{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9808917197452229,
  "eval_steps": 500,
  "global_step": 132,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 341.7267630440848,
      "epoch": 0.0074309978768577496,
      "grad_norm": 0.08086328746974374,
      "learning_rate": 0.0,
      "loss": -0.0006,
      "num_tokens": 3159359.0,
      "reward": 0.5170068130606696,
      "reward_std": 0.29308582487560453,
      "rewards/acc_reward_func": 0.5170068102223533,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "epoch": 0.014861995753715499,
      "grad_norm": 0.08087477809734007,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0006,
      "step": 2
    },
    {
      "clip_ratio": 0.0001838355625901992,
      "epoch": 0.022292993630573247,
      "grad_norm": 0.07945717618543717,
      "learning_rate": 1.4285714285714285e-07,
      "loss": -0.0006,
      "step": 3
    },
    {
      "clip_ratio": 0.0001572738398127036,
      "epoch": 0.029723991507430998,
      "grad_norm": 0.0826048402223115,
      "learning_rate": 2.1428571428571426e-07,
      "loss": -0.0006,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.3208690824963,
      "epoch": 0.037154989384288746,
      "grad_norm": 0.07979158138671237,
      "learning_rate": 2.857142857142857e-07,
      "loss": -0.0028,
      "num_tokens": 6223398.0,
      "reward": 0.5000000081601597,
      "reward_std": 0.30536195990585147,
      "rewards/acc_reward_func": 0.5000000081601597,
      "step": 5
    },
    {
      "clip_ratio": 0.0001808156685867635,
      "epoch": 0.044585987261146494,
      "grad_norm": 0.08024143057101281,
      "learning_rate": 3.5714285714285716e-07,
      "loss": -0.0028,
      "step": 6
    },
    {
      "clip_ratio": 0.00017867709487161067,
      "epoch": 0.05201698513800425,
      "grad_norm": 0.08005899196723062,
      "learning_rate": 4.285714285714285e-07,
      "loss": -0.0028,
      "step": 7
    },
    {
      "clip_ratio": 0.00019149975127047148,
      "epoch": 0.059447983014861996,
      "grad_norm": 0.0796285692450267,
      "learning_rate": 5e-07,
      "loss": -0.0028,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 336.19728597005206,
      "epoch": 0.06687898089171974,
      "grad_norm": 0.0809301416262032,
      "learning_rate": 5.714285714285714e-07,
      "loss": 0.0007,
      "num_tokens": 9136668.0,
      "reward": 0.49546485855465844,
      "reward_std": 0.3260552350963865,
      "rewards/acc_reward_func": 0.49546485855465844,
      "step": 9
    },
    {
      "clip_ratio": 0.00021897819568125887,
      "epoch": 0.07430997876857749,
      "grad_norm": 0.08109016776097855,
      "learning_rate": 6.428571428571429e-07,
      "loss": 0.0007,
      "step": 10
    },
    {
      "clip_ratio": 0.00020755232045693056,
      "epoch": 0.08174097664543524,
      "grad_norm": 0.08087779783507208,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.0007,
      "step": 11
    },
    {
      "clip_ratio": 0.00024463120367034294,
      "epoch": 0.08917197452229299,
      "grad_norm": 0.08067403590429052,
      "learning_rate": 7.857142857142856e-07,
      "loss": 0.0005,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 339.7154235839844,
      "epoch": 0.09660297239915075,
      "grad_norm": 0.0793808960257417,
      "learning_rate": 8.57142857142857e-07,
      "loss": -0.0056,
      "num_tokens": 12045721.0,
      "reward": 0.5907029594693866,
      "reward_std": 0.3145202554407574,
      "rewards/acc_reward_func": 0.5907029566310701,
      "step": 13
    },
    {
      "clip_ratio": 0.00019346000789526095,
      "epoch": 0.1040339702760085,
      "grad_norm": 0.07996587273324629,
      "learning_rate": 9.285714285714285e-07,
      "loss": -0.0056,
      "step": 14
    },
    {
      "clip_ratio": 0.0002197972238978504,
      "epoch": 0.11146496815286625,
      "grad_norm": 0.07844004513920859,
      "learning_rate": 1e-06,
      "loss": -0.0057,
      "step": 15
    },
    {
      "clip_ratio": 0.0002786624475750917,
      "epoch": 0.11889596602972399,
