{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5008,
  "eval_steps": 50,
  "global_step": 313,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.022786458333333325,
      "completions/max_length": 3982.8,
      "completions/max_terminated_length": 3982.8,
      "completions/mean_length": 517.1437744140625,
      "completions/mean_terminated_length": 529.1862243652344,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 2.8,
      "epoch": 0.008,
      "grad_norm": 0.0036133762914687395,
      "learning_rate": 3.9682539682539683e-07,
      "loss": 0.0046,
      "num_tokens": 6044560.0,
      "reward": 0.40889575481414797,
      "reward_std": 0.4367077946662903,
      "rewards/accuracy_reward": 0.24609375,
      "rewards/brier_reward": -0.025070433877408503,
      "rewards/confidence_one_or_zero": 0.341015625,
      "rewards/format_reward": 0.5967447876930236,
      "rewards/mean_confidence_reward": 0.8271564602851867,
      "step": 5
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.020442708333333347,
      "completions/max_length": 3987.8,
      "completions/max_terminated_length": 3987.8,
      "completions/mean_length": 502.8280090332031,
      "completions/mean_terminated_length": 513.294482421875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 4.4,
      "epoch": 0.016,
      "grad_norm": 0.042527128010988235,
      "learning_rate": 7.936507936507937e-07,
      "loss": -0.0035,
      "num_tokens": 11980487.0,
      "reward": 0.4451127886772156,
      "reward_std": 0.43396498560905455,
      "rewards/accuracy_reward": 0.2675781279802322,
      "rewards/brier_reward": -0.028027261979877947,
      "rewards/confidence_one_or_zero": 0.33984375,
      "rewards/format_reward": 0.6506510376930237,
      "rewards/mean_confidence_reward": 0.8417555093765259,
      "step": 10
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.015364583333333326,
      "completions/max_length": 3935.8,
      "completions/max_terminated_length": 3935.8,
      "completions/mean_length": 447.9428466796875,
      "completions/mean_terminated_length": 455.2278991699219,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 16.8,
      "epoch": 0.024,
      "grad_norm": 0.0023735298309475183,
      "learning_rate": 1.1904761904761906e-06,
      "loss": -0.0029,
      "num_tokens": 17470512.0,
      "reward": 0.5592861294746398,
      "reward_std": 0.42554730772972105,
      "rewards/accuracy_reward": 0.33515624403953553,
      "rewards/brier_reward": -0.03991579922439996,
      "rewards/confidence_one_or_zero": 0.35338541865348816,
      "rewards/format_reward": 0.8233072996139527,
      "rewards/mean_confidence_reward": 0.8678598046302796,
      "step": 15
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.011848958333333326,
      "completions/max_length": 3937.8,
      "completions/max_terminated_length": 3937.8,
      "completions/mean_length": 463.570068359375,
      "completions/mean_terminated_length": 469.2065063476563,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 66.6,
      "epoch": 0.032,
      "grad_norm": 0.002015794860199094,
      "learning_rate": 1.5873015873015873e-06,
      "loss": -0.0097,
      "num_tokens": 23091818.0,
      "reward": 0.6471682548522949,
      "reward_std": 0.39453019499778746,
      "rewards/accuracy_reward": 0.38802083134651183,
      "rewards/brier_reward": -0.04097447711974382,
      "rewards/confidence_one_or_zero": 0.33307291865348815,
      "rewards/format_reward": 0.947265625,
      "rewards/mean_confidence_reward": 0.8901778340339661,
      "step": 20
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.012109375,
      "completions/max_length": 3935.8,
      "completions/max_terminated_length": 3935.8,
      "completions/mean_length": 513.6576965332031,
      "completions/mean_terminated_length": 520.1555480957031,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 106.0,
      "epoch": 0.04,
      "grad_norm": 0.001320844516158104,
      "learning_rate": 1.984126984126984e-06,
      "loss": -0.0068,
      "num_tokens": 29117541.0,
      "reward": 0.7272847175598145,
      "reward_std": 0.3546645939350128,
      "rewards/accuracy_reward": 0.462890625,
      "rewards/brier_reward": 0.014180318638682365,
      "rewards/confidence_one_or_zero": 0.3223958253860474,
      "rewards/format_reward": 0.9774739623069764,
      "rewards/mean_confidence_reward": 0.9008969783782959,
      "step": 25
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.011718750000000023,
      "completions/max_length": 3553.6,
      "completions/max_terminated_length": 3553.6,
      "completions/mean_length": 564.8890747070312,
      "completions/mean_terminated_length": 571.5174133300782,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 139.6,
      "epoch": 0.048,
      "grad_norm": 0.0012930373195558786,
      "learning_rate": 2.380952380952381e-06,
      "loss": -0.0046,
      "num_tokens": 35529649.0,
      "reward": 0.8222924947738648,
      "reward_std": 0.3005819499492645,
      "rewards/accuracy_reward": 0.5587239503860474,
      "rewards/brier_reward": 0.10172324664890767,
      "rewards/confidence_one_or_zero": 0.25768229067325593,
      "rewards/format_reward": 0.9841145753860474,
      "rewards/mean_confidence_reward": 0.9016106605529786,
      "step": 30
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.011979166666666674,
      "completions/max_length": 4010.4,
      "completions/max_terminated_length": 4010.4,
      "completions/mean_length": 603.4971557617188,
      "completions/mean_terminated_length": 611.001708984375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 164.6,
      "epoch": 0.056,
      "grad_norm": 0.001176029210910201,
      "learning_rate": 2.7777777777777783e-06,
      "loss": -0.0003,
      "num_tokens": 42239899.0,
      "reward": 0.8979942679405213,
      "reward_std": 0.27076905965805054,
      "rewards/accuracy_reward": 0.6375,
      "rewards/brier_reward": 0.1740933060646057,
      "rewards/confidence_one_or_zero": 0.11796874850988388,
      "rewards/format_reward": 0.9843750119209289,
      "rewards/mean_confidence_reward": 0.8912721037864685,
      "step": 35
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.011197916666666651,
      "completions/max_length": 3596.0,
      "completions/max_terminated_length": 3596.0,
      "completions/mean_length": 612.3967651367187,
      "completions/mean_terminated_length": 619.3883911132813,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 170.0,
      "epoch": 0.064,
      "grad_norm": 0.0011552661890164018,
      "learning_rate": 3.1746031746031746e-06,
      "loss": -0.0042,
