{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 100,
  "global_step": 58,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 607.8761520385742,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.21163204312324524,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0347,
      "reward": 0.7064732536673546,
      "reward_std": 0.2907280754297972,
      "rewards/accuracy_reward": 0.6964286044239998,
      "rewards/format_reward": 0.010044643422588706,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.3354082107544,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.22014343738555908,
      "kl": 0.00016763806343078613,
      "learning_rate": 2.5e-06,
      "loss": 0.0443,
      "reward": 0.6668527107685804,
      "reward_std": 0.30428835609927773,
      "rewards/accuracy_reward": 0.6609933376312256,
      "rewards/format_reward": 0.005859375291038305,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.3207862854003,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.1867409348487854,
      "kl": 0.006086993217468262,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0561,
      "reward": 0.6991071760654449,
      "reward_std": 0.28411166220903394,
      "rewards/accuracy_reward": 0.6957589611411095,
      "rewards/format_reward": 0.0033482144586741923,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 591.975025177002,
      "epoch": 0.255863539445629,
      "grad_norm": 0.272446870803833,
      "kl": 0.006195259094238281,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0687,
      "reward": 0.7625000327825546,
      "reward_std": 0.21445324290543794,
      "rewards/accuracy_reward": 0.7611607477068901,
      "rewards/format_reward": 0.0013392857741564511,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.5134178161621,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.4638945162296295,
      "kl": 0.0028181076049804688,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0489,
      "reward": 0.7725446745753288,
      "reward_std": 0.1972122782841325,
      "rewards/accuracy_reward": 0.7714286059141159,
      "rewards/format_reward": 0.0011160714784637094,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 586.8551612854004,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.13729101419448853,
      "kl": 0.0033502578735351562,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0425,
      "reward": 0.7703125372529029,
      "reward_std": 0.19164941012859343,
      "rewards/accuracy_reward": 0.769866107404232,
      "rewards/format_reward": 0.00044642859138548373,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 589.1917663574219,
      "epoch": 0.511727078891258,
      "grad_norm": 0.10644034296274185,
      "kl": 0.004234695434570312,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.034,
      "reward": 0.7531250387430191,
      "reward_std": 0.18948373831808568,
      "rewards/accuracy_reward": 0.7511161044239998,
      "rewards/format_reward": 0.002008928661234677,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.0815017700195,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.09403481334447861,
      "kl": 0.004046249389648438,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0411,
      "reward": 0.7562500327825546,
      "reward_std": 0.18806953858584166,
      "rewards/accuracy_reward": 0.7495536088943482,
      "rewards/format_reward": 0.006696428894065321,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 564.9935516357422,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.22079423069953918,
      "kl": 0.00465545654296875,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0326,
      "reward": 0.7790178969502449,
      "reward_std": 0.18834841772913932,
      "rewards/accuracy_reward": 0.7680803954601287,
      "rewards/format_reward": 0.010937500628642738,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.787744140625,
      "epoch": 0.767590618336887,
      "grad_norm": 0.16064484417438507,
      "kl": 0.005054092407226563,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0423,
      "reward": 0.7665178939700127,
      "reward_std": 0.22603920232504607,
      "rewards/accuracy_reward": 0.7470982491970062,
      "rewards/format_reward": 0.019419643795117736,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.5013664245605,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.18539777398109436,
      "kl": 0.00699310302734375,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0301,
      "reward": 0.7915178954601287,
      "reward_std": 0.2188779940828681,
      "rewards/accuracy_reward": 0.755357176065445,
      "rewards/format_reward": 0.036160716065205635,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.8031463623047,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.2504195272922516,
      "kl": 0.009212875366210937,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.038,
      "reward": 0.8113839596509933,
      "reward_std": 0.2401964722201228,
      "rewards/accuracy_reward": 0.7790178954601288,
      "rewards/format_reward": 0.03236607303842902,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.5023994445801,
      "epoch": 0.9893390191897654,
      "kl": 0.011606852213541666,
      "reward": 0.8020833730697632,
      "reward_std": 0.23279494047164917,
      "rewards/accuracy_reward": 0.7678571765621504,
      "rewards/format_reward": 0.03422619208383063,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.04428564981910689,
      "train_runtime": 7921.619,
      "train_samples_per_second": 0.947,
      "train_steps_per_second": 0.007
    }
  ],
  "logging_steps": 5,
  "max_steps": 58,
  "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
}