{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24,
  "eval_steps": 500,
  "global_step": 450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 267.57032012939453,
      "epoch": 0.0010666666666666667,
      "grad_norm": 0.21315283923773715,
      "kl": 0.0,
      "learning_rate": 1.4285714285714285e-07,
      "loss": -0.0,
      "reward": 0.05208333465270698,
      "reward_std": 0.12187675526365638,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.05208333465270698,
      "step": 2
    },
    {
      "completion_length": 258.91667556762695,
      "epoch": 0.0021333333333333334,
      "grad_norm": 0.39624029123072774,
      "kl": 0.0005260705947875977,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.09114583535119891,
      "reward_std": 0.2148200012743473,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.08854166837409139,
      "step": 4
    },
    {
      "completion_length": 254.1770896911621,
      "epoch": 0.0032,
      "grad_norm": 0.2789454187627699,
      "kl": 0.0005211830139160156,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.08333333465270698,
      "reward_std": 0.20366852125152946,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.08072916860692203,
      "step": 6
    },
    {
      "completion_length": 266.03125858306885,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.1672287489912653,
      "kl": 0.0005216598510742188,
      "learning_rate": 5.714285714285714e-07,
      "loss": 0.0,
      "reward": 0.059895834885537624,
      "reward_std": 0.15923613868653774,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.059895834885537624,
      "step": 8
    },
    {
      "completion_length": 248.87240409851074,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.3959884726972001,
      "kl": 0.0006082057952880859,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.0,
      "reward": 0.09635416883975267,
      "reward_std": 0.21937653236091137,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.09114583488553762,
      "step": 10
    },
    {
      "completion_length": 245.3203182220459,
      "epoch": 0.0064,
      "grad_norm": 0.28583532187086624,
      "kl": 0.0007441043853759766,
      "learning_rate": 8.57142857142857e-07,
      "loss": 0.0,
      "reward": 0.0781250016298145,
      "reward_std": 0.2006211462430656,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.0781250016298145,
      "step": 12
    },
    {
      "completion_length": 232.69531726837158,
      "epoch": 0.007466666666666667,
      "grad_norm": 0.30125607821086126,
      "kl": 0.0013909339904785156,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.1015625016298145,
      "reward_std": 0.22314835852012038,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.09895833465270698,
      "step": 14
    },
    {
      "completion_length": 226.94271278381348,
      "epoch": 0.008533333333333334,
      "grad_norm": 0.3849252255439341,
      "kl": 0.00328826904296875,
      "learning_rate": 9.999480818449865e-07,
      "loss": 0.0,
      "reward": 0.14322916930541396,
      "reward_std": 0.2906528012827039,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.1380208362825215,
      "step": 16
    },
    {
      "completion_length": 185.41146278381348,
      "epoch": 0.0096,
      "grad_norm": 0.4848458043128939,
      "kl": 0.008299827575683594,
      "learning_rate": 9.997923381619255e-07,
      "loss": 0.0,
      "reward": 0.22395834000781178,
      "reward_std": 0.37577595096081495,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.21875000419095159,
      "step": 18
    },
    {
      "completion_length": 127.20573091506958,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.5597326437314926,
      "kl": 0.0244293212890625,
      "learning_rate": 9.995328012945157e-07,
      "loss": 0.0,
      "reward": 0.4505208469927311,
      "reward_std": 0.4992805514484644,
      "rewards/equation_reward_func": 0.01822916720993817,
      "rewards/format_reward_func": 0.4322916753590107,
      "step": 20
    },
    {
      "completion_length": 93.7682318687439,
      "epoch": 0.011733333333333333,
      "grad_norm": 0.5475672226425161,
      "kl": 0.048309326171875,
      "learning_rate": 9.991695251414583e-07,
      "loss": 0.0,
      "reward": 0.5755208488553762,
      "reward_std": 0.4716507475823164,
      "rewards/equation_reward_func": 0.007812500232830644,
      "rewards/format_reward_func": 0.5677083469927311,
      "step": 22
    },
    {
      "completion_length": 84.74739789962769,
      "epoch": 0.0128,
      "grad_norm": 0.6354342945220236,
      "kl": 0.0808868408203125,
      "learning_rate": 9.987025851452636e-07,
      "loss": 0.0001,
      "reward": 0.7291666865348816,
      "reward_std": 0.42415964510291815,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.7239583507180214,
      "step": 24
    },
    {
      "completion_length": 82.8072943687439,
      "epoch": 0.013866666666666666,
      "grad_norm": 0.5155554110023303,
      "kl": 0.128814697265625,
      "learning_rate": 9.981320782765846e-07,
      "loss": 0.0001,
      "reward": 0.8750000223517418,
      "reward_std": 0.270578539930284,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.8723958544433117,
      "step": 26
    },
    {
      "completion_length": 64.10937762260437,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.3480478612027264,
      "kl": 0.24591064453125,
      "learning_rate": 9.974581230140768e-07,
      "loss": 0.0002,
      "reward": 0.9713541902601719,
      "reward_std": 0.12162131909281015,
      "rewards/equation_reward_func": 0.007812500232830644,
      "rewards/format_reward_func": 0.9635416939854622,
      "step": 28
    },
    {
      "completion_length": 60.281251668930054,
      "epoch": 0.016,
      "grad_norm": 0.005562791609333587,
      "kl": 0.3465576171875,
      "learning_rate": 9.966808593197956e-07,
      "loss": 0.0003,
      "reward": 0.9869791753590107,
      "reward_std": 0.03682847833260894,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 30
    },
    {
      "completion_length": 45.51562637090683,
      "epoch": 0.017066666666666667,
      "grad_norm": 0.03097813949935321,
      "kl": 0.49072265625,
      "learning_rate": 9.958004486101293e-07,
      "loss": 0.0005,
      "reward": 0.9947916716337204,
      "reward_std": 0.014731391333043575,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 32
    },
    {
      "completion_length": 69.03385561704636,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.006598716445330455,
      "kl": 0.3668212890625,
      "learning_rate": 9.948170737222762e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 34
    },
    {
      "completion_length": 98.88541954755783,
      "epoch": 0.0192,
      "grad_norm": 0.00733244130846419,
      "kl": 0.3455810546875,
      "learning_rate": 9.937309388762758e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 36
    },
    {
      "completion_length": 79.19531428813934,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.002140189477159276,
      "kl": 0.337646484375,
      "learning_rate": 9.925422696325974e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 38
    },
    {
      "completion_length": 58.98437696695328,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.005041926134939414,
      "kl": 0.3271484375,
      "learning_rate": 9.912513128452973e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 40
    },
    {
      "completion_length": 83.58854389190674,
      "epoch": 0.0224,
      "grad_norm": 0.003080339075344411,
      "kl": 0.3336181640625,
      "learning_rate": 9.898583366107536e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 42
    },
    {