      "num_tokens": 49041922.0,
      "reward": 0.8792973160743713,
      "reward_std": 0.2753585934638977,
      "rewards/accuracy_reward": 0.613671886920929,
      "rewards/brier_reward": 0.15714388936758042,
      "rewards/confidence_one_or_zero": 0.046484375,
      "rewards/format_reward": 0.9877604126930237,
      "rewards/mean_confidence_reward": 0.8770272850990295,
      "step": 40
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.009114583333333348,
      "completions/max_length": 3943.6,
      "completions/max_terminated_length": 3943.6,
      "completions/mean_length": 639.211865234375,
      "completions/mean_terminated_length": 645.0858520507812,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 182.4,
      "epoch": 0.072,
      "grad_norm": 0.0007424323703162372,
      "learning_rate": 3.5714285714285718e-06,
      "loss": -0.0015,
      "num_tokens": 56011229.0,
      "reward": 0.9037358045578003,
      "reward_std": 0.2447569727897644,
      "rewards/accuracy_reward": 0.6365885376930237,
      "rewards/brier_reward": 0.1818029373884201,
      "rewards/confidence_one_or_zero": 0.02799479253590107,
      "rewards/format_reward": 0.9890625,
      "rewards/mean_confidence_reward": 0.85113023519516,
      "step": 45
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.016927083333333325,
      "completions/max_length": 3734.6,
      "completions/max_terminated_length": 3734.6,
      "completions/mean_length": 632.4301025390625,
      "completions/mean_terminated_length": 643.6194091796875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 193.0,
      "epoch": 0.08,
      "grad_norm": 0.000804801587946713,
      "learning_rate": 3.968253968253968e-06,
      "loss": -0.0049,
      "num_tokens": 62916484.0,
      "reward": 0.9066920757293702,
      "reward_std": 0.23956074118614196,
      "rewards/accuracy_reward": 0.638671875,
      "rewards/brier_reward": 0.19253399968147278,
      "rewards/confidence_one_or_zero": 0.02356770858168602,
      "rewards/format_reward": 0.9821614623069763,
      "rewards/mean_confidence_reward": 0.8142734408378601,
      "step": 50
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.019010416666666675,
      "completions/max_length": 3518.4,
      "completions/max_terminated_length": 3518.4,
      "completions/mean_length": 631.1892211914062,
      "completions/mean_terminated_length": 643.4000732421875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 201.8,
      "epoch": 0.088,
      "grad_norm": 0.0008334097219631076,
      "learning_rate": 4.365079365079366e-06,
      "loss": -0.0091,
      "num_tokens": 69818673.0,
      "reward": 0.9460753083229065,
      "reward_std": 0.22614575624465943,
      "rewards/accuracy_reward": 0.6798177123069763,
      "rewards/brier_reward": 0.23223904967308046,
      "rewards/confidence_one_or_zero": 0.0078124998603016135,
      "rewards/format_reward": 0.980078125,
      "rewards/mean_confidence_reward": 0.778917932510376,
      "step": 55
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.018489583333333327,
      "completions/max_length": 3826.4,
      "completions/max_terminated_length": 3826.4,
      "completions/mean_length": 643.7271118164062,
      "completions/mean_terminated_length": 655.7462646484375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 187.8,
      "epoch": 0.096,
      "grad_norm": 0.0008479825919494033,
      "learning_rate": 4.761904761904762e-06,
      "loss": -0.0088,
      "num_tokens": 76856609.0,
      "reward": 0.9268052816390991,
      "reward_std": 0.22018312215805053,
      "rewards/accuracy_reward": 0.6569010496139527,
      "rewards/brier_reward": 0.21648616194725037,
      "rewards/confidence_one_or_zero": 0.0006510416860692203,
      "rewards/format_reward": 0.9802083253860474,
      "rewards/mean_confidence_reward": 0.7498763203620911,
      "step": 60
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.019140625,
      "completions/max_length": 3656.0,
      "completions/max_terminated_length": 3656.0,
      "completions/mean_length": 607.9446899414063,
      "completions/mean_terminated_length": 619.7687255859375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 194.4,
      "epoch": 0.104,
      "grad_norm": 0.0009165199589915574,
      "learning_rate": 4.960000000000001e-06,
      "loss": -0.0094,
      "num_tokens": 83612600.0,
      "reward": 0.9750679016113282,
      "reward_std": 0.20449844002723694,
      "rewards/accuracy_reward": 0.7111979126930237,
      "rewards/brier_reward": 0.2588455408811569,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.980078125,
      "rewards/mean_confidence_reward": 0.7087564826011657,
      "step": 65
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.015234375,
      "completions/max_length": 3051.2,
      "completions/max_terminated_length": 3051.2,
      "completions/mean_length": 652.7509399414063,
      "completions/mean_terminated_length": 662.9979370117187,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 180.4,
      "epoch": 0.112,
      "grad_norm": 0.0007955668843351305,
      "learning_rate": 4.86e-06,
      "loss": -0.008,
      "num_tokens": 90710815.0,
      "reward": 0.886320424079895,
      "reward_std": 0.21232469975948334,
      "rewards/accuracy_reward": 0.604687488079071,
      "rewards/brier_reward": 0.18356477618217468,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.984375,
      "rewards/mean_confidence_reward": 0.6760807514190674,
      "step": 70
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.018619791666666653,
      "completions/max_length": 3539.6,
      "completions/max_terminated_length": 3539.6,
      "completions/mean_length": 593.0100463867187,
      "completions/mean_terminated_length": 604.2567993164063,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 165.6,
      "epoch": 0.12,
      "grad_norm": 0.0009145570220425725,
      "learning_rate": 4.76e-06,
      "loss": -0.012,
      "num_tokens": 97325932.0,
      "reward": 0.9395914673805237,
      "reward_std": 0.20647467374801637,
      "rewards/accuracy_reward": 0.6697916626930237,
      "rewards/brier_reward": 0.2282584637403488,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9811197757720947,
      "rewards/mean_confidence_reward": 0.6482942581176758,
      "step": 75
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.016276041666666675,
      "completions/max_length": 3118.2,
      "completions/max_terminated_length": 3118.2,
      "completions/mean_length": 609.7080810546875,
      "completions/mean_terminated_length": 619.7514404296875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 166.8,
