{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 36,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 3466.5059814453125,
      "epoch": 0.027972027972027972,
      "grad_norm": 0.8895487189292908,
      "kl": 0.0,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 0.4657738283276558,
      "reward_std": 0.11145408265292645,
      "rewards/accuracy_reward": 0.08630952658131719,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.379464291036129,
      "step": 1
    },
    {
      "completion_length": 3675.1995239257812,
      "epoch": 0.055944055944055944,
      "grad_norm": 0.5455259680747986,
      "kl": 0.0,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "reward": 0.473214291036129,
      "reward_std": 0.16795706376433372,
      "rewards/accuracy_reward": 0.09523809934034944,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.377976194024086,
      "step": 2
    },
    {
      "completion_length": 3365.479248046875,
      "epoch": 0.08391608391608392,
      "grad_norm": 0.5860073566436768,
      "kl": 0.0010356903076171875,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "reward": 0.5974702388048172,
      "reward_std": 0.14405668526887894,
      "rewards/accuracy_reward": 0.15476190694607794,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4427083432674408,
      "step": 3
    },
    {
      "completion_length": 3294.0059814453125,
      "epoch": 0.11188811188811189,
      "grad_norm": 0.9189674854278564,
      "kl": 0.012176513671875,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.0005,
      "reward": 0.5416666716337204,
      "reward_std": 0.17101533710956573,
      "rewards/accuracy_reward": 0.10119047877378762,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.440476194024086,
      "step": 4
    },
    {
      "completion_length": 3521.1310424804688,
      "epoch": 0.13986013986013987,
      "grad_norm": 0.6414484977722168,
      "kl": 0.028106689453125,
      "learning_rate": 2e-05,
      "loss": 0.0011,
      "reward": 0.5267857313156128,
      "reward_std": 0.13104644231498241,
      "rewards/accuracy_reward": 0.1428571492433548,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.383928582072258,
      "step": 5
    },
    {
      "completion_length": 3423.699462890625,
      "epoch": 0.16783216783216784,
      "grad_norm": 0.297514408826828,
      "kl": 0.07110595703125,
      "learning_rate": 1.995184726672197e-05,
      "loss": 0.0028,
      "reward": 0.5029762089252472,
      "reward_std": 0.21673721447587013,
      "rewards/accuracy_reward": 0.0922619067132473,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.410714291036129,
      "step": 6
    },
    {
      "completion_length": 3670.21435546875,
      "epoch": 0.1958041958041958,
      "grad_norm": 0.21352989971637726,
      "kl": 0.093505859375,
      "learning_rate": 1.9807852804032306e-05,
      "loss": 0.0037,
      "reward": 0.5223214328289032,
      "reward_std": 0.16733192279934883,
      "rewards/accuracy_reward": 0.1369047649204731,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.385416679084301,
      "step": 7
    },
    {
      "completion_length": 3450.9880981445312,
      "epoch": 0.22377622377622378,
      "grad_norm": 0.17398180067539215,
      "kl": 0.14306640625,
      "learning_rate": 1.956940335732209e-05,
      "loss": 0.0057,
      "reward": 0.547619067132473,
      "reward_std": 0.1912310989573598,
      "rewards/accuracy_reward": 0.14285714738070965,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4047619104385376,
      "step": 8
    },
    {
      "completion_length": 3476.6488647460938,
      "epoch": 0.2517482517482518,
      "grad_norm": 0.14767655730247498,
      "kl": 0.17724609375,
      "learning_rate": 1.9238795325112867e-05,
      "loss": 0.0071,
      "reward": 0.60863097012043,
      "reward_std": 0.13831812981516123,
      "rewards/accuracy_reward": 0.2023809552192688,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4062500074505806,
      "step": 9
    },
    {
      "completion_length": 3633.9048461914062,
      "epoch": 0.27972027972027974,
