{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9974847287100251,
  "eval_steps": 100,
  "global_step": 347,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 763.18623046875,
      "epoch": 0.01437297879985627,
      "grad_norm": 0.0765276625752449,
      "kl": -6.394833326339721e-06,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.0157,
      "reward": 0.17431640625,
      "reward_std": 0.23442449774593116,
      "rewards/accuracy_reward": 0.08994140625,
      "rewards/format_reward": 0.084375,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 698.62314453125,
      "epoch": 0.02874595759971254,
      "grad_norm": 0.11506624519824982,
      "kl": 0.00981593132019043,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.0584,
      "reward": 0.6732421875,
      "reward_std": 0.3674958860501647,
      "rewards/accuracy_reward": 0.07646484375,
      "rewards/format_reward": 0.59677734375,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 541.0048828125,
      "epoch": 0.04311893639956881,
      "grad_norm": 0.050542764365673065,
      "kl": 0.02561187744140625,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.0354,
      "reward": 1.036328125,
      "reward_std": 0.2127559134736657,
      "rewards/accuracy_reward": 0.102734375,
      "rewards/format_reward": 0.93359375,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.689453125,
      "epoch": 0.05749191519942508,
      "grad_norm": 0.03859843313694,
      "kl": 0.0311004638671875,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.0249,
      "reward": 1.1552734375,
      "reward_std": 0.23164508808404208,
      "rewards/accuracy_reward": 0.20224609375,
      "rewards/format_reward": 0.95302734375,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.31357421875,
      "epoch": 0.07186489399928135,
      "grad_norm": 0.046529632061719894,
      "kl": 0.0368988037109375,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 0.0151,
      "reward": 1.157421875,
      "reward_std": 0.20364541225135327,
      "rewards/accuracy_reward": 0.188671875,
      "rewards/format_reward": 0.96875,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 671.5869140625,
      "epoch": 0.08623787279913762,
      "grad_norm": 0.037584338337183,
      "kl": 0.03684234619140625,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 0.0213,
      "reward": 1.165234375,
      "reward_std": 0.24268896747380495,
      "rewards/accuracy_reward": 0.2177734375,
      "rewards/format_reward": 0.9474609375,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.013671875,
      "epoch": 0.1006108515989939,
      "grad_norm": 0.34336549043655396,
      "kl": 0.151519775390625,
      "learning_rate": 2e-05,
      "loss": 0.0359,
      "reward": 1.14091796875,
      "reward_std": 0.2653762998059392,
      "rewards/accuracy_reward": 0.2052734375,
      "rewards/format_reward": 0.93564453125,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 624.95908203125,
      "epoch": 0.11498383039885016,
      "grad_norm": 0.052160657942295074,
      "kl": 0.595263671875,
      "learning_rate": 1.9987329060020616e-05,
      "loss": 0.0668,
      "reward": 1.081640625,
      "reward_std": 0.3258050443604589,
      "rewards/accuracy_reward": 0.20439453125,
      "rewards/format_reward": 0.87724609375,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 691.025390625,
      "epoch": 0.12935680919870643,
      "grad_norm": 0.27032357454299927,
      "kl": 0.1677734375,
      "learning_rate": 1.9949348350626456e-05,
      "loss": 0.034,
      "reward": 0.9642578125,
      "reward_std": 0.4391048148274422,
      "rewards/accuracy_reward": 0.18740234375,
      "rewards/format_reward": 0.77685546875,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 440.34853515625,
      "epoch": 0.1437297879985627,
      "grad_norm": 0.6052369475364685,
      "kl": 0.800189208984375,
      "learning_rate": 1.9886154122075344e-05,
      "loss": 0.0919,
      "reward": 0.89814453125,
      "reward_std": 0.38281605690717696,
      "rewards/accuracy_reward": 0.11865234375,
      "rewards/format_reward": 0.7794921875,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.60771484375,
      "epoch": 0.15810276679841898,
      "grad_norm": 0.28784340620040894,
