{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.8,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 994.5480270385742,
      "epoch": 0.2,
      "grad_norm": 150.31954956054688,
      "kl": 0.0,
      "learning_rate": 6.000000000000001e-07,
      "loss": -0.0199,
      "reward": 0.7332589626312256,
      "reward_std": 0.41786693036556244,
      "rewards/accuracy_reward": 0.06808035937137902,
      "rewards/format_reward": 0.6651786044239998,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 990.802339553833,
      "epoch": 1.2,
      "grad_norm": 54.17798614501953,
      "kl": 0.3659858703613281,
      "learning_rate": 3e-06,
      "loss": -0.0063,
      "reward": 0.7120536062866449,
      "reward_std": 0.3840282931923866,
      "rewards/accuracy_reward": 0.04827009135624394,
      "rewards/format_reward": 0.6637835148721933,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1000.7384017944336,
      "epoch": 2.4,
      "grad_norm": 5.046395778656006,
      "kl": 2.1490234375,
      "learning_rate": 2.9095389311788626e-06,
      "loss": 0.0721,
      "reward": 0.7515625342726707,
      "reward_std": 0.3842003524303436,
      "rewards/accuracy_reward": 0.0529017882887274,
      "rewards/format_reward": 0.6986607432365417,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1002.4748840332031,
      "epoch": 3.6,
      "grad_norm": 0.10862202197313309,
      "kl": 2.658203125,
      "learning_rate": 2.649066664678467e-06,
      "loss": 0.1029,
      "reward": 0.8709821805357933,
      "reward_std": 0.3000884648412466,
      "rewards/accuracy_reward": 0.056250002211891116,
      "rewards/format_reward": 0.8147321805357933,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 965.2994049072265,
      "epoch": 4.8,
      "grad_norm": 0.10118918865919113,
      "kl": 3.084765625,
      "learning_rate": 2.25e-06,
      "loss": 0.1217,
      "reward": 0.9578125402331352,
      "reward_std": 0.21929518431425093,
      "rewards/accuracy_reward": 0.0584821458440274,
      "rewards/format_reward": 0.8993303999304771,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 943.4268981933594,
      "epoch": 6.2,
      "grad_norm": 0.08847808837890625,
      "kl": 3.2484375,
      "learning_rate": 1.7604722665003958e-06,
      "loss": 0.1278,
      "reward": 0.9870536178350449,
      "reward_std": 0.2045027593150735,
      "rewards/accuracy_reward": 0.06629464582074433,
      "rewards/format_reward": 0.9207589760422706,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 926.5740798950195,
      "epoch": 7.4,
      "grad_norm": 0.09610795229673386,
      "kl": 3.225390625,
      "learning_rate": 1.2395277334996047e-06,
      "loss": 0.1279,
      "reward": 1.0104911148548126,
      "reward_std": 0.17210464756935834,
      "rewards/accuracy_reward": 0.06875000335276127,
      "rewards/format_reward": 0.9417411088943481,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 941.6847366333008,
      "epoch": 8.6,
      "grad_norm": 0.08967000991106033,
      "kl": 3.17578125,
      "learning_rate": 7.500000000000003e-07,
      "loss": 0.1261,
      "reward": 1.0017857551574707,
      "reward_std": 0.19479605397209526,
      "rewards/accuracy_reward": 0.07098214563447983,
      "rewards/format_reward": 0.9308036118745804,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 943.0091369628906,
      "epoch": 9.8,
      "grad_norm": 0.08454571664333344,
      "kl": 3.143359375,
      "learning_rate": 3.5093333532153313e-07,
      "loss": 0.1251,
      "reward": 0.9921875491738319,
      "reward_std": 0.17412924002856017,
      "rewards/accuracy_reward": 0.05714286032598466,
      "rewards/format_reward": 0.9350446835160255,
      "step": 40
    },
    {
      "epoch": 9.8,
      "step": 40,
      "total_flos": 0.0,
      "train_loss": 0.09932380295358598,
      "train_runtime": 12056.1753,
      "train_samples_per_second": 0.523,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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
}