      "completion_length": 68.85677361488342,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.005847596515205166,
      "kl": 0.355712890625,
      "learning_rate": 9.88363630211991e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 44
    },
    {
      "completion_length": 81.716148853302,
      "epoch": 0.024533333333333334,
      "grad_norm": 0.006129286662886568,
      "kl": 0.351806640625,
      "learning_rate": 9.867675040586033e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 46
    },
    {
      "completion_length": 62.87239807844162,
      "epoch": 0.0256,
      "grad_norm": 0.0022686471037775217,
      "kl": 0.3529052734375,
      "learning_rate": 9.850702896222908e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 48
    },
    {
      "completion_length": 84.13802230358124,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.001575588022391806,
      "kl": 0.3443603515625,
      "learning_rate": 9.83272339368022e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 50
    },
    {
      "completion_length": 70.85937684774399,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.004893372014791935,
      "kl": 0.3424072265625,
      "learning_rate": 9.813740266808373e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 52
    },
    {
      "completion_length": 73.53125238418579,
      "epoch": 0.0288,
      "grad_norm": 0.0020037829167036995,
      "kl": 0.3336181640625,
      "learning_rate": 9.793757457883061e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 54
    },
    {
      "completion_length": 85.58073115348816,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.0052068637869011935,
      "kl": 0.3441162109375,
      "learning_rate": 9.772779116786567e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 56
    },
    {
      "completion_length": 62.17968964576721,
      "epoch": 0.030933333333333334,
      "grad_norm": 0.004204176702325925,
      "kl": 0.3392333984375,
      "learning_rate": 9.750809600145952e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 58
    },
    {
      "completion_length": 60.56250220537186,
      "epoch": 0.032,
      "grad_norm": 0.0015647291182938304,
      "kl": 0.35205078125,
      "learning_rate": 9.7278534704283e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 60
    },
    {
      "completion_length": 63.45312720537186,
      "epoch": 0.03306666666666667,
      "grad_norm": 0.017107550850573614,
      "kl": 0.3740234375,
      "learning_rate": 9.703915494993213e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 62
    },
    {
      "completion_length": 59.20312708616257,
      "epoch": 0.034133333333333335,
      "grad_norm": 0.002029246366701105,
      "kl": 0.344482421875,
      "learning_rate": 9.67900064510277e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 64
    },
    {
      "completion_length": 49.76041799783707,
      "epoch": 0.0352,
      "grad_norm": 0.1707108937849788,
      "kl": 0.3961181640625,
      "learning_rate": 9.653114094889126e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 66
    },
    {
      "completion_length": 66.42448157072067,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.004363766057327376,
      "kl": 0.3525390625,
      "learning_rate": 9.626261220279987e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 68
    },
    {
      "completion_length": 79.07031440734863,
      "epoch": 0.037333333333333336,
      "grad_norm": 0.002295566859094707,
      "kl": 0.3387451171875,
      "learning_rate": 9.598447597882179e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 70
    },
    {
      "completion_length": 85.9947943687439,
      "epoch": 0.0384,
      "grad_norm": 0.0028772515347710507,
      "kl": 0.3404541015625,
      "learning_rate": 9.56967900382354e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 72
    },
    {
      "completion_length": 118.54166984558105,
      "epoch": 0.039466666666666664,
      "grad_norm": 0.003804107622681421,
      "kl": 0.35693359375,
      "learning_rate": 9.539961412553374e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 74
    },
    {
      "completion_length": 132.24739825725555,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.0026976286800890377,
      "kl": 0.35357666015625,
      "learning_rate": 9.509300995601719e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 76
    },
    {
      "completion_length": 150.5286495089531,
      "epoch": 0.0416,
      "grad_norm": 0.0029703461859595446,
      "kl": 0.3365478515625,
      "learning_rate": 9.477704120297696e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 78
    },
    {
      "completion_length": 132.16927361488342,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.0042900161521961236,
      "kl": 0.347900390625,
      "learning_rate": 9.445177348447186e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 80
    },
    {
      "completion_length": 148.73958611488342,
      "epoch": 0.04373333333333333,
      "grad_norm": 0.009973192234001046,
      "kl": 0.3470458984375,
      "learning_rate": 9.41172743497012e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 82
    },
    {
      "completion_length": 160.5599009990692,
      "epoch": 0.0448,
      "grad_norm": 0.021317337320502894,
      "kl": 0.34527587890625,
      "learning_rate": 9.377361326497673e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 84
    },
    {
      "completion_length": 214.65365314483643,
      "epoch": 0.04586666666666667,
      "grad_norm": 0.004309834843107017,
      "kl": 0.30181884765625,
      "learning_rate": 9.342086159929629e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 86
    },
    {
      "completion_length": 297.28125953674316,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.004971116117055498,
      "kl": 0.263671875,
      "learning_rate": 9.305909260952254e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 88
    },
    {
      "completion_length": 306.1666736602783,
      "epoch": 0.048,
      "grad_norm": 0.00330346016739013,
      "kl": 0.24566650390625,
      "learning_rate": 9.268838142516943e-07,
      "loss": 0.0002,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 90
    },
    {
      "completion_length": 279.96355056762695,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.00318035450007201,
      "kl": 0.2532958984375,
      "learning_rate": 9.23088050327999e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 92
    },
    {
      "completion_length": 261.7994842529297,
      "epoch": 0.050133333333333335,
      "grad_norm": 0.0036946086220320164,
      "kl": 0.273681640625,
      "learning_rate": 9.192044226003788e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 94
    },
    {
      "completion_length": 275.30209159851074,
      "epoch": 0.0512,
      "grad_norm": 0.004650714378941982,
      "kl": 0.27423095703125,
      "learning_rate": 9.15233737591979e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 96
    },
    {
      "completion_length": 275.5442762374878,
      "epoch": 0.05226666666666667,
      "grad_norm": 0.005874752003273854,
      "kl": 0.283935546875,
      "learning_rate": 9.111768199053586e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 98
    },
    {
      "completion_length": 224.13542079925537,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.003767361371489657,
      "kl": 0.29339599609375,
      "learning_rate": 9.070345120512435e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 100
    },
    {
      "completion_length": 247.3671932220459,
      "epoch": 0.0544,
      "grad_norm": 0.21204929441134815,
      "kl": 0.5673828125,
      "learning_rate": 9.028076742735582e-07,
      "loss": 0.0006,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 102