      "kl": 2.12225341796875,
      "learning_rate": 1.979790652042268e-05,
      "loss": 0.1039,
      "reward": 0.85263671875,
      "reward_std": 0.4635654494166374,
      "rewards/accuracy_reward": 0.13447265625,
      "rewards/format_reward": 0.7181640625,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 718.88359375,
      "epoch": 0.17247574559827525,
      "grad_norm": 0.38119208812713623,
      "kl": 0.38172607421875,
      "learning_rate": 1.9684829181681236e-05,
      "loss": 0.0502,
      "reward": 1.06494140625,
      "reward_std": 0.3414448471739888,
      "rewards/accuracy_reward": 0.21650390625,
      "rewards/format_reward": 0.8484375,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.63818359375,
      "epoch": 0.18684872439813152,
      "grad_norm": 0.3849119246006012,
      "kl": 1.819970703125,
      "learning_rate": 1.954720866508546e-05,
      "loss": 0.1892,
      "reward": 0.9689453125,
      "reward_std": 0.4041255243122578,
      "rewards/accuracy_reward": 0.16826171875,
      "rewards/format_reward": 0.80068359375,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.92744140625,
      "epoch": 0.2012217031979878,
      "grad_norm": 0.16367273032665253,
      "kl": 0.688922119140625,
      "learning_rate": 1.9385393726896492e-05,
      "loss": 0.0581,
      "reward": 1.1560546875,
      "reward_std": 0.22550129257142543,
      "rewards/accuracy_reward": 0.19248046875,
      "rewards/format_reward": 0.96357421875,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 563.1029296875,
      "epoch": 0.21559468199784404,
      "grad_norm": 0.1713869571685791,
      "kl": 0.0900238037109375,
      "learning_rate": 1.9199794436588244e-05,
      "loss": 0.0071,
      "reward": 1.1892578125,
      "reward_std": 0.2032089052721858,
      "rewards/accuracy_reward": 0.21513671875,
      "rewards/format_reward": 0.97412109375,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.5634765625,
      "epoch": 0.2299676607977003,
      "grad_norm": 0.2464917004108429,
      "kl": 0.144158935546875,
      "learning_rate": 1.899088113765426e-05,
      "loss": 0.0189,
      "reward": 1.1546875,
      "reward_std": 0.2610320156440139,
      "rewards/accuracy_reward": 0.21083984375,
      "rewards/format_reward": 0.94384765625,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 652.05927734375,
      "epoch": 0.24434063959755659,
      "grad_norm": 0.2248377948999405,
      "kl": 0.716436767578125,
      "learning_rate": 1.875918325566888e-05,
      "loss": 0.0578,
      "reward": 1.06005859375,
      "reward_std": 0.33321408815681935,
      "rewards/accuracy_reward": 0.171484375,
      "rewards/format_reward": 0.88857421875,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 669.3861328125,
      "epoch": 0.25871361839741286,
      "grad_norm": 0.27829509973526,
      "kl": 0.617529296875,
      "learning_rate": 1.8505287956623298e-05,
      "loss": 0.0585,
      "reward": 1.14755859375,
      "reward_std": 0.2751380069181323,
      "rewards/accuracy_reward": 0.20859375,
      "rewards/format_reward": 0.93896484375,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.1236328125,
      "epoch": 0.27308659719726913,
      "grad_norm": 0.2939702868461609,
      "kl": 0.5397705078125,
      "learning_rate": 1.8229838658936566e-05,
      "loss": 0.0555,
      "reward": 1.137890625,
      "reward_std": 0.2469838338904083,
      "rewards/accuracy_reward": 0.1900390625,
      "rewards/format_reward": 0.9478515625,
      "step": 95
    },
    {
      "epoch": 0.2874595759971254,
      "grad_norm": 0.1728806495666504,
      "learning_rate": 1.7933533402912354e-05,
      "loss": 0.103,
      "step": 100
    },
    {
      "epoch": 0.2874595759971254,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 611.2384828951579,
      "eval_kl": 0.50033329778157,
      "eval_loss": 0.06100574508309364,
      "eval_reward": 1.1420381825938566,
      "eval_reward_std": 0.27033696519433437,
      "eval_rewards/accuracy_reward": 0.2020051194539249,
      "eval_rewards/format_reward": 0.9400330631399317,
      "eval_runtime": 16336.0108,
      "eval_samples_per_second": 0.287,
      "eval_steps_per_second": 0.002,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 629.018017578125,
