{
  "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": 3737.096435546875,
      "epoch": 0.027972027972027972,
      "grad_norm": 0.4781525433063507,
      "kl": 0.0,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 0.49642856419086456,
      "reward_std": 0.14968311414122581,
      "rewards/accuracy_reward": 0.09285714384168386,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4035714343190193,
      "step": 1
    },
    {
      "completion_length": 3919.4143676757812,
      "epoch": 0.055944055944055944,
      "grad_norm": 0.8725160360336304,
      "kl": 0.0,
      "learning_rate": 2.5e-06,
      "loss": 0.0,
      "reward": 0.41160714626312256,
      "reward_std": 0.1353183900937438,
      "rewards/accuracy_reward": 0.05000000260770321,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3616071417927742,
      "step": 2
    },
    {
      "completion_length": 3489.6964721679688,
      "epoch": 0.08391608391608392,
      "grad_norm": 0.5938112735748291,
      "kl": 0.000396728515625,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "reward": 0.5473214313387871,
      "reward_std": 0.15705320611596107,
      "rewards/accuracy_reward": 0.11785714374855161,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.42946429550647736,
      "step": 3
    },
    {
      "completion_length": 3473.7571411132812,
      "epoch": 0.11188811188811189,
      "grad_norm": 0.5854588747024536,
      "kl": 0.0013608932495117188,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.0001,
      "reward": 0.5562500059604645,
      "reward_std": 0.17247037403285503,
      "rewards/accuracy_reward": 0.14285714388825,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4133928641676903,
      "step": 4
    },
    {
      "completion_length": 3737.9000244140625,
      "epoch": 0.13986013986013987,
      "grad_norm": 8.259861946105957,
      "kl": 0.006000518798828125,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "reward": 0.5544642880558968,
      "reward_std": 0.19513687305152416,
      "rewards/accuracy_reward": 0.157142860814929,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3973214253783226,
      "step": 5
    },
    {
      "completion_length": 3746.992919921875,
      "epoch": 0.16783216783216784,
      "grad_norm": 0.48524144291877747,
      "kl": 0.0187225341796875,
      "learning_rate": 9.978331270024887e-06,
      "loss": 0.0007,
      "reward": 0.5526785850524902,
      "reward_std": 0.27357756346464157,
      "rewards/accuracy_reward": 0.12142857350409031,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4312500059604645,
      "step": 6
    },
    {
      "completion_length": 3918.8643188476562,
      "epoch": 0.1958041958041958,
      "grad_norm": 0.29106539487838745,
      "kl": 0.025238037109375,
      "learning_rate": 9.913533761814537e-06,
      "loss": 0.001,
      "reward": 0.5473214462399483,
      "reward_std": 0.2193572111427784,
      "rewards/accuracy_reward": 0.12857143231667578,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.41875000298023224,
      "step": 7
    },
    {
      "completion_length": 3901.175048828125,
      "epoch": 0.22377622377622378,
      "grad_norm": 0.34263718128204346,
      "kl": 0.0513916015625,
      "learning_rate": 9.80623151079494e-06,
      "loss": 0.0021,
      "reward": 0.4044642969965935,
      "reward_std": 0.12892552372068167,
      "rewards/accuracy_reward": 0.04642857238650322,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.35803572088479996,
      "step": 8
    },
    {
      "completion_length": 3889.1786499023438,
      "epoch": 0.2517482517482518,
      "grad_norm": 1.2209001779556274,
      "kl": 0.0618896484375,
      "learning_rate": 9.65745789630079e-06,
      "loss": 0.0025,
      "reward": 0.48303572088479996,
      "reward_std": 0.21392671391367912,
      "rewards/accuracy_reward": 0.11428571795113385,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3687500059604645,
      "step": 9
    },
    {
      "completion_length": 4003.5607299804688,
      "epoch": 0.27972027972027974,
      "grad_norm": 0.3369362950325012,