      "epoch": 0.128,
      "grad_norm": 0.0007646373705938458,
      "learning_rate": 4.66e-06,
      "loss": -0.0103,
      "num_tokens": 104054442.0,
      "reward": 0.9215831160545349,
      "reward_std": 0.19176682829856873,
      "rewards/accuracy_reward": 0.6473958373069764,
      "rewards/brier_reward": 0.21255435943603515,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.983203125,
      "rewards/mean_confidence_reward": 0.6473307490348816,
      "step": 80
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.026822916666666675,
      "completions/max_length": 3402.0,
      "completions/max_terminated_length": 3402.0,
      "completions/mean_length": 644.264990234375,
      "completions/mean_terminated_length": 661.9863647460937,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 194.8,
      "epoch": 0.136,
      "grad_norm": 0.0010355948470532894,
      "learning_rate": 4.56e-06,
      "loss": -0.0145,
      "num_tokens": 111063485.0,
      "reward": 0.9154714465141296,
      "reward_std": 0.2156262367963791,
      "rewards/accuracy_reward": 0.644921875,
      "rewards/brier_reward": 0.21387288570404053,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9721354126930237,
      "rewards/mean_confidence_reward": 0.6400976538658142,
      "step": 85
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.012890625,
      "completions/max_length": 3269.6,
      "completions/max_terminated_length": 3269.6,
      "completions/mean_length": 630.4013305664063,
      "completions/mean_terminated_length": 638.8163940429688,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 175.4,
      "epoch": 0.144,
      "grad_norm": 0.0007032503490336239,
      "learning_rate": 4.4600000000000005e-06,
      "loss": -0.0053,
      "num_tokens": 117975079.0,
      "reward": 0.9135956525802612,
      "reward_std": 0.19466035068035126,
      "rewards/accuracy_reward": 0.635546875,
      "rewards/brier_reward": 0.2046529918909073,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.9869791626930237,
      "rewards/mean_confidence_reward": 0.6112109422683716,
      "step": 90
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014322916666666675,
      "completions/max_length": 2962.0,
      "completions/max_terminated_length": 2962.0,
      "completions/mean_length": 616.5053588867188,
      "completions/mean_terminated_length": 625.3900756835938,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 165.6,
      "epoch": 0.152,
      "grad_norm": 0.000655663141515106,
      "learning_rate": 4.360000000000001e-06,
      "loss": -0.0075,
      "num_tokens": 124771408.0,
      "reward": 0.9443702101707458,
      "reward_std": 0.16352900862693787,
      "rewards/accuracy_reward": 0.6774739623069763,
      "rewards/brier_reward": 0.2255781203508377,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9856770873069763,
      "rewards/mean_confidence_reward": 0.5610416769981384,
      "step": 95
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.013281250000000022,
      "completions/max_length": 3385.6,
      "completions/max_terminated_length": 3385.6,
      "completions/mean_length": 618.2169555664062,
      "completions/mean_terminated_length": 626.5005737304688,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 144.6,
      "epoch": 0.16,
      "grad_norm": 0.0007609895546920598,
      "learning_rate": 4.26e-06,
      "loss": -0.0074,
      "num_tokens": 131564978.0,
      "reward": 0.9474413394927979,
      "reward_std": 0.1751583695411682,
      "rewards/accuracy_reward": 0.6798177003860474,
      "rewards/brier_reward": 0.22846483290195466,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9865885496139526,
      "rewards/mean_confidence_reward": 0.5765885591506958,
      "step": 100
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.02083333333333335,
      "completions/max_length": 3028.4,
      "completions/max_terminated_length": 3028.4,
      "completions/mean_length": 640.8221557617187,
      "completions/mean_terminated_length": 654.4121704101562,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 161.8,
      "epoch": 0.168,
      "grad_norm": 0.0007830454269424081,
      "learning_rate": 4.16e-06,
      "loss": -0.0135,
      "num_tokens": 138551452.0,
      "reward": 0.9133638978004456,
      "reward_std": 0.19001688957214355,
      "rewards/accuracy_reward": 0.639843738079071,
      "rewards/brier_reward": 0.20770572125911713,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9791666746139527,
      "rewards/mean_confidence_reward": 0.5819270849227905,
      "step": 105
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.013020833333333325,
      "completions/max_length": 3283.8,
      "completions/max_terminated_length": 3283.8,
      "completions/mean_length": 635.934130859375,
      "completions/mean_terminated_length": 644.3301391601562,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 170.6,
      "epoch": 0.176,
      "grad_norm": 0.0006950053502805531,
      "learning_rate": 4.060000000000001e-06,
      "loss": -0.0053,
      "num_tokens": 145476034.0,
      "reward": 0.9698919773101806,
      "reward_std": 0.1803193747997284,
      "rewards/accuracy_reward": 0.706640625,
      "rewards/brier_reward": 0.24628254175186157,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9868489503860474,
      "rewards/mean_confidence_reward": 0.5891406655311584,
      "step": 110
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.009244791666666674,
      "completions/max_length": 3327.4,
      "completions/max_terminated_length": 3327.4,
      "completions/mean_length": 657.1101684570312,
      "completions/mean_terminated_length": 663.3098876953125,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 183.8,
      "epoch": 0.184,
      "grad_norm": 0.0006515610148198903,
      "learning_rate": 3.96e-06,
      "loss": -0.0059,
      "num_tokens": 152559920.0,
      "reward": 0.9820420026779175,
      "reward_std": 0.15912432670593263,
      "rewards/accuracy_reward": 0.7200520753860473,
      "rewards/brier_reward": 0.2533950090408325,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.990625,
      "rewards/mean_confidence_reward": 0.5910872697830201,
      "step": 115
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.01809895833333335,
      "completions/max_length": 3138.4,
      "completions/max_terminated_length": 3138.4,
      "completions/mean_length": 695.5690307617188,
      "completions/mean_terminated_length": 708.5348388671875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 176.8,
      "epoch": 0.192,
      "grad_norm": 0.0007940581417642534,