      "epoch": 0.3018325547969817,
      "grad_norm": 0.1207083985209465,
      "kl": 1.06016845703125,
      "learning_rate": 1.761712308177359e-05,
      "loss": 0.1074,
      "reward": 1.059326171875,
      "reward_std": 0.35213989242911337,
      "rewards/accuracy_reward": 0.18974609375,
      "rewards/format_reward": 0.869580078125,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.68330078125,
      "epoch": 0.31620553359683795,
      "grad_norm": 0.12369602918624878,
      "kl": 2.13466796875,
      "learning_rate": 1.7281409538757886e-05,
      "loss": 0.1546,
      "reward": 1.06484375,
      "reward_std": 0.3502559883520007,
      "rewards/accuracy_reward": 0.1806640625,
      "rewards/format_reward": 0.8841796875,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.0666015625,
      "epoch": 0.3305785123966942,
      "grad_norm": 0.13101035356521606,
      "kl": 0.932763671875,
      "learning_rate": 1.6927243535095995e-05,
      "loss": 0.0856,
      "reward": 1.14521484375,
      "reward_std": 0.2656426582485437,
      "rewards/accuracy_reward": 0.20322265625,
      "rewards/format_reward": 0.9419921875,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.12802734375,
      "epoch": 0.3449514911965505,
      "grad_norm": 0.13193248212337494,
      "kl": 0.9656982421875,
      "learning_rate": 1.655552259402295e-05,
      "loss": 0.0881,
      "reward": 1.14560546875,
      "reward_std": 0.27462361557409165,
      "rewards/accuracy_reward": 0.21337890625,
      "rewards/format_reward": 0.9322265625,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 653.04599609375,
      "epoch": 0.35932446999640677,
      "grad_norm": 0.3534374535083771,
      "kl": 1.867626953125,
      "learning_rate": 1.6167188726285433e-05,
      "loss": 0.1558,
      "reward": 1.05126953125,
      "reward_std": 0.36074890177696944,
      "rewards/accuracy_reward": 0.18544921875,
      "rewards/format_reward": 0.8658203125,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.14736328125,
      "epoch": 0.37369744879626304,
      "grad_norm": 2.0081052780151367,
      "kl": 1.8935546875,
      "learning_rate": 1.5763226042909455e-05,
      "loss": 0.1105,
      "reward": 1.0998046875,
      "reward_std": 0.3096121703274548,
      "rewards/accuracy_reward": 0.18486328125,
      "rewards/format_reward": 0.91494140625,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.62197265625,
      "epoch": 0.3880704275961193,
      "grad_norm": 0.1118120476603508,
      "kl": 0.59337158203125,
      "learning_rate": 1.5344658261278013e-05,
      "loss": 0.031,
      "reward": 1.16611328125,
      "reward_std": 0.24496497269719839,
      "rewards/accuracy_reward": 0.21005859375,
      "rewards/format_reward": 0.9560546875,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.4009765625,
      "epoch": 0.4024434063959756,
      "grad_norm": 0.18786092102527618,
      "kl": 0.7201416015625,
      "learning_rate": 1.4912546110838775e-05,
      "loss": 0.0608,
      "reward": 1.1451171875,
      "reward_std": 0.2563774929381907,
      "rewards/accuracy_reward": 0.2021484375,
      "rewards/format_reward": 0.94296875,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 604.14072265625,
      "epoch": 0.41681638519583186,
      "grad_norm": 0.12442336976528168,
      "kl": 0.96689453125,
      "learning_rate": 1.4467984645016259e-05,
      "loss": 0.0834,
      "reward": 1.13984375,
      "reward_std": 0.2728093104436994,
      "rewards/accuracy_reward": 0.2001953125,
      "rewards/format_reward": 0.9396484375,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.0369140625,
      "epoch": 0.4311893639956881,
      "grad_norm": 0.17537765204906464,
      "kl": 0.687255859375,
      "learning_rate": 1.4012100466140579e-05,
      "loss": 0.0628,
      "reward": 1.12919921875,
      "reward_std": 0.24853361072018743,
      "rewards/accuracy_reward": 0.17646484375,
      "rewards/format_reward": 0.952734375,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.915234375,
      "epoch": 0.44556234279554435,
      "grad_norm": 0.11783521622419357,
      "kl": 0.83641357421875,
      "learning_rate": 1.3546048870425356e-05,
      "loss": 0.0734,
      "reward": 1.12666015625,
      "reward_std": 0.264958731085062,
