667,
      "grad_norm": 0.09659118998865841,
      "kl": 0.174560546875,
      "learning_rate": 3.1760001190287695e-07,
      "loss": 0.0002,
      "reward": 1.3697917014360428,
      "reward_std": 0.2577691958285868,
      "rewards/equation_reward_func": 0.38541667722165585,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 194
    },
    {
      "completion_length": 263.43750381469727,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.10853905143637654,
      "kl": 0.2979736328125,
      "learning_rate": 3.141245779747502e-07,
      "loss": 0.0003,
      "reward": 1.3072917088866234,
      "reward_std": 0.21552392421290278,
      "rewards/equation_reward_func": 0.3255208432674408,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 196
    },
    {
      "completion_length": 277.60938358306885,
      "epoch": 0.1056,
      "grad_norm": 0.0994297176227174,
      "kl": 0.17926025390625,
      "learning_rate": 3.106358271275056e-07,
      "loss": 0.0002,
      "reward": 1.2682291977107525,
      "reward_std": 0.23380355769768357,
      "rewards/equation_reward_func": 0.2838541707023978,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 198
    },
    {
      "completion_length": 286.0989685058594,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.09581836861635441,
      "kl": 0.16900634765625,
      "learning_rate": 3.0713448387917227e-07,
      "loss": 0.0002,
      "reward": 1.2760417088866234,
      "reward_std": 0.21799071412533522,
      "rewards/equation_reward_func": 0.2890625095460564,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 200
    },
    {
      "completion_length": 278.19792556762695,
      "epoch": 0.10773333333333333,
      "grad_norm": 0.09085167749204627,
      "kl": 0.16778564453125,
      "learning_rate": 3.0362127536287636e-07,
      "loss": 0.0002,
      "reward": 1.2604166939854622,
      "reward_std": 0.19841350940987468,
      "rewards/equation_reward_func": 0.28125000605359674,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 202
    },
    {
      "completion_length": 260.0130271911621,
      "epoch": 0.1088,
      "grad_norm": 0.10197204999448693,
      "kl": 0.17706298828125,
      "learning_rate": 3.0009693117583523e-07,
      "loss": 0.0002,
      "reward": 1.3593750298023224,
      "reward_std": 0.2952982089482248,
      "rewards/equation_reward_func": 0.3723958423361182,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 204
    },
    {
      "completion_length": 251.432297706604,
      "epoch": 0.10986666666666667,
      "grad_norm": 0.10432038403259797,
      "kl": 0.1810302734375,
      "learning_rate": 2.965621832278401e-07,
      "loss": 0.0002,
      "reward": 1.3880208805203438,
      "reward_std": 0.2258518268354237,
      "rewards/equation_reward_func": 0.3984375100117177,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 206
    },
    {
      "completion_length": 269.71354579925537,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.08880292797325792,
      "kl": 0.18695068359375,
      "learning_rate": 2.9301776558925875e-07,
      "loss": 0.0002,
      "reward": 1.3229166939854622,
      "reward_std": 0.22289922274649143,
      "rewards/equation_reward_func": 0.32812500884756446,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 208
    },
    {
      "completion_length": 277.0260543823242,
      "epoch": 0.112,
      "grad_norm": 0.09670496258985709,
      "kl": 0.18951416015625,
      "learning_rate": 2.894644143385885e-07,
      "loss": 0.0002,
      "reward": 1.289062537252903,
      "reward_std": 0.19973075529560447,
      "rewards/equation_reward_func": 0.30208334303461015,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 210
    },
    {
      "completion_length": 262.3463611602783,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.09450629356325996,
      "kl": 0.19244384765625,
      "learning_rate": 2.859028674095937e-07,
      "loss": 0.0002,
      "reward": 1.3567708656191826,
      "reward_std": 0.23411421943455935,
      "rewards/equation_reward_func": 0.3723958428017795,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 212
    },
    {
      "completion_length": 250.36198902130127,
      "epoch": 0.11413333333333334,
      "grad_norm": 0.10283327615954467,
      "kl": 0.19482421875,
      "learning_rate": 2.823338644380566e-07,
      "loss": 0.0002,
      "reward": 1.3984375298023224,
      "reward_std": 0.21787611162289977,
      "rewards/equation_reward_func": 0.4036458448972553,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 214
    },
    {
      "completion_length": 252.3880319595337,
      "epoch": 0.1152,
      "grad_norm": 0.0968619402675733,
      "kl": 0.21044921875,
      "learning_rate": 2.7875814660817504e-07,
      "loss": 0.0002,
      "reward": 1.330729216337204,
      "reward_std": 0.2707995134405792,
      "rewards/equation_reward_func": 0.34895834419876337,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 216
    },
    {
      "completion_length": 262.3906354904175,
      "epoch": 0.11626666666666667,
      "grad_norm": 0.10234953195139436,
      "kl": 0.1976318359375,
      "learning_rate": 2.751764564986396e-07,
      "loss": 0.0002,
      "reward": 1.3125000447034836,
      "reward_std": 0.21475645201280713,
      "rewards/equation_reward_func": 0.3307291753590107,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 218
    },
    {
      "completion_length": 253.4349021911621,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.0988359155618979,
      "kl": 0.23199462890625,
      "learning_rate": 2.715895379284194e-07,
      "loss": 0.0002,
      "reward": 1.3359375447034836,
      "reward_std": 0.27620820328593254,
      "rewards/equation_reward_func": 0.35416667512618005,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 220
    },
    {
      "completion_length": 256.5104236602783,
      "epoch": 0.1184,
      "grad_norm": 0.08785183961462481,
      "kl": 0.21124267578125,
      "learning_rate": 2.6799813580229174e-07,
      "loss": 0.0002,
      "reward": 1.2838542088866234,
      "reward_std": 0.20858342200517654,
      "rewards/equation_reward_func": 0.30468750675208867,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 222
    },
    {
      "completion_length": 264.7031354904175,
      "epoch": 0.11946666666666667,
      "grad_norm": 0.09106595411129506,
      "kl": 0.2265625,
      "learning_rate": 2.6440299595614606e-07,
      "loss": 0.0002,
      "reward": 1.3229166939854622,
      "reward_std": 0.26256967103108764,
      "rewards/equation_reward_func": 0.3359375095460564,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 224
    },
    {
      "completion_length": 244.47396755218506,
      "epoch": 0.12053333333333334,
      "grad_norm": 0.1322728306343847,
      "kl": 0.21575927734375,
      "learning_rate": 2.6080486500209347e-07,
      "loss": 0.0002,
