{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.21333333333333335,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 485.49220275878906,
      "epoch": 0.0010666666666666667,
      "grad_norm": 0.13112049806907955,
      "kl": 0.0,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0,
      "reward": 0.3281250111758709,
      "reward_std": 0.4913413915783167,
      "rewards/equation_reward_func": 0.05729166814126074,
      "rewards/format_reward_func": 0.27083334047347307,
      "step": 2
    },
    {
      "completion_length": 530.2265796661377,
      "epoch": 0.0021333333333333334,
      "grad_norm": 0.12198138838063947,
      "kl": 0.0003826618194580078,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0,
      "reward": 0.299479172565043,
      "reward_std": 0.44007157534360886,
      "rewards/equation_reward_func": 0.03385416744276881,
      "rewards/format_reward_func": 0.26562500977888703,
      "step": 4
    },
    {
      "completion_length": 496.8776264190674,
      "epoch": 0.0032,
      "grad_norm": 0.12162717174644112,
      "kl": 0.0003865957260131836,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0,
      "reward": 0.2916666781529784,
      "reward_std": 0.47719811648130417,
      "rewards/equation_reward_func": 0.05468750116415322,
      "rewards/format_reward_func": 0.23697917303070426,
      "step": 6
    },
    {
      "completion_length": 504.77865982055664,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.13232346241260942,
      "kl": 0.0003762245178222656,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.33593750558793545,
      "reward_std": 0.4751614350825548,
      "rewards/equation_reward_func": 0.04947916720993817,
      "rewards/format_reward_func": 0.28645834140479565,
      "step": 8
    },
    {
      "completion_length": 475.7057456970215,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.13843718324607834,
      "kl": 0.0003968477249145508,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0,
      "reward": 0.3828125102445483,
      "reward_std": 0.5227206833660603,
      "rewards/equation_reward_func": 0.0885416679084301,
      "rewards/format_reward_func": 0.2942708395421505,
      "step": 10
    },
    {
      "completion_length": 475.98699378967285,
      "epoch": 0.0064,
      "grad_norm": 0.14337833822484186,
      "kl": 0.0004818439483642578,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.33333334140479565,
      "reward_std": 0.4693184234201908,
      "rewards/equation_reward_func": 0.05468750139698386,
      "rewards/format_reward_func": 0.2786458395421505,
      "step": 12
    },
    {
      "completion_length": 472.8099060058594,
      "epoch": 0.007466666666666667,
      "grad_norm": 0.129867140491159,
      "kl": 0.0007684230804443359,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.45052084885537624,
      "reward_std": 0.5166866518557072,
      "rewards/equation_reward_func": 0.041666667675599456,
      "rewards/format_reward_func": 0.40885418094694614,
      "step": 14
    },
    {
      "completion_length": 464.46875762939453,
      "epoch": 0.008533333333333334,
      "grad_norm": 0.1279190002551803,
      "kl": 0.0013058185577392578,
      "learning_rate": 4.999740409224932e-07,
      "loss": 0.0,
      "reward": 0.5052083488553762,
      "reward_std": 0.5728582534939051,
      "rewards/equation_reward_func": 0.06510416814126074,
      "rewards/format_reward_func": 0.4401041753590107,
      "step": 16
    },
    {
      "completion_length": 480.2291717529297,
      "epoch": 0.0096,
      "grad_norm": 0.10647649710010443,
      "kl": 0.00380706787109375,
      "learning_rate": 4.998961690809627e-07,
      "loss": 0.0,
      "reward": 0.6588541902601719,
      "reward_std": 0.5287479311227798,
      "rewards/equation_reward_func": 0.05468750139698386,
      "rewards/format_reward_func": 0.6041666828095913,
      "step": 18
    },
    {
      "completion_length": 493.8073043823242,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.10522760864642984,
      "kl": 0.004913330078125,
      "learning_rate": 4.997664006472578e-07,
      "loss": 0.0,
      "reward": 0.7734375223517418,
      "reward_std": 0.4910791157744825,
      "rewards/equation_reward_func": 0.07031250116415322,
      "rewards/format_reward_func": 0.7031250186264515,
      "step": 20
    },
    {
      "completion_length": 455.6510524749756,
      "epoch": 0.011733333333333333,
      "grad_norm": 0.09917661844432689,
      "kl": 0.008411407470703125,
      "learning_rate": 4.995847625707292e-07,
      "loss": 0.0,
      "reward": 0.7812500186264515,
      "reward_std": 0.4674575887620449,
      "rewards/equation_reward_func": 0.0651041679084301,
      "rewards/format_reward_func": 0.7161458507180214,
      "step": 22
    },
    {
      "completion_length": 464.50782012939453,
      "epoch": 0.0128,
      "grad_norm": 0.10189992043189328,
      "kl": 0.0059833526611328125,
      "learning_rate": 4.993512925726318e-07,
      "loss": 0.0,
      "reward": 0.8619791865348816,
      "reward_std": 0.49650320410728455,
      "rewards/equation_reward_func": 0.08854166860692203,
      "rewards/format_reward_func": 0.7734375223517418,
      "step": 24
    },
    {
      "completion_length": 447.40626335144043,
      "epoch": 0.013866666666666666,
      "grad_norm": 0.09219816177682034,
      "kl": 0.006900787353515625,
      "learning_rate": 4.990660391382923e-07,
      "loss": 0.0,
      "reward": 0.960937537252903,
      "reward_std": 0.4377214591950178,
      "rewards/equation_reward_func": 0.11718750256113708,
      "rewards/format_reward_func": 0.8437500260770321,
      "step": 26
    },
    {
      "completion_length": 436.3099117279053,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.07907793746945187,
      "kl": 0.009281158447265625,
      "learning_rate": 4.987290615070384e-07,
      "loss": 0.0,
      "reward": 0.9713542014360428,
      "reward_std": 0.3975960807874799,
      "rewards/equation_reward_func": 0.09895833535119891,
      "rewards/format_reward_func": 0.872395858168602,
      "step": 28
    },
    {
      "completion_length": 428.1354293823242,
      "epoch": 0.016,
      "grad_norm": 0.0845150241699145,
      "kl": 0.011430740356445312,
      "learning_rate": 4.983404296598978e-07,
      "loss": 0.0,
      "reward": 0.9531250298023224,
      "reward_std": 0.359499204903841,
      "rewards/equation_reward_func": 0.0703125016298145,
      "rewards/format_reward_func": 0.8828125223517418,
      "step": 30
    },
    {
      "completion_length": 434.62500953674316,
      "epoch": 0.017066666666666667,
      "grad_norm": 0.08196142586101564,
      "kl": 0.010782241821289062,
      "learning_rate": 4.979002243050646e-07,
      "loss": 0.0,
      "reward": 1.0260416977107525,
      "reward_std": 0.30062979739159346,
      "rewards/equation_reward_func": 0.09635416860692203,
      "rewards/format_reward_func": 0.9296875260770321,
      "step": 32
    },
    {
      "completion_length": 438.08595085144043,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.08432790923176402,
      "kl": 0.012115478515625,
      "learning_rate": 4.974085368611381e-07,
      "loss": 0.0,
      "reward": 1.049479205161333,
      "reward_std": 0.3121222285553813,
      "rewards/equation_reward_func": 0.11197917000390589,
      "rewards/format_reward_func": 0.9375000223517418,
      "step": 34
    },
    {
      "completion_length": 421.16407203674316,
      "epoch": 0.0192,
      "grad_norm": 0.080639508238748,
      "kl": 0.01397705078125,
      "learning_rate": 4.968654694381379e-07,
      "loss": 0.0,
      "reward": 1.0598958618938923,
      "reward_std": 0.27779901027679443,
      "rewards/equation_reward_func": 0.10416666907258332,
      "rewards/format_reward_func": 0.9557291865348816,
      "step": 36
    },
    {
      "completion_length": 405.12240982055664,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.07681322754981268,
      "kl": 0.013866424560546875,
      "learning_rate": 4.962711348162987e-07,
      "loss": 0.0,
      "reward": 1.0390625409781933,
      "reward_std": 0.2664716215804219,
      "rewards/equation_reward_func": 0.08593750279396772,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 38
    },
    {
      "completion_length": 400.0104274749756,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.08198714493646655,
      "kl": 0.015293121337890625,
      "learning_rate": 4.956256564226487e-07,
      "loss": 0.0,
      "reward": 1.1067708730697632,
      "reward_std": 0.28682188084349036,
      "rewards/equation_reward_func": 0.14322917023673654,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 40
    },
    {
      "completion_length": 400.78386306762695,
      "epoch": 0.0224,
      "grad_norm": 0.0855015587257399,
      "kl": 0.018611907958984375,
      "learning_rate": 4.949291683053768e-07,
      "loss": 0.0,
      "reward": 1.0937500223517418,
      "reward_std": 0.24716421775519848,
      "rewards/equation_reward_func": 0.11197916930541396,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 42
    },
    {
      "completion_length": 408.34115409851074,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.07249139521720419,
      "kl": 0.017139434814453125,
      "learning_rate": 4.941818151059955e-07,
      "loss": 0.0,
      "reward": 1.0546875298023224,
      "reward_std": 0.24979113461449742,
      "rewards/equation_reward_func": 0.0963541695382446,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 44
    },
    {
      "completion_length": 388.62761306762695,
      "epoch": 0.024533333333333334,
      "grad_norm": 0.07507762465990464,
      "kl": 0.017795562744140625,
      "learning_rate": 4.933837520293017e-07,
      "loss": 0.0,
      "reward": 1.0781250484287739,
      "reward_std": 0.2523620016872883,
      "rewards/equation_reward_func": 0.11197917046956718,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 46
    },
    {
      "completion_length": 399.57032585144043,
      "epoch": 0.0256,
      "grad_norm": 0.06604121980517051,
      "kl": 0.017971038818359375,
      "learning_rate": 4.925351448111454e-07,
      "loss": 0.0,
      "reward": 1.0468750335276127,
      "reward_std": 0.21127380011603236,
      "rewards/equation_reward_func": 0.07552083511836827,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 48
    },
    {
      "completion_length": 392.4687557220459,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.0903310628098169,
      "kl": 0.0193939208984375,
      "learning_rate": 4.91636169684011e-07,
      "loss": 0.0,
      "reward": 1.1093750409781933,
      "reward_std": 0.29062134958803654,
