{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9956331877729258,
  "eval_steps": 500,
  "global_step": 57,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 611.8683166503906,
      "epoch": 0.017467248908296942,
      "grad_norm": 0.44490134716033936,
      "kl": 0.0,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.01,
      "reward": 0.1757812611758709,
      "reward_std": 0.32868771255016327,
      "rewards/format_reward": 0.03125000232830644,
      "rewards/math_reward": 0.0022321429569274187,
      "rewards/tag_count_reward": 0.1422991156578064,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.4286041259766,
      "epoch": 0.034934497816593885,
      "grad_norm": 0.367574542760849,
      "kl": 0.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.0233,
      "reward": 0.1356026828289032,
      "reward_std": 0.3039904832839966,
      "rewards/format_reward": 0.026785715948790312,
      "rewards/math_reward": 0.006696428870782256,
      "rewards/tag_count_reward": 0.102120541036129,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 696.294677734375,
      "epoch": 0.05240174672489083,
      "grad_norm": 0.38345867395401,
      "kl": 0.0002143383026123047,
      "learning_rate": 1e-05,
      "loss": 0.006,
      "reward": 0.12109375558793545,
      "reward_std": 0.29076967388391495,
      "rewards/format_reward": 0.020089286379516125,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.10100446827709675,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 695.0536041259766,
      "epoch": 0.06986899563318777,
      "grad_norm": 233.4116973876953,
      "kl": 5.1015625,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.2516,
      "reward": 0.5195312798023224,
      "reward_std": 0.6752565056085587,
      "rewards/format_reward": 0.1562500074505806,
      "rewards/math_reward": 0.006696428870782256,
      "rewards/tag_count_reward": 0.3565848395228386,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.7767944335938,
      "epoch": 0.08733624454148471,
      "grad_norm": 0.9780572652816772,
      "kl": 0.06536865234375,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0314,
      "reward": 0.8236607611179352,
      "reward_std": 0.7809945940971375,
      "rewards/format_reward": 0.2745535895228386,
      "rewards/math_reward": 0.015625000465661287,
      "rewards/tag_count_reward": 0.5334821715950966,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.350471496582,
      "epoch": 0.10480349344978165,
      "grad_norm": 0.4738185703754425,
      "kl": 0.049560546875,
      "learning_rate": 2e-05,
      "loss": 0.0527,
      "reward": 1.0898438096046448,
      "reward_std": 0.8620707839727402,
      "rewards/format_reward": 0.3861607313156128,
      "rewards/math_reward": 0.046875001629814506,
      "rewards/tag_count_reward": 0.6568080633878708,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.9687805175781,
      "epoch": 0.1222707423580786,
      "grad_norm": 0.3717152178287506,
      "kl": 0.03515625,
      "learning_rate": 1.9981033287370443e-05,
      "loss": 0.017,
      "reward": 1.471540242433548,
      "reward_std": 0.7875999510288239,
      "rewards/format_reward": 0.5691964626312256,
      "rewards/math_reward": 0.10937500488944352,
      "rewards/tag_count_reward": 0.7929687798023224,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.5692291259766,
      "epoch": 0.13973799126637554,
      "grad_norm": 3.1607468128204346,
      "kl": 0.07501220703125,
      "learning_rate": 1.992420509671936e-05,
      "loss": -0.015,
      "reward": 1.5334822237491608,
      "reward_std": 0.7315323203802109,
      "rewards/format_reward": 0.6294643133878708,
      "rewards/math_reward": 0.06473214784637094,
      "rewards/tag_count_reward": 0.839285746216774,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.6585083007812,
      "epoch": 0.1572052401746725,
      "grad_norm": 0.46272966265678406,
      "kl": 0.09783935546875,
      "learning_rate": 1.982973099683902e-05,
      "loss": 0.0339,
      "reward": 1.6484375596046448,
      "reward_std": 0.6518489569425583,
      "rewards/format_reward": 0.729910746216774,
      "rewards/math_reward": 0.04241071455180645,
      "rewards/tag_count_reward": 0.8761161118745804,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 587.7053909301758,
      "epoch": 0.17467248908296942,
      "grad_norm": 3.4270660877227783,
      "kl": 0.3292236328125,
      "learning_rate": 1.9697969360350098e-05,
      "loss": 0.0064,
      "reward": 1.860491156578064,
      "reward_std": 0.6494010090827942,
      "rewards/format_reward": 0.7633928954601288,
      "rewards/math_reward": 0.196428582072258,
      "rewards/tag_count_reward": 0.9006696790456772,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.0759124755859,
      "epoch": 0.19213973799126638,
      "grad_norm": 3.397077798843384,