    },
    {
      "completion_length": 213.47917222976685,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.003470673455225982,
      "kl": 0.3101806640625,
      "learning_rate": 8.984971843707787e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 104
    },
    {
      "completion_length": 175.9947967529297,
      "epoch": 0.05653333333333333,
      "grad_norm": 0.005812568326490043,
      "kl": 0.32379150390625,
      "learning_rate": 8.94103937513637e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 106
    },
    {
      "completion_length": 117.411461353302,
      "epoch": 0.0576,
      "grad_norm": 0.00354722170703675,
      "kl": 0.3551025390625,
      "learning_rate": 8.896288460592185e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 108
    },
    {
      "completion_length": 110.28125357627869,
      "epoch": 0.058666666666666666,
      "grad_norm": 0.003013321323384688,
      "kl": 0.339599609375,
      "learning_rate": 8.850728393614901e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 110
    },
    {
      "completion_length": 106.96875274181366,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.003463116039679215,
      "kl": 0.35595703125,
      "learning_rate": 8.804368635783002e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 112
    },
    {
      "completion_length": 85.82552325725555,
      "epoch": 0.0608,
      "grad_norm": 0.03677716489910903,
      "kl": 0.3817138671875,
      "learning_rate": 8.75721881474886e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 114
    },
    {
      "completion_length": 69.90885663032532,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.014502649071450845,
      "kl": 0.4462890625,
      "learning_rate": 8.709288722239342e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 116
    },
    {
      "completion_length": 60.21093875169754,
      "epoch": 0.06293333333333333,
      "grad_norm": 0.018319766426722457,
      "kl": 0.4246826171875,
      "learning_rate": 8.660588312022343e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 118
    },
    {
      "completion_length": 48.75000184774399,
      "epoch": 0.064,
      "grad_norm": 0.002819178233147015,
      "kl": 0.3572998046875,
      "learning_rate": 8.611127697839647e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 120
    },
    {
      "completion_length": 29.817708730697632,
      "epoch": 0.06506666666666666,
      "grad_norm": 0.048219241978314446,
      "kl": 0.6165771484375,
      "learning_rate": 8.560917151306592e-07,
      "loss": 0.0006,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 122
    },
    {
      "completion_length": 29.747396767139435,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.002491973715723316,
      "kl": 0.394287109375,
      "learning_rate": 8.509967099778933e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 124
    },
    {
      "completion_length": 22.74479252099991,
      "epoch": 0.0672,
      "grad_norm": 0.03159552463178033,
      "kl": 0.519287109375,
      "learning_rate": 8.458288124187358e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 126
    },
    {
      "completion_length": 18.752604603767395,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.0047626328583877,
      "kl": 0.4564208984375,
      "learning_rate": 8.405890956840135e-07,
      "loss": 0.0005,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 128
    },
    {
      "completion_length": 25.395833671092987,
      "epoch": 0.06933333333333333,
      "grad_norm": 0.007120551314141992,
      "kl": 0.4727783203125,
      "learning_rate": 8.352786479194287e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 130
    },
    {
      "completion_length": 18.677083611488342,
      "epoch": 0.0704,
      "grad_norm": 0.007749349171876843,
      "kl": 0.4617919921875,
      "learning_rate": 8.298985719595823e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 132
    },
    {
      "completion_length": 28.72395944595337,
      "epoch": 0.07146666666666666,
      "grad_norm": 0.01725236239408364,
      "kl": 0.512451171875,
      "learning_rate": 8.244499850989451e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 134
    },
    {
      "completion_length": 32.2265630364418,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.05304546123188986,
      "kl": 0.5926513671875,
      "learning_rate": 8.189340188598262e-07,
      "loss": 0.0006,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 136
    },
    {
      "completion_length": 23.41145896911621,
      "epoch": 0.0736,
      "grad_norm": 0.013190548453262672,
      "kl": 0.4632568359375,
      "learning_rate": 8.133518187573862e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 138
    },
    {
      "completion_length": 24.82552134990692,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.012046997929078053,
      "kl": 0.4591064453125,
      "learning_rate": 8.077045440617464e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 140
    },
    {
      "completion_length": 17.638021290302277,
      "epoch": 0.07573333333333333,
      "grad_norm": 0.0070439803543669265,
      "kl": 0.425048828125,
      "learning_rate": 8.019933675572388e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 142
    },
    {
      "completion_length": 19.304688096046448,
      "epoch": 0.0768,
      "grad_norm": 0.007119537656182656,
      "kl": 0.952880859375,
      "learning_rate": 7.962194752988518e-07,
      "loss": 0.001,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 144
    },
    {
      "completion_length": 23.434896171092987,
      "epoch": 0.07786666666666667,
      "grad_norm": 0.004351846417110043,
      "kl": 0.39892578125,
      "learning_rate": 7.903840663659184e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 146
    },
    {
      "completion_length": 17.36458384990692,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.014464883726819155,
      "kl": 0.427490234375,
      "learning_rate": 7.844883526131013e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 148
    },
    {
      "completion_length": 23.164062976837158,
      "epoch": 0.08,
      "grad_norm": 0.003214313828585045,
      "kl": 0.38671875,
      "learning_rate": 7.785335584187219e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 150
    },
    {
      "completion_length": 17.333333790302277,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.005708191232659279,
      "kl": 0.44482421875,
      "learning_rate": 7.725209204304928e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 152
    },
    {
      "completion_length": 18.481771171092987,
      "epoch": 0.08213333333333334,
      "grad_norm": 0.18097704136374604,
      "kl": 0.6190185546875,
      "learning_rate": 7.664516873086987e-07,
      "loss": 0.0006,
      "reward": 0.9947916716337204,
      "reward_std": 0.014731391333043575,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 154
    },
    {
      "completion_length": 20.242188155651093,
      "epoch": 0.0832,
      "grad_norm": 0.004715553205403204,
      "kl": 0.42333984375,
      "learning_rate": 7.603271194668835e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 156
    },
    {
      "completion_length": 18.898438036441803,
      "epoch": 0.08426666666666667,
      "grad_norm": 0.003966040759048578,
      "kl": 0.403076171875,
      "learning_rate": 7.541484888100973e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 158
    },
    {
      "completion_length": 16.60416716337204,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.002003261027905268,
      "kl": 0.34228515625,
      "learning_rate": 7.479170784707574e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 160
    },
    {