      "rewards/equation_reward_func": 0.13541667209938169,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 50
    },
    {
      "completion_length": 374.1067810058594,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.07170688038365744,
      "kl": 0.02114105224609375,
      "learning_rate": 4.906870133404186e-07,
      "loss": 0.0,
      "reward": 1.0833333693444729,
      "reward_std": 0.2505181049928069,
      "rewards/equation_reward_func": 0.11458333535119891,
      "rewards/format_reward_func": 0.9687500074505806,
      "step": 52
    },
    {
      "completion_length": 385.36980056762695,
      "epoch": 0.0288,
      "grad_norm": 0.08750349012682264,
      "kl": 0.02417755126953125,
      "learning_rate": 4.896878728941531e-07,
      "loss": 0.0,
      "reward": 1.1432292014360428,
      "reward_std": 0.3026517196558416,
      "rewards/equation_reward_func": 0.16666667186655104,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 54
    },
    {
      "completion_length": 382.93751335144043,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.08790417727425984,
      "kl": 0.018756866455078125,
      "learning_rate": 4.886389558393284e-07,
      "loss": 0.0,
      "reward": 1.1223958693444729,
      "reward_std": 0.2850013840943575,
      "rewards/equation_reward_func": 0.14322917023673654,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 56
    },
    {
      "completion_length": 403.513032913208,
      "epoch": 0.030933333333333334,
      "grad_norm": 0.07614614875477466,
      "kl": 0.02037811279296875,
      "learning_rate": 4.875404800072976e-07,
      "loss": 0.0,
      "reward": 1.1432292088866234,
      "reward_std": 0.2669796203263104,
      "rewards/equation_reward_func": 0.16145833977498114,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 58
    },
    {
      "completion_length": 382.6432418823242,
      "epoch": 0.032,
      "grad_norm": 0.0923539372935242,
      "kl": 0.02227783203125,
      "learning_rate": 4.86392673521415e-07,
      "loss": 0.0,
      "reward": 1.1536458693444729,
      "reward_std": 0.31879409588873386,
      "rewards/equation_reward_func": 0.1796875053551048,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 60
    },
    {
      "completion_length": 358.4349060058594,
      "epoch": 0.03306666666666667,
      "grad_norm": 0.10173008574938576,
      "kl": 0.02350616455078125,
      "learning_rate": 4.851957747496606e-07,
      "loss": 0.0,
      "reward": 1.1562500447034836,
      "reward_std": 0.2772155348211527,
      "rewards/equation_reward_func": 0.16927083814516664,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 62
    },
    {
      "completion_length": 364.18490409851074,
      "epoch": 0.034133333333333335,
      "grad_norm": 0.08496355305518047,
      "kl": 0.02630615234375,
      "learning_rate": 4.839500322551386e-07,
      "loss": 0.0,
      "reward": 1.1093750447034836,
      "reward_std": 0.24286148557439446,
      "rewards/equation_reward_func": 0.12760417093522847,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 64
    },
    {
      "completion_length": 373.833345413208,
      "epoch": 0.0352,
      "grad_norm": 0.09458816516251609,
      "kl": 0.02667999267578125,
      "learning_rate": 4.826557047444563e-07,
      "loss": 0.0,
      "reward": 1.1848958656191826,
      "reward_std": 0.3138170298188925,
      "rewards/equation_reward_func": 0.20572917233221233,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 66
    },
    {
      "completion_length": 352.2083435058594,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.08416786947789269,
      "kl": 0.03083038330078125,
      "learning_rate": 4.813130610139993e-07,
      "loss": 0.0,
      "reward": 1.1744792088866234,
      "reward_std": 0.2558550937101245,
      "rewards/equation_reward_func": 0.18489583861082792,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 68
    },
    {
      "completion_length": 377.62500953674316,
      "epoch": 0.037333333333333336,
      "grad_norm": 0.07539913544982475,
      "kl": 0.03000640869140625,
      "learning_rate": 4.799223798941089e-07,
      "loss": 0.0,
      "reward": 1.070312537252903,
      "reward_std": 0.17193882586434484,
      "rewards/equation_reward_func": 0.08072916814126074,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 70
    },
    {
      "completion_length": 370.4531364440918,
      "epoch": 0.0384,
      "grad_norm": 0.09187391682605524,
      "kl": 0.03436279296875,
      "learning_rate": 4.78483950191177e-07,
      "loss": 0.0,
      "reward": 1.1562500298023224,
      "reward_std": 0.2756755482405424,
      "rewards/equation_reward_func": 0.1770833362825215,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 72
    },
    {
      "completion_length": 373.04688453674316,
      "epoch": 0.039466666666666664,
      "grad_norm": 0.09537515327502834,
      "kl": 0.03740692138671875,
      "learning_rate": 4.769980706276687e-07,
      "loss": 0.0,
      "reward": 1.1354167088866234,
      "reward_std": 0.25582731096073985,
      "rewards/equation_reward_func": 0.15885417256504297,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 74
    },
    {
      "completion_length": 387.8567810058594,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.08000596022141122,
      "kl": 0.0389404296875,
      "learning_rate": 4.7546504978008595e-07,
      "loss": 0.0,
      "reward": 1.1328125335276127,
      "reward_std": 0.30721960263326764,
      "rewards/equation_reward_func": 0.1666666700039059,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 76
    },
    {
      "completion_length": 390.2448024749756,
      "epoch": 0.0416,
      "grad_norm": 0.07863626089779173,
      "kl": 0.0442047119140625,
      "learning_rate": 4.738852060148848e-07,
      "loss": 0.0,
      "reward": 1.127604205161333,
      "reward_std": 0.26901706866919994,
      "rewards/equation_reward_func": 0.15364583814516664,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 78
    },
    {
      "completion_length": 379.2161521911621,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.08601071973954262,
      "kl": 0.0434417724609375,
      "learning_rate": 4.722588674223593e-07,
      "loss": 0.0,
      "reward": 1.1380208656191826,
      "reward_std": 0.2812240272760391,
      "rewards/equation_reward_func": 0.164062503259629,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 80
    },
    {
      "completion_length": 371.59115409851074,
      "epoch": 0.04373333333333333,
      "grad_norm": 0.07568126962421509,
      "kl": 0.0497894287109375,
      "learning_rate": 4.70586371748506e-07,
      "loss": 0.0,
      "reward": 1.1380208656191826,
      "reward_std": 0.25603401800617576,
      "rewards/equation_reward_func": 0.15885417070239782,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 82
    },
    {
      "completion_length": 381.30469703674316,
      "epoch": 0.0448,
      "grad_norm": 0.08065111561901392,
      "kl": 0.0525665283203125,
      "learning_rate": 4.6886806632488363e-07,
      "loss": 0.0001,
      "reward": 1.169270858168602,
      "reward_std": 0.25962691847234964,
      "rewards/equation_reward_func": 0.19531250419095159,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 84
    },
    {
      "completion_length": 350.8020944595337,
      "epoch": 0.04586666666666667,
      "grad_norm": 0.09857801804683694,
      "kl": 0.0528717041015625,
      "learning_rate": 4.6710430799648143e-07,
      "loss": 0.0001,
      "reward": 1.1796875447034836,
      "reward_std": 0.2887058644555509,
      "rewards/equation_reward_func": 0.1979166748933494,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 86
    },
    {
      "completion_length": 353.0573024749756,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.09077630030450431,
      "kl": 0.059906005859375,
      "learning_rate": 4.652954630476127e-07,
      "loss": 0.0001,
      "reward": 1.2239583730697632,
      "reward_std": 0.30759388813748956,
      "rewards/equation_reward_func": 0.2500000069849193,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 88
    },
    {
      "completion_length": 359.1666784286499,
      "epoch": 0.048,
      "grad_norm": 0.09945645690703775,
      "kl": 0.0612030029296875,
      "learning_rate": 4.6344190712584713e-07,
      "loss": 0.0001,
      "reward": 1.2630208805203438,
      "reward_std": 0.27152396691963077,
      "rewards/equation_reward_func": 0.27604167466051877,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 90
    },
    {
      "completion_length": 349.7083396911621,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.08132855652754273,
      "kl": 0.06951904296875,
      "learning_rate": 4.615440251639995e-07,
      "loss": 0.0001,
      "reward": 1.2395833656191826,
      "reward_std": 0.23549415357410908,
      "rewards/equation_reward_func": 0.25520834024064243,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 92
    },
    {
      "completion_length": 340.52084159851074,
      "epoch": 0.050133333333333335,
      "grad_norm": 0.08537044318099979,
      "kl": 0.075592041015625,
      "learning_rate": 4.596022113001894e-07,
      "loss": 0.0001,
      "reward": 1.2916667014360428,
      "reward_std": 0.3320260518230498,
      "rewards/equation_reward_func": 0.312500006519258,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 94
    },
    {
      "completion_length": 327.13282012939453,
      "epoch": 0.0512,
      "grad_norm": 0.09240179037114699,
      "kl": 0.070465087890625,
      "learning_rate": 4.576168687959895e-07,
      "loss": 0.0001,
      "reward": 1.3255208805203438,
      "reward_std": 0.386608456261456,
      "rewards/equation_reward_func": 0.3567708469927311,
      "rewards/format_reward_func": 0.9687500074505806,
      "step": 96
    },
    {
      "completion_length": 364.68751335144043,
      "epoch": 0.05226666666666667,
      "grad_norm": 0.10186126690814491,
      "kl": 0.080108642578125,
      "learning_rate": 4.555884099526793e-07,
      "loss": 0.0001,
      "reward": 1.1770833656191826,
      "reward_std": 0.2261401410214603,
      "rewards/equation_reward_func": 0.19010417093522847,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 98
    },
    {
      "completion_length": 378.69271755218506,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.08283978901098998,
      "kl": 0.070831298828125,
      "learning_rate": 4.5351725602562174e-07,
      "loss": 0.0001,
      "reward": 1.2083333805203438,
      "reward_std": 0.2539973724633455,
      "rewards/equation_reward_func": 0.22656251001171768,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 100
    },