      "grad_norm": 0.16100460290908813,
      "kl": 0.22509765625,
      "learning_rate": 1.881921264348355e-05,
      "loss": 0.009,
      "reward": 0.5952381044626236,
      "reward_std": 0.15078665129840374,
      "rewards/accuracy_reward": 0.1934523843228817,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4017857238650322,
      "step": 10
    },
    {
      "completion_length": 3686.4376220703125,
      "epoch": 0.3076923076923077,
      "grad_norm": 0.17172181606292725,
      "kl": 0.275634765625,
      "learning_rate": 1.8314696123025456e-05,
      "loss": 0.011,
      "reward": 0.462053582072258,
      "reward_std": 0.09804809279739857,
      "rewards/accuracy_reward": 0.0892857164144516,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3727678656578064,
      "step": 11
    },
    {
      "completion_length": 3672.27392578125,
      "epoch": 0.3356643356643357,
      "grad_norm": 0.1871335655450821,
      "kl": 0.3203125,
      "learning_rate": 1.773010453362737e-05,
      "loss": 0.0128,
      "reward": 0.4568452462553978,
      "reward_std": 0.12699773162603378,
      "rewards/accuracy_reward": 0.08333333511836827,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3735119178891182,
      "step": 12
    },
    {
      "completion_length": 3601.2619018554688,
      "epoch": 0.36363636363636365,
      "grad_norm": 0.18339282274246216,
      "kl": 0.29345703125,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 0.0117,
      "reward": 0.6927083358168602,
      "reward_std": 0.1693347617983818,
      "rewards/accuracy_reward": 0.3125000111758709,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3802083358168602,
      "step": 13
    },
    {
      "completion_length": 3655.6220703125,
      "epoch": 0.3916083916083916,
      "grad_norm": 0.2542259693145752,
      "kl": 0.31201171875,
      "learning_rate": 1.6343932841636455e-05,
      "loss": 0.0125,
      "reward": 0.7633928656578064,
      "reward_std": 0.2515065036714077,
      "rewards/accuracy_reward": 0.3958333469927311,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3675595223903656,
      "step": 14
    },
    {
      "completion_length": 1954.6220397949219,
      "epoch": 0.4195804195804196,
      "grad_norm": 64.4857177734375,
      "kl": 1.3046875,
      "learning_rate": 1.5555702330196024e-05,
      "loss": 0.0522,
      "reward": 0.712053582072258,
      "reward_std": 0.3067754730582237,
      "rewards/accuracy_reward": 0.2440476194024086,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4680059552192688,
      "step": 15
    },
    {
      "completion_length": 3941.5655517578125,
      "epoch": 0.44755244755244755,
      "grad_norm": 0.17128266394138336,
      "kl": 0.34619140625,
      "learning_rate": 1.4713967368259981e-05,
      "loss": 0.0138,
      "reward": 0.6443452462553978,
      "reward_std": 0.24465123564004898,
      "rewards/accuracy_reward": 0.3005952388048172,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3437500074505806,
      "step": 16
    },
    {
      "completion_length": 3827.9702758789062,
      "epoch": 0.4755244755244755,
      "grad_norm": 13.349394798278809,
      "kl": 0.435546875,
      "learning_rate": 1.3826834323650899e-05,
      "loss": 0.0174,
      "reward": 0.6183035746216774,
      "reward_std": 0.22436635196208954,
      "rewards/accuracy_reward": 0.27678572852164507,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3415178582072258,
      "step": 17
    },
    {
      "completion_length": 3816.27685546875,
      "epoch": 0.5034965034965035,
      "grad_norm": 2.260338544845581,
      "kl": 0.365234375,
      "learning_rate": 1.2902846772544625e-05,
      "loss": 0.0146,
      "reward": 0.6138392984867096,
      "reward_std": 0.21694295294582844,
      "rewards/accuracy_reward": 0.2678571483120322,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3459821492433548,
      "step": 18
    },
    {
      "completion_length": 3398.71142578125,
      "epoch": 0.5314685314685315,
      "grad_norm": 0.17236056923866272,
      "kl": 0.34814453125,