      "completion_length": 24.88802134990692,
      "epoch": 0.0864,
      "grad_norm": 0.009211378308009072,
      "kl": 0.384521484375,
      "learning_rate": 7.416341825421753e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 162
    },
    {
      "completion_length": 16.401042103767395,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.005184187636080286,
      "kl": 0.3958740234375,
      "learning_rate": 7.353011058098103e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 164
    },
    {
      "completion_length": 21.299479722976685,
      "epoch": 0.08853333333333334,
      "grad_norm": 0.07832411386720319,
      "kl": 0.3946533203125,
      "learning_rate": 7.289191634803002e-07,
      "loss": 0.0004,
      "reward": 0.9947916716337204,
      "reward_std": 0.014731391333043575,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 166
    },
    {
      "completion_length": 21.127604603767395,
      "epoch": 0.0896,
      "grad_norm": 0.006189211303564496,
      "kl": 0.3651123046875,
      "learning_rate": 7.224896809083297e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 168
    },
    {
      "completion_length": 18.304687917232513,
      "epoch": 0.09066666666666667,
      "grad_norm": 0.002963052569393672,
      "kl": 0.345458984375,
      "learning_rate": 7.160139933213898e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 170
    },
    {
      "completion_length": 24.544271409511566,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.006809551131670742,
      "kl": 0.376708984375,
      "learning_rate": 7.094934455424888e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 172
    },
    {
      "completion_length": 23.06510454416275,
      "epoch": 0.0928,
      "grad_norm": 0.007169853628551621,
      "kl": 0.377685546875,
      "learning_rate": 7.029293917108677e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 174
    },
    {
      "completion_length": 29.372397005558014,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.0049609216072808585,
      "kl": 0.38330078125,
      "learning_rate": 6.963231950007844e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 176
    },
    {
      "completion_length": 22.596354842185974,
      "epoch": 0.09493333333333333,
      "grad_norm": 0.0033493670757974436,
      "kl": 0.3634033203125,
      "learning_rate": 6.896762273384178e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 178
    },
    {
      "completion_length": 21.070313334465027,
      "epoch": 0.096,
      "grad_norm": 0.007358426048838388,
      "kl": 0.374267578125,
      "learning_rate": 6.829898691169579e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 180
    },
    {
      "completion_length": 25.106771528720856,
      "epoch": 0.09706666666666666,
      "grad_norm": 0.002673392643388894,
      "kl": 0.3927001953125,
      "learning_rate": 6.762655089099353e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 182
    },
    {
      "completion_length": 23.947917222976685,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.009888137802143636,
      "kl": 0.4693603515625,
      "learning_rate": 6.695045431828524e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 184
    },
    {
      "completion_length": 20.893229484558105,
      "epoch": 0.0992,
      "grad_norm": 0.0022607662791007638,
      "kl": 0.3218994140625,
      "learning_rate": 6.627083760031754e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 186
    },
    {
      "completion_length": 19.361979722976685,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.0032574710279835964,
      "kl": 0.3485107421875,
      "learning_rate": 6.558784187487494e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 188
    },
    {
      "completion_length": 20.19270896911621,
      "epoch": 0.10133333333333333,
      "grad_norm": 0.003981197482172737,
      "kl": 0.3470458984375,
      "learning_rate": 6.490160898146918e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 190
    },
    {
      "completion_length": 19.901042342185974,
      "epoch": 0.1024,
      "grad_norm": 0.002514384073907443,
      "kl": 0.3548583984375,
      "learning_rate": 6.421228143188324e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 192
    },
    {
      "completion_length": 24.843750834465027,
      "epoch": 0.10346666666666667,
      "grad_norm": 0.007842781472841063,
      "kl": 0.3717041015625,
      "learning_rate": 6.352000238057539e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 194
    },
    {
      "completion_length": 21.859375715255737,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.0060397124228908665,
      "kl": 0.357666015625,
      "learning_rate": 6.282491559495004e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 196
    },
    {
      "completion_length": 22.700521767139435,
      "epoch": 0.1056,
      "grad_norm": 0.005604073044775416,
      "kl": 0.4561767578125,
      "learning_rate": 6.212716542550112e-07,
      "loss": 0.0005,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 198
    },
    {
      "completion_length": 22.901042222976685,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.0026562431855837636,
      "kl": 0.3526611328125,
      "learning_rate": 6.142689677583445e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 200
    },
    {
      "completion_length": 24.648437798023224,
      "epoch": 0.10773333333333333,
      "grad_norm": 0.002588639622294546,
      "kl": 0.3214111328125,
      "learning_rate": 6.072425507257527e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 202
    },
    {
      "completion_length": 22.382813036441803,
      "epoch": 0.1088,
      "grad_norm": 0.00323664281124744,
      "kl": 0.36767578125,
      "learning_rate": 6.001938623516705e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 204
    },
    {
      "completion_length": 23.174479961395264,
      "epoch": 0.10986666666666667,
      "grad_norm": 0.004194373433624107,
      "kl": 0.361572265625,
      "learning_rate": 5.931243664556802e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 206
    },
    {
      "completion_length": 22.494792342185974,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.0635254172279242,
      "kl": 0.389404296875,
      "learning_rate": 5.860355311785175e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 208
    },
    {
      "completion_length": 24.593750715255737,
      "epoch": 0.112,
      "grad_norm": 0.0040160843770763,
      "kl": 0.381103515625,
      "learning_rate": 5.78928828677177e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 210
    },
    {
      "completion_length": 19.640625715255737,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.004056894485336028,
      "kl": 0.3604736328125,
      "learning_rate": 5.718057348191874e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 212
    },
    {
      "completion_length": 26.078125655651093,
      "epoch": 0.11413333333333334,
      "grad_norm": 0.001426529705853739,
      "kl": 0.3330078125,
      "learning_rate": 5.646677288761132e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 214
    },
    {
      "completion_length": 18.17447954416275,
      "epoch": 0.1152,
      "grad_norm": 0.00410254063002437,
      "kl": 0.360107421875,
      "learning_rate": 5.575162932163501e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 216
    },
    {
      "completion_length": 19.231771647930145,
      "epoch": 0.11626666666666667,
      "grad_norm": 0.004454685182350467,
      "kl": 0.3358154296875,
      "learning_rate": 5.503529129972792e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 218
    },
    {
      "completion_length": 22.81250023841858,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.0033331903371956913,