    {
      "completion_length": 349.2812604904175,
      "epoch": 0.0544,
      "grad_norm": 0.10236681253344297,
      "kl": 0.0821533203125,
      "learning_rate": 4.514038371367791e-07,
      "loss": 0.0001,
      "reward": 1.3098958730697632,
      "reward_std": 0.3604734097607434,
      "rewards/equation_reward_func": 0.3437500107102096,
      "rewards/format_reward_func": 0.9661458432674408,
      "step": 102
    },
    {
      "completion_length": 357.3619899749756,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.09738517323229856,
      "kl": 0.086883544921875,
      "learning_rate": 4.4924859218538936e-07,
      "loss": 0.0001,
      "reward": 1.3046875447034836,
      "reward_std": 0.3255553734488785,
      "rewards/equation_reward_func": 0.33072917233221233,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 104
    },
    {
      "completion_length": 365.2760524749756,
      "epoch": 0.05653333333333333,
      "grad_norm": 0.08975705250496918,
      "kl": 0.09002685546875,
      "learning_rate": 4.470519687568185e-07,
      "loss": 0.0001,
      "reward": 1.2968750484287739,
      "reward_std": 0.33681244123727083,
      "rewards/equation_reward_func": 0.3203125132713467,
      "rewards/format_reward_func": 0.9765625074505806,
      "step": 106
    },
    {
      "completion_length": 380.40365505218506,
      "epoch": 0.0576,
      "grad_norm": 0.08944591293143872,
      "kl": 0.085113525390625,
      "learning_rate": 4.4481442302960923e-07,
      "loss": 0.0001,
      "reward": 1.2135417126119137,
      "reward_std": 0.2771795648150146,
      "rewards/equation_reward_func": 0.23697917629033327,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 108
    },
    {
      "completion_length": 389.94271659851074,
      "epoch": 0.058666666666666666,
      "grad_norm": 0.09619766220094576,
      "kl": 0.09130859375,
      "learning_rate": 4.4253641968074505e-07,
      "loss": 0.0001,
      "reward": 1.2500000335276127,
      "reward_std": 0.2892899289727211,
      "rewards/equation_reward_func": 0.28645833977498114,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 110
    },
    {
      "completion_length": 381.99219512939453,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.10498087313979376,
      "kl": 0.09423828125,
      "learning_rate": 4.402184317891501e-07,
      "loss": 0.0001,
      "reward": 1.3229167088866234,
      "reward_std": 0.32423597015440464,
      "rewards/equation_reward_func": 0.35156250768341124,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 112
    },
    {
      "completion_length": 391.36719608306885,
      "epoch": 0.0608,
      "grad_norm": 0.09626210263887228,
      "kl": 0.097381591796875,
      "learning_rate": 4.37860940737443e-07,
      "loss": 0.0001,
      "reward": 1.2500000298023224,
      "reward_std": 0.3341089729219675,
      "rewards/equation_reward_func": 0.28385417186655104,
      "rewards/format_reward_func": 0.9661458469927311,
      "step": 114
    },
    {
      "completion_length": 379.2812614440918,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.0919023125255689,
      "kl": 0.09576416015625,
      "learning_rate": 4.354644361119671e-07,
      "loss": 0.0001,
      "reward": 1.2968750447034836,
      "reward_std": 0.30213321885094047,
      "rewards/equation_reward_func": 0.3255208437331021,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 116
    },
    {
      "completion_length": 377.5989694595337,
      "epoch": 0.06293333333333333,
      "grad_norm": 0.1258891080084215,
      "kl": 0.1126708984375,
      "learning_rate": 4.3302941560111716e-07,
      "loss": 0.0001,
      "reward": 1.3880208656191826,
      "reward_std": 0.29082584474235773,
      "rewards/equation_reward_func": 0.4192708428017795,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 118
    },
    {
      "completion_length": 406.2239742279053,
      "epoch": 0.064,
      "grad_norm": 0.07341804992437591,
      "kl": 0.094940185546875,
      "learning_rate": 4.3055638489198236e-07,
      "loss": 0.0001,
      "reward": 1.3072916977107525,
      "reward_std": 0.3007106310687959,
      "rewards/equation_reward_func": 0.3359375118743628,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 120
    },
    {
      "completion_length": 440.3489694595337,
      "epoch": 0.06506666666666666,
      "grad_norm": 0.07202980759381214,
      "kl": 0.102081298828125,
      "learning_rate": 4.280458575653296e-07,
      "loss": 0.0001,
      "reward": 1.2942708879709244,
      "reward_std": 0.2781888456083834,
      "rewards/equation_reward_func": 0.34635417629033327,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 122
    },
    {
      "completion_length": 373.723970413208,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.12005799306967507,
      "kl": 0.109588623046875,
      "learning_rate": 4.2549835498894665e-07,
      "loss": 0.0001,
      "reward": 1.3619792126119137,
      "reward_std": 0.30349841713905334,
      "rewards/equation_reward_func": 0.39583334419876337,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 124
    },
    {
      "completion_length": 424.8880310058594,
      "epoch": 0.0672,
      "grad_norm": 0.08141812528362076,
      "kl": 0.11883544921875,
      "learning_rate": 4.229144062093679e-07,
      "loss": 0.0001,
      "reward": 1.3072916939854622,
      "reward_std": 0.34696589363738894,
      "rewards/equation_reward_func": 0.37760417303070426,
      "rewards/format_reward_func": 0.9296875186264515,
      "step": 126
    },
    {
      "completion_length": 425.73959732055664,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.08953036809534468,
      "kl": 0.10595703125,
      "learning_rate": 4.2029454784200675e-07,
      "loss": 0.0001,
      "reward": 1.3203125335276127,
      "reward_std": 0.2961498526856303,
      "rewards/equation_reward_func": 0.3697916716337204,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 128
    },
    {
      "completion_length": 456.8411560058594,
      "epoch": 0.06933333333333333,
      "grad_norm": 0.10202840286205975,
      "kl": 0.128875732421875,
      "learning_rate": 4.1763932395971433e-07,
      "loss": 0.0001,
      "reward": 1.2786458767950535,
      "reward_std": 0.3274143426679075,
      "rewards/equation_reward_func": 0.3463541774544865,
      "rewards/format_reward_func": 0.9322916902601719,
      "step": 130
    },
    {
      "completion_length": 391.82552909851074,
      "epoch": 0.0704,
      "grad_norm": 0.09343731649452018,
      "kl": 0.116180419921875,
      "learning_rate": 4.1494928597979117e-07,
      "loss": 0.0001,
      "reward": 1.4427083693444729,
      "reward_std": 0.2739125872030854,
      "rewards/equation_reward_func": 0.48958334792405367,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 132
    },
    {
      "completion_length": 431.3073043823242,
      "epoch": 0.07146666666666666,
      "grad_norm": 0.09948838006778335,
      "kl": 0.113861083984375,
      "learning_rate": 4.122249925494726e-07,
      "loss": 0.0001,
      "reward": 1.3281250447034836,
      "reward_std": 0.267287774477154,
      "rewards/equation_reward_func": 0.37239584675990045,
      "rewards/format_reward_func": 0.955729179084301,
      "step": 134
    },
    {
      "completion_length": 422.07032203674316,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.07206798558431624,
      "kl": 0.13238525390625,
      "learning_rate": 4.094670094299131e-07,
      "loss": 0.0001,
      "reward": 1.3750000409781933,
      "reward_std": 0.2928238473832607,
      "rewards/equation_reward_func": 0.42447917675599456,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 136
    },
    {
      "completion_length": 443.54688358306885,
      "epoch": 0.0736,
      "grad_norm": 0.10976069905088891,
      "kl": 0.109039306640625,
      "learning_rate": 4.066759093786931e-07,
      "loss": 0.0001,
      "reward": 1.2630208693444729,
      "reward_std": 0.2776200850494206,
      "rewards/equation_reward_func": 0.33593750884756446,
      "rewards/format_reward_func": 0.927083358168602,
      "step": 138
    },
    {
      "completion_length": 398.4505367279053,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.084796835919473,
      "kl": 0.12408447265625,
      "learning_rate": 4.038522720308732e-07,
      "loss": 0.0001,
      "reward": 1.4375000484287739,
      "reward_std": 0.2496197698637843,
      "rewards/equation_reward_func": 0.4739583432674408,
      "rewards/format_reward_func": 0.9635416902601719,
      "step": 140
    },
    {
      "completion_length": 375.6875104904175,
      "epoch": 0.07573333333333333,
      "grad_norm": 0.06578890547258132,
      "kl": 0.134796142578125,
      "learning_rate": 4.009966837786194e-07,
      "loss": 0.0001,
      "reward": 1.4114583730697632,
      "reward_std": 0.2610441828146577,
      "rewards/equation_reward_func": 0.4453125111758709,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 142
    },
    {
      "completion_length": 368.48959255218506,
      "epoch": 0.0768,
      "grad_norm": 0.10902428052363637,
      "kl": 0.136627197265625,
      "learning_rate": 3.981097376494259e-07,
      "loss": 0.0001,
      "reward": 1.4687500521540642,
      "reward_std": 0.25434603728353977,
      "rewards/equation_reward_func": 0.49218752002343535,
      "rewards/format_reward_func": 0.9765625223517418,
      "step": 144
    },
    {
      "completion_length": 406.8750114440918,
      "epoch": 0.07786666666666667,
      "grad_norm": 0.1079461409291905,
      "kl": 0.129119873046875,
      "learning_rate": 3.951920331829592e-07,
      "loss": 0.0001,
      "reward": 1.3281250484287739,
      "reward_std": 0.25655436515808105,
      "rewards/equation_reward_func": 0.372395841171965,
      "rewards/format_reward_func": 0.9557291902601719,
      "step": 146
    },
    {
      "completion_length": 401.2213668823242,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.1067111189421742,
      "kl": 0.132171630859375,
      "learning_rate": 3.922441763065506e-07,
      "loss": 0.0001,
      "reward": 1.3906250447034836,
      "reward_std": 0.249761619605124,
      "rewards/equation_reward_func": 0.4270833439659327,
      "rewards/format_reward_func": 0.9635416902601719,
      "step": 148
    },
    {
      "completion_length": 450.89844512939453,
      "epoch": 0.08,
      "grad_norm": 0.07018564082166065,
      "kl": 0.112640380859375,
      "learning_rate": 3.8926677920936093e-07,
      "loss": 0.0001,
      "reward": 1.1901042126119137,
      "reward_std": 0.21367743890732527,
      "rewards/equation_reward_func": 0.23177083814516664,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 150
