metadata.json

{
  "name": "Camoniana",
  "description": "Portuguese literary corpus prepared for character-level language modeling.",
  "language": "pt",
  "modality": "text",
  "tokenization": "character-level",
  "num_characters": 1046084,
  "num_lines": 34241,
  "vocab_size": 119,
  "train_characters": 941475,
  "validation_characters": 104609,
  "split_method": "contiguous 90/10 split by character offset",
  "license": "other",
  "license_status": "Authorial works are public-domain by age in many jurisdictions; transcription and edition lineage for this release has not been fully reconstructed."
}