      "kl": 0.3734130859375,
      "learning_rate": 5.431790758568388e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 220
    },
    {
      "completion_length": 33.567709386348724,
      "epoch": 0.1184,
      "grad_norm": 0.15156820624659995,
      "kl": 0.3978271484375,
      "learning_rate": 5.359962716045835e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 222
    },
    {
      "completion_length": 23.721354722976685,
      "epoch": 0.11946666666666667,
      "grad_norm": 0.1302812080137746,
      "kl": 0.3497314453125,
      "learning_rate": 5.288059919122921e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 224
    },
    {
      "completion_length": 23.40104216337204,
      "epoch": 0.12053333333333334,
      "grad_norm": 0.10213989334337155,
      "kl": 0.3587646484375,
      "learning_rate": 5.216097300041869e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 226
    },
    {
      "completion_length": 28.752604722976685,
      "epoch": 0.1216,
      "grad_norm": 0.00405575423941108,
      "kl": 0.356689453125,
      "learning_rate": 5.144089803468332e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 228
    },
    {
      "completion_length": 21.56770896911621,
      "epoch": 0.12266666666666666,
      "grad_norm": 0.006688124816156981,
      "kl": 0.38134765625,
      "learning_rate": 5.072052383387786e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 230
    },
    {
      "completion_length": 30.83333396911621,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.003894854329196007,
      "kl": 0.36962890625,
      "learning_rate": 5e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 232
    },
    {
      "completion_length": 38.3385426402092,
      "epoch": 0.1248,
      "grad_norm": 0.003321341352419865,
      "kl": 0.339111328125,
      "learning_rate": 4.927947616612215e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 234
    },
    {
      "completion_length": 25.40104252099991,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.003665106813183815,
      "kl": 0.347900390625,
      "learning_rate": 4.855910196531669e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 236
    },
    {
      "completion_length": 32.97656351327896,
      "epoch": 0.12693333333333334,
      "grad_norm": 0.0029113458934489096,
      "kl": 0.3333740234375,
      "learning_rate": 4.783902699958129e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 238
    },
    {
      "completion_length": 32.789063453674316,
      "epoch": 0.128,
      "grad_norm": 0.2144030619978656,
      "kl": 0.98974609375,
      "learning_rate": 4.711940080877079e-07,
      "loss": 0.001,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 240
    },
    {
      "completion_length": 27.617188334465027,
      "epoch": 0.12906666666666666,
      "grad_norm": 0.004925288676087789,
      "kl": 0.355712890625,
      "learning_rate": 4.6400372839541647e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 242
    },
    {
      "completion_length": 27.859375417232513,
      "epoch": 0.13013333333333332,
      "grad_norm": 0.004078504522837724,
      "kl": 0.3455810546875,
      "learning_rate": 4.568209241431614e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 244
    },
    {
      "completion_length": 38.55729287862778,
      "epoch": 0.1312,
      "grad_norm": 0.003946730384617455,
      "kl": 0.3524169921875,
      "learning_rate": 4.4964708700272086e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 246
    },
    {
      "completion_length": 36.60937589406967,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.0025911319683286625,
      "kl": 0.3311767578125,
      "learning_rate": 4.424837067836499e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 248
    },
    {
      "completion_length": 34.45312637090683,
      "epoch": 0.13333333333333333,
      "grad_norm": 0.02082559848620119,
      "kl": 0.3614501953125,
      "learning_rate": 4.353322711238869e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 250
    },
    {
      "completion_length": 44.54427218437195,
      "epoch": 0.1344,
      "grad_norm": 0.004583784786433975,
      "kl": 0.3564453125,
      "learning_rate": 4.2819426518081256e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 252
    },
    {
      "completion_length": 32.47656333446503,
      "epoch": 0.13546666666666668,
      "grad_norm": 0.2516579496285295,
      "kl": 0.348388671875,
      "learning_rate": 4.21071171322823e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 254
    },
    {
      "completion_length": 33.247396886348724,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.003750844211497512,
      "kl": 0.336669921875,
      "learning_rate": 4.139644688214826e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 256
    },
    {
      "completion_length": 40.09375071525574,
      "epoch": 0.1376,
      "grad_norm": 0.0056285121180003365,
      "kl": 0.3736572265625,
      "learning_rate": 4.068756335443198e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 258
    },
    {
      "completion_length": 32.55468821525574,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.19004745287506442,
      "kl": 0.3536376953125,
      "learning_rate": 3.998061376483297e-07,
      "loss": 0.0004,
      "reward": 0.9947916716337204,
      "reward_std": 0.014731391333043575,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 260
    },
    {
      "completion_length": 41.53125083446503,
      "epoch": 0.13973333333333332,
      "grad_norm": 0.005613467195859889,
      "kl": 0.3878173828125,
      "learning_rate": 3.9275744927424723e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 262
    },
    {
      "completion_length": 41.14323008060455,
      "epoch": 0.1408,
      "grad_norm": 0.004079597325625193,
      "kl": 0.341552734375,
      "learning_rate": 3.8573103224165547e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 264
    },
    {
      "completion_length": 45.320313930511475,
      "epoch": 0.14186666666666667,
      "grad_norm": 0.004450670587730382,
      "kl": 0.3409423828125,
      "learning_rate": 3.787283457449889e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 266
    },
    {
      "completion_length": 34.77083444595337,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.005059899644254497,
      "kl": 0.3505859375,
      "learning_rate": 3.717508440504997e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 268
    },
    {
      "completion_length": 34.54948043823242,
      "epoch": 0.144,
      "grad_norm": 0.23991193127694865,
      "kl": 0.3560791015625,
      "learning_rate": 3.64799976194246e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 270
    },
    {
      "completion_length": 25.007812976837158,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.004586704746066067,
      "kl": 0.342529296875,
      "learning_rate": 3.5787718568116757e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 272
    },
    {
      "completion_length": 33.093751192092896,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.07458441934892074,
      "kl": 0.3690185546875,
      "learning_rate": 3.5098391018530813e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 274
    },
    {
      "completion_length": 26.036458909511566,
      "epoch": 0.1472,
      "grad_norm": 0.0018656461831637952,
      "kl": 0.3399658203125,
      "learning_rate": 3.4412158125125073e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 276
    },
    {
      "completion_length": 27.08854240179062,
      "epoch": 0.14826666666666666,
      "grad_norm": 0.008897084345492313,
      "kl": 0.3919677734375,
      "learning_rate": 3.372916239968245e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 278
