data/processed/split_summary.json

{
  "seed": 42,
  "total_rows": 3414,
  "ratios": {
    "train": 0.7,
    "val": 0.1,
    "test": 0.2
  },
  "splits": {
    "train": {
      "rows": 2389,
      "label_distribution": {
        "non-clickbait": 1644,
        "clickbait": 745
      }
    },
    "val": {
      "rows": 342,
      "label_distribution": {
        "non-clickbait": 235,
        "clickbait": 107
      }
    },
    "test": {
      "rows": 683,
      "label_distribution": {
        "non-clickbait": 470,
        "clickbait": 213
      }
    }
  }
}