      "rewards/accuracy_reward": 0.18427734375,
      "rewards/format_reward": 0.9423828125,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.70517578125,
      "epoch": 0.4599353215954006,
      "grad_norm": 0.13742466270923615,
      "kl": 0.7468505859375,
      "learning_rate": 1.3071010920229909e-05,
      "loss": 0.0682,
      "reward": 1.122265625,
      "reward_std": 0.2766525615006685,
      "rewards/accuracy_reward": 0.18798828125,
      "rewards/format_reward": 0.93427734375,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.36552734375,
      "epoch": 0.4743083003952569,
      "grad_norm": 0.4238876700401306,
      "kl": 1.381640625,
      "learning_rate": 1.2588190451025209e-05,
      "loss": 0.1039,
      "reward": 1.13544921875,
      "reward_std": 0.31343956142663953,
      "rewards/accuracy_reward": 0.2201171875,
      "rewards/format_reward": 0.91533203125,
      "step": 165
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 672.3779296875,
      "epoch": 0.48868127919511317,
      "grad_norm": 0.13015827536582947,
      "kl": 1.4199462890625,
      "learning_rate": 1.2098811020648475e-05,
      "loss": 0.0989,
      "reward": 1.11416015625,
      "reward_std": 0.3195471292361617,
      "rewards/accuracy_reward": 0.208203125,
      "rewards/format_reward": 0.90595703125,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.84326171875,
      "epoch": 0.5030542579949695,
      "grad_norm": 0.2257327437400818,
      "kl": 1.1652099609375,
      "learning_rate": 1.1604112808577603e-05,
      "loss": 0.101,
      "reward": 1.1236328125,
      "reward_std": 0.30357036273926497,
      "rewards/accuracy_reward": 0.211328125,
      "rewards/format_reward": 0.9123046875,
      "step": 175
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.3265625,
      "epoch": 0.5174272367948257,
      "grad_norm": 0.11806362867355347,
      "kl": 0.7406005859375,
      "learning_rate": 1.11053494730832e-05,
      "loss": 0.0699,
      "reward": 1.1373046875,
      "reward_std": 0.25564199751242994,
      "rewards/accuracy_reward": 0.19658203125,
      "rewards/format_reward": 0.94072265625,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 652.4876953125,
      "epoch": 0.531800215594682,
      "grad_norm": 0.12807710468769073,
      "kl": 0.58621826171875,
      "learning_rate": 1.0603784974222862e-05,
      "loss": 0.0587,
      "reward": 1.173046875,
      "reward_std": 0.26026681158691645,
      "rewards/accuracy_reward": 0.2248046875,
      "rewards/format_reward": 0.9482421875,
      "step": 185
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.433203125,
      "epoch": 0.5461731943945383,
      "grad_norm": 0.10217402130365372,
      "kl": 0.9344970703125,
      "learning_rate": 1.0100690370728756e-05,
      "loss": 0.0809,
      "reward": 1.1609375,
      "reward_std": 0.2667428271844983,
      "rewards/accuracy_reward": 0.2150390625,
      "rewards/format_reward": 0.9458984375,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.68701171875,
      "epoch": 0.5605461731943946,
      "grad_norm": 0.13498954474925995,
      "kl": 0.67510986328125,
      "learning_rate": 9.597340598905851e-06,
      "loss": 0.0603,
      "reward": 1.1654296875,
      "reward_std": 0.25683426298201084,
      "rewards/accuracy_reward": 0.21796875,
      "rewards/format_reward": 0.9474609375,
      "step": 195
    },
    {
      "epoch": 0.5749191519942508,
      "grad_norm": 0.1882268339395523,
      "learning_rate": 9.095011241703623e-06,
      "loss": 0.0719,
      "step": 200
    },
    {
      "epoch": 0.5749191519942508,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 658.4643835907503,
      "eval_kl": 0.7806433980375427,
      "eval_loss": 0.06092459335923195,
      "eval_reward": 1.149637372013652,
      "eval_reward_std": 0.27955490747409467,
      "eval_rewards/accuracy_reward": 0.2150170648464164,
      "eval_rewards/format_reward": 0.9346203071672355,
      "eval_runtime": 16414.3395,
      "eval_samples_per_second": 0.286,
      "eval_steps_per_second": 0.002,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 652.56220703125,
      "epoch": 0.589292130794107,
      "grad_norm": 0.1547040194272995,
      "kl": 0.93699951171875,