      "grad_norm": 0.07872371102981156,
      "learning_rate": 1e-06,
      "loss": -0.0058,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 343.09864443824404,
      "epoch": 0.12632696390658174,
      "grad_norm": 0.06834766428561097,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "num_tokens": 15049348.0,
      "reward": 0.5997732579708099,
      "reward_std": 0.2505793724031675,
      "rewards/acc_reward_func": 0.5997732522941771,
      "step": 17
    },
    {
      "clip_ratio": 0.00017599576330255893,
      "epoch": 0.1337579617834395,
      "grad_norm": 0.06874704664521386,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "step": 18
    },
    {
      "clip_ratio": 0.0003469426301307976,
      "epoch": 0.14118895966029724,
      "grad_norm": 0.06727013413601377,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "step": 19
    },
    {
      "clip_ratio": 0.000489247930956827,
      "epoch": 0.14861995753715498,
      "grad_norm": 0.06670999385159201,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 350.2063555036272,
      "epoch": 0.15605095541401273,
      "grad_norm": 0.06650745108644616,
      "learning_rate": 1e-06,
      "loss": -0.0001,
      "num_tokens": 18191946.0,
      "reward": 0.6530612352348509,
      "reward_std": 0.24400150775909424,
      "rewards/acc_reward_func": 0.6530612295582181,
      "step": 21
    },
    {
      "clip_ratio": 0.00013373352599696124,
      "epoch": 0.16348195329087048,
      "grad_norm": 0.06586079666953687,
      "learning_rate": 1e-06,
      "loss": -0.0001,
      "step": 22
    },
    {
      "clip_ratio": 0.0001875036643815249,
      "epoch": 0.17091295116772823,
      "grad_norm": 0.06545454216834161,
      "learning_rate": 1e-06,
      "loss": -0.0002,
      "step": 23
    },
    {
      "clip_ratio": 0.0003731138775557546,
      "epoch": 0.17834394904458598,
      "grad_norm": 0.06428724973892952,
      "learning_rate": 1e-06,
      "loss": -0.0003,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 352.0895749046689,
      "epoch": 0.18577494692144372,
      "grad_norm": 0.06446210353107223,
      "learning_rate": 1e-06,
      "loss": -0.0037,
      "num_tokens": 21159481.0,
      "reward": 0.5736961549236661,
      "reward_std": 0.2319913577465784,
      "rewards/acc_reward_func": 0.5736961492470333,
      "step": 25
    },
    {
      "clip_ratio": 0.0001779778396580479,
      "epoch": 0.1932059447983015,
      "grad_norm": 0.06458451199765729,
      "learning_rate": 1e-06,
      "loss": -0.0038,
      "step": 26
    },
    {
      "clip_ratio": 0.00031768538021770796,
      "epoch": 0.20063694267515925,
      "grad_norm": 0.06403775486903479,
      "learning_rate": 1e-06,
      "loss": -0.0039,
      "step": 27
    },
    {
      "clip_ratio": 0.0004978579201547074,
      "epoch": 0.208067940552017,
      "grad_norm": 0.06384970547978061,
      "learning_rate": 1e-06,
      "loss": -0.004,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 353.5793733142671,
      "epoch": 0.21549893842887474,
      "grad_norm": 0.07091531570599886,
      "learning_rate": 1e-06,
      "loss": -0.0007,
      "num_tokens": 24284078.0,
      "reward": 0.5782313063031151,
      "reward_std": 0.27138930842989967,
      "rewards/acc_reward_func": 0.5782313006264823,
      "step": 29
    },
    {
      "clip_ratio": 0.00023922019664453166,
      "epoch": 0.2229299363057325,
      "grad_norm": 0.07207676160001003,
      "learning_rate": 1e-06,
      "loss": -0.0008,
      "step": 30
    },
    {
      "clip_ratio": 0.00028778909801899103,
      "epoch": 0.23036093418259024,
      "grad_norm": 0.08311384778120834,
      "learning_rate": 1e-06,
      "loss": -0.001,
      "step": 31
    },
    {
      "clip_ratio": 0.0005288766468376187,
      "epoch": 0.23779193205944799,
      "grad_norm": 0.07065550775183596,
      "learning_rate": 1e-06,
      "loss": -0.0012,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 374.4387802850632,
      "epoch": 0.24522292993630573,