      "kl": 0.068359375,
      "learning_rate": 9.468645689567599e-06,
      "loss": 0.0027,
      "reward": 0.46339286118745804,
      "reward_std": 0.22164541110396385,
      "rewards/accuracy_reward": 0.09285714570432901,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3705357164144516,
      "step": 10
    },
    {
      "completion_length": 3949.378662109375,
      "epoch": 0.3076923076923077,
      "grad_norm": 0.7158792018890381,
      "kl": 0.0810546875,
      "learning_rate": 9.241613255361455e-06,
      "loss": 0.0032,
      "reward": 0.4375000074505806,
      "reward_std": 0.1808728687465191,
      "rewards/accuracy_reward": 0.08214285899884999,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3553571403026581,
      "step": 11
    },
    {
      "completion_length": 4065.3965454101562,
      "epoch": 0.3356643356643357,
      "grad_norm": 0.26943907141685486,
      "kl": 0.0992431640625,
      "learning_rate": 8.978547040132317e-06,
      "loss": 0.004,
      "reward": 0.35625000298023224,
      "reward_std": 0.11728819366544485,
      "rewards/accuracy_reward": 0.0321428575553,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3241071403026581,
      "step": 12
    },
    {
      "completion_length": 3880.08203125,
      "epoch": 0.36363636363636365,
      "grad_norm": 0.3461170792579651,
      "kl": 0.1202392578125,
      "learning_rate": 8.681980515339464e-06,
      "loss": 0.0048,
      "reward": 0.6062500029802322,
      "reward_std": 0.22258323803544044,
      "rewards/accuracy_reward": 0.2607142962515354,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3455357179045677,
      "step": 13
    },
    {
      "completion_length": 3993.267822265625,
      "epoch": 0.3916083916083916,
      "grad_norm": 0.32656392455101013,
      "kl": 0.13916015625,
      "learning_rate": 8.354769778736407e-06,
      "loss": 0.0056,
      "reward": 0.6285714283585548,
      "reward_std": 0.29084962233901024,
      "rewards/accuracy_reward": 0.29285714589059353,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.33571428060531616,
      "step": 14
    },
    {
      "completion_length": 4049.6571044921875,
      "epoch": 0.4195804195804196,
      "grad_norm": 0.3071815073490143,
      "kl": 0.1650390625,
      "learning_rate": 8.00006604858821e-06,
      "loss": 0.0066,
      "reward": 0.5178571417927742,
      "reward_std": 0.17362720984965563,
      "rewards/accuracy_reward": 0.20714286155998707,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3107142820954323,
      "step": 15
    },
    {
      "completion_length": 4045.789306640625,
      "epoch": 0.44755244755244755,
      "grad_norm": 0.40854963660240173,
      "kl": 0.2138671875,
      "learning_rate": 7.621285315716991e-06,
      "loss": 0.0085,
      "reward": 0.5482142865657806,
      "reward_std": 0.2475740760564804,
      "rewards/accuracy_reward": 0.22500000521540642,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3232142925262451,
      "step": 16
    },
    {
      "completion_length": 3979.9750366210938,
      "epoch": 0.4755244755244755,
      "grad_norm": 0.48029232025146484,
      "kl": 0.250732421875,
      "learning_rate": 7.222075445642904e-06,
      "loss": 0.01,
      "reward": 0.5223214402794838,
      "reward_std": 0.2250591404736042,
      "rewards/accuracy_reward": 0.20357143506407738,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3187500014901161,
      "step": 17
    },
    {
      "completion_length": 3790.7999877929688,
      "epoch": 0.5034965034965035,
      "grad_norm": 0.59281986951828,
      "kl": 0.26953125,
      "learning_rate": 6.80628104764508e-06,
      "loss": 0.0108,
      "reward": 0.5714285746216774,
      "reward_std": 0.20503208599984646,
      "rewards/accuracy_reward": 0.21428572107106447,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3571428582072258,
      "step": 18
    },
    {
      "completion_length": 3469.2178344726562,
      "epoch": 0.5314685314685315,
      "grad_norm": 5.099720478057861,
      "kl": 0.376953125,
      "learning_rate": 6.377906449072578e-06,