    },
    {
      "completion_length": 319.31250762939453,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.09931506831300466,
      "kl": 0.16046142578125,
      "learning_rate": 3.862604602152464e-07,
      "loss": 0.0002,
      "reward": 1.5156250521540642,
      "reward_std": 0.17254623072221875,
      "rewards/equation_reward_func": 0.5416666828095913,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 152
    },
    {
      "completion_length": 372.6979274749756,
      "epoch": 0.08213333333333334,
      "grad_norm": 0.08423289509476255,
      "kl": 0.1458740234375,
      "learning_rate": 3.8322584365434934e-07,
      "loss": 0.0001,
      "reward": 1.4036458805203438,
      "reward_std": 0.23620562674477696,
      "rewards/equation_reward_func": 0.43229168234393,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 154
    },
    {
      "completion_length": 383.4323024749756,
      "epoch": 0.0832,
      "grad_norm": 0.11684907563681693,
      "kl": 0.13134765625,
      "learning_rate": 3.8016355973344173e-07,
      "loss": 0.0001,
      "reward": 1.3619792014360428,
      "reward_std": 0.2366077760234475,
      "rewards/equation_reward_func": 0.39062501303851604,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 156
    },
    {
      "completion_length": 421.96876335144043,
      "epoch": 0.08426666666666667,
      "grad_norm": 0.1641382478416818,
      "kl": 0.124755859375,
      "learning_rate": 3.7707424440504863e-07,
      "loss": 0.0001,
      "reward": 1.296875037252903,
      "reward_std": 0.23676540749147534,
      "rewards/equation_reward_func": 0.3333333421032876,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 158
    },
    {
      "completion_length": 365.9166793823242,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.07817184043452526,
      "kl": 0.1395263671875,
      "learning_rate": 3.739585392353787e-07,
      "loss": 0.0001,
      "reward": 1.3880208767950535,
      "reward_std": 0.19910774566233158,
      "rewards/equation_reward_func": 0.4140625100117177,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 160
    },
    {
      "completion_length": 336.37500953674316,
      "epoch": 0.0864,
      "grad_norm": 0.08666435211660832,
      "kl": 0.155364990234375,
      "learning_rate": 3.7081709127108767e-07,
      "loss": 0.0002,
      "reward": 1.5182291939854622,
      "reward_std": 0.20631011482328176,
      "rewards/equation_reward_func": 0.5312500149011612,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 162
    },
    {
      "completion_length": 360.15625953674316,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.07995177005809427,
      "kl": 0.134307861328125,
      "learning_rate": 3.6765055290490513e-07,
      "loss": 0.0001,
      "reward": 1.367187537252903,
      "reward_std": 0.23405077820643783,
      "rewards/equation_reward_func": 0.3776041774544865,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 164
    },
    {
      "completion_length": 316.8932418823242,
      "epoch": 0.08853333333333334,
      "grad_norm": 0.15304933704233734,
      "kl": 0.155242919921875,
      "learning_rate": 3.644595817401501e-07,
      "loss": 0.0002,
      "reward": 1.533854216337204,
      "reward_std": 0.25381680950522423,
      "rewards/equation_reward_func": 0.5442708469927311,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 166
    },
    {
      "completion_length": 388.1224002838135,
      "epoch": 0.0896,
      "grad_norm": 0.08385643880585995,
      "kl": 0.141143798828125,
      "learning_rate": 3.6124484045416483e-07,
      "loss": 0.0001,
      "reward": 1.3723958767950535,
      "reward_std": 0.24668778479099274,
      "rewards/equation_reward_func": 0.40885417466051877,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 168
    },
    {
      "completion_length": 351.93750858306885,
      "epoch": 0.09066666666666667,
      "grad_norm": 0.07697038389931672,
      "kl": 0.14190673828125,
      "learning_rate": 3.580069966606949e-07,
      "loss": 0.0001,
      "reward": 1.4479166939854622,
      "reward_std": 0.2045988291501999,
      "rewards/equation_reward_func": 0.4557291753590107,
      "rewards/format_reward_func": 0.9921875037252903,
      "step": 170
    },
    {
      "completion_length": 334.4765729904175,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.0735056203449723,
      "kl": 0.144317626953125,
      "learning_rate": 3.547467227712444e-07,
      "loss": 0.0001,
      "reward": 1.4973958656191826,
      "reward_std": 0.19876712281256914,
      "rewards/equation_reward_func": 0.5130208474583924,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 172
    },
    {
      "completion_length": 327.29948711395264,
      "epoch": 0.0928,
      "grad_norm": 0.12548498564602414,
      "kl": 0.1490478515625,
      "learning_rate": 3.5146469585543386e-07,
      "loss": 0.0001,
      "reward": 1.4947917088866234,
      "reward_std": 0.22728270338848233,
      "rewards/equation_reward_func": 0.5078125204890966,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 174
    },
    {
      "completion_length": 398.49220085144043,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.08293639399157755,
      "kl": 0.149139404296875,
      "learning_rate": 3.481615975003922e-07,
      "loss": 0.0001,
      "reward": 1.3489583730697632,
      "reward_std": 0.20350094605237246,
      "rewards/equation_reward_func": 0.3645833428017795,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 176
    },
    {
      "completion_length": 324.3489713668823,
      "epoch": 0.09493333333333333,
      "grad_norm": 0.09550896801162942,
      "kl": 0.155364990234375,
      "learning_rate": 3.448381136692089e-07,
      "loss": 0.0002,
      "reward": 1.4583333730697632,
      "reward_std": 0.1914014257490635,
      "rewards/equation_reward_func": 0.4661458458285779,
      "rewards/format_reward_func": 0.9921875037252903,
      "step": 178
    },
    {
      "completion_length": 363.1067838668823,
      "epoch": 0.096,
      "grad_norm": 0.06570582721768603,
      "kl": 0.131988525390625,
      "learning_rate": 3.4149493455847897e-07,
      "loss": 0.0001,
      "reward": 1.4010417014360428,
      "reward_std": 0.15116061177104712,
      "rewards/equation_reward_func": 0.41927084419876337,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 180
    },
    {
      "completion_length": 350.994797706604,
      "epoch": 0.09706666666666666,
      "grad_norm": 0.08114840800090561,
      "kl": 0.165863037109375,
      "learning_rate": 3.3813275445496766e-07,
      "loss": 0.0002,
      "reward": 1.4375000447034836,
      "reward_std": 0.2402082341723144,
      "rewards/equation_reward_func": 0.45572917768731713,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 182
    },
    {
      "completion_length": 341.2786521911621,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.09604192137372529,
      "kl": 0.140167236328125,
      "learning_rate": 3.347522715914262e-07,
      "loss": 0.0001,
      "reward": 1.4739583805203438,
      "reward_std": 0.2517684092745185,
      "rewards/equation_reward_func": 0.48958334303461015,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 184
    },
    {
      "completion_length": 334.8333444595337,
      "epoch": 0.0992,
      "grad_norm": 0.06092803443628096,
      "kl": 0.15545654296875,
      "learning_rate": 3.313541880015877e-07,
      "loss": 0.0002,
      "reward": 1.4869792014360428,
      "reward_std": 0.17553172213956714,
      "rewards/equation_reward_func": 0.49218751210719347,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 186
    },
    {
      "completion_length": 296.9895906448364,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.08698608388272158,
      "kl": 0.15386962890625,
      "learning_rate": 3.279392093743747e-07,
      "loss": 0.0002,
      "reward": 1.484375037252903,
      "reward_std": 0.18876954959705472,
      "rewards/equation_reward_func": 0.5130208488553762,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 188
    },
    {
      "completion_length": 369.6171989440918,
      "epoch": 0.10133333333333333,
      "grad_norm": 0.07100886524919588,
      "kl": 0.1356201171875,
      "learning_rate": 3.245080449073459e-07,
      "loss": 0.0001,
      "reward": 1.3932291939854622,
      "reward_std": 0.22231243178248405,
      "rewards/equation_reward_func": 0.40885417722165585,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 190
    },
    {
      "completion_length": 340.66928005218506,
      "epoch": 0.1024,
      "grad_norm": 0.06938271896378453,
      "kl": 0.164703369140625,
      "learning_rate": 3.210614071594162e-07,
      "loss": 0.0002,
      "reward": 1.4322917088866234,
      "reward_std": 0.21930959541350603,
      "rewards/equation_reward_func": 0.44791667885147035,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 192
    },
    {
      "completion_length": 366.10157108306885,
      "epoch": 0.10346666666666667,
      "grad_norm": 0.0640261989037358,
      "kl": 0.1395263671875,
      "learning_rate": 3.1760001190287695e-07,
      "loss": 0.0001,
      "reward": 1.3567708730697632,
      "reward_std": 0.14457962242886424,
      "rewards/equation_reward_func": 0.3671875111758709,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 194
    },
    {
      "completion_length": 335.52344608306885,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.07693618401025876,
      "kl": 0.152862548828125,
      "learning_rate": 3.141245779747502e-07,
      "loss": 0.0002,
      "reward": 1.3723958805203438,
      "reward_std": 0.19283229811117053,
      "rewards/equation_reward_func": 0.3828125139698386,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 196
    },
    {
      "completion_length": 317.8932409286499,
      "epoch": 0.1056,
      "grad_norm": 0.08196224183163763,
      "kl": 0.158447265625,
      "learning_rate": 3.106358271275056e-07,
      "loss": 0.0002,
      "reward": 1.4947917126119137,
      "reward_std": 0.19350500591099262,
      "rewards/equation_reward_func": 0.5104166797827929,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 198
    },
    {
      "completion_length": 317.8619861602783,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.09403807304980984,
      "kl": 0.15289306640625,
      "learning_rate": 3.0713448387917227e-07,
      "loss": 0.0002,
      "reward": 1.4609375223517418,
      "reward_std": 0.18717782059684396,
      "rewards/equation_reward_func": 0.46354167303070426,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 200
    },
    {
      "completion_length": 349.03907012939453,