      "reward": 1.4036458730697632,
      "reward_std": 0.27088421024382114,
      "rewards/equation_reward_func": 0.4192708428017795,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 226
    },
    {
      "completion_length": 245.83334159851074,
      "epoch": 0.1216,
      "grad_norm": 0.11379963074576292,
      "kl": 0.21893310546875,
      "learning_rate": 2.572044901734166e-07,
      "loss": 0.0002,
      "reward": 1.3020833730697632,
      "reward_std": 0.2311198292300105,
      "rewards/equation_reward_func": 0.31510417931713164,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 228
    },
    {
      "completion_length": 257.00000381469727,
      "epoch": 0.12266666666666666,
      "grad_norm": 0.12307948001244351,
      "kl": 0.2279052734375,
      "learning_rate": 2.536026191693893e-07,
      "loss": 0.0002,
      "reward": 1.2421875409781933,
      "reward_std": 0.2563111218623817,
      "rewards/equation_reward_func": 0.2656250074505806,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 230
    },
    {
      "completion_length": 245.31771278381348,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.1293557912538195,
      "kl": 0.2315673828125,
      "learning_rate": 2.5e-07,
      "loss": 0.0002,
      "reward": 1.3020833805203438,
      "reward_std": 0.24902541749179363,
      "rewards/equation_reward_func": 0.3177083423361182,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 232
    },
    {
      "completion_length": 242.99219512939453,
      "epoch": 0.1248,
      "grad_norm": 0.11767176490254756,
      "kl": 0.250732421875,
      "learning_rate": 2.4639738083061073e-07,
      "loss": 0.0003,
      "reward": 1.2682291939854622,
      "reward_std": 0.2607572884298861,
      "rewards/equation_reward_func": 0.283854172565043,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 234
    },
    {
      "completion_length": 238.18490505218506,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.09332728459883043,
      "kl": 0.2532958984375,
      "learning_rate": 2.4279550982658345e-07,
      "loss": 0.0003,
      "reward": 1.3255208805203438,
      "reward_std": 0.23872209014371037,
      "rewards/equation_reward_func": 0.33854167512618005,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 236
    },
    {
      "completion_length": 233.143235206604,
      "epoch": 0.12693333333333334,
      "grad_norm": 0.11331636950947853,
      "kl": 0.2747802734375,
      "learning_rate": 2.3919513499790646e-07,
      "loss": 0.0003,
      "reward": 1.3307292088866234,
      "reward_std": 0.22626318270340562,
      "rewards/equation_reward_func": 0.3489583448972553,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 238
    },
    {
      "completion_length": 235.2890682220459,
      "epoch": 0.128,
      "grad_norm": 0.104056938618789,
      "kl": 0.263916015625,
      "learning_rate": 2.3559700404385394e-07,
      "loss": 0.0003,
      "reward": 1.3333333805203438,
      "reward_std": 0.2541612219065428,
      "rewards/equation_reward_func": 0.3463541802484542,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 240
    },
    {
      "completion_length": 227.68750858306885,
      "epoch": 0.12906666666666666,
      "grad_norm": 0.08311002502749948,
      "kl": 0.25885009765625,
      "learning_rate": 2.3200186419770823e-07,
      "loss": 0.0003,
      "reward": 1.2395833618938923,
      "reward_std": 0.16169963357970119,
      "rewards/equation_reward_func": 0.25260417349636555,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 242
    },
    {
      "completion_length": 219.63542366027832,
      "epoch": 0.13013333333333332,
      "grad_norm": 0.12211002931078498,
      "kl": 0.259521484375,
      "learning_rate": 2.284104620715807e-07,
      "loss": 0.0003,
      "reward": 1.3567708656191826,
      "reward_std": 0.28411849960684776,
      "rewards/equation_reward_func": 0.3854166744276881,
      "rewards/format_reward_func": 0.9713541902601719,
      "step": 244
    },
    {
      "completion_length": 240.791672706604,
      "epoch": 0.1312,
      "grad_norm": 0.10287478218748722,
      "kl": 0.27532958984375,
      "learning_rate": 2.2482354350136043e-07,
      "loss": 0.0003,
      "reward": 1.2083333693444729,
      "reward_std": 0.2309797713533044,
      "rewards/equation_reward_func": 0.22656250768341124,
      "rewards/format_reward_func": 0.9817708395421505,
      "step": 246
    },
    {
      "completion_length": 219.9010467529297,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.11156569898091559,
      "kl": 0.28363037109375,
      "learning_rate": 2.2124185339182496e-07,
      "loss": 0.0003,
      "reward": 1.2942708656191826,
      "reward_std": 0.23296030843630433,
      "rewards/equation_reward_func": 0.30989584093913436,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 248
    },
    {
      "completion_length": 235.89063358306885,
      "epoch": 0.13333333333333333,
      "grad_norm": 0.10404156435534162,
      "kl": 0.3035888671875,
      "learning_rate": 2.1766613556194344e-07,
      "loss": 0.0003,
      "reward": 1.2760417014360428,
      "reward_std": 0.2603864520788193,
      "rewards/equation_reward_func": 0.31770834093913436,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 250
    },
    {
      "completion_length": 199.33333778381348,
      "epoch": 0.1344,
      "grad_norm": 0.12893130763177774,
      "kl": 0.27520751953125,
      "learning_rate": 2.1409713259040628e-07,
      "loss": 0.0003,
      "reward": 1.4661458656191826,
      "reward_std": 0.2550729913637042,
      "rewards/equation_reward_func": 0.49218750931322575,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 252
    },
    {
      "completion_length": 235.2213592529297,
      "epoch": 0.13546666666666668,
      "grad_norm": 0.098316148163009,
      "kl": 0.324462890625,
      "learning_rate": 2.105355856614115e-07,
      "loss": 0.0003,
      "reward": 1.2291666977107525,
      "reward_std": 0.217855678871274,
      "rewards/equation_reward_func": 0.2604166716337204,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 254
    },
    {
      "completion_length": 221.35937976837158,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.09611054923720826,
      "kl": 0.31512451171875,
      "learning_rate": 2.069822344107413e-07,
      "loss": 0.0003,
      "reward": 1.2187500521540642,
      "reward_std": 0.1683449512347579,
      "rewards/equation_reward_func": 0.24739584093913436,
      "rewards/format_reward_func": 0.9713541902601719,
      "step": 256
    },
    {
      "completion_length": 209.40104866027832,
      "epoch": 0.1376,
      "grad_norm": 0.12532736489907845,
      "kl": 0.320068359375,
      "learning_rate": 2.034378167721599e-07,
      "loss": 0.0003,
      "reward": 1.361979216337204,
      "reward_std": 0.21956727374345064,