      "learning_rate": 8.594975296149076e-06,
      "loss": 0.0647,
      "reward": 1.1623046875,
      "reward_std": 0.28741056518629193,
      "rewards/accuracy_reward": 0.23125,
      "rewards/format_reward": 0.9310546875,
      "step": 205
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 674.5626953125,
      "epoch": 0.6036651095939634,
      "grad_norm": 0.25151509046554565,
      "kl": 0.9999267578125,
      "learning_rate": 8.098499947332935e-06,
      "loss": 0.0775,
      "reward": 1.1466796875,
      "reward_std": 0.30369703844189644,
      "rewards/accuracy_reward": 0.22509765625,
      "rewards/format_reward": 0.92158203125,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 676.977734375,
      "epoch": 0.6180380883938196,
      "grad_norm": 0.20043928921222687,
      "kl": 0.7748779296875,
      "learning_rate": 7.606843357124426e-06,
      "loss": 0.0573,
      "reward": 1.15302734375,
      "reward_std": 0.28829708844423296,
      "rewards/accuracy_reward": 0.2244140625,
      "rewards/format_reward": 0.92861328125,
      "step": 215
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.955078125,
      "epoch": 0.6324110671936759,
      "grad_norm": 0.12682239711284637,
      "kl": 0.7095947265625,
      "learning_rate": 7.12125147575254e-06,
      "loss": 0.0548,
      "reward": 1.1763671875,
      "reward_std": 0.25821941047906877,
      "rewards/accuracy_reward": 0.23046875,
      "rewards/format_reward": 0.9458984375,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 647.839453125,
      "epoch": 0.6467840459935321,
      "grad_norm": 0.13890360295772552,
      "kl": 0.63245849609375,
      "learning_rate": 6.6429548843339554e-06,
      "loss": 0.0502,
      "reward": 1.1654296875,
      "reward_std": 0.2512395134195685,
      "rewards/accuracy_reward": 0.21337890625,
      "rewards/format_reward": 0.95205078125,
      "step": 225
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.53623046875,
      "epoch": 0.6611570247933884,
      "grad_norm": 0.15598197281360626,
      "kl": 0.87559814453125,
      "learning_rate": 6.173165676349103e-06,
      "loss": 0.0703,
      "reward": 1.155078125,
      "reward_std": 0.2729664742946625,
      "rewards/accuracy_reward": 0.213671875,
      "rewards/format_reward": 0.94140625,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.5361328125,
      "epoch": 0.6755300035932447,
      "grad_norm": 0.15446113049983978,
      "kl": 0.77437744140625,
      "learning_rate": 5.713074385969457e-06,
      "loss": 0.0688,
      "reward": 1.16953125,
      "reward_std": 0.28331395238637924,
      "rewards/accuracy_reward": 0.2296875,
      "rewards/format_reward": 0.93984375,
      "step": 235
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 647.3654296875,
      "epoch": 0.689902982393101,
      "grad_norm": 0.2089157998561859,
      "kl": 1.21328125,
      "learning_rate": 5.263846971020108e-06,
      "loss": 0.1016,
      "reward": 1.116796875,
      "reward_std": 0.31174491699784995,
      "rewards/accuracy_reward": 0.2029296875,
      "rewards/format_reward": 0.9138671875,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.523828125,
      "epoch": 0.7042759611929572,
      "grad_norm": 0.16784484684467316,
      "kl": 0.791552734375,
      "learning_rate": 4.826621858223431e-06,
      "loss": 0.0734,
      "reward": 1.143359375,
      "reward_std": 0.28859285488724706,
      "rewards/accuracy_reward": 0.2154296875,
      "rewards/format_reward": 0.9279296875,
      "step": 245
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.52958984375,
      "epoch": 0.7186489399928135,
      "grad_norm": 0.1753949671983719,
      "kl": 0.98125,
      "learning_rate": 4.40250705821178e-06,
      "loss": 0.0812,
      "reward": 1.1546875,
      "reward_std": 0.2736880548298359,
      "rewards/accuracy_reward": 0.2154296875,
      "rewards/format_reward": 0.9392578125,
      "step": 250
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 635.1673828125,
      "epoch": 0.7330219187926698,
      "grad_norm": 0.20336733758449554,
      "kl": 0.55863037109375,
      "learning_rate": 3.99257735762021e-06,
      "loss": 0.0458,
      "reward": 1.17392578125,
      "reward_std": 0.23981231823563576,
      "rewards/accuracy_reward": 0.21728515625,