      "kl": 0.296875,
      "learning_rate": 1.9529420004271568e-05,
      "loss": 0.1006,
      "reward": 1.2667411267757416,
      "reward_std": 0.8302740603685379,
      "rewards/format_reward": 0.4977678805589676,
      "rewards/math_reward": 0.0535714291036129,
      "rewards/tag_count_reward": 0.7154018133878708,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.763427734375,
      "epoch": 0.2096069868995633,
      "grad_norm": 1.088446855545044,
      "kl": 0.34130859375,
      "learning_rate": 1.932472229404356e-05,
      "loss": 0.1301,
      "reward": 0.9073661118745804,
      "reward_std": 0.8017744421958923,
      "rewards/format_reward": 0.305803582072258,
      "rewards/math_reward": 0.02455357206054032,
      "rewards/tag_count_reward": 0.5770089626312256,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 697.044677734375,
      "epoch": 0.22707423580786026,
      "grad_norm": 35.29081344604492,
      "kl": 3.5400390625,
      "learning_rate": 1.9084652718195237e-05,
      "loss": 0.0956,
      "reward": 0.8772321939468384,
      "reward_std": 0.8175660222768784,
      "rewards/format_reward": 0.294642873108387,
      "rewards/math_reward": 0.02901785750873387,
      "rewards/tag_count_reward": 0.5535714700818062,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 714.5536041259766,
      "epoch": 0.2445414847161572,
      "grad_norm": 6.326420307159424,
      "kl": 1.01171875,
      "learning_rate": 1.8810121942857848e-05,
      "loss": 0.1219,
      "reward": 0.6640625298023224,
      "reward_std": 0.7399376779794693,
      "rewards/format_reward": 0.1785714402794838,
      "rewards/math_reward": 0.04017857299186289,
      "rewards/tag_count_reward": 0.4453125298023224,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 732.5803985595703,
      "epoch": 0.26200873362445415,
      "grad_norm": 67.30467224121094,
      "kl": 2.271484375,
      "learning_rate": 1.8502171357296144e-05,
      "loss": 0.1286,
      "reward": 0.6646205633878708,
      "reward_std": 0.7443441152572632,
      "rewards/format_reward": 0.2053571529686451,
      "rewards/math_reward": 0.013392857741564512,
      "rewards/tag_count_reward": 0.4458705559372902,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 720.5268096923828,
      "epoch": 0.2794759825327511,
      "grad_norm": 5.1966753005981445,
      "kl": 1.416015625,
      "learning_rate": 1.816196912356222e-05,
      "loss": 0.0392,
      "reward": 0.683035746216774,
      "reward_std": 0.7767317444086075,
      "rewards/format_reward": 0.232142873108387,
      "rewards/math_reward": 0.02455357275903225,
      "rewards/tag_count_reward": 0.4263393059372902,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.1183471679688,
      "epoch": 0.29694323144104806,
      "grad_norm": 1619.3175048828125,
      "kl": 8.3203125,
      "learning_rate": 1.7790805745256703e-05,
      "loss": 0.5223,
      "reward": 0.4051339477300644,
      "reward_std": 0.5999295711517334,
      "rewards/format_reward": 0.11607143096625805,
      "rewards/math_reward": 0.01562500116415322,
      "rewards/tag_count_reward": 0.2734375074505806,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 555.8906631469727,
      "epoch": 0.314410480349345,
      "grad_norm": 5.135190486907959,
      "kl": 4.6875,
      "learning_rate": 1.7390089172206594e-05,
      "loss": -0.1531,
      "reward": 0.2232142984867096,
      "reward_std": 0.39768724143505096,
      "rewards/format_reward": 0.05580357275903225,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1674107201397419,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 390.2209930419922,
      "epoch": 0.3318777292576419,
      "grad_norm": 1.996619462966919,
      "kl": 2.265625,
      "learning_rate": 1.696133945962927e-05,
      "loss": -0.3963,
      "reward": 0.12667411379516125,
      "reward_std": 0.2982850968837738,
      "rewards/format_reward": 0.03348214388824999,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.0931919664144516,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 278.82814025878906,
      "epoch": 0.34934497816593885,
      "grad_norm": 1.6652499437332153,
      "kl": 2.1640625,
      "learning_rate": 1.650618300204242e-05,
      "loss": -0.7212,
      "reward": 0.1657366119325161,
      "reward_std": 0.3662828281521797,
      "rewards/format_reward": 0.040178573690354824,
      "rewards/math_reward": 0.008928572060540318,
      "rewards/tag_count_reward": 0.1166294701397419,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 375.1116256713867,
      "epoch": 0.36681222707423583,
      "grad_norm": 1.9009864330291748,
      "kl": 1.830078125,
      "learning_rate": 1.6026346363792565e-05,
      "loss": -0.5307,
      "reward": 0.2220982201397419,
      "reward_std": 0.43379347771406174,
      "rewards/format_reward": 0.040178573690354824,