      "rewards/equation_reward_func": 0.3828125153668225,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 258
    },
    {
      "completion_length": 219.97136116027832,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.113037756435864,
      "kl": 0.3076171875,
      "learning_rate": 1.9990306882416485e-07,
      "loss": 0.0003,
      "reward": 1.3046875447034836,
      "reward_std": 0.21061105513945222,
      "rewards/equation_reward_func": 0.33072918001562357,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 260
    },
    {
      "completion_length": 209.12500381469727,
      "epoch": 0.13973333333333332,
      "grad_norm": 0.10909571017775514,
      "kl": 0.28814697265625,
      "learning_rate": 1.9637872463712362e-07,
      "loss": 0.0003,
      "reward": 1.390625026077032,
      "reward_std": 0.2569071822799742,
      "rewards/equation_reward_func": 0.4270833421032876,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 262
    },
    {
      "completion_length": 220.04948329925537,
      "epoch": 0.1408,
      "grad_norm": 0.13107792080020514,
      "kl": 0.3321533203125,
      "learning_rate": 1.9286551612082773e-07,
      "loss": 0.0003,
      "reward": 1.2682292237877846,
      "reward_std": 0.23067627055570483,
      "rewards/equation_reward_func": 0.3098958460614085,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 264
    },
    {
      "completion_length": 223.55729866027832,
      "epoch": 0.14186666666666667,
      "grad_norm": 0.11293904370379487,
      "kl": 0.3299560546875,
      "learning_rate": 1.8936417287249446e-07,
      "loss": 0.0003,
      "reward": 1.2552083656191826,
      "reward_std": 0.18667185213416815,
      "rewards/equation_reward_func": 0.28385417559184134,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 266
    },
    {
      "completion_length": 222.44532012939453,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.10500319283294889,
      "kl": 0.31304931640625,
      "learning_rate": 1.8587542202524985e-07,
      "loss": 0.0003,
      "reward": 1.2656250409781933,
      "reward_std": 0.24632562743499875,
      "rewards/equation_reward_func": 0.3098958421032876,
      "rewards/format_reward_func": 0.9557291865348816,
      "step": 268
    },
    {
      "completion_length": 232.02605056762695,
      "epoch": 0.144,
      "grad_norm": 0.1361334206215253,
      "kl": 0.3328857421875,
      "learning_rate": 1.82399988097123e-07,
      "loss": 0.0003,
      "reward": 1.221354205161333,
      "reward_std": 0.28732520481571555,
      "rewards/equation_reward_func": 0.26562500675208867,
      "rewards/format_reward_func": 0.9557291939854622,
      "step": 270
    },
    {
      "completion_length": 233.33073616027832,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.12856640865780958,
      "kl": 0.42327880859375,
      "learning_rate": 1.7893859284058378e-07,
      "loss": 0.0004,
      "reward": 1.2708333693444729,
      "reward_std": 0.29480867367237806,
      "rewards/equation_reward_func": 0.33854167466051877,
      "rewards/format_reward_func": 0.9322916902601719,
      "step": 272
    },
    {
      "completion_length": 240.84896278381348,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.14218114335423593,
      "kl": 0.358642578125,
      "learning_rate": 1.7549195509265407e-07,
      "loss": 0.0004,
      "reward": 1.1979167200624943,
      "reward_std": 0.3153935894370079,
      "rewards/equation_reward_func": 0.2421875053551048,
      "rewards/format_reward_func": 0.9557291939854622,
      "step": 274
    },
    {
      "completion_length": 212.5859432220459,
      "epoch": 0.1472,
      "grad_norm": 0.15425213130457013,
      "kl": 0.3331298828125,
      "learning_rate": 1.7206079062562536e-07,
      "loss": 0.0003,
      "reward": 1.2942708618938923,
      "reward_std": 0.28991915099322796,
      "rewards/equation_reward_func": 0.3411458428017795,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 276
    },
    {
      "completion_length": 210.10417222976685,
      "epoch": 0.14826666666666666,
      "grad_norm": 0.1176238080060518,
      "kl": 0.3582763671875,
      "learning_rate": 1.6864581199841226e-07,
      "loss": 0.0004,
      "reward": 1.2864583767950535,
      "reward_std": 0.25504604540765285,
      "rewards/equation_reward_func": 0.3333333423361182,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 278
    },
    {
      "completion_length": 211.26563167572021,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.13044509335491464,
      "kl": 0.3336181640625,
      "learning_rate": 1.6524772840857388e-07,
      "loss": 0.0003,
      "reward": 1.291666705161333,
      "reward_std": 0.2607252886518836,
      "rewards/equation_reward_func": 0.34895834024064243,
      "rewards/format_reward_func": 0.9427083469927311,
      "step": 280
    },
    {
      "completion_length": 211.22396564483643,
      "epoch": 0.1504,
      "grad_norm": 0.10582943007679689,
      "kl": 0.3338623046875,
      "learning_rate": 1.6186724554503237e-07,
      "loss": 0.0003,
      "reward": 1.3229166939854622,
      "reward_std": 0.2122802771627903,
      "rewards/equation_reward_func": 0.34895834093913436,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 282
    },
    {
      "completion_length": 201.869797706604,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.11240228454116408,
      "kl": 0.33392333984375,
      "learning_rate": 1.5850506544152103e-07,
      "loss": 0.0003,
      "reward": 1.4453125298023224,
      "reward_std": 0.2747932760976255,
      "rewards/equation_reward_func": 0.4817708432674408,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 284
    },
    {
      "completion_length": 207.7682342529297,
      "epoch": 0.15253333333333333,
      "grad_norm": 0.13714067033287647,
      "kl": 0.346923828125,
      "learning_rate": 1.5516188633079107e-07,
      "loss": 0.0003,
      "reward": 1.3567708618938923,
      "reward_std": 0.25441598519682884,
      "rewards/equation_reward_func": 0.38802084024064243,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 286
    },
    {
      "completion_length": 202.4375057220459,
      "epoch": 0.1536,
      "grad_norm": 0.13406177536267386,
      "kl": 0.3206787109375,
      "learning_rate": 1.5183840249960784e-07,
      "loss": 0.0003,
      "reward": 1.4375000335276127,
      "reward_std": 0.25232651783153415,
      "rewards/equation_reward_func": 0.4713541753590107,
      "rewards/format_reward_func": 0.9661458469927311,
      "step": 288
    },
    {
      "completion_length": 204.04948711395264,
      "epoch": 0.15466666666666667,
      "grad_norm": 0.16984473117009352,
      "kl": 0.343017578125,
      "learning_rate": 1.4853530414456612e-07,
      "loss": 0.0003,
      "reward": 1.3619792014360428,
      "reward_std": 0.289654694031924,
      "rewards/equation_reward_func": 0.3984375095460564,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 290
