{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9491525423728814,
  "eval_steps": 100,
  "global_step": 14,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 494.59716796875,
      "epoch": 0.06779661016949153,
      "grad_norm": 1.2378678233303615,
      "kl": 0.0,
      "learning_rate": 1.5e-06,
      "loss": 0.0199,
      "num_tokens": 2637734.0,
      "reward": 0.58935546875,
      "reward_std": 0.4038642942905426,
      "rewards/accuracy_reward": 0.17529296875,
      "rewards/format_reward": 0.4140625,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.5352783203125,
      "epoch": 0.3389830508474576,
      "grad_norm": 0.1876558781807472,
      "kl": 0.0034079551696777344,
      "learning_rate": 2.5606601717798212e-06,
      "loss": 0.0182,
      "num_tokens": 12816784.0,
      "reward": 1.02020263671875,
      "reward_std": 0.35889170970767736,
      "rewards/accuracy_reward": 0.2501220703125,
      "rewards/format_reward": 0.77008056640625,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.808447265625,
      "epoch": 0.6779661016949152,
      "grad_norm": 0.09218154890901796,
      "kl": 0.02796630859375,
      "learning_rate": 7.500000000000003e-07,
      "loss": 0.0271,
      "num_tokens": 26676837.0,
      "reward": 1.59814453125,
      "reward_std": 0.2891646586358547,
      "rewards/accuracy_reward": 0.701904296875,
      "rewards/format_reward": 0.896240234375,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 473.8798494338989,
      "epoch": 0.9491525423728814,
      "kl": 0.029388427734375,
      "num_tokens": 36845879.0,
      "reward": 1.6905517578125,
      "reward_std": 0.22085796762257814,
      "rewards/accuracy_reward": 0.72515869140625,
      "rewards/format_reward": 0.96539306640625,
      "step": 14,
      "total_flos": 0.0,
      "train_loss": 0.021999110600778034,
      "train_runtime": 3853.2833,
      "train_samples_per_second": 1.946,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 14,
  "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
}