      "grad_norm": 0.0689708415519406,
      "learning_rate": 1e-06,
      "loss": 0.003,
      "num_tokens": 27446489.0,
      "reward": 0.585034022728602,
      "reward_std": 0.2801268689689182,
      "rewards/acc_reward_func": 0.5850340198902857,
      "step": 33
    },
    {
      "clip_ratio": 0.00021597234375630726,
      "epoch": 0.2526539278131635,
      "grad_norm": 0.06897131457392458,
      "learning_rate": 1e-06,
      "loss": 0.0029,
      "step": 34
    },
    {
      "clip_ratio": 0.0003070946047609184,
      "epoch": 0.26008492569002123,
      "grad_norm": 0.0674766183603089,
      "learning_rate": 1e-06,
      "loss": 0.0028,
      "step": 35
    },
    {
      "clip_ratio": 0.0005742841105010095,
      "epoch": 0.267515923566879,
      "grad_norm": 0.06745873238088902,
      "learning_rate": 1e-06,
      "loss": 0.0026,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 379.7687145414807,
      "epoch": 0.2749469214437367,
      "grad_norm": 0.061592227582679786,
      "learning_rate": 1e-06,
      "loss": 0.0005,
      "num_tokens": 30450929.0,
      "reward": 0.596371889823959,
      "reward_std": 0.2418635892016547,
      "rewards/acc_reward_func": 0.5963718841473261,
      "step": 37
    },
    {
      "clip_ratio": 0.00013849885224425678,
      "epoch": 0.2823779193205945,
      "grad_norm": 0.06152688747932998,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "step": 38
    },
    {
      "clip_ratio": 0.00022306023882785148,
      "epoch": 0.2898089171974522,
      "grad_norm": 0.06152633134972845,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "step": 39
    },
    {
      "clip_ratio": 0.00031142526720595056,
      "epoch": 0.29723991507430997,
      "grad_norm": 0.0626993089929395,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 390.8605521065848,
      "epoch": 0.3046709129511677,
      "grad_norm": 0.0629758366767815,
      "learning_rate": 1e-06,
      "loss": 0.0036,
      "num_tokens": 33596816.0,
      "reward": 0.636054433527447,
      "reward_std": 0.25603189283893224,
      "rewards/acc_reward_func": 0.6360544250124976,
      "step": 41
    },
    {
      "clip_ratio": 0.00010988019805933748,
      "epoch": 0.31210191082802546,
      "grad_norm": 0.06378608109398301,
      "learning_rate": 1e-06,
      "loss": 0.0035,
      "step": 42
    },
    {
      "clip_ratio": 0.00025688905103985843,
      "epoch": 0.3195329087048832,
      "grad_norm": 0.06325047212115749,
      "learning_rate": 1e-06,
      "loss": 0.0034,
      "step": 43
    },
    {
      "clip_ratio": 0.00039352324078901715,
      "epoch": 0.32696390658174096,
      "grad_norm": 0.06227219214519032,
      "learning_rate": 1e-06,
      "loss": 0.0032,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 380.9455857049851,
      "epoch": 0.3343949044585987,
      "grad_norm": 0.06916759874959395,
      "learning_rate": 1e-06,
      "loss": 0.005,
      "num_tokens": 36816368.0,
      "reward": 0.5861678095090956,
      "reward_std": 0.2711287704961641,
      "rewards/acc_reward_func": 0.5861678066707793,
      "step": 45
    },
    {
      "clip_ratio": 0.00013988154855074493,
      "epoch": 0.34182590233545646,
      "grad_norm": 0.06865178228849304,
      "learning_rate": 1e-06,
      "loss": 0.0049,
      "step": 46
    },
    {
      "clip_ratio": 0.00035448711389021595,
      "epoch": 0.3492569002123142,
      "grad_norm": 0.0674984372138465,
      "learning_rate": 1e-06,
      "loss": 0.0047,
      "step": 47
    },
    {
      "clip_ratio": 0.000534125243402308,
      "epoch": 0.35668789808917195,
      "grad_norm": 0.06713323621764225,
      "learning_rate": 1e-06,
      "loss": 0.0045,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 366.5612284342448,
      "epoch": 0.3641188959660297,
      "grad_norm": 0.0628565554992099,
      "learning_rate": 1e-06,
      "loss": -0.0007,
      "num_tokens": 39788825.0,
      "reward": 0.6405895806494213,
      "reward_std": 0.23035428832684243,
      "rewards/acc_reward_func": 0.6405895692961556,