      "epoch": 0.10773333333333333,
      "grad_norm": 0.0618629735347176,
      "kl": 0.14788818359375,
      "learning_rate": 3.0362127536287636e-07,
      "loss": 0.0001,
      "reward": 1.3906250447034836,
      "reward_std": 0.20409536687657237,
      "rewards/equation_reward_func": 0.4088541797827929,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 202
    },
    {
      "completion_length": 280.3645906448364,
      "epoch": 0.1088,
      "grad_norm": 0.06963964620642316,
      "kl": 0.168182373046875,
      "learning_rate": 3.0009693117583523e-07,
      "loss": 0.0002,
      "reward": 1.5182292088866234,
      "reward_std": 0.12625272339209914,
      "rewards/equation_reward_func": 0.5234375149011612,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 204
    },
    {
      "completion_length": 307.3724031448364,
      "epoch": 0.10986666666666667,
      "grad_norm": 0.07039292185299532,
      "kl": 0.175628662109375,
      "learning_rate": 2.965621832278401e-07,
      "loss": 0.0002,
      "reward": 1.486979205161333,
      "reward_std": 0.15833014715462923,
      "rewards/equation_reward_func": 0.49739584419876337,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 206
    },
    {
      "completion_length": 341.1041774749756,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.06143705194536295,
      "kl": 0.154327392578125,
      "learning_rate": 2.9301776558925875e-07,
      "loss": 0.0002,
      "reward": 1.4218750447034836,
      "reward_std": 0.15025723353028297,
      "rewards/equation_reward_func": 0.4375000144354999,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 208
    },
    {
      "completion_length": 334.9270935058594,
      "epoch": 0.112,
      "grad_norm": 0.0858887137019435,
      "kl": 0.15777587890625,
      "learning_rate": 2.894644143385885e-07,
      "loss": 0.0002,
      "reward": 1.4739583730697632,
      "reward_std": 0.24965603277087212,
      "rewards/equation_reward_func": 0.4895833469927311,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 210
    },
    {
      "completion_length": 322.8906316757202,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.0904494730503501,
      "kl": 0.16424560546875,
      "learning_rate": 2.859028674095937e-07,
      "loss": 0.0002,
      "reward": 1.4479167088866234,
      "reward_std": 0.17492430424317718,
      "rewards/equation_reward_func": 0.4635416748933494,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 212
    },
    {
      "completion_length": 296.30469512939453,
      "epoch": 0.11413333333333334,
      "grad_norm": 0.08632111011343167,
      "kl": 0.1639404296875,
      "learning_rate": 2.823338644380566e-07,
      "loss": 0.0002,
      "reward": 1.5156250521540642,
      "reward_std": 0.14193214289844036,
      "rewards/equation_reward_func": 0.5338541865348816,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 214
    },
    {
      "completion_length": 306.0208387374878,
      "epoch": 0.1152,
      "grad_norm": 0.08745683085319692,
      "kl": 0.159393310546875,
      "learning_rate": 2.7875814660817504e-07,
      "loss": 0.0002,
      "reward": 1.5364583805203438,
      "reward_std": 0.22032672306522727,
      "rewards/equation_reward_func": 0.557291679084301,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 216
    },
    {
      "completion_length": 351.677095413208,
      "epoch": 0.11626666666666667,
      "grad_norm": 0.0850761586470191,
      "kl": 0.1666259765625,
      "learning_rate": 2.751764564986396e-07,
      "loss": 0.0002,
      "reward": 1.3958333656191826,
      "reward_std": 0.28290195716544986,
      "rewards/equation_reward_func": 0.4062500118743628,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 218
    },
    {
      "completion_length": 308.17448806762695,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.07684504845856611,
      "kl": 0.173187255859375,
      "learning_rate": 2.715895379284194e-07,
      "loss": 0.0002,
      "reward": 1.5104166977107525,
      "reward_std": 0.18647652165964246,
      "rewards/equation_reward_func": 0.5390625081490725,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 220
    },
    {
      "completion_length": 328.5859479904175,
      "epoch": 0.1184,
      "grad_norm": 0.11203271187490091,
      "kl": 0.16510009765625,
      "learning_rate": 2.6799813580229174e-07,
      "loss": 0.0002,
      "reward": 1.4557292014360428,
      "reward_std": 0.2715247953310609,
      "rewards/equation_reward_func": 0.4895833469927311,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 222
    },
    {
      "completion_length": 388.0573043823242,
      "epoch": 0.11946666666666667,
      "grad_norm": 0.06225243124003096,
      "kl": 0.14501953125,
      "learning_rate": 2.6440299595614606e-07,
      "loss": 0.0001,
      "reward": 1.3333333730697632,
      "reward_std": 0.22618821496143937,
      "rewards/equation_reward_func": 0.35937500931322575,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 224
    },
    {
      "completion_length": 303.80209159851074,
      "epoch": 0.12053333333333334,
      "grad_norm": 0.08582274740530478,
      "kl": 0.18316650390625,
      "learning_rate": 2.6080486500209347e-07,
      "loss": 0.0002,
      "reward": 1.5052083618938923,
      "reward_std": 0.1736113135702908,
      "rewards/equation_reward_func": 0.5234375176951289,
      "rewards/format_reward_func": 0.9817708395421505,
      "step": 226
    },
    {
      "completion_length": 366.50261306762695,
      "epoch": 0.1216,
      "grad_norm": 0.08747456062041993,
      "kl": 0.17315673828125,
      "learning_rate": 2.572044901734166e-07,
      "loss": 0.0002,
      "reward": 1.3489583805203438,
      "reward_std": 0.2056382312439382,
      "rewards/equation_reward_func": 0.3880208423361182,
      "rewards/format_reward_func": 0.9609375260770321,
      "step": 228
    },
    {
      "completion_length": 335.09636306762695,
      "epoch": 0.12266666666666666,
      "grad_norm": 0.34141856790128705,
      "kl": 0.56512451171875,
      "learning_rate": 2.536026191693893e-07,
      "loss": 0.0006,
      "reward": 1.3645833805203438,
      "reward_std": 0.15194532042369246,
      "rewards/equation_reward_func": 0.3906250118743628,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 230
    },
    {
      "completion_length": 338.17969703674316,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.06484061977942189,
      "kl": 0.17218017578125,
      "learning_rate": 2.5e-07,
      "loss": 0.0002,
      "reward": 1.4270833693444729,
      "reward_std": 0.15773996990174055,
      "rewards/equation_reward_func": 0.4479166741948575,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 232
    },
    {
      "completion_length": 369.4713649749756,
      "epoch": 0.1248,
      "grad_norm": 0.06533716217495784,
      "kl": 0.172607421875,
      "learning_rate": 2.4639738083061073e-07,
      "loss": 0.0002,
      "reward": 1.3229167014360428,
      "reward_std": 0.1891809026710689,
      "rewards/equation_reward_func": 0.3619791816454381,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 234
    },
    {
      "completion_length": 326.32552909851074,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.11951442156862668,
      "kl": 0.16986083984375,
      "learning_rate": 2.4279550982658345e-07,
      "loss": 0.0002,
      "reward": 1.4348958805203438,
      "reward_std": 0.20815222803503275,
      "rewards/equation_reward_func": 0.45833334419876337,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 236
    },
    {
      "completion_length": 288.45052909851074,
      "epoch": 0.12693333333333334,
      "grad_norm": 0.08691549270327721,
      "kl": 0.19256591796875,
      "learning_rate": 2.3919513499790646e-07,
      "loss": 0.0002,
      "reward": 1.5260417088866234,
      "reward_std": 0.1767690358683467,
      "rewards/equation_reward_func": 0.5520833525806665,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 238
    },
    {
      "completion_length": 286.3697991371155,
      "epoch": 0.128,
      "grad_norm": 0.07559248926700043,
      "kl": 0.19488525390625,
      "learning_rate": 2.3559700404385394e-07,
      "loss": 0.0002,
      "reward": 1.528645858168602,
      "reward_std": 0.22208327893167734,
      "rewards/equation_reward_func": 0.5520833469927311,
      "rewards/format_reward_func": 0.9765625037252903,
      "step": 240
    },
    {
      "completion_length": 340.2682399749756,
      "epoch": 0.12906666666666666,
      "grad_norm": 0.10530634140083081,
      "kl": 0.189453125,
      "learning_rate": 2.3200186419770823e-07,
      "loss": 0.0002,
      "reward": 1.3984375149011612,
      "reward_std": 0.23324821423739195,
      "rewards/equation_reward_func": 0.4348958395421505,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 242
    },
    {
      "completion_length": 362.66927909851074,
      "epoch": 0.13013333333333332,
      "grad_norm": 1.3010084453655495,
      "kl": 0.183837890625,
      "learning_rate": 2.284104620715807e-07,
      "loss": 0.0002,
      "reward": 1.2968750298023224,
      "reward_std": 0.2373127401806414,
      "rewards/equation_reward_func": 0.3515625090803951,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 244
    },
    {
      "completion_length": 340.72917652130127,
      "epoch": 0.1312,
      "grad_norm": 0.06959692685962483,
      "kl": 0.18695068359375,
      "learning_rate": 2.2482354350136043e-07,
      "loss": 0.0002,
      "reward": 1.354166716337204,
      "reward_std": 0.19205195363610983,
      "rewards/equation_reward_func": 0.3802083437331021,
      "rewards/format_reward_func": 0.9739583544433117,
      "step": 246
    },
    {
      "completion_length": 340.98438453674316,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.0765593787053769,
      "kl": 0.202880859375,
      "learning_rate": 2.2124185339182496e-07,
      "loss": 0.0002,
      "reward": 1.3203125447034836,
      "reward_std": 0.20337719656527042,
      "rewards/equation_reward_func": 0.36979167396202683,
      "rewards/format_reward_func": 0.950520858168602,
      "step": 248
    },
    {
      "completion_length": 324.2135548591614,
      "epoch": 0.13333333333333333,
      "grad_norm": 0.06964123521038948,
      "kl": 0.191162109375,
      "learning_rate": 2.1766613556194344e-07,
      "loss": 0.0002,
      "reward": 1.4114583618938923,
      "reward_std": 0.1772644561715424,
      "rewards/equation_reward_func": 0.44010417722165585,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 250
    },
    {
      "completion_length": 345.8020935058594,
      "epoch": 0.1344,
      "grad_norm": 0.07570707850632788,