    },
    {
      "completion_length": 29.460938274860382,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.0038377037869509346,
      "kl": 0.3658447265625,
      "learning_rate": 3.3049545681714775e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 280
    },
    {
      "completion_length": 20.976563274860382,
      "epoch": 0.1504,
      "grad_norm": 0.005935196775123734,
      "kl": 0.365234375,
      "learning_rate": 3.2373449109006474e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 282
    },
    {
      "completion_length": 28.750000536441803,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.006185835967833735,
      "kl": 0.3516845703125,
      "learning_rate": 3.1701013088304206e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 284
    },
    {
      "completion_length": 33.2838551402092,
      "epoch": 0.15253333333333333,
      "grad_norm": 0.004000349539922501,
      "kl": 0.3592529296875,
      "learning_rate": 3.1032377266158214e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 286
    },
    {
      "completion_length": 25.153646647930145,
      "epoch": 0.1536,
      "grad_norm": 0.006473305213296596,
      "kl": 0.3848876953125,
      "learning_rate": 3.036768049992157e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 288
    },
    {
      "completion_length": 29.770834028720856,
      "epoch": 0.15466666666666667,
      "grad_norm": 0.003973211467908967,
      "kl": 0.3634033203125,
      "learning_rate": 2.9707060828913224e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 290
    },
    {
      "completion_length": 22.49479216337204,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.00485962798897511,
      "kl": 0.3812255859375,
      "learning_rate": 2.9050655445751137e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 292
    },
    {
      "completion_length": 20.432292222976685,
      "epoch": 0.1568,
      "grad_norm": 0.0763280002987439,
      "kl": 0.359619140625,
      "learning_rate": 2.839860066786103e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 294
    },
    {
      "completion_length": 22.606771171092987,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.007131272098647573,
      "kl": 0.407958984375,
      "learning_rate": 2.7751031909167045e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 296
    },
    {
      "completion_length": 26.085938036441803,
      "epoch": 0.15893333333333334,
      "grad_norm": 0.006348176701408371,
      "kl": 0.374755859375,
      "learning_rate": 2.710808365197e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 298
    },
    {
      "completion_length": 28.302084028720856,
      "epoch": 0.16,
      "grad_norm": 0.00329994798018086,
      "kl": 0.357421875,
      "learning_rate": 2.646988941901898e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 300
    },
    {
      "completion_length": 24.33333396911621,
      "epoch": 0.16106666666666666,
      "grad_norm": 0.0041357777851463405,
      "kl": 0.3416748046875,
      "learning_rate": 2.583658174578247e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 302
    },
    {
      "completion_length": 21.351562976837158,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.004744062593804727,
      "kl": 0.3682861328125,
      "learning_rate": 2.520829215292426e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 304
    },
    {
      "completion_length": 17.56510454416275,
      "epoch": 0.1632,
      "grad_norm": 0.004071588499335152,
      "kl": 0.3665771484375,
      "learning_rate": 2.4585151118990285e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 306
    },
    {
      "completion_length": 25.093750953674316,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.003447178186189921,
      "kl": 0.3665771484375,
      "learning_rate": 2.396728805331167e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 308
    },
    {
      "completion_length": 39.61198019981384,
      "epoch": 0.16533333333333333,
      "grad_norm": 0.15778923141355317,
      "kl": 0.3824462890625,
      "learning_rate": 2.3354831269130132e-07,
      "loss": 0.0004,
      "reward": 0.9947916716337204,
      "reward_std": 0.014731391333043575,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 310
    },
    {
      "completion_length": 20.757813096046448,
      "epoch": 0.1664,
      "grad_norm": 0.006322854264740659,
      "kl": 0.3919677734375,
      "learning_rate": 2.2747907956950707e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 312
    },
    {
      "completion_length": 18.778646290302277,
      "epoch": 0.16746666666666668,
      "grad_norm": 0.004277715160452868,
      "kl": 0.3692626953125,
      "learning_rate": 2.2146644158127826e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 314
    },
    {
      "completion_length": 22.914063036441803,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.005925839314806503,
      "kl": 0.3868408203125,
      "learning_rate": 2.1551164738689892e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 316
    },
    {
      "completion_length": 19.528646647930145,
      "epoch": 0.1696,
      "grad_norm": 0.004034463075140326,
      "kl": 0.3629150390625,
      "learning_rate": 2.0961593363408154e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 318
    },
    {
      "completion_length": 26.062500655651093,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.003874147172074953,
      "kl": 0.3551025390625,
      "learning_rate": 2.037805247011482e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 320
    },
    {
      "completion_length": 22.450521290302277,
      "epoch": 0.17173333333333332,
      "grad_norm": 0.0036545323789303995,
      "kl": 0.3665771484375,
      "learning_rate": 1.9800663244276127e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 322
    },
    {
      "completion_length": 33.1458340883255,
      "epoch": 0.1728,
      "grad_norm": 0.006124961025482053,
      "kl": 0.358642578125,
      "learning_rate": 1.9229545593825363e-07,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 324
    },
    {
      "completion_length": 25.63541728258133,
      "epoch": 0.17386666666666667,
      "grad_norm": 0.004330409449251065,
      "kl": 0.38134765625,
      "learning_rate": 1.8664818124261373e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 326
    },
    {
      "completion_length": 30.52343863248825,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.004429155190479566,
      "kl": 0.3779296875,
      "learning_rate": 1.8106598114017397e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 328
    },
    {
      "completion_length": 29.343750715255737,
      "epoch": 0.176,
      "grad_norm": 0.004794753409205765,
      "kl": 0.361083984375,
      "learning_rate": 1.7555001490105486e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 330
    },
    {
      "completion_length": 28.692708909511566,
      "epoch": 0.17706666666666668,
      "grad_norm": 0.005714723884457156,
      "kl": 0.3719482421875,
      "learning_rate": 1.7010142804041783e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 332
    },
    {
      "completion_length": 24.77864646911621,
      "epoch": 0.17813333333333334,
      "grad_norm": 0.004745602217539158,
      "kl": 0.35888671875,
      "learning_rate": 1.6472135208057125e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 334
    },
    {
      "completion_length": 30.609375655651093,
      "epoch": 0.1792,
      "grad_norm": 0.004920986126883945,
      "kl": 0.380126953125,
      "learning_rate": 1.5941090431598653e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 336
    },
    {
      "completion_length": 32.19010490179062,