      "learning_rate": 3.86e-06,
      "loss": -0.0103,
      "num_tokens": 159942178.0,
      "reward": 0.9360029339790344,
      "reward_std": 0.18510036766529084,
      "rewards/accuracy_reward": 0.6666666626930237,
      "rewards/brier_reward": 0.22355664074420928,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9817708373069763,
      "rewards/mean_confidence_reward": 0.5853255391120911,
      "step": 120
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.015234375,
      "completions/max_length": 3468.4,
      "completions/max_terminated_length": 3468.4,
      "completions/mean_length": 657.5879150390625,
      "completions/mean_terminated_length": 667.9752075195313,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 201.2,
      "epoch": 0.2,
      "grad_norm": 0.0007764480542391539,
      "learning_rate": 3.7600000000000004e-06,
      "loss": -0.0088,
      "num_tokens": 167048389.0,
      "reward": 0.9672588467597961,
      "reward_std": 0.17604605853557587,
      "rewards/accuracy_reward": 0.704296875,
      "rewards/brier_reward": 0.24544334709644317,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.984765625,
      "rewards/mean_confidence_reward": 0.5908463835716248,
      "step": 125
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014192708333333349,
      "completions/max_length": 2982.8,
      "completions/max_terminated_length": 2982.8,
      "completions/mean_length": 646.93935546875,
      "completions/mean_terminated_length": 656.359033203125,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 176.0,
      "epoch": 0.208,
      "grad_norm": 0.0008264482603408396,
      "learning_rate": 3.66e-06,
      "loss": -0.0081,
      "num_tokens": 174077587.0,
      "reward": 0.9522372364997864,
      "reward_std": 0.1730486124753952,
      "rewards/accuracy_reward": 0.685156238079071,
      "rewards/brier_reward": 0.2334990918636322,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9858072996139526,
      "rewards/mean_confidence_reward": 0.5902611374855041,
      "step": 130
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.015885416666666673,
      "completions/max_length": 3613.0,
      "completions/max_terminated_length": 3613.0,
      "completions/mean_length": 618.9064086914062,
      "completions/mean_terminated_length": 628.990087890625,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 136.8,
      "epoch": 0.216,
      "grad_norm": 0.0010248884791508317,
      "learning_rate": 3.5600000000000002e-06,
      "loss": -0.0097,
      "num_tokens": 180870020.0,
      "reward": 0.9591554403305054,
      "reward_std": 0.16766308546066283,
      "rewards/accuracy_reward": 0.69453125,
      "rewards/brier_reward": 0.24004367589950562,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9837239742279053,
      "rewards/mean_confidence_reward": 0.596709668636322,
      "step": 135
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.010546875,
      "completions/max_length": 3258.0,
      "completions/max_terminated_length": 3258.0,
      "completions/mean_length": 633.2195556640625,
      "completions/mean_terminated_length": 639.9541259765625,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 136.2,
      "epoch": 0.224,
      "grad_norm": 0.0006863352609798312,
      "learning_rate": 3.46e-06,
      "loss": -0.004,
      "num_tokens": 187792570.0,
      "reward": 0.9510619521141053,
      "reward_std": 0.16896803975105285,
      "rewards/accuracy_reward": 0.6817708373069763,
      "rewards/brier_reward": 0.23140884935855865,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9889322876930237,
      "rewards/mean_confidence_reward": 0.5977343916893005,
      "step": 140
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.019270833333333348,
      "completions/max_length": 2914.2,
      "completions/max_terminated_length": 2914.2,
      "completions/mean_length": 633.729052734375,
      "completions/mean_terminated_length": 646.0748168945313,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 149.2,
      "epoch": 0.232,
      "grad_norm": 0.0008178013376891613,
      "learning_rate": 3.3600000000000004e-06,
      "loss": -0.0088,
      "num_tokens": 194738425.0,
      "reward": 0.9498031616210938,
      "reward_std": 0.15397230684757232,
      "rewards/accuracy_reward": 0.6842447876930237,
      "rewards/brier_reward": 0.23475064039230348,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9805989742279053,
      "rewards/mean_confidence_reward": 0.5970442771911622,
      "step": 145
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.017708333333333326,
      "completions/max_length": 3197.8,
      "completions/max_terminated_length": 3197.8,
      "completions/mean_length": 639.7275024414063,
      "completions/mean_terminated_length": 651.0954467773438,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 177.2,
      "epoch": 0.24,
      "grad_norm": 0.0007018765318207443,
      "learning_rate": 3.2600000000000006e-06,
      "loss": -0.0103,
      "num_tokens": 201704044.0,
      "reward": 0.9639625310897827,
      "reward_std": 0.1720540165901184,
      "rewards/accuracy_reward": 0.698046875,
      "rewards/brier_reward": 0.24757291674613952,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9822916626930237,
      "rewards/mean_confidence_reward": 0.6781770825386048,
      "step": 150
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.017838541666666652,
      "completions/max_length": 2620.8,
      "completions/max_terminated_length": 2620.8,
      "completions/mean_length": 589.9436401367187,
      "completions/mean_terminated_length": 600.5923095703125,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 155.0,
      "epoch": 0.248,
      "grad_norm": 0.0008456971263512969,
      "learning_rate": 3.1600000000000002e-06,
      "loss": -0.009,
      "num_tokens": 208277339.0,
      "reward": 0.9775675773620606,
      "reward_std": 0.1654069423675537,
      "rewards/accuracy_reward": 0.7130208253860474,
      "rewards/brier_reward": 0.26006900370121,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.98203125,
      "rewards/mean_confidence_reward": 0.7014974117279053,
      "step": 155
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.02291666666666665,
      "completions/max_length": 3518.4,
      "completions/max_terminated_length": 3518.4,
      "completions/mean_length": 632.367333984375,
      "completions/mean_terminated_length": 647.7365600585938,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 143.0,
      "epoch": 0.256,
      "grad_norm": 0.0006815266679041088,
      "learning_rate": 3.0600000000000003e-06,
      "loss": -0.011,