      "step": 49
    },
    {
      "clip_ratio": 0.00017204703319640387,
      "epoch": 0.37154989384288745,
      "grad_norm": 0.060413657444765324,
      "learning_rate": 1e-06,
      "loss": -0.0008,
      "step": 50
    },
    {
      "clip_ratio": 0.00028196911201424274,
      "epoch": 0.37898089171974525,
      "grad_norm": 0.060029825820231635,
      "learning_rate": 1e-06,
      "loss": -0.0009,
      "step": 51
    },
    {
      "clip_ratio": 0.0005388078487677765,
      "epoch": 0.386411889596603,
      "grad_norm": 0.05859097094352694,
      "learning_rate": 1e-06,
      "loss": -0.0011,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 370.1700730096726,
      "epoch": 0.39384288747346075,
      "grad_norm": 0.062489857926374105,
      "learning_rate": 1e-06,
      "loss": -0.002,
      "num_tokens": 42977507.0,
      "reward": 0.6337868599664598,
      "reward_std": 0.2415066155649367,
      "rewards/acc_reward_func": 0.6337868571281433,
      "step": 53
    },
    {
      "clip_ratio": 0.00015676757123271403,
      "epoch": 0.4012738853503185,
      "grad_norm": 0.062323008512970825,
      "learning_rate": 1e-06,
      "loss": -0.0021,
      "step": 54
    },
    {
      "clip_ratio": 0.0002369500718833435,
      "epoch": 0.40870488322717624,
      "grad_norm": 0.06229025216025021,
      "learning_rate": 1e-06,
      "loss": -0.0023,
      "step": 55
    },
    {
      "clip_ratio": 0.00043342224342354926,
      "epoch": 0.416135881104034,
      "grad_norm": 0.061363769616563854,
      "learning_rate": 1e-06,
      "loss": -0.0025,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 377.649664015997,
      "epoch": 0.42356687898089174,
      "grad_norm": 0.060693382900306965,
      "learning_rate": 1e-06,
      "loss": -0.0069,
      "num_tokens": 45992960.0,
      "reward": 0.6213152082193465,
      "reward_std": 0.19496617785521916,
      "rewards/acc_reward_func": 0.6213151997043973,
      "step": 57
    },
    {
      "clip_ratio": 0.00011746683992983197,
      "epoch": 0.4309978768577495,
      "grad_norm": 0.057906127015709595,
      "learning_rate": 1e-06,
      "loss": -0.007,
      "step": 58
    },
    {
      "clip_ratio": 0.00015394135230703147,
      "epoch": 0.43842887473460723,
      "grad_norm": 0.05770813992767556,
      "learning_rate": 1e-06,
      "loss": -0.0071,
      "step": 59
    },
    {
      "clip_ratio": 0.0002990371741318432,
      "epoch": 0.445859872611465,
      "grad_norm": 0.057169748017192656,
      "learning_rate": 1e-06,
      "loss": -0.0073,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 385.07256789434524,
      "epoch": 0.45329087048832273,
      "grad_norm": 0.06503664456534936,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "num_tokens": 48962556.0,
      "reward": 0.6394557903210322,
      "reward_std": 0.2468004703876518,
      "rewards/acc_reward_func": 0.6394557846443993,
      "step": 61
    },
    {
      "clip_ratio": 0.0001142212568083778,
      "epoch": 0.4607218683651805,
      "grad_norm": 0.07260193604541537,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "step": 62
    },
    {
      "clip_ratio": 0.0002252710536205476,
      "epoch": 0.4681528662420382,
      "grad_norm": 0.06540521622742039,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "step": 63
    },
    {
      "clip_ratio": 0.00048679712857674096,
      "epoch": 0.47558386411889597,
      "grad_norm": 0.06392595201565687,
      "learning_rate": 1e-06,
      "loss": 0.0001,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 369.952386765253,
      "epoch": 0.4830148619957537,
      "grad_norm": 0.0631263236426717,
      "learning_rate": 1e-06,
      "loss": 0.0013,
      "num_tokens": 52081410.0,
      "reward": 0.5975056723469779,
      "reward_std": 0.24312191580732664,
      "rewards/acc_reward_func": 0.5975056723469779,
      "step": 65
    },
    {
      "clip_ratio": 0.0001545405653554813,
      "epoch": 0.49044585987261147,
      "grad_norm": 0.06318597960704911,
      "learning_rate": 1e-06,