      "epoch": 0.18026666666666666,
      "grad_norm": 0.004022967836188304,
      "kl": 0.3558349609375,
      "learning_rate": 1.5417118758126408e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 338
    },
    {
      "completion_length": 28.093750953674316,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.003832900467363359,
      "kl": 0.3629150390625,
      "learning_rate": 1.490032900221068e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 340
    },
    {
      "completion_length": 28.463542699813843,
      "epoch": 0.1824,
      "grad_norm": 0.005487525138346315,
      "kl": 0.3526611328125,
      "learning_rate": 1.4390828486934058e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 342
    },
    {
      "completion_length": 24.825521528720856,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.0031850458946458297,
      "kl": 0.36962890625,
      "learning_rate": 1.3888723021603526e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 344
    },
    {
      "completion_length": 42.6744801402092,
      "epoch": 0.18453333333333333,
      "grad_norm": 0.0035741821948086553,
      "kl": 0.34521484375,
      "learning_rate": 1.3394116879776567e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 346
    },
    {
      "completion_length": 32.55729299783707,
      "epoch": 0.1856,
      "grad_norm": 0.0047430778934828615,
      "kl": 0.3699951171875,
      "learning_rate": 1.2907112777606576e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 348
    },
    {
      "completion_length": 32.25260508060455,
      "epoch": 0.18666666666666668,
      "grad_norm": 0.003936651676485424,
      "kl": 0.3486328125,
      "learning_rate": 1.2427811852511395e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 350
    },
    {
      "completion_length": 30.427084147930145,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.0023807576550031397,
      "kl": 0.3360595703125,
      "learning_rate": 1.1956313642169973e-07,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 352
    },
    {
      "completion_length": 34.77864670753479,
      "epoch": 0.1888,
      "grad_norm": 0.15089837821656296,
      "kl": 0.34228515625,
      "learning_rate": 1.1492716063850971e-07,
      "loss": 0.0003,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 354
    },
    {
      "completion_length": 25.59895896911621,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.004901087575209092,
      "kl": 0.3863525390625,
      "learning_rate": 1.1037115394078162e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 356
    },
    {
      "completion_length": 27.411459028720856,
      "epoch": 0.19093333333333334,
      "grad_norm": 0.2515997335919245,
      "kl": 0.415283203125,
      "learning_rate": 1.058960624863629e-07,
      "loss": 0.0004,
      "reward": 0.9947916716337204,
      "reward_std": 0.014731391333043575,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 358
    },
    {
      "completion_length": 24.864584028720856,
      "epoch": 0.192,
      "grad_norm": 0.0046814573951274975,
      "kl": 0.3677978515625,
      "learning_rate": 1.015028156292212e-07,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 360
    },
    {
      "completion_length": 26.3255215883255,
      "epoch": 0.19306666666666666,
      "grad_norm": 0.003523171914321313,
      "kl": 0.3526611328125,
      "learning_rate": 9.719232572644187e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 362
    },
    {
      "completion_length": 17.119792222976685,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.0033418329191399116,
      "kl": 0.3404541015625,
      "learning_rate": 9.296548794875658e-08,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 364
    },
    {
      "completion_length": 23.22135478258133,
      "epoch": 0.1952,
      "grad_norm": 0.003827547965209838,
      "kl": 0.3699951171875,
      "learning_rate": 8.882318009464123e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 366
    },
    {
      "completion_length": 25.372396528720856,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.005771301327474302,
      "kl": 0.390869140625,
      "learning_rate": 8.476626240802099e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 368
    },
    {
      "completion_length": 28.466146528720856,
      "epoch": 0.19733333333333333,
      "grad_norm": 0.005304014054373385,
      "kl": 0.3919677734375,
      "learning_rate": 8.079557739962128e-08,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 370
    },
    {
      "completion_length": 28.109375596046448,
      "epoch": 0.1984,
      "grad_norm": 0.008746822168976117,
      "kl": 0.381591796875,
      "learning_rate": 7.691194967200098e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 372
    },
    {
      "completion_length": 29.783854722976685,
      "epoch": 0.19946666666666665,
      "grad_norm": 0.003852702379103747,
      "kl": 0.3544921875,
      "learning_rate": 7.311618574830569e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 374
    },
    {
      "completion_length": 37.625001072883606,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.006486759121452462,
      "kl": 0.382568359375,
      "learning_rate": 6.940907390477457e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 376
    },
    {
      "completion_length": 22.640625834465027,
      "epoch": 0.2016,
      "grad_norm": 0.005855790645544301,
      "kl": 0.3614501953125,
      "learning_rate": 6.579138400703715e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 378
    },
    {
      "completion_length": 33.35937559604645,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.0045352649165850195,
      "kl": 0.3572998046875,
      "learning_rate": 6.22638673502327e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 380
    },
    {
      "completion_length": 37.419271528720856,
      "epoch": 0.20373333333333332,
      "grad_norm": 0.0055192004314618985,
      "kl": 0.3720703125,
      "learning_rate": 5.882725650298787e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 382
    },
    {
      "completion_length": 35.075521647930145,
      "epoch": 0.2048,
      "grad_norm": 0.00906841179693692,
      "kl": 0.388671875,
      "learning_rate": 5.548226515528132e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 384
    },
    {
      "completion_length": 44.68489694595337,
      "epoch": 0.20586666666666667,
      "grad_norm": 0.005997242826384812,
      "kl": 0.363037109375,
      "learning_rate": 5.222958797023036e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 386
    },
    {
      "completion_length": 36.622397005558014,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.10183351032451872,
      "kl": 0.375244140625,
      "learning_rate": 4.9069900439828115e-08,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 388
    },
    {
      "completion_length": 27.940105140209198,
      "epoch": 0.208,
      "grad_norm": 0.006355436428529637,
      "kl": 0.3638916015625,
      "learning_rate": 4.600385874466256e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 390
    },
    {
      "completion_length": 26.825521230697632,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.005474495697015851,
      "kl": 0.3861083984375,
      "learning_rate": 4.303209961764587e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 392
    },
    {
      "completion_length": 43.97395968437195,
      "epoch": 0.21013333333333334,
      "grad_norm": 0.0034248967357224185,
      "kl": 0.3330078125,
      "learning_rate": 4.015524021178196e-08,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 394
    },
    {
      "completion_length": 33.33333444595337,
      "epoch": 0.2112,
      "grad_norm": 0.006434693223212768,