    },
    {
      "completion_length": 217.69531965255737,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.12975704642789157,
      "kl": 0.34332275390625,
      "learning_rate": 1.4525327722875568e-07,
      "loss": 0.0003,
      "reward": 1.3281250335276127,
      "reward_std": 0.2559014759026468,
      "rewards/equation_reward_func": 0.3776041779201478,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 292
    },
    {
      "completion_length": 217.88542366027832,
      "epoch": 0.1568,
      "grad_norm": 0.1447740844459516,
      "kl": 0.35040283203125,
      "learning_rate": 1.4199300333930515e-07,
      "loss": 0.0004,
      "reward": 1.3072917014360428,
      "reward_std": 0.2723052576184273,
      "rewards/equation_reward_func": 0.356770837912336,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 294
    },
    {
      "completion_length": 264.39844512939453,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.1548489136971411,
      "kl": 0.3575439453125,
      "learning_rate": 1.3875515954583523e-07,
      "loss": 0.0004,
      "reward": 1.236979216337204,
      "reward_std": 0.40396298840641975,
      "rewards/equation_reward_func": 0.33854167303070426,
      "rewards/format_reward_func": 0.8984375260770321,
      "step": 296
    },
    {
      "completion_length": 241.82292366027832,
      "epoch": 0.15893333333333334,
      "grad_norm": 0.12585452418366477,
      "kl": 0.344970703125,
      "learning_rate": 1.3554041825985e-07,
      "loss": 0.0003,
      "reward": 1.268229205161333,
      "reward_std": 0.34716328978538513,
      "rewards/equation_reward_func": 0.3437500079162419,
      "rewards/format_reward_func": 0.9244791865348816,
      "step": 298
    },
    {
      "completion_length": 229.080735206604,
      "epoch": 0.16,
      "grad_norm": 0.16255294926931252,
      "kl": 0.3646240234375,
      "learning_rate": 1.323494470950949e-07,
      "loss": 0.0004,
      "reward": 1.2708333730697632,
      "reward_std": 0.3114820602349937,
      "rewards/equation_reward_func": 0.33593751140870154,
      "rewards/format_reward_func": 0.9348958507180214,
      "step": 300
    },
    {
      "completion_length": 236.8880262374878,
      "epoch": 0.16106666666666666,
      "grad_norm": 0.10225194248172664,
      "kl": 0.35693359375,
      "learning_rate": 1.2918290872891236e-07,
      "loss": 0.0004,
      "reward": 1.2500000409781933,
      "reward_std": 0.26974589098244905,
      "rewards/equation_reward_func": 0.3098958421032876,
      "rewards/format_reward_func": 0.9401041828095913,
      "step": 302
    },
    {
      "completion_length": 226.41406631469727,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.10873028298363663,
      "kl": 0.354736328125,
      "learning_rate": 1.260414607646213e-07,
      "loss": 0.0004,
      "reward": 1.2916667014360428,
      "reward_std": 0.31566831516101956,
      "rewards/equation_reward_func": 0.356770841171965,
      "rewards/format_reward_func": 0.9348958544433117,
      "step": 304
    },
    {
      "completion_length": 214.26302528381348,
      "epoch": 0.1632,
      "grad_norm": 0.14569928954768213,
      "kl": 0.3468017578125,
      "learning_rate": 1.2292575559495143e-07,
      "loss": 0.0003,
      "reward": 1.3932291939854622,
      "reward_std": 0.3335285438224673,
      "rewards/equation_reward_func": 0.4557291807141155,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 306
    },
    {
      "completion_length": 239.88802909851074,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.12106578538476627,
      "kl": 0.3829345703125,
      "learning_rate": 1.1983644026655835e-07,
      "loss": 0.0004,
      "reward": 1.2317708767950535,
      "reward_std": 0.3285315982066095,
      "rewards/equation_reward_func": 0.3151041779201478,
      "rewards/format_reward_func": 0.9166666865348816,
      "step": 308
    },
    {
      "completion_length": 228.510422706604,
      "epoch": 0.16533333333333333,
      "grad_norm": 0.13496613978502955,
      "kl": 0.395751953125,
      "learning_rate": 1.1677415634565066e-07,
      "loss": 0.0004,
      "reward": 1.2968750409781933,
      "reward_std": 0.3311383193358779,
      "rewards/equation_reward_func": 0.348958341171965,
      "rewards/format_reward_func": 0.9479166902601719,
      "step": 310
    },
    {
      "completion_length": 214.658860206604,
      "epoch": 0.1664,
      "grad_norm": 0.11732455077535138,
      "kl": 0.352783203125,
      "learning_rate": 1.1373953978475353e-07,
      "loss": 0.0004,
      "reward": 1.3333333693444729,
      "reward_std": 0.24432587856426835,
      "rewards/equation_reward_func": 0.4010416781529784,
      "rewards/format_reward_func": 0.9322916865348816,
      "step": 312
    },
    {
      "completion_length": 235.30469417572021,
      "epoch": 0.16746666666666668,
      "grad_norm": 0.1252191603595379,
      "kl": 0.37646484375,
      "learning_rate": 1.1073322079063913e-07,
      "loss": 0.0004,
      "reward": 1.2500000298023224,
      "reward_std": 0.2639185069128871,
      "rewards/equation_reward_func": 0.3020833421032876,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 314
    },
    {
      "completion_length": 238.377610206604,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.14125725531034033,
      "kl": 0.3873291015625,
      "learning_rate": 1.0775582369344946e-07,
      "loss": 0.0004,
      "reward": 1.2994791977107525,
      "reward_std": 0.2981275082565844,
      "rewards/equation_reward_func": 0.3671875074505806,
      "rewards/format_reward_func": 0.9322916865348816,
      "step": 316
    },
    {
      "completion_length": 213.09114933013916,
      "epoch": 0.1696,
      "grad_norm": 0.2908710342105865,
      "kl": 0.4017333984375,
      "learning_rate": 1.0480796681704077e-07,
      "loss": 0.0004,
      "reward": 1.3333333656191826,
      "reward_std": 0.27439625281840563,
      "rewards/equation_reward_func": 0.3802083460614085,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 318
    },
    {
      "completion_length": 241.1015682220459,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.1531930365995129,
      "kl": 0.4034423828125,
      "learning_rate": 1.018902623505741e-07,
      "loss": 0.0004,
      "reward": 1.2291666939854622,
      "reward_std": 0.28229701425880194,
      "rewards/equation_reward_func": 0.29947917512618005,
      "rewards/format_reward_func": 0.9296875074505806,
      "step": 320
    },
    {
      "completion_length": 216.5885467529297,
      "epoch": 0.17173333333333332,
      "grad_norm": 0.11781884256770431,
      "kl": 0.388427734375,
      "learning_rate": 9.900331622138063e-08,
      "loss": 0.0004,
      "reward": 1.3359375447034836,
      "reward_std": 0.30114804534241557,
      "rewards/equation_reward_func": 0.3984375111758709,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 322