      "loss": 0.0012,
      "step": 66
    },
    {
      "clip_ratio": 0.0002261997837021703,
      "epoch": 0.4978768577494692,
      "grad_norm": 0.0629106951108031,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "step": 67
    },
    {
      "clip_ratio": 0.00041743737771563855,
      "epoch": 0.505307855626327,
      "grad_norm": 0.0632657220715682,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 364.53175281343005,
      "epoch": 0.5127388535031847,
      "grad_norm": 0.06136012395478247,
      "learning_rate": 1e-06,
      "loss": -0.0006,
      "num_tokens": 55109647.0,
      "reward": 0.6020408258551643,
      "reward_std": 0.2137400745635941,
      "rewards/acc_reward_func": 0.6020408201785314,
      "step": 69
    },
    {
      "clip_ratio": 0.00010235635393958849,
      "epoch": 0.5201698513800425,
      "grad_norm": 0.06243874921426781,
      "learning_rate": 1e-06,
      "loss": -0.0007,
      "step": 70
    },
    {
      "clip_ratio": 0.00016079150147907924,
      "epoch": 0.5276008492569002,
      "grad_norm": 0.06340802042111414,
      "learning_rate": 1e-06,
      "loss": -0.0008,
      "step": 71
    },
    {
      "clip_ratio": 0.0002864431884344889,
      "epoch": 0.535031847133758,
      "grad_norm": 0.060899777207080064,
      "learning_rate": 1e-06,
      "loss": -0.0011,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 372.2551051548549,
      "epoch": 0.5424628450106157,
      "grad_norm": 0.06521913849953302,
      "learning_rate": 1e-06,
      "loss": 0.0014,
      "num_tokens": 58184698.0,
      "reward": 0.5793650916644505,
      "reward_std": 0.23960900750188602,
      "rewards/acc_reward_func": 0.5793650888261341,
      "step": 73
    },
    {
      "clip_ratio": 0.00015685576034316217,
      "epoch": 0.5498938428874734,
      "grad_norm": 0.0650407164221492,
      "learning_rate": 1e-06,
      "loss": 0.0013,
      "step": 74
    },
    {
      "clip_ratio": 0.0002204552958054202,
      "epoch": 0.5573248407643312,
      "grad_norm": 0.06495775379721254,
      "learning_rate": 1e-06,
      "loss": 0.0011,
      "step": 75
    },
    {
      "clip_ratio": 0.0004874699469447868,
      "epoch": 0.564755838641189,
      "grad_norm": 0.06475707873807389,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 383.8061305454799,
      "epoch": 0.5721868365180467,
      "grad_norm": 0.06331746668722547,
      "learning_rate": 1e-06,
      "loss": 0.0031,
      "num_tokens": 61126891.0,
      "reward": 0.6473922999132247,
      "reward_std": 0.2356209299039273,
      "rewards/acc_reward_func": 0.6473922942365918,
      "step": 77
    },
    {
      "clip_ratio": 0.00015557293539002006,
      "epoch": 0.5796178343949044,
      "grad_norm": 0.06358059963704424,
      "learning_rate": 1e-06,
      "loss": 0.003,
      "step": 78
    },
    {
      "clip_ratio": 0.0002651862686477779,
      "epoch": 0.5870488322717622,
      "grad_norm": 0.06242581352023444,
      "learning_rate": 1e-06,
      "loss": 0.0028,
      "step": 79
    },
    {
      "clip_ratio": 0.0004893070893428687,
      "epoch": 0.5944798301486199,
      "grad_norm": 0.0609064485591393,
      "learning_rate": 1e-06,
      "loss": 0.0026,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 350.4886692592076,
      "epoch": 0.6019108280254777,
      "grad_norm": 0.06314986431723936,
      "learning_rate": 1e-06,
      "loss": 0.0029,
      "num_tokens": 64278222.0,
      "reward": 0.6802721172571182,
      "reward_std": 0.22581943603498594,
      "rewards/acc_reward_func": 0.6802721059038526,
      "step": 81
    },
    {
      "clip_ratio": 0.0001433593038416889,
      "epoch": 0.6093418259023354,
      "grad_norm": 0.06290949597812766,
      "learning_rate": 1e-06,
      "loss": 0.0028,
      "step": 82
    },
    {
      "clip_ratio": 0.0002783082295320041,
      "epoch": 0.6167728237791932,
      "grad_norm": 0.06287850239253823,
      "learning_rate": 1e-06,
      "loss": 0.0026,
      "step": 83
    },
    {
      "clip_ratio": 0.0005158363062600117,