      "kl": 0.3935546875,
      "learning_rate": 3.7373877972001255e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 396
    },
    {
      "completion_length": 32.841146647930145,
      "epoch": 0.21226666666666666,
      "grad_norm": 0.004929606396670415,
      "kl": 0.3687744140625,
      "learning_rate": 3.46885905110873e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 398
    },
    {
      "completion_length": 25.093750596046448,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.005791877214539501,
      "kl": 0.3695068359375,
      "learning_rate": 3.20999354897229e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 400
    },
    {
      "completion_length": 32.02604269981384,
      "epoch": 0.2144,
      "grad_norm": 0.005202107306989119,
      "kl": 0.3798828125,
      "learning_rate": 2.9608450500678562e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 402
    },
    {
      "completion_length": 27.648438274860382,
      "epoch": 0.21546666666666667,
      "grad_norm": 0.005371398662773084,
      "kl": 0.3856201171875,
      "learning_rate": 2.721465295716996e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 404
    },
    {
      "completion_length": 27.742188096046448,
      "epoch": 0.21653333333333333,
      "grad_norm": 0.006378137301419084,
      "kl": 0.3594970703125,
      "learning_rate": 2.4919039985404622e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 406
    },
    {
      "completion_length": 24.539063334465027,
      "epoch": 0.2176,
      "grad_norm": 0.006280450050690689,
      "kl": 0.3812255859375,
      "learning_rate": 2.2722088321343258e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 408
    },
    {
      "completion_length": 29.023438096046448,
      "epoch": 0.21866666666666668,
      "grad_norm": 0.007309421154996187,
      "kl": 0.390380859375,
      "learning_rate": 2.0624254211693894e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 410
    },
    {
      "completion_length": 27.057292222976685,
      "epoch": 0.21973333333333334,
      "grad_norm": 0.004269729933642793,
      "kl": 0.3533935546875,
      "learning_rate": 1.8625973319162602e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 412
    },
    {
      "completion_length": 29.026042342185974,
      "epoch": 0.2208,
      "grad_norm": 0.0076460713410529055,
      "kl": 0.4102783203125,
      "learning_rate": 1.672766063197789e-08,
      "loss": 0.0004,
      "reward": 0.9973958358168602,
      "reward_std": 0.007365695666521788,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 414
    },
    {
      "completion_length": 39.75000101327896,
      "epoch": 0.22186666666666666,
      "grad_norm": 0.00506495764972648,
      "kl": 0.365234375,
      "learning_rate": 1.492971037770924e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 416
    },
    {
      "completion_length": 24.554688572883606,
      "epoch": 0.22293333333333334,
      "grad_norm": 0.003635383901655059,
      "kl": 0.356689453125,
      "learning_rate": 1.3232495941396637e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 418
    },
    {
      "completion_length": 27.872396171092987,
      "epoch": 0.224,
      "grad_norm": 0.005264483567820246,
      "kl": 0.374755859375,
      "learning_rate": 1.1636369788008971e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 420
    },
    {
      "completion_length": 26.59114670753479,
      "epoch": 0.22506666666666666,
      "grad_norm": 0.0037470230331634537,
      "kl": 0.3553466796875,
      "learning_rate": 1.014166338924627e-08,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 422
    },
    {
      "completion_length": 26.52343863248825,
      "epoch": 0.22613333333333333,
      "grad_norm": 0.0050309925811448716,
      "kl": 0.371826171875,
      "learning_rate": 8.748687154702672e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 424
    },
    {
      "completion_length": 42.177085161209106,
      "epoch": 0.2272,
      "grad_norm": 0.004690685442083978,
      "kl": 0.3511962890625,
      "learning_rate": 7.457730367402549e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 426
    },
    {
      "completion_length": 33.70052194595337,
      "epoch": 0.22826666666666667,
      "grad_norm": 0.0037686388761955696,
      "kl": 0.35205078125,
      "learning_rate": 6.269061123724162e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 428
    },
    {
      "completion_length": 27.914063453674316,
      "epoch": 0.22933333333333333,
      "grad_norm": 0.003330651432955464,
      "kl": 0.354736328125,
      "learning_rate": 5.182926277723821e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 430
    },
    {
      "completion_length": 34.72656321525574,
      "epoch": 0.2304,
      "grad_norm": 0.004691669166337136,
      "kl": 0.3543701171875,
      "learning_rate": 4.199551389870659e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 432
    },
    {
      "completion_length": 21.466146230697632,
      "epoch": 0.23146666666666665,
      "grad_norm": 0.006847054731283313,
      "kl": 0.3829345703125,
      "learning_rate": 3.3191406802041688e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 434
    },
    {
      "completion_length": 24.791667580604553,
      "epoch": 0.23253333333333334,
      "grad_norm": 0.004002274972123922,
      "kl": 0.4224853515625,
      "learning_rate": 2.541876985923119e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 436
    },
    {
      "completion_length": 26.74479252099991,
      "epoch": 0.2336,
      "grad_norm": 0.004841844168761783,
      "kl": 0.3985595703125,
      "learning_rate": 1.867921723415433e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 438
    },
    {
      "completion_length": 30.26562613248825,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.0037147860959137894,
      "kl": 0.3653564453125,
      "learning_rate": 1.2974148547362228e-09,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 440
    },
    {
      "completion_length": 39.00781399011612,
      "epoch": 0.23573333333333332,
      "grad_norm": 0.0036142159951702123,
      "kl": 0.3477783203125,
      "learning_rate": 8.304748585417076e-10,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 442
    },
    {
      "completion_length": 22.03385466337204,
      "epoch": 0.2368,
      "grad_norm": 0.004359450323671437,
      "kl": 0.3602294921875,
      "learning_rate": 4.671987054842841e-10,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 444
    },
    {
      "completion_length": 25.75520932674408,
      "epoch": 0.23786666666666667,
      "grad_norm": 0.0055438604826414635,
      "kl": 0.366455078125,
      "learning_rate": 2.076618380744133e-10,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 446
    },
    {
      "completion_length": 29.468751192092896,
      "epoch": 0.23893333333333333,
      "grad_norm": 0.004686764994007478,
      "kl": 0.3499755859375,
      "learning_rate": 5.191815501343066e-11,
      "loss": 0.0003,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 448
    },
    {
      "completion_length": 23.330729842185974,
      "epoch": 0.24,
      "grad_norm": 0.002691527304301135,
      "kl": 0.359375,
      "learning_rate": 0.0,
      "loss": 0.0004,
      "reward": 1.0,
      "reward_std": 0.0,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 1.0,
      "step": 450
    },
    {
      "epoch": 0.24,
      "step": 450,
      "total_flos": 0.0,
      "train_loss": 0.00035568679777363087,
      "train_runtime": 8958.7654,
      "train_samples_per_second": 1.206,
      "train_steps_per_second": 0.05
    }
  ],
  "logging_steps": 2,
  "max_steps": 450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}