      "rewards/format_reward": 0.956640625,
      "step": 255
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.9087890625,
      "epoch": 0.7473948975925261,
      "grad_norm": 0.16080701351165771,
      "kl": 0.696923828125,
      "learning_rate": 3.5978715953751207e-06,
      "loss": 0.0567,
      "reward": 1.1685546875,
      "reward_std": 0.24907034020870925,
      "rewards/accuracy_reward": 0.21376953125,
      "rewards/format_reward": 0.95478515625,
      "step": 260
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.18583984375,
      "epoch": 0.7617678763923823,
      "grad_norm": 0.18338614702224731,
      "kl": 1.0648681640625,
      "learning_rate": 3.2193900300810908e-06,
      "loss": 0.0778,
      "reward": 1.151953125,
      "reward_std": 0.26931764371693134,
      "rewards/accuracy_reward": 0.210546875,
      "rewards/format_reward": 0.94140625,
      "step": 265
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 634.9572265625,
      "epoch": 0.7761408551922386,
      "grad_norm": 0.13022945821285248,
      "kl": 0.7796142578125,
      "learning_rate": 2.8580918051775542e-06,
      "loss": 0.065,
      "reward": 1.165625,
      "reward_std": 0.27459610607475043,
      "rewards/accuracy_reward": 0.2244140625,
      "rewards/format_reward": 0.9412109375,
      "step": 270
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.88232421875,
      "epoch": 0.7905138339920948,
      "grad_norm": 0.1719103306531906,
      "kl": 0.8088623046875,
      "learning_rate": 2.514892518288988e-06,
      "loss": 0.0696,
      "reward": 1.15087890625,
      "reward_std": 0.2822716049849987,
      "rewards/accuracy_reward": 0.21640625,
      "rewards/format_reward": 0.93447265625,
      "step": 275
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.853125,
      "epoch": 0.8048868127919512,
      "grad_norm": 0.21337589621543884,
      "kl": 0.9040283203125,
      "learning_rate": 2.190661900928426e-06,
      "loss": 0.0753,
      "reward": 1.1412109375,
      "reward_std": 0.2784146698191762,
      "rewards/accuracy_reward": 0.2029296875,
      "rewards/format_reward": 0.93828125,
      "step": 280
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.5638671875,
      "epoch": 0.8192597915918074,
      "grad_norm": 0.1362425535917282,
      "kl": 0.95645751953125,
      "learning_rate": 1.8862216144342692e-06,
      "loss": 0.0749,
      "reward": 1.14130859375,
      "reward_std": 0.2679125562310219,
      "rewards/accuracy_reward": 0.20546875,
      "rewards/format_reward": 0.93583984375,
      "step": 285
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.32958984375,
      "epoch": 0.8336327703916637,
      "grad_norm": 0.13494881987571716,
      "kl": 0.8051513671875,
      "learning_rate": 1.6023431677260215e-06,
      "loss": 0.0684,
      "reward": 1.16240234375,
      "reward_std": 0.26225354727357625,
      "rewards/accuracy_reward": 0.21396484375,
      "rewards/format_reward": 0.9484375,
      "step": 290
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.16083984375,
      "epoch": 0.8480057491915199,
      "grad_norm": 0.16026277840137482,
      "kl": 0.8697265625,
      "learning_rate": 1.339745962155613e-06,
      "loss": 0.0712,
      "reward": 1.15966796875,
      "reward_std": 0.2733839010819793,
      "rewards/accuracy_reward": 0.21552734375,
      "rewards/format_reward": 0.944140625,
      "step": 295
    },
    {
      "epoch": 0.8623787279913762,
      "grad_norm": 0.155064195394516,
      "learning_rate": 1.099095468409156e-06,
      "loss": 0.0785,
      "step": 300
    },
    {
      "epoch": 0.8623787279913762,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 628.6954391531569,
      "eval_kl": 0.8880319432593856,
      "eval_loss": 0.07323075085878372,
      "eval_reward": 1.1617160836177474,
      "eval_reward_std": 0.2670084892838888,
      "eval_rewards/accuracy_reward": 0.21819005972696245,
      "eval_rewards/format_reward": 0.943526023890785,
      "eval_runtime": 16336.911,
      "eval_samples_per_second": 0.287,
      "eval_steps_per_second": 0.002,
      "step": 300
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.594287109375,
      "epoch": 0.8767517067912325,
      "grad_norm": 0.1458193063735962,
