{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 746.608283996582,
      "epoch": 0.0128,
      "grad_norm": 0.5155316591262817,
      "kl": 0.0,
      "learning_rate": 3.75e-07,
      "loss": 0.0,
      "reward": 0.14843750838190317,
      "reward_std": 0.16338601242750883,
      "rewards/accuracy_reward": 0.14843750838190317,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 727.0990829467773,
      "epoch": 0.064,
      "grad_norm": 1.084052324295044,
      "kl": 0.0003186464309692383,
      "learning_rate": 1.875e-06,
      "loss": 0.0,
      "reward": 0.1143973259604536,
      "reward_std": 0.12805407610721886,
      "rewards/accuracy_reward": 0.1143973259604536,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "completion_length": 736.6721374511719,
      "epoch": 0.128,
      "grad_norm": 0.36135947704315186,
      "kl": 0.0037824630737304686,
      "learning_rate": 2.993961440992859e-06,
      "loss": 0.0002,
      "reward": 0.1305803638882935,
      "reward_std": 0.12962731290608645,
      "rewards/accuracy_reward": 0.1305803638882935,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 779.8250381469727,
      "epoch": 0.192,
      "grad_norm": 1.732832670211792,
      "kl": 0.010242462158203125,
      "learning_rate": 2.9265847744427307e-06,
      "loss": 0.0004,
      "reward": 0.1544642923399806,
      "reward_std": 0.1452927845530212,
      "rewards/accuracy_reward": 0.1544642923399806,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "completion_length": 793.7176681518555,
      "epoch": 0.256,
      "grad_norm": 0.45336416363716125,
      "kl": 0.05513496398925781,
      "learning_rate": 2.7876731904027993e-06,
      "loss": 0.0022,
      "reward": 0.1392857213038951,
      "reward_std": 0.1322711819317192,
      "rewards/accuracy_reward": 0.1392857213038951,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 772.2924423217773,
      "epoch": 0.32,
      "grad_norm": 1.4548314809799194,
      "kl": 0.008623123168945312,
      "learning_rate": 2.584192295741087e-06,
      "loss": 0.0003,
      "reward": 0.17254465222358703,
      "reward_std": 0.1551555698737502,
      "rewards/accuracy_reward": 0.17254465222358703,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 758.8147674560547,
      "epoch": 0.384,
      "grad_norm": 0.43251317739486694,
      "kl": 0.04488029479980469,
      "learning_rate": 2.3263454721781537e-06,
      "loss": 0.0018,
      "reward": 0.18013393748551607,
      "reward_std": 0.15790632385760545,
      "rewards/accuracy_reward": 0.18013393748551607,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 759.0868667602539,
      "epoch": 0.448,
      "grad_norm": 2.017112970352173,
      "kl": 0.01151580810546875,
      "learning_rate": 2.027062236122014e-06,
      "loss": 0.0005,
      "reward": 0.16049108020961284,
      "reward_std": 0.14527264153584837,
      "rewards/accuracy_reward": 0.16049108020961284,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "completion_length": 762.179719543457,
      "epoch": 0.512,
      "grad_norm": 2.426039934158325,
      "kl": 0.0172943115234375,
      "learning_rate": 1.7013498987264833e-06,
      "loss": 0.0007,
      "reward": 0.17991072311997414,
      "reward_std": 0.15640948014333844,
      "rewards/accuracy_reward": 0.17991072311997414,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 754.7252563476562,
      "epoch": 0.576,
      "grad_norm": 1.071414828300476,
      "kl": 0.008959579467773437,
      "learning_rate": 1.3655410366448499e-06,
      "loss": 0.0004,
      "reward": 0.20066965445876123,
      "reward_std": 0.15480793919414282,
      "rewards/accuracy_reward": 0.20066965445876123,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "completion_length": 765.8024856567383,
      "epoch": 0.64,
      "grad_norm": 0.8965884447097778,
      "kl": 0.008967971801757813,
      "learning_rate": 1.036474508437579e-06,
      "loss": 0.0004,
      "reward": 0.18415179559960962,
      "reward_std": 0.17140543824061752,
      "rewards/accuracy_reward": 0.18415179559960962,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 740.766552734375,
      "epoch": 0.704,
      "grad_norm": 27.7662296295166,
      "kl": 0.016036224365234376,
      "learning_rate": 7.30651083891141e-07,
      "loss": 0.0006,
      "reward": 0.1348214357160032,
      "reward_std": 0.1273266337811947,
      "rewards/accuracy_reward": 0.1348214357160032,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "completion_length": 741.5098526000977,
      "epoch": 0.768,
      "grad_norm": 0.12588420510292053,
      "kl": 0.011777877807617188,
      "learning_rate": 4.63406026519703e-07,
      "loss": 0.0005,
      "reward": 0.19330358118750154,
      "reward_std": 0.15862846123054625,
      "rewards/accuracy_reward": 0.19330358118750154,
      "rewards/format_reward": 0.0,
      "step": 60
    },
    {
      "completion_length": 734.3199005126953,
      "epoch": 0.832,
      "grad_norm": 4.302868843078613,
      "kl": 0.07119407653808593,
      "learning_rate": 2.48140119418046e-07,
      "loss": 0.0029,
      "reward": 0.16562500977888703,
      "reward_std": 0.14798311032354833,
      "rewards/accuracy_reward": 0.16562500977888703,
      "rewards/format_reward": 0.0,
      "step": 65
    },
    {
      "completion_length": 737.8087387084961,
      "epoch": 0.896,
      "grad_norm": 1.6815804243087769,
      "kl": 0.00724334716796875,
      "learning_rate": 9.564769404039419e-08,
      "loss": 0.0003,
      "reward": 0.17589286509901286,
      "reward_std": 0.14854512372985482,
      "rewards/accuracy_reward": 0.17589286509901286,
      "rewards/format_reward": 0.0,
      "step": 70
    },
    {
      "completion_length": 740.4969161987304,
      "epoch": 0.96,
      "grad_norm": 0.14792795479297638,
      "kl": 0.034131622314453124,
      "learning_rate": 1.357535734809795e-08,
      "loss": 0.0014,
      "reward": 0.16607143636792898,
      "reward_std": 0.14445759197697045,
      "rewards/accuracy_reward": 0.16607143636792898,
      "rewards/format_reward": 0.0,
      "step": 75
    },
    {
      "completion_length": 740.5067342122396,
      "epoch": 0.9984,
      "kl": 0.02994537353515625,
      "reward": 0.17708334047347307,
      "reward_std": 0.1565772045093278,
      "rewards/accuracy_reward": 0.17708334047347307,
      "rewards/format_reward": 0.0,
      "step": 78,
      "total_flos": 0.0,
      "train_loss": 0.0006319970209619013,
      "train_runtime": 19204.1352,
      "train_samples_per_second": 0.521,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 78,
  "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
}