      "loss": 0.0151,
      "reward": 0.6991071403026581,
      "reward_std": 0.2952312082052231,
      "rewards/accuracy_reward": 0.30000000819563866,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.3991071358323097,
      "step": 19
    },
    {
      "completion_length": 3258.9036254882812,
      "epoch": 0.5594405594405595,
      "grad_norm": 0.476639062166214,
      "kl": 0.423828125,
      "learning_rate": 5.9410771314830255e-06,
      "loss": 0.017,
      "reward": 0.7660714238882065,
      "reward_std": 0.3425598815083504,
      "rewards/accuracy_reward": 0.3107142932713032,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4553571492433548,
      "step": 20
    },
    {
      "completion_length": 2936.7607421875,
      "epoch": 0.5874125874125874,
      "grad_norm": 0.9111078977584839,
      "kl": 0.4521484375,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.0181,
      "reward": 0.7901785671710968,
      "reward_std": 0.2908039838075638,
      "rewards/accuracy_reward": 0.30714286491274834,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.48303571343421936,
      "step": 21
    },
    {
      "completion_length": 2737.4607543945312,
      "epoch": 0.6153846153846154,
      "grad_norm": 0.5476343035697937,
      "kl": 0.54345703125,
      "learning_rate": 5.0589228685169776e-06,
      "loss": 0.0217,
      "reward": 0.7124999910593033,
      "reward_std": 0.22632914781570435,
      "rewards/accuracy_reward": 0.22500000917352736,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.48750000447034836,
      "step": 22
    },
    {
      "completion_length": 2191.2464599609375,
      "epoch": 0.6433566433566433,
      "grad_norm": 0.6787258386611938,
      "kl": 0.5205078125,
      "learning_rate": 4.622093550927423e-06,
      "loss": 0.0208,
      "reward": 0.8017857074737549,
      "reward_std": 0.24978942424058914,
      "rewards/accuracy_reward": 0.32500000670552254,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.47678571939468384,
      "step": 23
    },
    {
      "completion_length": 2145.5035705566406,
      "epoch": 0.6713286713286714,
      "grad_norm": 0.9464374780654907,
      "kl": 0.6494140625,
      "learning_rate": 4.193718952354921e-06,
      "loss": 0.026,
      "reward": 0.6982142925262451,
      "reward_std": 0.2445763349533081,
      "rewards/accuracy_reward": 0.19642857555299997,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5017857104539871,
      "step": 24
    },
    {
      "completion_length": 1999.6785888671875,
      "epoch": 0.6993006993006993,
      "grad_norm": 0.7832499742507935,
      "kl": 0.626953125,
      "learning_rate": 3.777924554357096e-06,
      "loss": 0.025,
      "reward": 0.7687499970197678,
      "reward_std": 0.2781074270606041,
      "rewards/accuracy_reward": 0.2678571492433548,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.500892847776413,
      "step": 25
    },
    {
      "completion_length": 2141.8035888671875,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.8674513697624207,
      "kl": 0.765625,
      "learning_rate": 3.378714684283011e-06,
      "loss": 0.0306,
      "reward": 0.7142857015132904,
      "reward_std": 0.226736880838871,
      "rewards/accuracy_reward": 0.22500000149011612,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4892857298254967,
      "step": 26
    },
    {
      "completion_length": 1950.5678100585938,
      "epoch": 0.7552447552447552,
      "grad_norm": 0.7500644326210022,
      "kl": 0.65234375,
      "learning_rate": 2.9999339514117913e-06,
      "loss": 0.0261,
      "reward": 0.7374999821186066,
      "reward_std": 0.25540266558527946,
      "rewards/accuracy_reward": 0.19285714533179998,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.5446428656578064,
      "step": 27
    },
    {
      "completion_length": 2502.7821655273438,
      "epoch": 0.7832167832167832,
      "grad_norm": 1.129561185836792,
      "kl": 0.869140625,
      "learning_rate": 2.645230221263596e-06,
      "loss": 0.0348,
      "reward": 0.5919642895460129,
      "reward_std": 0.18119085393846035,