      "kl": 0.95950927734375,
      "learning_rate": 8.810015400790994e-07,
      "loss": 0.0809,
      "reward": 1.16162109375,
      "reward_std": 0.26864673662930727,
      "rewards/accuracy_reward": 0.2203125,
      "rewards/format_reward": 0.94130859375,
      "step": 305
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.09326171875,
      "epoch": 0.8911246855910887,
      "grad_norm": 0.14581456780433655,
      "kl": 0.82933349609375,
      "learning_rate": 6.860168681805946e-07,
      "loss": 0.0661,
      "reward": 1.16982421875,
      "reward_std": 0.26240854635834693,
      "rewards/accuracy_reward": 0.2216796875,
      "rewards/format_reward": 0.94814453125,
      "step": 310
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.76943359375,
      "epoch": 0.905497664390945,
      "grad_norm": 0.16072359681129456,
      "kl": 0.80216064453125,
      "learning_rate": 5.146355805285452e-07,
      "loss": 0.0637,
      "reward": 1.17431640625,
      "reward_std": 0.2672739554196596,
      "rewards/accuracy_reward": 0.22734375,
      "rewards/format_reward": 0.94697265625,
      "step": 315
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.221484375,
      "epoch": 0.9198706431908013,
      "grad_norm": 0.1719951331615448,
      "kl": 0.84737548828125,
      "learning_rate": 3.6729198952483725e-07,
      "loss": 0.0748,
      "reward": 1.158203125,
      "reward_std": 0.2642348381690681,
      "rewards/accuracy_reward": 0.2169921875,
      "rewards/format_reward": 0.9412109375,
      "step": 320
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.33896484375,
      "epoch": 0.9342436219906576,
      "grad_norm": 0.20107921957969666,
      "kl": 0.87275390625,
      "learning_rate": 2.4435949152906144e-07,
      "loss": 0.0757,
      "reward": 1.15966796875,
      "reward_std": 0.27580115627497437,
      "rewards/accuracy_reward": 0.22021484375,
      "rewards/format_reward": 0.939453125,
      "step": 325
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 643.75712890625,
      "epoch": 0.9486166007905138,
      "grad_norm": 0.14510348439216614,
      "kl": 0.821826171875,
      "learning_rate": 1.4614962060194303e-07,
      "loss": 0.0658,
      "reward": 1.140625,
      "reward_std": 0.2549537133425474,
      "rewards/accuracy_reward": 0.1978515625,
      "rewards/format_reward": 0.9427734375,
      "step": 330
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 643.530078125,
      "epoch": 0.9629895795903701,
      "grad_norm": 0.14030759036540985,
      "kl": 0.77998046875,
      "learning_rate": 7.291125901946027e-08,
      "loss": 0.0701,
      "reward": 1.1693359375,
      "reward_std": 0.2593334957957268,
      "rewards/accuracy_reward": 0.22529296875,
      "rewards/format_reward": 0.94404296875,
      "step": 335
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.19736328125,
      "epoch": 0.9773625583902263,
      "grad_norm": 0.16929689049720764,
      "kl": 0.835546875,
      "learning_rate": 2.4830006558373975e-08,
      "loss": 0.0697,
      "reward": 1.162109375,
      "reward_std": 0.26842295806854966,
      "rewards/accuracy_reward": 0.2173828125,
      "rewards/format_reward": 0.9447265625,
      "step": 340
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.66806640625,
      "epoch": 0.9917355371900827,
      "grad_norm": 0.17319317162036896,
      "kl": 0.85125732421875,
      "learning_rate": 2.0277101514987184e-09,
      "loss": 0.0724,
      "reward": 1.15966796875,
      "reward_std": 0.2777851399034262,
      "rewards/accuracy_reward": 0.21982421875,
      "rewards/format_reward": 0.93984375,
      "step": 345
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 643.1118812561035,
      "epoch": 0.9974847287100251,
      "kl": 0.8223876953125,
      "reward": 1.182861328125,
      "reward_std": 0.2758036791346967,
      "rewards/accuracy_reward": 0.25390625,
      "rewards/format_reward": 0.928955078125,
      "step": 347,
      "total_flos": 0.0,
      "train_loss": 0.0704507840104852,
      "train_runtime": 435678.5475,
      "train_samples_per_second": 0.204,
      "train_steps_per_second": 0.001
    }
  ],
  "logging_steps": 5,
  "max_steps": 347,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}