      "epoch": 0.6242038216560509,
      "grad_norm": 0.06275939403952029,
      "learning_rate": 1e-06,
      "loss": 0.0024,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 359.168942406064,
      "epoch": 0.6316348195329087,
      "grad_norm": 0.07881948540281701,
      "learning_rate": 1e-06,
      "loss": -0.0073,
      "num_tokens": 67289105.0,
      "reward": 0.6598639573369708,
      "reward_std": 0.24390507791013943,
      "rewards/acc_reward_func": 0.659863951660338,
      "step": 85
    },
    {
      "clip_ratio": 0.00013236766764228896,
      "epoch": 0.6390658174097664,
      "grad_norm": 0.0653084592600677,
      "learning_rate": 1e-06,
      "loss": -0.0074,
      "step": 86
    },
    {
      "clip_ratio": 0.00017881063554557928,
      "epoch": 0.6464968152866242,
      "grad_norm": 0.06524851177812671,
      "learning_rate": 1e-06,
      "loss": -0.0077,
      "step": 87
    },
    {
      "clip_ratio": 0.0002778974991261272,
      "epoch": 0.6539278131634819,
      "grad_norm": 0.06451115804588689,
      "learning_rate": 1e-06,
      "loss": -0.0079,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 357.64512852260043,
      "epoch": 0.6613588110403397,
      "grad_norm": 0.06630152283194597,
      "learning_rate": 1e-06,
      "loss": -0.0,
      "num_tokens": 70432648.0,
      "reward": 0.6712018279802232,
      "reward_std": 0.24537649963583266,
      "rewards/acc_reward_func": 0.6712018194652739,
      "step": 89
    },
    {
      "clip_ratio": 0.0001349434973062238,
      "epoch": 0.6687898089171974,
      "grad_norm": 0.06559670392653466,
      "learning_rate": 1e-06,
      "loss": -0.0001,
      "step": 90
    },
    {
      "clip_ratio": 0.0002572811865233927,
      "epoch": 0.6762208067940552,
      "grad_norm": 0.06520494024347283,
      "learning_rate": 1e-06,
      "loss": -0.0004,
      "step": 91
    },
    {
      "clip_ratio": 0.0007034737652810734,
      "epoch": 0.6836518046709129,
      "grad_norm": 0.06315058857400183,
      "learning_rate": 1e-06,
      "loss": -0.0006,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 353.4535217285156,
      "epoch": 0.6910828025477707,
      "grad_norm": 0.05958332825443139,
      "learning_rate": 1e-06,
      "loss": 0.0038,
      "num_tokens": 73419374.0,
      "reward": 0.6451247235139211,
      "reward_std": 0.20120730180115926,
      "rewards/acc_reward_func": 0.6451247235139211,
      "step": 93
    },
    {
      "clip_ratio": 0.00012057262124255344,
      "epoch": 0.6985138004246284,
      "grad_norm": 0.060331181041075654,
      "learning_rate": 1e-06,
      "loss": 0.0037,
      "step": 94
    },
    {
      "clip_ratio": 0.00019830298922551308,
      "epoch": 0.7059447983014862,
      "grad_norm": 0.059953496435375425,
      "learning_rate": 1e-06,
      "loss": 0.0035,
      "step": 95
    },
    {
      "clip_ratio": 0.0004421481661709203,
      "epoch": 0.7133757961783439,
      "grad_norm": 0.059328760204792295,
      "learning_rate": 1e-06,
      "loss": 0.0032,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 351.6281244187128,
      "epoch": 0.7208067940552016,
      "grad_norm": 0.06476127865871668,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "num_tokens": 76658560.0,
      "reward": 0.6360544292699724,
      "reward_std": 0.21411728167108127,
      "rewards/acc_reward_func": 0.6360544264316559,
      "step": 97
    },
    {
      "clip_ratio": 0.0001179320069717332,
      "epoch": 0.7282377919320594,
      "grad_norm": 0.06488361533451398,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "step": 98
    },
    {
      "clip_ratio": 0.00018475095047116547,
      "epoch": 0.7356687898089171,
      "grad_norm": 0.06472419917345307,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "step": 99
    },
    {
      "clip_ratio": 0.000287527184853042,
      "epoch": 0.7430997876857749,
      "grad_norm": 0.0639335874652328,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.3061276390439,
      "epoch": 0.7505307855626328,