      "num_tokens": 215204320.0,
      "reward": 0.9521210074424744,
      "reward_std": 0.172203129529953,
      "rewards/accuracy_reward": 0.6856770992279053,
      "rewards/brier_reward": 0.2417280375957489,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9768229126930237,
      "rewards/mean_confidence_reward": 0.6928971409797668,
      "step": 160
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.019661458333333347,
      "completions/max_length": 3060.4,
      "completions/max_terminated_length": 3060.4,
      "completions/mean_length": 585.7721557617188,
      "completions/mean_terminated_length": 597.5513793945313,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 160.8,
      "epoch": 0.264,
      "grad_norm": 0.0007917583570815623,
      "learning_rate": 2.96e-06,
      "loss": -0.012,
      "num_tokens": 221775434.0,
      "reward": 0.9763261079788208,
      "reward_std": 0.16924993693828583,
      "rewards/accuracy_reward": 0.7118489503860473,
      "rewards/brier_reward": 0.26097134947776796,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9798177123069763,
      "rewards/mean_confidence_reward": 0.7043619632720948,
      "step": 165
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.02291666666666665,
      "completions/max_length": 3014.2,
      "completions/max_terminated_length": 3014.2,
      "completions/mean_length": 579.3830932617187,
      "completions/mean_terminated_length": 592.9825561523437,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 153.4,
      "epoch": 0.272,
      "grad_norm": 0.0008060517720878124,
      "learning_rate": 2.86e-06,
      "loss": -0.0122,
      "num_tokens": 228282408.0,
      "reward": 0.9391168594360352,
      "reward_std": 0.18324347734451293,
      "rewards/accuracy_reward": 0.6692708492279053,
      "rewards/brier_reward": 0.23186497688293456,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.9770833253860474,
      "rewards/mean_confidence_reward": 0.7257981657981872,
      "step": 170
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.013932291666666674,
      "completions/max_length": 3355.2,
      "completions/max_terminated_length": 3355.2,
      "completions/mean_length": 559.7354370117188,
      "completions/mean_terminated_length": 567.7333129882812,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 139.6,
      "epoch": 0.28,
      "grad_norm": 0.0007986059063114226,
      "learning_rate": 2.7600000000000003e-06,
      "loss": -0.0046,
      "num_tokens": 234631064.0,
      "reward": 0.994887912273407,
      "reward_std": 0.1719746321439743,
      "rewards/accuracy_reward": 0.730078125,
      "rewards/brier_reward": 0.2736147344112396,
      "rewards/confidence_one_or_zero": 0.00026041667442768814,
      "rewards/format_reward": 0.9860677123069763,
      "rewards/mean_confidence_reward": 0.7623633146286011,
      "step": 175
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.021223958333333327,
      "completions/max_length": 2695.0,
      "completions/max_terminated_length": 2695.0,
      "completions/mean_length": 575.9768432617187,
      "completions/mean_terminated_length": 588.6129516601562,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 191.8,
      "epoch": 0.288,
      "grad_norm": 0.0006665615946985781,
      "learning_rate": 2.6600000000000004e-06,
      "loss": -0.0118,
      "num_tokens": 241108422.0,
      "reward": 0.940597414970398,
      "reward_std": 0.1621477097272873,
      "rewards/accuracy_reward": 0.6713541626930237,
      "rewards/brier_reward": 0.23117966949939728,
      "rewards/confidence_one_or_zero": 0.0009114583488553762,
      "rewards/format_reward": 0.9786458373069763,
      "rewards/mean_confidence_reward": 0.7551562786102295,
      "step": 180
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.016666666666666673,
      "completions/max_length": 2449.8,
      "completions/max_terminated_length": 2449.8,
      "completions/mean_length": 566.245849609375,
      "completions/mean_terminated_length": 575.84814453125,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 138.8,
      "epoch": 0.296,
      "grad_norm": 0.0007533118478022516,
      "learning_rate": 2.56e-06,
      "loss": -0.0121,
      "num_tokens": 247544262.0,
      "reward": 0.9880049824714661,
      "reward_std": 0.17469136714935302,
      "rewards/accuracy_reward": 0.7209635376930237,
      "rewards/brier_reward": 0.27169823050498965,
      "rewards/confidence_one_or_zero": 0.0006510416860692203,
      "rewards/format_reward": 0.9833333373069764,
      "rewards/mean_confidence_reward": 0.7388398289680481,
      "step": 185
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.03932291666666667,
      "completions/max_length": 3461.4,
      "completions/max_terminated_length": 3461.4,
      "completions/mean_length": 593.4655029296875,
      "completions/mean_terminated_length": 617.8613647460937,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 125.4,
      "epoch": 0.304,
      "grad_norm": 0.0008256498840637505,
      "learning_rate": 2.46e-06,
      "loss": -0.0209,
      "num_tokens": 254165373.0,
      "reward": 0.9393493890762329,
      "reward_std": 0.1917530208826065,
      "rewards/accuracy_reward": 0.674609363079071,
      "rewards/brier_reward": 0.24352831542491912,
      "rewards/confidence_one_or_zero": 0.002734374941792339,
      "rewards/format_reward": 0.960546875,
      "rewards/mean_confidence_reward": 0.7073220014572144,
      "step": 190
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.02213541666666665,
      "completions/max_length": 2661.2,
      "completions/max_terminated_length": 2661.2,
      "completions/mean_length": 607.9656494140625,
      "completions/mean_terminated_length": 621.8783081054687,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 154.4,
      "epoch": 0.312,
      "grad_norm": 0.0007380797178484499,
      "learning_rate": 2.3600000000000003e-06,
      "loss": -0.011,
      "num_tokens": 260926645.0,
      "reward": 0.9378235578536988,
      "reward_std": 0.1884024739265442,
      "rewards/accuracy_reward": 0.6649739384651184,
      "rewards/brier_reward": 0.23292421996593476,
      "rewards/confidence_one_or_zero": 0.0010416666977107526,
      "rewards/format_reward": 0.977734363079071,
      "rewards/mean_confidence_reward": 0.7236497402191162,
      "step": 195
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.017708333333333347,
      "completions/max_length": 2874.8,
      "completions/max_terminated_length": 2874.8,
      "completions/mean_length": 596.4682495117188,
      "completions/mean_terminated_length": 607.2115966796875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 130.4,
