Datasets:

magyar-nlp-szine-java
/

youtube

Modalities:

Formats:

Languages:

Size:

Libraries:

License:

Dataset Viewer

The dataset viewer is not available for this dataset.

Unexpected token '<', "<html> <h"... is not valid JSON

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

YouTube Dataset (Semantic Chunks)

Hungarian YouTube comments dataset preprocessed with semantic chunking.


Rows	629,767
Tokens	21,339,568
Tokenizer	`magyar-nlp-szine-java/exotic_modernbert_128k_tokenizer_modified`