---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: authors
    sequence: string
  - name: title
    dtype: string
  - name: published
    dtype: int64
  - name: description
    dtype: string
  - name: tags
    dtype: string
  - name: serie
    dtype: string
  - name: serieseq
    dtype: string
  - name: id
    dtype: int64
  splits:
  - name: train
    num_bytes: 44717507254
    num_examples: 98388
  - name: validation
    num_bytes: 1374274701
    num_examples: 3043
  download_size: 28254000311
  dataset_size: 46091781955
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
language:
- es
license: unknown
pretty_name: gromenaaue
---

# GROMENAUER Dataset

The **GROMENAUER** (**G**ran **R**epertorio **O**riginal de **M**anuscritos **E**lectrónicos **N**o **A**utorizados, **U**surpados y **E**scandalosamente **R**obados) dataset is a collection of approximately 100,000 Spanish books. 

**Description:**
This dataset comprises a vast array of literary works in Spanish, spanning various genres, epochs, and authors. From classic literature to contemporary fiction, it offers a comprehensive insight into the Spanish literary landscape. The texts included are digitized versions of original manuscripts, some of which might have been unauthorized, usurped, or scandalously stolen, contributing to the dataset's unique character and historical value.

The training dataset contains ~7.5 billion words, or ~14.5 billion tokens (Mistral-7b-v0.1 tokenizer)

**Usage:**
Researchers, linguists, historians, and enthusiasts can utilize this dataset for a wide range of purposes, including but not limited to:
- Linguistic analysis
- Literary studies
- Historical research
- Cultural exploration
- Machine learning applications, such as natural language processing (NLP) and text generation

**Ethical Considerations:**
While the dataset contains texts that may have been acquired through questionable means, users are encouraged to approach its utilization ethically and responsibly. Acknowledgment of authors and respect for intellectual property rights should be maintained, even in cases where the origins of the manuscripts are uncertain.

**Note:**
Due to the nature of the dataset, users should exercise caution and discretion in handling and analyzing its contents, recognizing the potential legal and ethical implications associated with the acquisition and dissemination of unauthorized literary works.