    },
    {
      "completion_length": 214.03907012939453,
      "epoch": 0.1728,
      "grad_norm": 0.10954877745998995,
      "kl": 0.368896484375,
      "learning_rate": 9.614772796912681e-08,
      "loss": 0.0004,
      "reward": 1.3671875447034836,
      "reward_std": 0.22091607144102454,
      "rewards/equation_reward_func": 0.41406251210719347,
      "rewards/format_reward_func": 0.9531250223517418,
      "step": 324
    },
    {
      "completion_length": 230.15365314483643,
      "epoch": 0.17386666666666667,
      "grad_norm": 0.13872024524911714,
      "kl": 0.3896484375,
      "learning_rate": 9.332409062130686e-08,
      "loss": 0.0004,
      "reward": 1.3151042088866234,
      "reward_std": 0.3028682228177786,
      "rewards/equation_reward_func": 0.37500000977888703,
      "rewards/format_reward_func": 0.9401041865348816,
      "step": 326
    },
    {
      "completion_length": 222.4427146911621,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.11028673754906193,
      "kl": 0.3843994140625,
      "learning_rate": 9.053299057008699e-08,
      "loss": 0.0004,
      "reward": 1.3489583618938923,
      "reward_std": 0.270692175719887,
      "rewards/equation_reward_func": 0.4010416781529784,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 328
    },
    {
      "completion_length": 231.38282012939453,
      "epoch": 0.176,
      "grad_norm": 2.1383311810316163,
      "kl": 0.735107421875,
      "learning_rate": 8.777500745052743e-08,
      "loss": 0.0007,
      "reward": 1.278645858168602,
      "reward_std": 0.32529697054997087,
      "rewards/equation_reward_func": 0.33333334513008595,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 330
    },
    {
      "completion_length": 232.697922706604,
      "epoch": 0.17706666666666668,
      "grad_norm": 0.14187859687757484,
      "kl": 0.3770751953125,
      "learning_rate": 8.505071402020892e-08,
      "loss": 0.0004,
      "reward": 1.2604167014360428,
      "reward_std": 0.247287486679852,
      "rewards/equation_reward_func": 0.3281250090803951,
      "rewards/format_reward_func": 0.9322916828095913,
      "step": 332
    },
    {
      "completion_length": 214.4010467529297,
      "epoch": 0.17813333333333334,
      "grad_norm": 0.14833746575850393,
      "kl": 0.406005859375,
      "learning_rate": 8.236067604028562e-08,
      "loss": 0.0004,
      "reward": 1.2630208805203438,
      "reward_std": 0.2821256769821048,
      "rewards/equation_reward_func": 0.30468750838190317,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 334
    },
    {
      "completion_length": 213.20834016799927,
      "epoch": 0.1792,
      "grad_norm": 0.12802075982093267,
      "kl": 0.4344482421875,
      "learning_rate": 7.970545215799327e-08,
      "loss": 0.0004,
      "reward": 1.3828125298023224,
      "reward_std": 0.25213046092540026,
      "rewards/equation_reward_func": 0.44531251210719347,
      "rewards/format_reward_func": 0.9375000223517418,
      "step": 336
    },
    {
      "completion_length": 202.57031726837158,
      "epoch": 0.18026666666666666,
      "grad_norm": 0.13303012649631915,
      "kl": 0.371337890625,
      "learning_rate": 7.708559379063204e-08,
      "loss": 0.0004,
      "reward": 1.4036458693444729,
      "reward_std": 0.23924620263278484,
      "rewards/equation_reward_func": 0.43489584466442466,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 338
    },
    {
      "completion_length": 221.41927909851074,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.1402363674174963,
      "kl": 0.3974609375,
      "learning_rate": 7.45016450110534e-08,
      "loss": 0.0004,
      "reward": 1.3463542014360428,
      "reward_std": 0.28338290052488446,
      "rewards/equation_reward_func": 0.40364584419876337,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 340
    },
    {
      "completion_length": 207.67188358306885,
      "epoch": 0.1824,
      "grad_norm": 0.11802543114470954,
      "kl": 0.3690185546875,
      "learning_rate": 7.195414243467029e-08,
      "loss": 0.0004,
      "reward": 1.3020833730697632,
      "reward_std": 0.23917974904179573,
      "rewards/equation_reward_func": 0.3567708432674408,
      "rewards/format_reward_func": 0.9453125223517418,
      "step": 342
    },
    {
      "completion_length": 183.59375286102295,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.14195230576888693,
      "kl": 0.3905029296875,
      "learning_rate": 6.944361510801763e-08,
      "loss": 0.0004,
      "reward": 1.4505208656191826,
      "reward_std": 0.19811600586399436,
      "rewards/equation_reward_func": 0.4739583458285779,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 344
    },
    {
      "completion_length": 216.69792318344116,
      "epoch": 0.18453333333333333,
      "grad_norm": 0.10573381707872133,
      "kl": 0.390380859375,
      "learning_rate": 6.697058439888283e-08,
      "loss": 0.0004,
      "reward": 1.3203125596046448,
      "reward_std": 0.21469871560111642,
      "rewards/equation_reward_func": 0.35416667675599456,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 346
    },
    {
      "completion_length": 233.00781726837158,
      "epoch": 0.1856,
      "grad_norm": 0.13820862378578067,
      "kl": 0.395263671875,
      "learning_rate": 6.453556388803288e-08,
      "loss": 0.0004,
      "reward": 1.281250037252903,
      "reward_std": 0.27478035958483815,
      "rewards/equation_reward_func": 0.3437500111758709,
      "rewards/format_reward_func": 0.9375000074505806,
      "step": 348
    },
    {
      "completion_length": 219.3515682220459,
      "epoch": 0.18666666666666668,
      "grad_norm": 0.12258952059086844,
      "kl": 0.3804931640625,
      "learning_rate": 6.213905926255697e-08,
      "loss": 0.0004,
      "reward": 1.252604216337204,
      "reward_std": 0.23317333636805415,
      "rewards/equation_reward_func": 0.29687500838190317,
      "rewards/format_reward_func": 0.9557291865348816,
      "step": 350
    },
    {
      "completion_length": 212.2656307220459,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.11958751629837379,
      "kl": 0.4261474609375,
      "learning_rate": 5.978156821084987e-08,
      "loss": 0.0004,
      "reward": 1.343750037252903,
      "reward_std": 0.2985822893679142,
      "rewards/equation_reward_func": 0.3828125102445483,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 352
    },
    {
      "completion_length": 220.0442762374878,
      "epoch": 0.1888,
      "grad_norm": 0.12118729628849659,
      "kl": 0.376953125,
      "learning_rate": 5.7463580319254853e-08,
      "loss": 0.0004,
      "reward": 1.3750000484287739,
      "reward_std": 0.20456179324537516,
      "rewards/equation_reward_func": 0.40885417931713164,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 354
    },
    {