      "rewards/math_reward": 0.004464285913854837,
      "rewards/tag_count_reward": 0.1774553656578064,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 447.9419860839844,
      "epoch": 0.38427947598253276,
      "grad_norm": 1.1156517267227173,
      "kl": 1.85546875,
      "learning_rate": 1.552364972960506e-05,
      "loss": -0.4126,
      "reward": 0.2204241156578064,
      "reward_std": 0.377070814371109,
      "rewards/format_reward": 0.03125000209547579,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1891741119325161,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 526.6406478881836,
      "epoch": 0.4017467248908297,
      "grad_norm": 2.6996726989746094,
      "kl": 2.9921875,
      "learning_rate": 1.5000000000000002e-05,
      "loss": -0.3377,
      "reward": 0.176339291036129,
      "reward_std": 0.2370295263826847,
      "rewards/format_reward": 0.004464285913854837,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1718750074505806,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.6183319091797,
      "epoch": 0.4192139737991266,
      "grad_norm": 1.0993598699569702,
      "kl": 2.52734375,
      "learning_rate": 1.4457383557765385e-05,
      "loss": -0.3695,
      "reward": 0.1322544738650322,
      "reward_std": 0.14882266148924828,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1322544738650322,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 736.7522735595703,
      "epoch": 0.4366812227074236,
      "grad_norm": 179.8589324951172,
      "kl": 3.986328125,
      "learning_rate": 1.3897858732926794e-05,
      "loss": -0.2141,
      "reward": 0.1523437574505806,
      "reward_std": 0.1332450620830059,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1523437574505806,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 678.4129943847656,
      "epoch": 0.45414847161572053,
      "grad_norm": 2.871249198913574,
      "kl": 3.12109375,
      "learning_rate": 1.3323547994796597e-05,
      "loss": -0.2502,
      "reward": 0.1439732238650322,
      "reward_std": 0.1400768794119358,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1439732238650322,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.0134124755859,
      "epoch": 0.47161572052401746,
      "grad_norm": 3.0205397605895996,
      "kl": 1.2001953125,
      "learning_rate": 1.2736629900720832e-05,
      "loss": -0.2349,
      "reward": 0.1171875037252903,
      "reward_std": 0.12809346616268158,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1171875037252903,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.3660888671875,
      "epoch": 0.4890829694323144,
      "grad_norm": 14.793048858642578,
      "kl": 1.2587890625,
      "learning_rate": 1.2139330832064975e-05,
      "loss": -0.2831,
      "reward": 0.12276786006987095,
      "reward_std": 0.1267802119255066,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.12276786006987095,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.0401992797852,
      "epoch": 0.5065502183406113,
      "grad_norm": 1.9193952083587646,
      "kl": 1.31640625,
      "learning_rate": 1.1533916548786856e-05,
      "loss": -0.2434,
      "reward": 0.11607143469154835,
      "reward_std": 0.1321311704814434,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.11607143469154835,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.2522583007812,
      "epoch": 0.5240174672489083,
      "grad_norm": 1.2092971801757812,
      "kl": 1.2158203125,
      "learning_rate": 1.092268359463302e-05,
      "loss": -0.3131,
      "reward": 0.1160714328289032,
      "reward_std": 0.13385851308703423,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1160714328289032,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.825927734375,
      "epoch": 0.5414847161572053,
      "grad_norm": 2.0326733589172363,
      "kl": 2.9453125,
      "learning_rate": 1.0307950585561705e-05,
      "loss": -0.2521,
      "reward": 0.1132812574505806,
      "reward_std": 0.1328713335096836,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1132812574505806,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.2344055175781,
      "epoch": 0.5589519650655022,
      "grad_norm": 0.9160346984863281,
      "kl": 1.41796875,
      "learning_rate": 9.692049414438298e-06,
      "loss": -0.197,
      "reward": 0.11328125558793545,
      "reward_std": 0.13070819526910782,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.11328125558793545,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 561.7076110839844,
      "epoch": 0.5764192139737991,
      "grad_norm": 20.22221565246582,
      "kl": 2.130859375,
      "learning_rate": 9.07731640536698e-06,
      "loss": -0.1843,
      "reward": 0.1188616156578064,
      "reward_std": 0.13797599077224731,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1188616156578064,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.8482360839844,