      "grad_norm": 0.06715329563723746,
      "learning_rate": 1e-06,
      "loss": -0.0025,
      "num_tokens": 79817980.0,
      "reward": 0.624716560045878,
      "reward_std": 0.24032448941753023,
      "rewards/acc_reward_func": 0.6247165515309289,
      "step": 101
    },
    {
      "clip_ratio": 0.00013254733875371692,
      "epoch": 0.7579617834394905,
      "grad_norm": 0.06714369960521104,
      "learning_rate": 1e-06,
      "loss": -0.0026,
      "step": 102
    },
    {
      "clip_ratio": 0.00020292648419161283,
      "epoch": 0.7653927813163482,
      "grad_norm": 0.06738540272745804,
      "learning_rate": 1e-06,
      "loss": -0.0029,
      "step": 103
    },
    {
      "clip_ratio": 0.00046159424389424243,
      "epoch": 0.772823779193206,
      "grad_norm": 0.06683308540913054,
      "learning_rate": 1e-06,
      "loss": -0.0032,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.98299589611236,
      "epoch": 0.7802547770700637,
      "grad_norm": 0.06213033347416386,
      "learning_rate": 1e-06,
      "loss": 0.0061,
      "num_tokens": 82797505.0,
      "reward": 0.7063492139180502,
      "reward_std": 0.18706076290635837,
      "rewards/acc_reward_func": 0.7063492082414173,
      "step": 105
    },
    {
      "clip_ratio": 9.041415337595113e-05,
      "epoch": 0.7876857749469215,
      "grad_norm": 0.06177639834948758,
      "learning_rate": 1e-06,
      "loss": 0.0061,
      "step": 106
    },
    {
      "clip_ratio": 0.0001607231185646794,
      "epoch": 0.7951167728237792,
      "grad_norm": 0.06109816753456417,
      "learning_rate": 1e-06,
      "loss": 0.0059,
      "step": 107
    },
    {
      "clip_ratio": 0.0002942822845527969,
      "epoch": 0.802547770700637,
      "grad_norm": 0.05990890192520818,
      "learning_rate": 1e-06,
      "loss": 0.0056,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 331.62925792875745,
      "epoch": 0.8099787685774947,
      "grad_norm": 0.0705484973603166,
      "learning_rate": 1e-06,
      "loss": -0.0006,
      "num_tokens": 85780678.0,
      "reward": 0.6575963837759835,
      "reward_std": 0.22970955535059884,
      "rewards/acc_reward_func": 0.6575963809376671,
      "step": 109
    },
    {
      "clip_ratio": 0.00013627276140531258,
      "epoch": 0.8174097664543525,
      "grad_norm": 0.07009285424608136,
      "learning_rate": 1e-06,
      "loss": -0.0007,
      "step": 110
    },
    {
      "clip_ratio": 0.0002046585505013354,
      "epoch": 0.8248407643312102,
      "grad_norm": 0.0700883025067068,
      "learning_rate": 1e-06,
      "loss": -0.001,
      "step": 111
    },
    {
      "clip_ratio": 0.0005059108720853969,
      "epoch": 0.832271762208068,
      "grad_norm": 0.06888801243376852,
      "learning_rate": 1e-06,
      "loss": -0.0013,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 322.01588076636904,
      "epoch": 0.8397027600849257,
      "grad_norm": 0.06536166070879086,
      "learning_rate": 1e-06,
      "loss": 0.0023,
      "num_tokens": 88940424.0,
      "reward": 0.742630402247111,
      "reward_std": 0.18359530522000223,
      "rewards/acc_reward_func": 0.7426303908938453,
      "step": 113
    },
    {
      "clip_ratio": 0.0001239214994018853,
      "epoch": 0.8471337579617835,
      "grad_norm": 0.06527450868376303,
      "learning_rate": 1e-06,
      "loss": 0.0022,
      "step": 114
    },
    {
      "clip_ratio": 0.00020105073227092536,
      "epoch": 0.8545647558386412,
      "grad_norm": 0.06460779643206258,
      "learning_rate": 1e-06,
      "loss": 0.002,
      "step": 115
    },
    {
      "clip_ratio": 0.00045002524607947895,
      "epoch": 0.861995753715499,
      "grad_norm": 0.06392727815491918,
      "learning_rate": 1e-06,
      "loss": 0.0017,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 311.251706077939,
      "epoch": 0.8694267515923567,
      "grad_norm": 0.07018788315451174,
      "learning_rate": 1e-06,
      "loss": -0.0027,
      "num_tokens": 91757766.0,
      "reward": 0.6519274456160409,