      "rewards/accuracy_reward": 0.10714286030270159,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4848214387893677,
      "step": 28
    },
    {
      "completion_length": 2735.4535522460938,
      "epoch": 0.8111888111888111,
      "grad_norm": 0.673635721206665,
      "kl": 0.8076171875,
      "learning_rate": 2.3180194846605367e-06,
      "loss": 0.0323,
      "reward": 0.6232142895460129,
      "reward_std": 0.23368354886770248,
      "rewards/accuracy_reward": 0.1428571476135403,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4803571552038193,
      "step": 29
    },
    {
      "completion_length": 2827.32861328125,
      "epoch": 0.8391608391608392,
      "grad_norm": 1.255906343460083,
      "kl": 0.8408203125,
      "learning_rate": 2.021452959867684e-06,
      "loss": 0.0336,
      "reward": 0.5776785612106323,
      "reward_std": 0.23459363728761673,
      "rewards/accuracy_reward": 0.08928571734577417,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4883928596973419,
      "step": 30
    },
    {
      "completion_length": 3130.0,
      "epoch": 0.8671328671328671,
      "grad_norm": 0.8778608441352844,
      "kl": 0.8134765625,
      "learning_rate": 1.7583867446385461e-06,
      "loss": 0.0325,
      "reward": 0.5687500089406967,
      "reward_std": 0.21473033353686333,
      "rewards/accuracy_reward": 0.11785714444704354,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4508928656578064,
      "step": 31
    },
    {
      "completion_length": 3108.8750610351562,
      "epoch": 0.8951048951048951,
      "grad_norm": 0.703495979309082,
      "kl": 0.6640625,
      "learning_rate": 1.531354310432403e-06,
      "loss": 0.0266,
      "reward": 0.6116071492433548,
      "reward_std": 0.24049308896064758,
      "rewards/accuracy_reward": 0.15000000223517418,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4616071507334709,
      "step": 32
    },
    {
      "completion_length": 3385.7571411132812,
      "epoch": 0.9230769230769231,
      "grad_norm": 0.9375730752944946,
      "kl": 0.7919921875,
      "learning_rate": 1.3425421036992098e-06,
      "loss": 0.0316,
      "reward": 0.4883928596973419,
      "reward_std": 0.172358563169837,
      "rewards/accuracy_reward": 0.06428571604192257,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4241071343421936,
      "step": 33
    },
    {
      "completion_length": 3170.0820922851562,
      "epoch": 0.951048951048951,
      "grad_norm": 1.1337316036224365,
      "kl": 0.58447265625,
      "learning_rate": 1.1937684892050606e-06,
      "loss": 0.0234,
      "reward": 0.6321428567171097,
      "reward_std": 0.288823913782835,
      "rewards/accuracy_reward": 0.17857143469154835,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.4535714313387871,
      "step": 34
    },
    {
      "completion_length": 3635.4571533203125,
      "epoch": 0.9790209790209791,
      "grad_norm": 0.7671982049942017,
      "kl": 0.62109375,
      "learning_rate": 1.0864662381854632e-06,
      "loss": 0.0248,
      "reward": 0.557142861187458,
      "reward_std": 0.18924793601036072,
      "rewards/accuracy_reward": 0.14642857760190964,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.410714291036129,
      "step": 35
    },
    {
      "completion_length": 3688.6334635416665,
      "epoch": 1.0,
      "grad_norm": 0.7671982049942017,
      "kl": 0.62109375,
      "learning_rate": 1.0216687299751146e-06,
      "loss": 0.0186,
      "reward": 0.5226190686225891,
      "reward_std": 0.2011700620253881,
      "rewards/accuracy_reward": 0.10952381292978923,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.41309523582458496,
      "step": 36
    },
    {
      "epoch": 1.0,
      "step": 36,
      "total_flos": 0.0,
      "train_loss": 0.014487237919349916,
      "train_runtime": 13938.4252,
      "train_samples_per_second": 0.072,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 1,
  "max_steps": 36,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "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": 10,
  "trial_name": null,
  "trial_params": null
}