      "epoch": 0.32,
      "grad_norm": 0.0008115461678244174,
      "learning_rate": 2.2600000000000004e-06,
      "loss": -0.0088,
      "num_tokens": 267572673.0,
      "reward": 0.9656657576560974,
      "reward_std": 0.1814965397119522,
      "rewards/accuracy_reward": 0.6955729126930237,
      "rewards/brier_reward": 0.25358234643936156,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.9821614742279052,
      "rewards/mean_confidence_reward": 0.7382291555404663,
      "step": 200
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.01888020833333335,
      "completions/max_length": 3335.4,
      "completions/max_terminated_length": 3335.4,
      "completions/mean_length": 572.5916870117187,
      "completions/mean_terminated_length": 583.6033447265625,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 156.2,
      "epoch": 0.328,
      "grad_norm": 0.0007804105989634991,
      "learning_rate": 2.16e-06,
      "loss": -0.012,
      "num_tokens": 274018497.0,
      "reward": 0.9807969927787781,
      "reward_std": 0.18046627938747406,
      "rewards/accuracy_reward": 0.7140625,
      "rewards/brier_reward": 0.26639663875103,
      "rewards/confidence_one_or_zero": 0.0007812500116415322,
      "rewards/format_reward": 0.9811197996139527,
      "rewards/mean_confidence_reward": 0.7495768427848816,
      "step": 205
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.028385416666666653,
      "completions/max_length": 2882.0,
      "completions/max_terminated_length": 2882.0,
      "completions/mean_length": 593.1833618164062,
      "completions/mean_terminated_length": 610.51943359375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 152.6,
      "epoch": 0.336,
      "grad_norm": 0.000877701211720705,
      "learning_rate": 2.06e-06,
      "loss": -0.0189,
      "num_tokens": 280657921.0,
      "reward": 0.9234808564186097,
      "reward_std": 0.1844559758901596,
      "rewards/accuracy_reward": 0.651171863079071,
      "rewards/brier_reward": 0.22429133653640748,
      "rewards/confidence_one_or_zero": 0.00039062501164153216,
      "rewards/format_reward": 0.971484363079071,
      "rewards/mean_confidence_reward": 0.7052604198455811,
      "step": 210
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.021875,
      "completions/max_length": 2389.4,
      "completions/max_terminated_length": 2389.4,
      "completions/mean_length": 559.9268310546875,
      "completions/mean_terminated_length": 572.4205932617188,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 150.0,
      "epoch": 0.344,
      "grad_norm": 0.0007080782670527697,
      "learning_rate": 1.9600000000000003e-06,
      "loss": -0.011,
      "num_tokens": 286990543.0,
      "reward": 0.9907698631286621,
      "reward_std": 0.16264156401157379,
      "rewards/accuracy_reward": 0.7279947876930237,
      "rewards/brier_reward": 0.2754066318273544,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.978125,
      "rewards/mean_confidence_reward": 0.6646796941757203,
      "step": 215
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.015234374999999977,
      "completions/max_length": 2514.4,
      "completions/max_terminated_length": 2514.4,
      "completions/mean_length": 620.4054931640625,
      "completions/mean_terminated_length": 629.9885131835938,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 173.8,
      "epoch": 0.352,
      "grad_norm": 0.0006695219781249762,
      "learning_rate": 1.8600000000000002e-06,
      "loss": -0.0094,
      "num_tokens": 293826777.0,
      "reward": 0.9660478711128235,
      "reward_std": 0.14799184501171112,
      "rewards/accuracy_reward": 0.6959635496139527,
      "rewards/brier_reward": 0.25135333836078644,
      "rewards/confidence_one_or_zero": 0.00026041667442768814,
      "rewards/format_reward": 0.9847656130790711,
      "rewards/mean_confidence_reward": 0.6576888203620911,
      "step": 220
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014713541666666673,
      "completions/max_length": 3235.2,
      "completions/max_terminated_length": 3235.2,
      "completions/mean_length": 644.6151245117187,
      "completions/mean_terminated_length": 653.9868896484375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 168.6,
      "epoch": 0.36,
      "grad_norm": 0.0008919989923015237,
      "learning_rate": 1.76e-06,
      "loss": -0.0092,
      "num_tokens": 300848461.0,
      "reward": 0.9726941227912903,
      "reward_std": 0.1757771283388138,
      "rewards/accuracy_reward": 0.702734375,
      "rewards/brier_reward": 0.2573541611433029,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.9852864623069764,
      "rewards/mean_confidence_reward": 0.6574609637260437,
      "step": 225
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.013802083333333326,
      "completions/max_length": 2613.0,
      "completions/max_terminated_length": 2613.0,
      "completions/mean_length": 584.2405029296875,
      "completions/mean_terminated_length": 592.501171875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 170.0,
      "epoch": 0.368,
      "grad_norm": 0.0010123475221917033,
      "learning_rate": 1.6600000000000002e-06,
      "loss": -0.0075,
      "num_tokens": 307402660.0,
      "reward": 1.0088900566101073,
      "reward_std": 0.15745406746864318,
      "rewards/accuracy_reward": 0.7428385257720947,
      "rewards/brier_reward": 0.2887296438217163,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9861979126930237,
      "rewards/mean_confidence_reward": 0.6941731929779053,
      "step": 230
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014453125,
      "completions/max_length": 3235.2,
      "completions/max_terminated_length": 3235.2,
      "completions/mean_length": 618.46591796875,
      "completions/mean_terminated_length": 627.46552734375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 173.8,
      "epoch": 0.376,
      "grad_norm": 0.0008414699113927782,
      "learning_rate": 1.56e-06,
      "loss": -0.0063,
      "num_tokens": 314222270.0,
      "reward": 0.9931760191917419,
      "reward_std": 0.16961570382118224,
      "rewards/accuracy_reward": 0.7246093869209289,
      "rewards/brier_reward": 0.2761809766292572,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.985546875,
      "rewards/mean_confidence_reward": 0.73552086353302,
      "step": 235
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.017317708333333327,
      "completions/max_length": 2330.2,
      "completions/max_terminated_length": 2330.2,
      "completions/mean_length": 612.2080932617188,
