Spanish Wikipedia Articles Preprocessed for Language Model Training

A preprocessed version of the Spanish Wikipedia dump from May 2026, totaling 8.4 GB. The dataset was created by raj2708 and includes articles parsed from wikitext to plain text. It is intended for use in large language model pretraining.

Use Cases

• Pretrain Spanish language models based on the cleaned article text.
• Fine-tune models for Spanish text generation tasks based on the encyclopedia content.
• Benchmark text processing or filtering algorithms based on the preprocessed corpus.
• Analyze Spanish language patterns and knowledge representation based on Wikipedia articles.

Strengths

• The dataset is explicitly preprocessed and cleaned, ready for immediate use.
• The total size of 8.4 GB indicates a substantial text corpus.
• It has a clear license (CC-BY-SA 4.0) and source (es.wikipedia.org).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• The description metadata is limited; actual data quality requires manual inspection after download.
• The row count is unknown, which may limit suitability assessment.

Provenance

Source

es.wikipedia.org

Collection Method

Parsed from a full Spanish Wikipedia XML BZ2 dump using mwparserfromhell, with redirects and short articles filtered.

Time Range

Dump date is May 2026.

Freshness

Last updated 2026-06-01 11:44:03; freshness should be verified.