      "learning_rate": 1.1950903220161286e-05,
      "loss": 0.0139,
      "reward": 0.7968750149011612,
      "reward_std": 0.2352231778204441,
      "rewards/accuracy_reward": 0.4196428582072258,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3772321492433548,
      "step": 19
    },
    {
      "completion_length": 3874.4227294921875,
      "epoch": 0.5594405594405595,
      "grad_norm": 0.1826857030391693,
      "kl": 0.37646484375,
      "learning_rate": 1.098017140329561e-05,
      "loss": 0.015,
      "reward": 0.6227678656578064,
      "reward_std": 0.2774837426841259,
      "rewards/accuracy_reward": 0.2738095261156559,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3489583358168602,
      "step": 20
    },
    {
      "completion_length": 3617.1904907226562,
      "epoch": 0.5874125874125874,
      "grad_norm": 1.202433466911316,
      "kl": 0.333984375,
      "learning_rate": 1e-05,
      "loss": 0.0134,
      "reward": 0.7008928805589676,
      "reward_std": 0.2767525836825371,
      "rewards/accuracy_reward": 0.3392857201397419,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3616071492433548,
      "step": 21
    },
    {
      "completion_length": 3817.994140625,
      "epoch": 0.6153846153846154,
      "grad_norm": 1.7046478986740112,
      "kl": 0.34375,
      "learning_rate": 9.019828596704394e-06,
      "loss": 0.0138,
      "reward": 0.654017873108387,
      "reward_std": 0.24345579743385315,
      "rewards/accuracy_reward": 0.29761905409395695,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3563988134264946,
      "step": 22
    },
    {
      "completion_length": 3474.8810424804688,
      "epoch": 0.6433566433566433,
      "grad_norm": 3.3308699131011963,
      "kl": 0.3984375,
      "learning_rate": 8.04909677983872e-06,
      "loss": 0.0159,
      "reward": 0.8549107313156128,
      "reward_std": 0.251787431538105,
      "rewards/accuracy_reward": 0.476190485060215,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3787202462553978,
      "step": 23
    },
    {
      "completion_length": 3292.071533203125,
      "epoch": 0.6713286713286714,
      "grad_norm": 0.8111428022384644,
      "kl": 0.34228515625,
      "learning_rate": 7.097153227455379e-06,
      "loss": 0.0137,
      "reward": 0.8050595372915268,
      "reward_std": 0.2745497487485409,
      "rewards/accuracy_reward": 0.416666679084301,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3883928656578064,
      "step": 24
    },
    {
      "completion_length": 3325.571533203125,
      "epoch": 0.6993006993006993,
      "grad_norm": 0.2051524817943573,
      "kl": 0.33544921875,
      "learning_rate": 6.173165676349103e-06,
      "loss": 0.0134,
      "reward": 0.8095238208770752,
      "reward_std": 0.2951691634953022,
      "rewards/accuracy_reward": 0.4166666716337204,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3928571492433548,
      "step": 25
    },
    {
      "completion_length": 3417.4406127929688,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.1402244120836258,
      "kl": 0.3544921875,
      "learning_rate": 5.286032631740023e-06,
      "loss": 0.0142,
      "reward": 0.723958358168602,
      "reward_std": 0.15768472477793694,
      "rewards/accuracy_reward": 0.3720238171517849,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3519345298409462,
      "step": 26
    },
    {
      "completion_length": 2988.1697387695312,
      "epoch": 0.7552447552447552,
      "grad_norm": 0.15998531877994537,
      "kl": 0.33447265625,
      "learning_rate": 4.444297669803981e-06,
      "loss": 0.0134,
      "reward": 0.79613097012043,
      "reward_std": 0.17738395184278488,
      "rewards/accuracy_reward": 0.377976194024086,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.418154776096344,
      "step": 27
    },
    {
      "completion_length": 3657.1666870117188,
      "epoch": 0.7832167832167832,
      "grad_norm": 0.173067107796669,
      "kl": 0.361328125,
      "learning_rate": 3.6560671583635467e-06,
      "loss": 0.0145,
      "reward": 0.5111607238650322,
      "reward_std": 0.11157247237861156,
