{
  "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": [
    {
      "completion_length": 616.6462249755859,
      "epoch": 0.017057569296375266,
      "grad_norm": 1.7272135019302368,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0357,
      "reward": 0.6808036044239998,
      "reward_std": 0.31602950021624565,
      "rewards/accuracy_reward": 0.6662946790456772,
      "rewards/format_reward": 0.014508929336443543,
      "step": 1
    },
    {
      "completion_length": 611.7843494415283,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.3145679533481598,
      "kl": 0.00017917156219482422,
      "learning_rate": 2.5e-06,
      "loss": 0.0441,
      "reward": 0.6612723525613546,
      "reward_std": 0.322592549957335,
      "rewards/accuracy_reward": 0.6476004794239998,
      "rewards/format_reward": 0.013671875844011083,
      "step": 5
    },
    {
      "completion_length": 606.3022621154785,
      "epoch": 0.17057569296375266,
      "grad_norm": 0.9135303497314453,
      "kl": 0.002117562294006348,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.0486,
      "reward": 0.7031250327825547,
      "reward_std": 0.3129757303744555,
      "rewards/accuracy_reward": 0.6863839581608773,
      "rewards/format_reward": 0.01674107222352177,
      "step": 10
    },
    {
      "completion_length": 589.8219047546387,
      "epoch": 0.255863539445629,
      "grad_norm": 0.9820574522018433,
      "kl": 0.01598987579345703,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0522,
      "reward": 0.9247768327593804,
      "reward_std": 0.3940876517444849,
      "rewards/accuracy_reward": 0.7486607536673546,
      "rewards/format_reward": 0.1761160797206685,
      "step": 15
    },
    {
      "completion_length": 623.1513641357421,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.30078014731407166,
      "kl": 0.0150848388671875,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0499,
      "reward": 1.178348271548748,
      "reward_std": 0.4842135950922966,
      "rewards/accuracy_reward": 0.7214286014437675,
      "rewards/format_reward": 0.45691966339945794,
      "step": 20
    },
    {
      "completion_length": 673.8540496826172,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.27627527713775635,
      "kl": 0.023541259765625,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0483,
      "reward": 1.3975447028875352,
      "reward_std": 0.46302561461925507,
      "rewards/accuracy_reward": 0.7002232432365417,
      "rewards/format_reward": 0.6973214596509933,
      "step": 25
    },
    {
      "completion_length": 645.6444488525391,
      "epoch": 0.511727078891258,
      "grad_norm": 0.15549008548259735,
      "kl": 0.01860809326171875,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.051,
      "reward": 1.4763393461704255,
      "reward_std": 0.4491942249238491,
      "rewards/accuracy_reward": 0.7241071775555611,
      "rewards/format_reward": 0.7522321805357933,
      "step": 30
    },
    {
      "completion_length": 638.6131942749023,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.14530694484710693,
      "kl": 0.018096923828125,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0398,
      "reward": 1.5024554222822188,
      "reward_std": 0.41087508872151374,
      "rewards/accuracy_reward": 0.7180803894996644,
      "rewards/format_reward": 0.7843750357627869,
      "step": 35
    },
    {
      "completion_length": 644.3710052490235,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.21266834437847137,
      "kl": 0.02000579833984375,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.03,
      "reward": 1.574776855111122,
      "reward_std": 0.3656768787652254,
      "rewards/accuracy_reward": 0.7404018193483353,
      "rewards/format_reward": 0.8343750387430191,
      "step": 40
    },
    {
      "completion_length": 660.5759185791015,
      "epoch": 0.767590618336887,
      "grad_norm": 0.19263683259487152,
      "kl": 0.021728515625,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0311,
      "reward": 1.5725447177886962,
      "reward_std": 0.3758295692503452,
      "rewards/accuracy_reward": 0.7430803939700127,
      "rewards/format_reward": 0.829464316368103,
      "step": 45
    },
    {
      "completion_length": 651.923243713379,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.24893562495708466,
      "kl": 0.023681640625,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.028,
      "reward": 1.5631697088479997,
      "reward_std": 0.37856253087520597,
      "rewards/accuracy_reward": 0.7366071760654449,
      "rewards/format_reward": 0.8265625387430191,
      "step": 50
    },
    {
      "completion_length": 644.7035942077637,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.13216549158096313,
      "kl": 0.0205230712890625,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0207,
      "reward": 1.619196492433548,
      "reward_std": 0.3612879175692797,
      "rewards/accuracy_reward": 0.7718750357627868,
      "rewards/format_reward": 0.847321467101574,
      "step": 55
    },
    {
      "completion_length": 639.9285990397135,
      "epoch": 0.9893390191897654,
      "kl": 0.049023946126302086,
      "reward": 1.606026868025462,
      "reward_std": 0.3578349606444438,
      "rewards/accuracy_reward": 0.750000037252903,
      "rewards/format_reward": 0.8560268258055052,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.03837096903087764,
      "train_runtime": 7408.2441,
      "train_samples_per_second": 1.012,
      "train_steps_per_second": 0.008
    }
  ],
  "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
}