      "kl": 0.1973876953125,
      "learning_rate": 2.1409713259040628e-07,
      "loss": 0.0002,
      "reward": 1.3802083656191826,
      "reward_std": 0.29142854968085885,
      "rewards/equation_reward_func": 0.4427083432674408,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 252
    },
    {
      "completion_length": 320.135422706604,
      "epoch": 0.13546666666666668,
      "grad_norm": 0.09393829326895221,
      "kl": 0.2281494140625,
      "learning_rate": 2.105355856614115e-07,
      "loss": 0.0002,
      "reward": 1.3932292126119137,
      "reward_std": 0.26339731831103563,
      "rewards/equation_reward_func": 0.4479166797827929,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 254
    },
    {
      "completion_length": 296.93751096725464,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.10209820757740125,
      "kl": 0.21038818359375,
      "learning_rate": 2.069822344107413e-07,
      "loss": 0.0002,
      "reward": 1.458333384245634,
      "reward_std": 0.2565371445380151,
      "rewards/equation_reward_func": 0.518229179084301,
      "rewards/format_reward_func": 0.9401041902601719,
      "step": 256
    },
    {
      "completion_length": 349.19271659851074,
      "epoch": 0.1376,
      "grad_norm": 0.08766590131943175,
      "kl": 0.20196533203125,
      "learning_rate": 2.034378167721599e-07,
      "loss": 0.0002,
      "reward": 1.3046875409781933,
      "reward_std": 0.26743903663009405,
      "rewards/equation_reward_func": 0.369791679084301,
      "rewards/format_reward_func": 0.9348958544433117,
      "step": 258
    },
    {
      "completion_length": 313.64844703674316,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.06846377481636261,
      "kl": 0.21148681640625,
      "learning_rate": 1.9990306882416485e-07,
      "loss": 0.0002,
      "reward": 1.4401041865348816,
      "reward_std": 0.18869125936180353,
      "rewards/equation_reward_func": 0.47916667885147035,
      "rewards/format_reward_func": 0.9609375260770321,
      "step": 260
    },
    {
      "completion_length": 364.0937614440918,
      "epoch": 0.13973333333333332,
      "grad_norm": 0.09910364824859053,
      "kl": 0.19482421875,
      "learning_rate": 1.9637872463712362e-07,
      "loss": 0.0002,
      "reward": 1.3229167088866234,
      "reward_std": 0.28169540874660015,
      "rewards/equation_reward_func": 0.40364584513008595,
      "rewards/format_reward_func": 0.9192708507180214,
      "step": 262
    },
    {
      "completion_length": 299.85417461395264,
      "epoch": 0.1408,
      "grad_norm": 0.08446825243053117,
      "kl": 0.21746826171875,
      "learning_rate": 1.9286551612082773e-07,
      "loss": 0.0002,
      "reward": 1.4270833618938923,
      "reward_std": 0.254016081802547,
      "rewards/equation_reward_func": 0.4921875118743628,
      "rewards/format_reward_func": 0.9348958469927311,
      "step": 264
    },
    {
      "completion_length": 310.96875762939453,
      "epoch": 0.14186666666666667,
      "grad_norm": 0.08911856620012026,
      "kl": 0.20257568359375,
      "learning_rate": 1.8936417287249446e-07,
      "loss": 0.0002,
      "reward": 1.4843750521540642,
      "reward_std": 0.2710961364209652,
      "rewards/equation_reward_func": 0.5364583488553762,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 266
    },
    {
      "completion_length": 370.25782585144043,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.09626183381346134,
      "kl": 0.20556640625,
      "learning_rate": 1.8587542202524985e-07,
      "loss": 0.0002,
      "reward": 1.2786458618938923,
      "reward_std": 0.3155774394981563,
      "rewards/equation_reward_func": 0.3567708437331021,
      "rewards/format_reward_func": 0.9218750186264515,
      "step": 268
    },
    {
      "completion_length": 332.5781316757202,
      "epoch": 0.144,
      "grad_norm": 0.08565923133416391,
      "kl": 0.19805908203125,
      "learning_rate": 1.82399988097123e-07,
      "loss": 0.0002,
      "reward": 1.4322917088866234,
      "reward_std": 0.2691522967070341,
      "rewards/equation_reward_func": 0.4713541753590107,
      "rewards/format_reward_func": 0.9609375111758709,
      "step": 270
    },
    {
      "completion_length": 319.74740505218506,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.08562597807288291,
      "kl": 0.20941162109375,
      "learning_rate": 1.7893859284058378e-07,
      "loss": 0.0002,
      "reward": 1.437500026077032,
      "reward_std": 0.15510809421539307,
      "rewards/equation_reward_func": 0.48177084047347307,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 272
    },
    {
      "completion_length": 330.8229260444641,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.09727298428810927,
      "kl": 0.21234130859375,
      "learning_rate": 1.7549195509265407e-07,
      "loss": 0.0002,
      "reward": 1.3671875335276127,
      "reward_std": 0.2629071534611285,
      "rewards/equation_reward_func": 0.4348958428017795,
      "rewards/format_reward_func": 0.932291679084301,
      "step": 274
    },
    {
      "completion_length": 326.29167795181274,
      "epoch": 0.1472,
      "grad_norm": 0.08109494925679199,
      "kl": 652.1973876953125,
      "learning_rate": 1.7206079062562536e-07,
      "loss": 0.6501,
      "reward": 1.388020858168602,
      "reward_std": 0.2561843590810895,
      "rewards/equation_reward_func": 0.45052084513008595,
      "rewards/format_reward_func": 0.9375000223517418,
      "step": 276
    },
    {
      "completion_length": 272.9349036216736,
      "epoch": 0.14826666666666666,
      "grad_norm": 0.13757973385264483,
      "kl": 0.20831298828125,
      "learning_rate": 1.6864581199841226e-07,
      "loss": 0.0002,
      "reward": 1.4973958730697632,
      "reward_std": 0.25729979015886784,
      "rewards/equation_reward_func": 0.5338541837409139,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 278
    },
    {
      "completion_length": 315.5833430290222,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.08374109208415109,
      "kl": 0.2430419921875,
      "learning_rate": 1.6524772840857388e-07,
      "loss": 0.0002,
      "reward": 1.4166667014360428,
      "reward_std": 0.29498640540987253,
      "rewards/equation_reward_func": 0.4869791779201478,
      "rewards/format_reward_func": 0.9296875149011612,
      "step": 280
    },
    {
      "completion_length": 315.5911560058594,
      "epoch": 0.1504,
      "grad_norm": 0.10137195504364034,
      "kl": 0.20843505859375,
      "learning_rate": 1.6186724554503237e-07,
      "loss": 0.0002,
      "reward": 1.4010417088866234,
      "reward_std": 0.28228656528517604,
      "rewards/equation_reward_func": 0.46354168001562357,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 282
    },
    {
      "completion_length": 290.7968854904175,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.09905811066649715,
      "kl": 0.2109375,
      "learning_rate": 1.5850506544152103e-07,
      "loss": 0.0002,
      "reward": 1.4635417014360428,
      "reward_std": 0.22374329762533307,
      "rewards/equation_reward_func": 0.5000000149011612,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 284
    },
    {
      "completion_length": 320.25261402130127,
      "epoch": 0.15253333333333333,
      "grad_norm": 0.09912686512752485,
      "kl": 0.2152099609375,
      "learning_rate": 1.5516188633079107e-07,
      "loss": 0.0002,
      "reward": 1.3958333767950535,
      "reward_std": 0.2674556761048734,
      "rewards/equation_reward_func": 0.4531250149011612,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 286
    },
    {
      "completion_length": 344.84896755218506,
      "epoch": 0.1536,
      "grad_norm": 0.10445550891293186,
      "kl": 0.220947265625,
      "learning_rate": 1.5183840249960784e-07,
      "loss": 0.0002,
      "reward": 1.2552083767950535,
      "reward_std": 0.3165745525620878,
      "rewards/equation_reward_func": 0.32291667559184134,
      "rewards/format_reward_func": 0.9322916865348816,
      "step": 288
    },
    {
      "completion_length": 356.31511402130127,
      "epoch": 0.15466666666666667,
      "grad_norm": 2.7822894146719332,
      "kl": 4.81689453125,
      "learning_rate": 1.4853530414456612e-07,
      "loss": 0.0048,
      "reward": 1.322916705161333,
      "reward_std": 0.2766005671583116,
      "rewards/equation_reward_func": 0.38541667675599456,
      "rewards/format_reward_func": 0.9375000260770321,
      "step": 290
    },
    {
      "completion_length": 333.315110206604,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.06794837428942249,
      "kl": 0.20697021484375,
      "learning_rate": 1.4525327722875568e-07,
      "loss": 0.0002,
      "reward": 1.3854167126119137,
      "reward_std": 0.23598269233480096,
      "rewards/equation_reward_func": 0.4531250102445483,
      "rewards/format_reward_func": 0.9322916865348816,
      "step": 292
    },
    {
      "completion_length": 345.68230056762695,
      "epoch": 0.1568,
      "grad_norm": 0.7164246784066609,
      "kl": 0.811279296875,
      "learning_rate": 1.4199300333930515e-07,
      "loss": 0.0008,
      "reward": 1.4036458805203438,
      "reward_std": 0.2555072852410376,
      "rewards/equation_reward_func": 0.4661458469927311,
      "rewards/format_reward_func": 0.9375000260770321,
      "step": 294
    },
    {
      "completion_length": 347.70834255218506,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.10342795992816751,
      "kl": 0.212646484375,
      "learning_rate": 1.3875515954583523e-07,
      "loss": 0.0002,
      "reward": 1.338541705161333,
      "reward_std": 0.23221872048452497,
      "rewards/equation_reward_func": 0.40364584419876337,
      "rewards/format_reward_func": 0.934895858168602,
      "step": 296
    },
    {
      "completion_length": 364.0052127838135,
      "epoch": 0.15893333333333334,
      "grad_norm": 0.08244692798356867,
      "kl": 0.215087890625,
      "learning_rate": 1.3554041825985e-07,
      "loss": 0.0002,
      "reward": 1.338541705161333,
      "reward_std": 0.2457712898030877,
      "rewards/equation_reward_func": 0.4062500100117177,
      "rewards/format_reward_func": 0.9322916865348816,
      "step": 298
    },
    {
      "completion_length": 331.8255319595337,
      "epoch": 0.16,
      "grad_norm": 0.09987532231840503,
      "kl": 0.21429443359375,
      "learning_rate": 1.323494470950949e-07,
      "loss": 0.0002,
      "reward": 1.346354216337204,
      "reward_std": 0.2734911320731044,
      "rewards/equation_reward_func": 0.419270841171965,
      "rewards/format_reward_func": 0.9270833507180214,
      "step": 300
    },
    {