      "rewards/accuracy_reward": 0.14583333395421505,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3653273805975914,
      "step": 28
    },
    {
      "completion_length": 3481.666748046875,
      "epoch": 0.8111888111888111,
      "grad_norm": 0.18258774280548096,
      "kl": 0.31201171875,
      "learning_rate": 2.9289321881345257e-06,
      "loss": 0.0125,
      "reward": 0.7008928805589676,
      "reward_std": 0.20082013495266438,
      "rewards/accuracy_reward": 0.2976190522313118,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4032738208770752,
      "step": 29
    },
    {
      "completion_length": 3570.7173461914062,
      "epoch": 0.8391608391608392,
      "grad_norm": 4.07529354095459,
      "kl": 0.39990234375,
      "learning_rate": 2.26989546637263e-06,
      "loss": 0.016,
      "reward": 0.534226194024086,
      "reward_std": 0.15385975502431393,
      "rewards/accuracy_reward": 0.16071429196745157,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3735119104385376,
      "step": 30
    },
    {
      "completion_length": 3432.7529907226562,
      "epoch": 0.8671328671328671,
      "grad_norm": 0.19004768133163452,
      "kl": 0.3251953125,
      "learning_rate": 1.6853038769745466e-06,
      "loss": 0.013,
      "reward": 0.6971726417541504,
      "reward_std": 0.21272188052535057,
      "rewards/accuracy_reward": 0.2976190522313118,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.399553582072258,
      "step": 31
    },
    {
      "completion_length": 3168.482177734375,
      "epoch": 0.8951048951048951,
      "grad_norm": 0.175730898976326,
      "kl": 0.31787109375,
      "learning_rate": 1.1807873565164507e-06,
      "loss": 0.0127,
      "reward": 0.6726190447807312,
      "reward_std": 0.1167123094201088,
      "rewards/accuracy_reward": 0.2648809589445591,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4077381044626236,
      "step": 32
    },
    {
      "completion_length": 3504.7440795898438,
      "epoch": 0.9230769230769231,
      "grad_norm": 0.2087661623954773,
      "kl": 0.37939453125,
      "learning_rate": 7.612046748871327e-07,
      "loss": 0.0152,
      "reward": 0.674107164144516,
      "reward_std": 0.19641772098839283,
      "rewards/accuracy_reward": 0.28571428544819355,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3883928656578064,
      "step": 33
    },
    {
      "completion_length": 2965.9286499023438,
      "epoch": 0.951048951048951,
      "grad_norm": 0.19867944717407227,
      "kl": 0.3359375,
      "learning_rate": 4.305966426779118e-07,
      "loss": 0.0134,
      "reward": 0.7485119253396988,
      "reward_std": 0.204582080245018,
      "rewards/accuracy_reward": 0.3244047649204731,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4241071492433548,
      "step": 34
    },
    {
      "completion_length": 3074.5208740234375,
      "epoch": 0.9790209790209791,
      "grad_norm": 0.19133983552455902,
      "kl": 0.32958984375,
      "learning_rate": 1.921471959676957e-07,
      "loss": 0.0132,
      "reward": 0.7343750149011612,
      "reward_std": 0.15512866899371147,
      "rewards/accuracy_reward": 0.3214285783469677,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4129464402794838,
      "step": 35
    },
    {
      "completion_length": 3781.7938639322915,
      "epoch": 1.0,
      "grad_norm": 0.19133983552455902,
      "kl": 0.392578125,
      "learning_rate": 4.815273327803183e-08,
      "loss": 0.0118,
      "reward": 0.4583333532015483,
      "reward_std": 0.11528908833861351,
      "rewards/accuracy_reward": 0.1071428582072258,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3511904776096344,
      "step": 36
    },
    {
      "epoch": 1.0,
      "step": 36,
      "total_flos": 0.0,
      "train_loss": 0.011807907519011864,
      "train_runtime": 16456.6289,
      "train_samples_per_second": 0.061,
      "train_steps_per_second": 0.002
    }
  ],
  "logging_steps": 1,
  "max_steps": 36,
  "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": 12,
  "trial_name": null,
  "trial_params": null
}