      "completions/mean_terminated_length": 622.9677734375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 162.2,
      "epoch": 0.384,
      "grad_norm": 0.0008407265413552523,
      "learning_rate": 1.46e-06,
      "loss": -0.008,
      "num_tokens": 320982012.0,
      "reward": 0.960022795200348,
      "reward_std": 0.17092747390270233,
      "rewards/accuracy_reward": 0.689453125,
      "rewards/brier_reward": 0.248025381565094,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9825520873069763,
      "rewards/mean_confidence_reward": 0.7460026144981384,
      "step": 240
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.019661458333333347,
      "completions/max_length": 3335.2,
      "completions/max_terminated_length": 3335.2,
      "completions/mean_length": 649.0472900390625,
      "completions/mean_terminated_length": 662.4468383789062,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 185.6,
      "epoch": 0.392,
      "grad_norm": 0.0007866204250603914,
      "learning_rate": 1.3600000000000001e-06,
      "loss": -0.0067,
      "num_tokens": 328065415.0,
      "reward": 0.936725127696991,
      "reward_std": 0.1595529317855835,
      "rewards/accuracy_reward": 0.662890636920929,
      "rewards/brier_reward": 0.23033641576766967,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9802083373069763,
      "rewards/mean_confidence_reward": 0.7382617235183716,
      "step": 245
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.011718750000000023,
      "completions/max_length": 2999.8,
      "completions/max_terminated_length": 2999.8,
      "completions/mean_length": 656.2833618164062,
      "completions/mean_terminated_length": 664.0399291992187,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 158.6,
      "epoch": 0.4,
      "grad_norm": 0.0008523129508830607,
      "learning_rate": 1.26e-06,
      "loss": -0.0052,
      "num_tokens": 335168647.0,
      "reward": 0.9867102146148682,
      "reward_std": 0.16842909455299376,
      "rewards/accuracy_reward": 0.7174479126930237,
      "rewards/brier_reward": 0.26780647337436675,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9881510496139526,
      "rewards/mean_confidence_reward": 0.7479882836341858,
      "step": 250
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.01015625,
      "completions/max_length": 2904.0,
      "completions/max_terminated_length": 2904.0,
      "completions/mean_length": 630.57294921875,
      "completions/mean_terminated_length": 636.9603881835938,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 212.2,
      "epoch": 0.408,
      "grad_norm": 0.0009290463058277965,
      "learning_rate": 1.1600000000000001e-06,
      "loss": -0.005,
      "num_tokens": 342078007.0,
      "reward": 1.0194429516792298,
      "reward_std": 0.14954735934734345,
      "rewards/accuracy_reward": 0.753125,
      "rewards/brier_reward": 0.2959021717309952,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.98984375,
      "rewards/mean_confidence_reward": 0.7490299463272094,
      "step": 255
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.012500000000000022,
      "completions/max_length": 3052.0,
      "completions/max_terminated_length": 3052.0,
      "completions/mean_length": 666.5530151367187,
      "completions/mean_terminated_length": 674.8721313476562,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 180.4,
      "epoch": 0.416,
      "grad_norm": 0.0007056944887153804,
      "learning_rate": 1.06e-06,
      "loss": -0.0091,
      "num_tokens": 349275278.0,
      "reward": 1.0045409798622131,
      "reward_std": 0.15459776818752288,
      "rewards/accuracy_reward": 0.7372395753860473,
      "rewards/brier_reward": 0.2845878809690475,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9872395873069764,
      "rewards/mean_confidence_reward": 0.7450390815734863,
      "step": 260
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.016536458333333327,
      "completions/max_length": 3251.2,
      "completions/max_terminated_length": 3251.2,
      "completions/mean_length": 660.6612182617188,
      "completions/mean_terminated_length": 671.8212524414063,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 188.0,
      "epoch": 0.424,
      "grad_norm": 0.0007217851816676557,
      "learning_rate": 9.600000000000001e-07,
      "loss": -0.0072,
      "num_tokens": 356405508.0,
      "reward": 0.9873118281364441,
      "reward_std": 0.17308388352394105,
      "rewards/accuracy_reward": 0.722265613079071,
      "rewards/brier_reward": 0.26953068375587463,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9828125,
      "rewards/mean_confidence_reward": 0.7402786254882813,
      "step": 265
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.013151041666666674,
      "completions/max_length": 3198.4,
      "completions/max_terminated_length": 3198.4,
      "completions/mean_length": 647.463427734375,
      "completions/mean_terminated_length": 656.1833862304687,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 216.4,
      "epoch": 0.432,
      "grad_norm": 0.0007142944377847016,
      "learning_rate": 8.6e-07,
      "loss": -0.0073,
      "num_tokens": 363451467.0,
      "reward": 0.9884425401687622,
      "reward_std": 0.1667942523956299,
      "rewards/accuracy_reward": 0.719921875,
      "rewards/brier_reward": 0.27009943425655364,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9868489623069763,
      "rewards/mean_confidence_reward": 0.7386263132095336,
      "step": 270
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.010416666666666675,
      "completions/max_length": 3087.4,
      "completions/max_terminated_length": 3087.4,
      "completions/mean_length": 651.3906494140625,
      "completions/mean_terminated_length": 658.2500732421875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 175.6,
      "epoch": 0.44,
      "grad_norm": 0.0006245598196983337,
      "learning_rate": 7.6e-07,
      "loss": -0.0035,
      "num_tokens": 370501987.0,
      "reward": 0.9973590612411499,
      "reward_std": 0.15688675343990327,
      "rewards/accuracy_reward": 0.7296875,
      "rewards/brier_reward": 0.2755631357431412,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.989453125,
      "rewards/mean_confidence_reward": 0.7199348926544189,
      "step": 275
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.012630208333333327,
      "completions/max_length": 3619.0,
      "completions/max_terminated_length": 3619.0,
      "completions/mean_length": 661.80068359375,
      "completions/mean_terminated_length": 670.2946411132813,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 185.6,
      "epoch": 0.448,