      "completion_length": 307.08594703674316,
      "epoch": 0.16106666666666666,
      "grad_norm": 0.1125414635061611,
      "kl": 0.23858642578125,
      "learning_rate": 1.2918290872891236e-07,
      "loss": 0.0002,
      "reward": 1.4453125521540642,
      "reward_std": 0.21242208778858185,
      "rewards/equation_reward_func": 0.5000000167638063,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 302
    },
    {
      "completion_length": 328.62761211395264,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.09030164033344525,
      "kl": 0.2169189453125,
      "learning_rate": 1.260414607646213e-07,
      "loss": 0.0002,
      "reward": 1.3593750223517418,
      "reward_std": 0.23800909472629428,
      "rewards/equation_reward_func": 0.4322916760575026,
      "rewards/format_reward_func": 0.9270833469927311,
      "step": 304
    },
    {
      "completion_length": 327.34115505218506,
      "epoch": 0.1632,
      "grad_norm": 0.084605395523196,
      "kl": 0.2135009765625,
      "learning_rate": 1.2292575559495143e-07,
      "loss": 0.0002,
      "reward": 1.4036458805203438,
      "reward_std": 0.22087143966928124,
      "rewards/equation_reward_func": 0.46875000884756446,
      "rewards/format_reward_func": 0.9348958469927311,
      "step": 306
    },
    {
      "completion_length": 351.6718807220459,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.07930872672272365,
      "kl": 0.20831298828125,
      "learning_rate": 1.1983644026655835e-07,
      "loss": 0.0002,
      "reward": 1.2734375484287739,
      "reward_std": 0.2579080620780587,
      "rewards/equation_reward_func": 0.3489583432674408,
      "rewards/format_reward_func": 0.9244791939854622,
      "step": 308
    },
    {
      "completion_length": 319.75261306762695,
      "epoch": 0.16533333333333333,
      "grad_norm": 0.08243203802642829,
      "kl": 0.60150146484375,
      "learning_rate": 1.1677415634565066e-07,
      "loss": 0.0006,
      "reward": 1.3984375298023224,
      "reward_std": 0.2197747863829136,
      "rewards/equation_reward_func": 0.4765625111758709,
      "rewards/format_reward_func": 0.9218750223517418,
      "step": 310
    },
    {
      "completion_length": 323.96094512939453,
      "epoch": 0.1664,
      "grad_norm": 0.11257163282471587,
      "kl": 0.22088623046875,
      "learning_rate": 1.1373953978475353e-07,
      "loss": 0.0002,
      "reward": 1.4427083805203438,
      "reward_std": 0.24971541296690702,
      "rewards/equation_reward_func": 0.4921875118743628,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 312
    },
    {
      "completion_length": 318.0625071525574,
      "epoch": 0.16746666666666668,
      "grad_norm": 0.13460921785647972,
      "kl": 0.208740234375,
      "learning_rate": 1.1073322079063913e-07,
      "loss": 0.0002,
      "reward": 1.408854205161333,
      "reward_std": 0.24213178781792521,
      "rewards/equation_reward_func": 0.4635416781529784,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 314
    },
    {
      "completion_length": 319.69532012939453,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.06764575402513974,
      "kl": 0.19781494140625,
      "learning_rate": 1.0775582369344946e-07,
      "loss": 0.0002,
      "reward": 1.3697917275130749,
      "reward_std": 0.19613029761239886,
      "rewards/equation_reward_func": 0.41145834838971496,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 316
    },
    {
      "completion_length": 326.3984441757202,
      "epoch": 0.1696,
      "grad_norm": 0.09588186769190044,
      "kl": 0.20263671875,
      "learning_rate": 1.0480796681704077e-07,
      "loss": 0.0002,
      "reward": 1.3697916939854622,
      "reward_std": 0.2743687550537288,
      "rewards/equation_reward_func": 0.4218750102445483,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 318
    },
    {
      "completion_length": 259.299485206604,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.11223918390782062,
      "kl": 0.21942138671875,
      "learning_rate": 1.018902623505741e-07,
      "loss": 0.0002,
      "reward": 1.6145833730697632,
      "reward_std": 0.23305691732093692,
      "rewards/equation_reward_func": 0.6510416809469461,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 320
    },
    {
      "completion_length": 284.2135534286499,
      "epoch": 0.17173333333333332,
      "grad_norm": 0.08458399836341833,
      "kl": 0.20318603515625,
      "learning_rate": 9.900331622138063e-08,
      "loss": 0.0002,
      "reward": 1.5026042014360428,
      "reward_std": 0.19581556040793657,
      "rewards/equation_reward_func": 0.5494791748933494,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 322
    },
    {
      "completion_length": 292.7161521911621,
      "epoch": 0.1728,
      "grad_norm": 0.08816589688045734,
      "kl": 0.2020263671875,
      "learning_rate": 9.614772796912681e-08,
      "loss": 0.0002,
      "reward": 1.5286458656191826,
      "reward_std": 0.19881888292729855,
      "rewards/equation_reward_func": 0.580729179084301,
      "rewards/format_reward_func": 0.9479166753590107,
      "step": 324
    },
    {
      "completion_length": 293.8697986602783,
      "epoch": 0.17386666666666667,
      "grad_norm": 0.10028674350174616,
      "kl": 0.199462890625,
      "learning_rate": 9.332409062130686e-08,
      "loss": 0.0002,
      "reward": 1.3984375558793545,
      "reward_std": 0.19026571605354548,
      "rewards/equation_reward_func": 0.42708334350027144,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 326
    },
    {
      "completion_length": 279.73438262939453,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.1228325067153052,
      "kl": 0.2115478515625,
      "learning_rate": 9.053299057008699e-08,
      "loss": 0.0002,
      "reward": 1.4843750447034836,
      "reward_std": 0.2166175846941769,
      "rewards/equation_reward_func": 0.5312500144354999,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 328
    },
    {
      "completion_length": 340.1328229904175,
      "epoch": 0.176,
      "grad_norm": 0.060609232054425886,
      "kl": 0.201171875,
      "learning_rate": 8.777500745052743e-08,
      "loss": 0.0002,
      "reward": 1.2682292014360428,
      "reward_std": 0.18949319841340184,
      "rewards/equation_reward_func": 0.3098958395421505,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 330
    },
    {
      "completion_length": 296.57292556762695,
      "epoch": 0.17706666666666668,
      "grad_norm": 0.724192154071362,
      "kl": 2.50811767578125,
      "learning_rate": 8.505071402020892e-08,
      "loss": 0.0025,
      "reward": 1.473958384245634,
      "reward_std": 0.2744275564327836,
      "rewards/equation_reward_func": 0.5182291809469461,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 332
    },
    {
      "completion_length": 314.63542556762695,
      "epoch": 0.17813333333333334,
      "grad_norm": 0.07744283195119217,
      "kl": 0.21636962890625,
      "learning_rate": 8.236067604028562e-08,
      "loss": 0.0002,
      "reward": 1.385416716337204,
      "reward_std": 0.24492605542764068,
      "rewards/equation_reward_func": 0.4453125149011612,
      "rewards/format_reward_func": 0.9401041939854622,
      "step": 334
    },
    {
      "completion_length": 283.77605056762695,
      "epoch": 0.1792,
      "grad_norm": 0.09014034608663256,
      "kl": 0.21405029296875,
      "learning_rate": 7.970545215799327e-08,
      "loss": 0.0002,
      "reward": 1.4609375409781933,
      "reward_std": 0.2396250907331705,
      "rewards/equation_reward_func": 0.5208333446644247,
      "rewards/format_reward_func": 0.9401041902601719,
      "step": 336
    },
    {
      "completion_length": 326.2578239440918,
      "epoch": 0.18026666666666666,
      "grad_norm": 0.08812085562876558,
      "kl": 0.203857421875,
      "learning_rate": 7.708559379063204e-08,
      "loss": 0.0002,
      "reward": 1.377604216337204,
      "reward_std": 0.2484951955266297,
      "rewards/equation_reward_func": 0.45572918234393,
      "rewards/format_reward_func": 0.9218750186264515,
      "step": 338
    },
    {
      "completion_length": 290.2395906448364,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.12096313121694759,
      "kl": 0.96221923828125,
      "learning_rate": 7.45016450110534e-08,
      "loss": 0.001,
      "reward": 1.4557292126119137,
      "reward_std": 0.17546244524419308,
      "rewards/equation_reward_func": 0.5078125062864274,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 340
    },
    {
      "completion_length": 277.1953191757202,
      "epoch": 0.1824,
      "grad_norm": 0.10079414583051031,
      "kl": 0.2108154296875,
      "learning_rate": 7.195414243467029e-08,
      "loss": 0.0002,
      "reward": 1.5416667088866234,
      "reward_std": 0.21704197628423572,
      "rewards/equation_reward_func": 0.5781250186264515,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 342
    },
    {
      "completion_length": 323.82813262939453,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.09237849268052975,
      "kl": 0.20794677734375,
      "learning_rate": 6.944361510801763e-08,
      "loss": 0.0002,
      "reward": 1.367187537252903,
      "reward_std": 0.27509107533842325,
      "rewards/equation_reward_func": 0.43229168257676065,
      "rewards/format_reward_func": 0.9348958544433117,
      "step": 344
    },
    {
      "completion_length": 325.0052185058594,
      "epoch": 0.18453333333333333,
      "grad_norm": 0.11420262344533028,
      "kl": 0.19384765625,
      "learning_rate": 6.697058439888283e-08,
      "loss": 0.0002,
      "reward": 1.3619792088866234,
      "reward_std": 0.22549985628575087,
      "rewards/equation_reward_func": 0.40625001210719347,
      "rewards/format_reward_func": 0.955729179084301,
      "step": 346
    },
    {
      "completion_length": 281.9427156448364,
      "epoch": 0.1856,
      "grad_norm": 0.06077341740641138,
      "kl": 0.2020263671875,
      "learning_rate": 6.453556388803288e-08,
      "loss": 0.0002,
      "reward": 1.505208358168602,
      "reward_std": 0.2182565974071622,
      "rewards/equation_reward_func": 0.5546875118743628,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 348
    },
    {
      "completion_length": 277.53646516799927,
      "epoch": 0.18666666666666668,
      "grad_norm": 0.07280350150631636,
      "kl": 0.217041015625,
      "learning_rate": 6.213905926255697e-08,
      "loss": 0.0002,
      "reward": 1.531250026077032,
      "reward_std": 0.20672065950930119,
      "rewards/equation_reward_func": 0.5677083488553762,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 350
    },
    {