      "completion_length": 213.0312557220459,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.12175659264450302,
      "kl": 0.396484375,
      "learning_rate": 5.518557697039081e-08,
      "loss": 0.0004,
      "reward": 1.3593750447034836,
      "reward_std": 0.20982732716947794,
      "rewards/equation_reward_func": 0.39062501466833055,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 356
    },
    {
      "completion_length": 214.0520887374878,
      "epoch": 0.19093333333333334,
      "grad_norm": 0.11008803039882169,
      "kl": 0.38201904296875,
      "learning_rate": 5.294803124318145e-08,
      "loss": 0.0004,
      "reward": 1.359375037252903,
      "reward_std": 0.25468964176252484,
      "rewards/equation_reward_func": 0.4088541760575026,
      "rewards/format_reward_func": 0.9505208469927311,
      "step": 358
    },
    {
      "completion_length": 209.78125619888306,
      "epoch": 0.192,
      "grad_norm": 0.10722836188171843,
      "kl": 0.3824462890625,
      "learning_rate": 5.07514078146106e-08,
      "loss": 0.0004,
      "reward": 1.3750000447034836,
      "reward_std": 0.1821616035886109,
      "rewards/equation_reward_func": 0.4218750118743628,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 360
    },
    {
      "completion_length": 209.42448616027832,
      "epoch": 0.19306666666666666,
      "grad_norm": 0.13457651208761368,
      "kl": 0.384521484375,
      "learning_rate": 4.859616286322094e-08,
      "loss": 0.0004,
      "reward": 1.2968750484287739,
      "reward_std": 0.2525270893238485,
      "rewards/equation_reward_func": 0.3359375102445483,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 362
    },
    {
      "completion_length": 205.18229579925537,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.10020777868736853,
      "kl": 0.3636474609375,
      "learning_rate": 4.648274397437829e-08,
      "loss": 0.0004,
      "reward": 1.3541667014360428,
      "reward_std": 0.20756287965923548,
      "rewards/equation_reward_func": 0.3906250100117177,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 364
    },
    {
      "completion_length": 219.6822967529297,
      "epoch": 0.1952,
      "grad_norm": 0.11935949150349823,
      "kl": 0.4034423828125,
      "learning_rate": 4.4411590047320617e-08,
      "loss": 0.0004,
      "reward": 1.2630208618938923,
      "reward_std": 0.29622318036854267,
      "rewards/equation_reward_func": 0.3098958460614085,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 366
    },
    {
      "completion_length": 230.60156965255737,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.13258533859783664,
      "kl": 0.3961181640625,
      "learning_rate": 4.2383131204010494e-08,
      "loss": 0.0004,
      "reward": 1.2786458879709244,
      "reward_std": 0.2918337839655578,
      "rewards/equation_reward_func": 0.33072917629033327,
      "rewards/format_reward_func": 0.9479166902601719,
      "step": 368
    },
    {
      "completion_length": 203.557297706604,
      "epoch": 0.19733333333333333,
      "grad_norm": 0.1304869275544946,
      "kl": 0.502197265625,
      "learning_rate": 4.039778869981064e-08,
      "loss": 0.0005,
      "reward": 1.4296875521540642,
      "reward_std": 0.3036642442457378,
      "rewards/equation_reward_func": 0.4713541849050671,
      "rewards/format_reward_func": 0.9583333618938923,
      "step": 370
    },
    {
      "completion_length": 211.19531726837158,
      "epoch": 0.1984,
      "grad_norm": 1.144924514294268,
      "kl": 2.046875,
      "learning_rate": 3.845597483600049e-08,
      "loss": 0.002,
      "reward": 1.2447917126119137,
      "reward_std": 0.16396735096350312,
      "rewards/equation_reward_func": 0.2838541741948575,
      "rewards/format_reward_func": 0.9609375260770321,
      "step": 372
    },
    {
      "completion_length": 195.34375524520874,
      "epoch": 0.19946666666666665,
      "grad_norm": 0.10968517021101469,
      "kl": 0.382080078125,
      "learning_rate": 3.655809287415284e-08,
      "loss": 0.0004,
      "reward": 1.4739583618938923,
      "reward_std": 0.20083042420446873,
      "rewards/equation_reward_func": 0.5104166716337204,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 374
    },
    {
      "completion_length": 190.3776092529297,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.12082076973998813,
      "kl": 0.379638671875,
      "learning_rate": 3.4704536952387285e-08,
      "loss": 0.0004,
      "reward": 1.3932292088866234,
      "reward_std": 0.24945232598111033,
      "rewards/equation_reward_func": 0.4244791779201478,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 376
    },
    {
      "completion_length": 203.86719226837158,
      "epoch": 0.2016,
      "grad_norm": 0.12653378929717862,
      "kl": 0.38720703125,
      "learning_rate": 3.2895692003518575e-08,
      "loss": 0.0004,
      "reward": 1.304687537252903,
      "reward_std": 0.2262433897703886,
      "rewards/equation_reward_func": 0.33854167768731713,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 378
    },
    {
      "completion_length": 212.7161512374878,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.11145532156837913,
      "kl": 0.392333984375,
      "learning_rate": 3.113193367511635e-08,
      "loss": 0.0004,
      "reward": 1.3151042088866234,
      "reward_std": 0.22319858288392425,
      "rewards/equation_reward_func": 0.3489583474583924,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 380
    },
    {
      "completion_length": 209.70052814483643,
      "epoch": 0.20373333333333332,
      "grad_norm": 0.12915431757327603,
      "kl": 0.3778076171875,
      "learning_rate": 2.9413628251493934e-08,
      "loss": 0.0004,
      "reward": 1.3463542126119137,
      "reward_std": 0.2786430944688618,
      "rewards/equation_reward_func": 0.39322917349636555,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 382
    },
    {
      "completion_length": 189.9661512374878,
      "epoch": 0.2048,
      "grad_norm": 0.123129591961198,
      "kl": 0.41717529296875,
      "learning_rate": 2.774113257764066e-08,
      "loss": 0.0004,
      "reward": 1.4088541939854622,
      "reward_std": 0.1688971514813602,
      "rewards/equation_reward_func": 0.4322916769888252,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 384
    },
    {
      "completion_length": 205.56250381469727,
      "epoch": 0.20586666666666667,
      "grad_norm": 0.0939102325622787,
      "kl": 0.3865966796875,
      "learning_rate": 2.611479398511518e-08,
      "loss": 0.0004,
      "reward": 1.3177083618938923,
      "reward_std": 0.22996263904497027,
      "rewards/equation_reward_func": 0.3567708469927311,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 386
    },
    {
      "completion_length": 210.1015682220459,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.12154037298414708,
      "kl": 0.39013671875,