      "epoch": 0.5938864628820961,
      "grad_norm": 16.17409896850586,
      "kl": 2.591796875,
      "learning_rate": 8.466083451213145e-06,
      "loss": -0.0359,
      "reward": 0.10993303917348385,
      "reward_std": 0.1322279404848814,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.10993303917348385,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 458.41297149658203,
      "epoch": 0.611353711790393,
      "grad_norm": 1.844921588897705,
      "kl": 1.6328125,
      "learning_rate": 7.860669167935028e-06,
      "loss": -0.0605,
      "reward": 0.10825893469154835,
      "reward_std": 0.13310172408819199,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.10825893469154835,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 408.12947845458984,
      "epoch": 0.62882096069869,
      "grad_norm": 3.107694149017334,
      "kl": 1.501953125,
      "learning_rate": 7.263370099279173e-06,
      "loss": -0.1977,
      "reward": 0.12667411379516125,
      "reward_std": 0.14315131679177284,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.12667411379516125,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 397.15403747558594,
      "epoch": 0.6462882096069869,
      "grad_norm": 3.9368510246276855,
      "kl": 2.07421875,
      "learning_rate": 6.6764520052034054e-06,
      "loss": -0.1921,
      "reward": 0.13950893469154835,
      "reward_std": 0.15414584055542946,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.13950893469154835,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 419.35269927978516,
      "epoch": 0.6637554585152838,
      "grad_norm": 1.3178058862686157,
      "kl": 1.17578125,
      "learning_rate": 6.102141267073207e-06,
      "loss": -0.178,
      "reward": 0.141183041036129,
      "reward_std": 0.16641410812735558,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.141183041036129,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 438.5692138671875,
      "epoch": 0.6812227074235808,
      "grad_norm": 2.4538495540618896,
      "kl": 0.7109375,
      "learning_rate": 5.542616442234618e-06,
      "loss": -0.0449,
      "reward": 0.1612723283469677,
      "reward_std": 0.15356499329209328,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1612723283469677,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 460.3995819091797,
      "epoch": 0.6986899563318777,
      "grad_norm": 0.8929517865180969,
      "kl": 0.7744140625,
      "learning_rate": 5.000000000000003e-06,
      "loss": -0.2372,
      "reward": 0.1568080447614193,
      "reward_std": 0.17059754580259323,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1568080447614193,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 496.6250305175781,
      "epoch": 0.7161572052401747,
      "grad_norm": 1.5635244846343994,
      "kl": 0.75,
      "learning_rate": 4.476350270394942e-06,
      "loss": -0.1742,
      "reward": 0.1674107201397419,
      "reward_std": 0.1668030135333538,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1674107201397419,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 544.7656707763672,
      "epoch": 0.7336244541484717,
      "grad_norm": 1.9236276149749756,
      "kl": 0.7568359375,
      "learning_rate": 3.973653636207437e-06,
      "loss": -0.1978,
      "reward": 0.1841517947614193,
      "reward_std": 0.17343608289957047,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1841517947614193,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.763427734375,
      "epoch": 0.7510917030567685,
      "grad_norm": 1.9667587280273438,
      "kl": 1.2421875,
      "learning_rate": 3.493816997957582e-06,
      "loss": -0.1336,
      "reward": 0.1830357238650322,
      "reward_std": 0.17494625598192215,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1830357238650322,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 575.8772735595703,
      "epoch": 0.7685589519650655,
      "grad_norm": 1.682228922843933,
      "kl": 1.52734375,
      "learning_rate": 3.0386605403707347e-06,
      "loss": -0.0949,
      "reward": 0.1908482238650322,
      "reward_std": 0.17441384866833687,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1908482238650322,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.6518096923828,
      "epoch": 0.7860262008733624,
      "grad_norm": 1.8132331371307373,
      "kl": 1.7109375,
      "learning_rate": 2.6099108277934105e-06,
      "loss": -0.095,
      "reward": 0.1958705447614193,
      "reward_std": 0.17568127438426018,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1958705447614193,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.0089569091797,
      "epoch": 0.8034934497816594,
      "grad_norm": 1.854785442352295,
      "kl": 1.494140625,
      "learning_rate": 2.209194254743295e-06,
      "loss": -0.118,