      "completion_length": 312.04167127609253,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.1087978791138447,
      "kl": 0.2020263671875,
      "learning_rate": 5.978156821084987e-08,
      "loss": 0.0002,
      "reward": 1.4140625447034836,
      "reward_std": 0.28301327722147107,
      "rewards/equation_reward_func": 0.466145847691223,
      "rewards/format_reward_func": 0.947916679084301,
      "step": 352
    },
    {
      "completion_length": 317.19793128967285,
      "epoch": 0.1888,
      "grad_norm": 0.07417618395975106,
      "kl": 0.2265625,
      "learning_rate": 5.7463580319254853e-08,
      "loss": 0.0002,
      "reward": 1.3203125298023224,
      "reward_std": 0.26750255608931184,
      "rewards/equation_reward_func": 0.38281251210719347,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 354
    },
    {
      "completion_length": 293.9921941757202,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.07468400690550668,
      "kl": 0.20404052734375,
      "learning_rate": 5.518557697039081e-08,
      "loss": 0.0002,
      "reward": 1.4609375447034836,
      "reward_std": 0.1771034407429397,
      "rewards/equation_reward_func": 0.5026041809469461,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 356
    },
    {
      "completion_length": 280.30730056762695,
      "epoch": 0.19093333333333334,
      "grad_norm": 0.10781941802386472,
      "kl": 0.21038818359375,
      "learning_rate": 5.294803124318145e-08,
      "loss": 0.0002,
      "reward": 1.4713541939854622,
      "reward_std": 0.25494540203362703,
      "rewards/equation_reward_func": 0.5130208432674408,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 358
    },
    {
      "completion_length": 282.8307361602783,
      "epoch": 0.192,
      "grad_norm": 0.07731622217050184,
      "kl": 0.234375,
      "learning_rate": 5.07514078146106e-08,
      "loss": 0.0002,
      "reward": 1.455729190260172,
      "reward_std": 0.1868862328119576,
      "rewards/equation_reward_func": 0.5052083507180214,
      "rewards/format_reward_func": 0.950520858168602,
      "step": 360
    },
    {
      "completion_length": 282.46094703674316,
      "epoch": 0.19306666666666666,
      "grad_norm": 0.1170891309726923,
      "kl": 0.21356201171875,
      "learning_rate": 4.859616286322094e-08,
      "loss": 0.0002,
      "reward": 1.4296875558793545,
      "reward_std": 0.25303210131824017,
      "rewards/equation_reward_func": 0.47916667722165585,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 362
    },
    {
      "completion_length": 292.97917556762695,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.13632866604679422,
      "kl": 0.21417236328125,
      "learning_rate": 4.648274397437829e-08,
      "loss": 0.0002,
      "reward": 1.4192708656191826,
      "reward_std": 0.2526449547149241,
      "rewards/equation_reward_func": 0.4765625155996531,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 364
    },
    {
      "completion_length": 255.17970085144043,
      "epoch": 0.1952,
      "grad_norm": 0.09399601024713211,
      "kl": 0.19903564453125,
      "learning_rate": 4.4411590047320617e-08,
      "loss": 0.0002,
      "reward": 1.6093750186264515,
      "reward_std": 0.17325606709346175,
      "rewards/equation_reward_func": 0.632812513737008,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 366
    },
    {
      "completion_length": 305.31511306762695,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.09591492950535363,
      "kl": 0.20355224609375,
      "learning_rate": 4.2383131204010494e-08,
      "loss": 0.0002,
      "reward": 1.3203125335276127,
      "reward_std": 0.27175394957885146,
      "rewards/equation_reward_func": 0.3723958432674408,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 368
    },
    {
      "completion_length": 274.8046979904175,
      "epoch": 0.19733333333333333,
      "grad_norm": 0.07448380610650404,
      "kl": 0.22686767578125,
      "learning_rate": 4.039778869981064e-08,
      "loss": 0.0002,
      "reward": 1.408854205161333,
      "reward_std": 0.1444006934762001,
      "rewards/equation_reward_func": 0.4427083423361182,
      "rewards/format_reward_func": 0.9661458432674408,
      "step": 370
    },
    {
      "completion_length": 295.8073043823242,
      "epoch": 0.1984,
      "grad_norm": 0.08046455188518299,
      "kl": 0.20953369140625,
      "learning_rate": 3.845597483600049e-08,
      "loss": 0.0002,
      "reward": 1.3984375335276127,
      "reward_std": 0.2535385196097195,
      "rewards/equation_reward_func": 0.4505208421032876,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 372
    },
    {
      "completion_length": 281.377610206604,
      "epoch": 0.19946666666666665,
      "grad_norm": 0.1231734086836591,
      "kl": 0.2137451171875,
      "learning_rate": 3.655809287415284e-08,
      "loss": 0.0002,
      "reward": 1.4322916939854622,
      "reward_std": 0.2503278241492808,
      "rewards/equation_reward_func": 0.48958334792405367,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 374
    },
    {
      "completion_length": 254.83854961395264,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.08564861254444861,
      "kl": 0.20831298828125,
      "learning_rate": 3.4704536952387285e-08,
      "loss": 0.0002,
      "reward": 1.5598958618938923,
      "reward_std": 0.18554534064605832,
      "rewards/equation_reward_func": 0.5937500074505806,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 376
    },
    {
      "completion_length": 322.82292556762695,
      "epoch": 0.2016,
      "grad_norm": 0.13685877620537756,
      "kl": 0.1998291015625,
      "learning_rate": 3.2895692003518575e-08,
      "loss": 0.0002,
      "reward": 1.3541667088866234,
      "reward_std": 0.24434165563434362,
      "rewards/equation_reward_func": 0.40625001350417733,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 378
    },
    {
      "completion_length": 267.2968854904175,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.10129770118592953,
      "kl": 0.20574951171875,
      "learning_rate": 3.113193367511635e-08,
      "loss": 0.0002,
      "reward": 1.5260417014360428,
      "reward_std": 0.1942262640222907,
      "rewards/equation_reward_func": 0.570312513038516,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 380
    },
    {
      "completion_length": 299.36719369888306,
      "epoch": 0.20373333333333332,
      "grad_norm": 0.14156955034396038,
      "kl": 0.23870849609375,
      "learning_rate": 2.9413628251493934e-08,
      "loss": 0.0002,
      "reward": 1.4817708730697632,
      "reward_std": 0.2450651628896594,
      "rewards/equation_reward_func": 0.5286458479240537,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 382
    },
    {
      "completion_length": 305.47396659851074,
      "epoch": 0.2048,
      "grad_norm": 0.1385039320261207,
      "kl": 0.21966552734375,
      "learning_rate": 2.774113257764066e-08,
      "loss": 0.0002,
      "reward": 1.3437500298023224,
      "reward_std": 0.14954586559906602,
      "rewards/equation_reward_func": 0.39322917722165585,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 384
    },
    {
      "completion_length": 283.994797706604,
      "epoch": 0.20586666666666667,
      "grad_norm": 0.08046202628694818,
      "kl": 0.20733642578125,
      "learning_rate": 2.611479398511518e-08,
      "loss": 0.0002,
      "reward": 1.429687526077032,
      "reward_std": 0.20751357544213533,
      "rewards/equation_reward_func": 0.4739583395421505,
      "rewards/format_reward_func": 0.955729179084301,
      "step": 386
    },
    {
      "completion_length": 299.1197986602783,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.09252027726323123,
      "kl": 0.20440673828125,
      "learning_rate": 2.4534950219914057e-08,
      "loss": 0.0002,
      "reward": 1.4296875447034836,
      "reward_std": 0.28528111474588513,
      "rewards/equation_reward_func": 0.48958334559574723,
      "rewards/format_reward_func": 0.9401041828095913,
      "step": 388
    },
    {
      "completion_length": 304.86459159851074,
      "epoch": 0.208,
      "grad_norm": 0.1258367102892705,
      "kl": 0.209228515625,
      "learning_rate": 2.300192937233128e-08,
      "loss": 0.0002,
      "reward": 1.4218750298023224,
      "reward_std": 0.25267517613247037,
      "rewards/equation_reward_func": 0.4843750155996531,
      "rewards/format_reward_func": 0.9375000149011612,
      "step": 390
    },
    {
      "completion_length": 258.6849012374878,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.0745793407744394,
      "kl": 0.21881103515625,
      "learning_rate": 2.1516049808822935e-08,
      "loss": 0.0002,
      "reward": 1.5729167088866234,
      "reward_std": 0.18264030944555998,
      "rewards/equation_reward_func": 0.6250000149011612,
      "rewards/format_reward_func": 0.947916679084301,
      "step": 392
    },
    {
      "completion_length": 315.58073902130127,
      "epoch": 0.21013333333333334,
      "grad_norm": 0.08894574865715046,
      "kl": 0.212890625,
      "learning_rate": 2.007762010589098e-08,
      "loss": 0.0002,
      "reward": 1.375000026077032,
      "reward_std": 0.20314896712079644,
      "rewards/equation_reward_func": 0.4244791748933494,
      "rewards/format_reward_func": 0.9505208469927311,
      "step": 394
    },
    {
      "completion_length": 285.76042652130127,
      "epoch": 0.2112,
      "grad_norm": 0.0978921012598675,
      "kl": 0.21484375,
      "learning_rate": 1.8686938986000627e-08,
      "loss": 0.0002,
      "reward": 1.4947917312383652,
      "reward_std": 0.22043491480872035,
      "rewards/equation_reward_func": 0.5234375174622983,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 396
    },
    {
      "completion_length": 249.42448902130127,
      "epoch": 0.21226666666666666,
      "grad_norm": 0.11183349657491888,
      "kl": 0.24554443359375,
      "learning_rate": 1.734429525554365e-08,
      "loss": 0.0002,
      "reward": 1.5598958730697632,
      "reward_std": 0.22952570766210556,
      "rewards/equation_reward_func": 0.6145833507180214,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 398
    },
    {
      "completion_length": 305.55209159851074,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.1522170775819166,
      "kl": 0.2047119140625,
      "learning_rate": 1.604996774486145e-08,
      "loss": 0.0002,
      "reward": 1.440104205161333,
      "reward_std": 0.25300154415890574,
      "rewards/equation_reward_func": 0.494791679084301,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 400
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}