      "grad_norm": 0.0009292034083046019,
      "learning_rate": 6.6e-07,
      "loss": -0.0071,
      "num_tokens": 377668072.0,
      "reward": 0.9768817186355591,
      "reward_std": 0.18133719861507416,
      "rewards/accuracy_reward": 0.7075520873069763,
      "rewards/brier_reward": 0.2588273108005524,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9873697876930236,
      "rewards/mean_confidence_reward": 0.7144075512886048,
      "step": 280
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.010807291666666673,
      "completions/max_length": 3593.2,
      "completions/max_terminated_length": 3593.2,
      "completions/mean_length": 645.8926025390625,
      "completions/mean_terminated_length": 652.9858520507812,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 184.0,
      "epoch": 0.456,
      "grad_norm": 0.0008182061137631536,
      "learning_rate": 5.6e-07,
      "loss": -0.0036,
      "num_tokens": 384670223.0,
      "reward": 1.014344823360443,
      "reward_std": 0.1618893563747406,
      "rewards/accuracy_reward": 0.7499999880790711,
      "rewards/brier_reward": 0.2896124541759491,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9890625,
      "rewards/mean_confidence_reward": 0.7284570455551147,
      "step": 285
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014713541666666696,
      "completions/max_length": 3342.6,
      "completions/max_terminated_length": 3342.6,
      "completions/mean_length": 677.264599609375,
      "completions/mean_terminated_length": 687.6894653320312,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 174.4,
      "epoch": 0.464,
      "grad_norm": 0.0005594675894826651,
      "learning_rate": 4.6000000000000004e-07,
      "loss": -0.0087,
      "num_tokens": 391929279.0,
      "reward": 0.9603251934051513,
      "reward_std": 0.15469011068344116,
      "rewards/accuracy_reward": 0.687890636920929,
      "rewards/brier_reward": 0.2475892573595047,
      "rewards/confidence_one_or_zero": 0.00013020833721384407,
      "rewards/format_reward": 0.985156261920929,
      "rewards/mean_confidence_reward": 0.7115533709526062,
      "step": 290
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014583333333333347,
      "completions/max_length": 2386.2,
      "completions/max_terminated_length": 2386.2,
      "completions/mean_length": 650.3267211914062,
      "completions/mean_terminated_length": 660.108154296875,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 160.8,
      "epoch": 0.472,
      "grad_norm": 0.0006979788304306567,
      "learning_rate": 3.6e-07,
      "loss": -0.0072,
      "num_tokens": 398957644.0,
      "reward": 0.9805139303207397,
      "reward_std": 0.14907170832157135,
      "rewards/accuracy_reward": 0.7131510496139526,
      "rewards/brier_reward": 0.2625761777162552,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9852864623069764,
      "rewards/mean_confidence_reward": 0.7069010376930237,
      "step": 295
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.009635416666666652,
      "completions/max_length": 3024.2,
      "completions/max_terminated_length": 3024.2,
      "completions/mean_length": 677.6005493164063,
      "completions/mean_terminated_length": 684.2862182617188,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 193.2,
      "epoch": 0.48,
      "grad_norm": 0.0013121222145855427,
      "learning_rate": 2.6e-07,
      "loss": -0.0045,
      "num_tokens": 406218768.0,
      "reward": 0.9787513613700867,
      "reward_std": 0.15693043768405915,
      "rewards/accuracy_reward": 0.7069010496139526,
      "rewards/brier_reward": 0.2602229714393616,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9903645753860474,
      "rewards/mean_confidence_reward": 0.7023437142372131,
      "step": 300
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.013281250000000022,
      "completions/max_length": 2630.8,
      "completions/max_terminated_length": 2630.8,
      "completions/mean_length": 657.7256713867188,
      "completions/mean_terminated_length": 666.553271484375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 179.8,
      "epoch": 0.488,
      "grad_norm": 0.0007174566271714866,
      "learning_rate": 1.6e-07,
      "loss": -0.0079,
      "num_tokens": 413315925.0,
      "reward": 1.0456790566444396,
      "reward_std": 0.1455691635608673,
      "rewards/accuracy_reward": 0.7888020873069763,
      "rewards/brier_reward": 0.3159534573554993,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9865885496139526,
      "rewards/mean_confidence_reward": 0.7023958563804626,
      "step": 305
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.014843750000000022,
      "completions/max_length": 3217.8,
      "completions/max_terminated_length": 3217.8,
      "completions/mean_length": 659.7379150390625,
      "completions/mean_terminated_length": 669.87099609375,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 161.6,
      "epoch": 0.496,
      "grad_norm": 0.0006406163447536528,
      "learning_rate": 6.000000000000001e-08,
      "loss": -0.0099,
      "num_tokens": 420434712.0,
      "reward": 0.9986423373222351,
      "reward_std": 0.1562984451651573,
      "rewards/accuracy_reward": 0.7326822876930237,
      "rewards/brier_reward": 0.2798227518796921,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.984765636920929,
      "rewards/mean_confidence_reward": 0.6980143189430237,
      "step": 310
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.01801215277777779,
      "completions/max_length": 3332.0,
      "completions/max_terminated_length": 3332.0,
      "completions/mean_length": 705.877197265625,
      "completions/mean_terminated_length": 718.6570434570312,
      "completions/min_length": 0.0,
      "completions/min_terminated_length": 172.33333333333334,
      "epoch": 0.5008,
      "num_tokens": 424935810.0,
      "reward": 0.9911277294158936,
      "reward_std": 0.17695151766141257,
      "rewards/accuracy_reward": 0.7261284788449606,
      "rewards/brier_reward": 0.27412516872088116,
      "rewards/confidence_one_or_zero": 0.0,
      "rewards/format_reward": 0.9819878538449606,
      "rewards/mean_confidence_reward": 0.6981228590011597,
      "step": 313,
      "total_flos": 0.0,
      "train_loss": -0.008061443609497941,
      "train_runtime": 61085.0472,
      "train_samples_per_second": 0.246,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 5,
  "max_steps": 313,
  "num_input_tokens_seen": 424935810,
  "num_train_epochs": 1,
  "save_steps": 50,
  "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": 6,
  "trial_name": null,
  "trial_params": null
}