      "reward": 0.1791294738650322,
      "reward_std": 0.17129607126116753,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1791294738650322,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.0737152099609,
      "epoch": 0.8209606986899564,
      "grad_norm": 1.340461015701294,
      "kl": 1.1064453125,
      "learning_rate": 1.8380308764377841e-06,
      "loss": -0.1276,
      "reward": 0.2148437611758709,
      "reward_std": 0.1898629516363144,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2148437611758709,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.2924346923828,
      "epoch": 0.8384279475982532,
      "grad_norm": 1.8552510738372803,
      "kl": 0.98046875,
      "learning_rate": 1.4978286427038602e-06,
      "loss": -0.1139,
      "reward": 0.2053571529686451,
      "reward_std": 0.193530160933733,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2053571529686451,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.4910888671875,
      "epoch": 0.8558951965065502,
      "grad_norm": 3.160490036010742,
      "kl": 0.767578125,
      "learning_rate": 1.1898780571421554e-06,
      "loss": -0.0985,
      "reward": 0.1914062574505806,
      "reward_std": 0.18279264867305756,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1914062574505806,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.294677734375,
      "epoch": 0.8733624454148472,
      "grad_norm": 2.6852455139160156,
      "kl": 0.8779296875,
      "learning_rate": 9.153472818047627e-07,
      "loss": -0.1051,
      "reward": 0.1986607238650322,
      "reward_std": 0.184061449021101,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1986607238650322,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.8460083007812,
      "epoch": 0.8908296943231441,
      "grad_norm": 0.8228152394294739,
      "kl": 0.8427734375,
      "learning_rate": 6.752777059564431e-07,
      "loss": -0.0832,
      "reward": 0.2159598357975483,
      "reward_std": 0.20024307817220688,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2159598357975483,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.6049346923828,
      "epoch": 0.9082969432314411,
      "grad_norm": 1.3566585779190063,
      "kl": 0.7158203125,
      "learning_rate": 4.7057999572843516e-07,
      "loss": -0.1733,
      "reward": 0.1975446529686451,
      "reward_std": 0.18785196170210838,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.1975446529686451,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.1339416503906,
      "epoch": 0.925764192139738,
      "grad_norm": 2.234724283218384,
      "kl": 0.6767578125,
      "learning_rate": 3.020306396499062e-07,
      "loss": -0.0538,
      "reward": 0.2193080447614193,
      "reward_std": 0.19074836000800133,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2193080447614193,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.1317291259766,
      "epoch": 0.9432314410480349,
      "grad_norm": 1.5052021741867065,
      "kl": 0.810546875,
      "learning_rate": 1.7026900316098217e-07,
      "loss": -0.0754,
      "reward": 0.2243303656578064,
      "reward_std": 0.18597954511642456,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2243303656578064,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 550.7991333007812,
      "epoch": 0.9606986899563319,
      "grad_norm": 1.7257078886032104,
      "kl": 0.822265625,
      "learning_rate": 7.579490328064265e-08,
      "loss": -0.07,
      "reward": 0.2031250074505806,
      "reward_std": 0.19252333417534828,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2031250074505806,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.982177734375,
      "epoch": 0.9781659388646288,
      "grad_norm": 1.599360704421997,
      "kl": 0.736328125,
      "learning_rate": 1.896671262955896e-08,
      "loss": -0.0884,
      "reward": 0.188616082072258,
      "reward_std": 0.18718989938497543,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.188616082072258,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.84375,
      "epoch": 0.9956331877729258,
      "grad_norm": 1.2704195976257324,
      "kl": 0.8037109375,
      "learning_rate": 0.0,
      "loss": -0.0482,
      "reward": 0.2237723357975483,
      "reward_std": 0.20235323160886765,
      "rewards/format_reward": 0.0,
      "rewards/math_reward": 0.0,
      "rewards/tag_count_reward": 0.2237723357975483,
      "step": 57
    },
    {
      "epoch": 0.9956331877729258,
      "step": 57,
      "total_flos": 0.0,
      "train_loss": -0.11113634665512986,
      "train_runtime": 2287.8706,
      "train_samples_per_second": 0.699,
      "train_steps_per_second": 0.025
    }
  ],
  "logging_steps": 1,
  "max_steps": 57,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "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": 16,
  "trial_name": null,
  "trial_params": null
}