      "learning_rate": 2.4534950219914057e-08,
      "loss": 0.0004,
      "reward": 1.3723958730697632,
      "reward_std": 0.28099459456279874,
      "rewards/equation_reward_func": 0.41927084419876337,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 388
    },
    {
      "completion_length": 210.7968807220459,
      "epoch": 0.208,
      "grad_norm": 0.14427015934487517,
      "kl": 0.3916015625,
      "learning_rate": 2.300192937233128e-08,
      "loss": 0.0004,
      "reward": 1.3411458730697632,
      "reward_std": 0.24223684472963214,
      "rewards/equation_reward_func": 0.37760418001562357,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 390
    },
    {
      "completion_length": 189.89583730697632,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.11214385635704058,
      "kl": 0.37567138671875,
      "learning_rate": 2.1516049808822935e-08,
      "loss": 0.0004,
      "reward": 1.5130208730697632,
      "reward_std": 0.207259779330343,
      "rewards/equation_reward_func": 0.5416666772216558,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 392
    },
    {
      "completion_length": 218.88542366027832,
      "epoch": 0.21013333333333334,
      "grad_norm": 0.1023812335324068,
      "kl": 0.4085693359375,
      "learning_rate": 2.007762010589098e-08,
      "loss": 0.0004,
      "reward": 1.2421875335276127,
      "reward_std": 0.21409294102340937,
      "rewards/equation_reward_func": 0.2760416760575026,
      "rewards/format_reward_func": 0.9661458432674408,
      "step": 394
    },
    {
      "completion_length": 209.28386116027832,
      "epoch": 0.2112,
      "grad_norm": 0.10974356057655374,
      "kl": 0.416748046875,
      "learning_rate": 1.8686938986000627e-08,
      "loss": 0.0004,
      "reward": 1.3333333730697632,
      "reward_std": 0.22778501268476248,
      "rewards/equation_reward_func": 0.36197917675599456,
      "rewards/format_reward_func": 0.9713541902601719,
      "step": 396
    },
    {
      "completion_length": 192.30208730697632,
      "epoch": 0.21226666666666666,
      "grad_norm": 0.11760308419492767,
      "kl": 0.37054443359375,
      "learning_rate": 1.734429525554365e-08,
      "loss": 0.0004,
      "reward": 1.4036458656191826,
      "reward_std": 0.22485763859003782,
      "rewards/equation_reward_func": 0.43489584466442466,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 398
    },
    {
      "completion_length": 209.5078182220459,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.10236539080609225,
      "kl": 0.42138671875,
      "learning_rate": 1.604996774486145e-08,
      "loss": 0.0004,
      "reward": 1.3255208767950535,
      "reward_std": 0.2059581303037703,
      "rewards/equation_reward_func": 0.3645833428017795,
      "rewards/format_reward_func": 0.9609375223517418,
      "step": 400
    },
    {
      "completion_length": 216.28385734558105,
      "epoch": 0.2144,
      "grad_norm": 0.10293935285191148,
      "kl": 0.3819580078125,
      "learning_rate": 1.4804225250339281e-08,
      "loss": 0.0004,
      "reward": 1.244791705161333,
      "reward_std": 0.19649146730080247,
      "rewards/equation_reward_func": 0.28645834093913436,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 402
    },
    {
      "completion_length": 207.4218807220459,
      "epoch": 0.21546666666666667,
      "grad_norm": 0.15265266622020815,
      "kl": 0.37109375,
      "learning_rate": 1.360732647858498e-08,
      "loss": 0.0004,
      "reward": 1.3255208730697632,
      "reward_std": 0.24445909913629293,
      "rewards/equation_reward_func": 0.364583341171965,
      "rewards/format_reward_func": 0.9609375260770321,
      "step": 404
    },
    {
      "completion_length": 210.13802528381348,
      "epoch": 0.21653333333333333,
      "grad_norm": 0.14981850219369797,
      "kl": 0.39166259765625,
      "learning_rate": 1.2459519992702311e-08,
      "loss": 0.0004,
      "reward": 1.3489583879709244,
      "reward_std": 0.22357356874272227,
      "rewards/equation_reward_func": 0.3750000090803951,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 406
    },
    {
      "completion_length": 235.8932342529297,
      "epoch": 0.2176,
      "grad_norm": 0.10799575484597196,
      "kl": 0.4716796875,
      "learning_rate": 1.1361044160671629e-08,
      "loss": 0.0005,
      "reward": 1.221354205161333,
      "reward_std": 0.20188527600839734,
      "rewards/equation_reward_func": 0.2786458421032876,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 408
    },
    {
      "completion_length": 219.68750667572021,
      "epoch": 0.21866666666666668,
      "grad_norm": 0.1133082426664694,
      "kl": 0.3914794921875,
      "learning_rate": 1.0312127105846947e-08,
      "loss": 0.0004,
      "reward": 1.3020833805203438,
      "reward_std": 0.2558655394241214,
      "rewards/equation_reward_func": 0.3437500107102096,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 410
    },
    {
      "completion_length": 211.38802814483643,
      "epoch": 0.21973333333333334,
      "grad_norm": 0.13471396481390774,
      "kl": 0.35931396484375,
      "learning_rate": 9.312986659581301e-09,
      "loss": 0.0004,
      "reward": 1.3515625409781933,
      "reward_std": 0.24611902190372348,
      "rewards/equation_reward_func": 0.39843750884756446,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 412
    },
    {
      "completion_length": 201.41146564483643,
      "epoch": 0.2208,
      "grad_norm": 0.11335542913941801,
      "kl": 0.3677978515625,
      "learning_rate": 8.363830315988945e-09,
      "loss": 0.0004,
      "reward": 1.3906250298023224,
      "reward_std": 0.18594739492982626,
      "rewards/equation_reward_func": 0.421875006519258,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 414
    },
    {
      "completion_length": 210.43490362167358,
      "epoch": 0.22186666666666666,
      "grad_norm": 0.14868071239640504,
      "kl": 0.3895263671875,
      "learning_rate": 7.46485518885462e-09,
      "loss": 0.0004,
      "reward": 1.367187537252903,
      "reward_std": 0.24797423044219613,
      "rewards/equation_reward_func": 0.4088541786186397,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 416
    },
    {
      "completion_length": 195.0286521911621,
      "epoch": 0.22293333333333334,
      "grad_norm": 0.30752969562363436,
      "kl": 0.83837890625,
      "learning_rate": 6.616247970698319e-09,
      "loss": 0.0008,
      "reward": 1.3411458730697632,
      "reward_std": 0.19077761005610228,
      "rewards/equation_reward_func": 0.37500000838190317,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 418
    },
    {
      "completion_length": 198.98438119888306,
      "epoch": 0.224,
      "grad_norm": 0.08154005592688214,
      "kl": 0.36297607421875,
      "learning_rate": 5.8181848940044855e-09,
      "loss": 0.0004,
      "reward": 1.4296875447034836,