      "reward_std": 0.20784893667414076,
      "rewards/acc_reward_func": 0.6519274342627752,
      "step": 117
    },
    {
      "clip_ratio": 0.000157653278403727,
      "epoch": 0.8768577494692145,
      "grad_norm": 0.0690532669370979,
      "learning_rate": 1e-06,
      "loss": -0.0028,
      "step": 118
    },
    {
      "clip_ratio": 0.00033690567943267524,
      "epoch": 0.8842887473460722,
      "grad_norm": 0.06835107752563815,
      "learning_rate": 1e-06,
      "loss": -0.0031,
      "step": 119
    },
    {
      "clip_ratio": 0.000745917763283831,
      "epoch": 0.89171974522293,
      "grad_norm": 0.06660313306459224,
      "learning_rate": 1e-06,
      "loss": -0.0034,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 324.4727943057106,
      "epoch": 0.8991507430997877,
      "grad_norm": 0.06970834157267367,
      "learning_rate": 1e-06,
      "loss": -0.0038,
      "num_tokens": 94686801.0,
      "reward": 0.6337868613856179,
      "reward_std": 0.19493895627203442,
      "rewards/acc_reward_func": 0.6337868528706687,
      "step": 121
    },
    {
      "clip_ratio": 0.00011362713467817576,
      "epoch": 0.9065817409766455,
      "grad_norm": 0.06999963278535759,
      "learning_rate": 1e-06,
      "loss": -0.004,
      "step": 122
    },
    {
      "clip_ratio": 0.00021538332068649608,
      "epoch": 0.9140127388535032,
      "grad_norm": 0.06972968350918567,
      "learning_rate": 1e-06,
      "loss": -0.0043,
      "step": 123
    },
    {
      "clip_ratio": 0.00044560064478511255,
      "epoch": 0.921443736730361,
      "grad_norm": 0.06882950080744477,
      "learning_rate": 1e-06,
      "loss": -0.0046,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 325.98413231259303,
      "epoch": 0.9288747346072187,
      "grad_norm": 0.07128773867618274,
      "learning_rate": 1e-06,
      "loss": 0.001,
      "num_tokens": 97858699.0,
      "reward": 0.6553288144724709,
      "reward_std": 0.20388960643183618,
      "rewards/acc_reward_func": 0.6553288059575217,
      "step": 125
    },
    {
      "clip_ratio": 0.0001051227392010679,
      "epoch": 0.9363057324840764,
      "grad_norm": 0.07130043132021105,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "step": 126
    },
    {
      "clip_ratio": 0.00021789341259309802,
      "epoch": 0.9437367303609342,
      "grad_norm": 0.08439067946992163,
      "learning_rate": 1e-06,
      "loss": 0.0006,
      "step": 127
    },
    {
      "clip_ratio": 0.00045677864164601835,
      "epoch": 0.9511677282377919,
      "grad_norm": 0.0729135580640395,
      "learning_rate": 1e-06,
      "loss": 0.0002,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 330.68141392299106,
      "epoch": 0.9585987261146497,
      "grad_norm": 0.07242470586815654,
      "learning_rate": 1e-06,
      "loss": 0.0012,
      "num_tokens": 100697360.0,
      "reward": 0.6746031840642294,
      "reward_std": 0.21919804066419601,
      "rewards/acc_reward_func": 0.6746031812259129,
      "step": 129
    },
    {
      "clip_ratio": 0.00016543883005700385,
      "epoch": 0.9660297239915074,
      "grad_norm": 0.07263658018223751,
      "learning_rate": 1e-06,
      "loss": 0.001,
      "step": 130
    },
    {
      "clip_ratio": 0.00029576754535637085,
      "epoch": 0.9734607218683652,
      "grad_norm": 0.07191823196569669,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "step": 131
    },
    {
      "clip_ratio": 0.0006197429174790159,
      "epoch": 0.9808917197452229,
      "grad_norm": 0.07137316493917,
      "learning_rate": 1e-06,
      "loss": 0.0004,
      "step": 132
    },
    {
      "epoch": 0.9808917197452229,
      "step": 132,
      "total_flos": 0.0,
      "train_loss": -0.00028816385895769406,
      "train_runtime": 58537.3897,
      "train_samples_per_second": 0.084,
      "train_steps_per_second": 0.002
    }
  ],
  "logging_steps": 1,
  "max_steps": 134,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 250,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}