      "reward_std": 0.1724507100880146,
      "rewards/equation_reward_func": 0.4557291816454381,
      "rewards/format_reward_func": 0.973958358168602,
      "step": 420
    },
    {
      "completion_length": 196.791672706604,
      "epoch": 0.22506666666666666,
      "grad_norm": 0.11472474901171029,
      "kl": 0.3736572265625,
      "learning_rate": 5.070831694623135e-09,
      "loss": 0.0004,
      "reward": 1.4218750447034836,
      "reward_std": 0.18971921829506755,
      "rewards/equation_reward_func": 0.4531250153668225,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 422
    },
    {
      "completion_length": 214.48177433013916,
      "epoch": 0.22613333333333333,
      "grad_norm": 0.19796695153042362,
      "kl": 0.5394287109375,
      "learning_rate": 4.374343577351336e-09,
      "loss": 0.0005,
      "reward": 1.3437500149011612,
      "reward_std": 0.26268601790070534,
      "rewards/equation_reward_func": 0.37760417722165585,
      "rewards/format_reward_func": 0.9661458469927311,
      "step": 424
    },
    {
      "completion_length": 212.32292461395264,
      "epoch": 0.2272,
      "grad_norm": 0.13068267211176063,
      "kl": 0.3795166015625,
      "learning_rate": 3.7288651837012745e-09,
      "loss": 0.0004,
      "reward": 1.3645833730697632,
      "reward_std": 0.2530872863717377,
      "rewards/equation_reward_func": 0.39583334559574723,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 426
    },
    {
      "completion_length": 226.4791717529297,
      "epoch": 0.22826666666666667,
      "grad_norm": 0.14424359988915694,
      "kl": 0.4056396484375,
      "learning_rate": 3.134530561862081e-09,
      "loss": 0.0004,
      "reward": 1.2395833805203438,
      "reward_std": 0.3454265049658716,
      "rewards/equation_reward_func": 0.2994791753590107,
      "rewards/format_reward_func": 0.9401041939854622,
      "step": 428
    },
    {
      "completion_length": 204.4713592529297,
      "epoch": 0.22933333333333333,
      "grad_norm": 0.1330205901178611,
      "kl": 0.366455078125,
      "learning_rate": 2.5914631388619103e-09,
      "loss": 0.0004,
      "reward": 1.3776042014360428,
      "reward_std": 0.22500443970784545,
      "rewards/equation_reward_func": 0.4192708458285779,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 430
    },
    {
      "completion_length": 209.04687976837158,
      "epoch": 0.2304,
      "grad_norm": 0.09979880993143887,
      "kl": 0.391357421875,
      "learning_rate": 2.0997756949353297e-09,
      "loss": 0.0004,
      "reward": 1.2786458730697632,
      "reward_std": 0.2494813478551805,
      "rewards/equation_reward_func": 0.3151041727978736,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 432
    },
    {
      "completion_length": 217.9817762374878,
      "epoch": 0.23146666666666665,
      "grad_norm": 0.10595848082640444,
      "kl": 0.3980712890625,
      "learning_rate": 1.6595703401020844e-09,
      "loss": 0.0004,
      "reward": 1.2473958618938923,
      "reward_std": 0.24186762887984514,
      "rewards/equation_reward_func": 0.2890625062864274,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 434
    },
    {
      "completion_length": 216.22656726837158,
      "epoch": 0.23253333333333334,
      "grad_norm": 0.17534721876727652,
      "kl": 0.3792724609375,
      "learning_rate": 1.2709384929615596e-09,
      "loss": 0.0004,
      "reward": 1.315104205161333,
      "reward_std": 0.2798662828281522,
      "rewards/equation_reward_func": 0.34895833884365857,
      "rewards/format_reward_func": 0.9661458469927311,
      "step": 436
    },
    {
      "completion_length": 191.87239933013916,
      "epoch": 0.2336,
      "grad_norm": 0.11732733986679211,
      "kl": 0.3629150390625,
      "learning_rate": 9.339608617077165e-10,
      "loss": 0.0004,
      "reward": 1.4375000409781933,
      "reward_std": 0.18146159406751394,
      "rewards/equation_reward_func": 0.4661458486225456,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 438
    },
    {
      "completion_length": 218.74219226837158,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.1658444401737881,
      "kl": 0.392333984375,
      "learning_rate": 6.487074273681114e-10,
      "loss": 0.0004,
      "reward": 1.257812537252903,
      "reward_std": 0.2663137102499604,
      "rewards/equation_reward_func": 0.31250000838190317,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 440
    },
    {
      "completion_length": 201.56250667572021,
      "epoch": 0.23573333333333332,
      "grad_norm": 0.12929764070884708,
      "kl": 0.39117431640625,
      "learning_rate": 4.152374292708538e-10,
      "loss": 0.0004,
      "reward": 1.3906250447034836,
      "reward_std": 0.2424735059030354,
      "rewards/equation_reward_func": 0.4322916816454381,
      "rewards/format_reward_func": 0.958333358168602,
      "step": 442
    },
    {
      "completion_length": 220.77084064483643,
      "epoch": 0.2368,
      "grad_norm": 0.09650589876778566,
      "kl": 0.421142578125,
      "learning_rate": 2.3359935274214204e-10,
      "loss": 0.0004,
      "reward": 1.286458384245634,
      "reward_std": 0.18493427569046617,
      "rewards/equation_reward_func": 0.32031250884756446,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 444
    },
    {
      "completion_length": 201.27604961395264,
      "epoch": 0.23786666666666667,
      "grad_norm": 0.09761306843622496,
      "kl": 0.372802734375,
      "learning_rate": 1.0383091903720665e-10,
      "loss": 0.0004,
      "reward": 1.3567708767950535,
      "reward_std": 0.1694594444707036,
      "rewards/equation_reward_func": 0.3854166795499623,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 446
    },
    {
      "completion_length": 200.41146278381348,
      "epoch": 0.23893333333333333,
      "grad_norm": 0.12671602239291313,
      "kl": 0.3585205078125,
      "learning_rate": 2.595907750671533e-11,
      "loss": 0.0004,
      "reward": 1.3880208618938923,
      "reward_std": 0.1734816818498075,
      "rewards/equation_reward_func": 0.4140625111758709,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 448
    },
    {
      "completion_length": 214.627610206604,
      "epoch": 0.24,
      "grad_norm": 0.1096617550559703,
      "kl": 0.471435546875,
      "learning_rate": 0.0,
      "loss": 0.0005,
      "reward": 1.2604167200624943,
      "reward_std": 0.20810852525755763,
      "rewards/equation_reward_func": 0.30208334350027144,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 450
    },
    {
      "epoch": 0.24,
      "step": 450,
      "total_flos": 0.0,
      "train_loss": 0.00024176122421092967,
      "train_runtime": 39285.3137,
      "train_samples_per_second": 0.275,
      "train_steps_per_second": 0.011
    }
  ],
  "logging_steps": 2,
  "max_steps": 450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}