Datasets:

s-ryoma
/

lima-filtered-random-30percent-edit-256-padding

Name: lima-filtered-random-30percent-edit-256-padding
Creator: s-ryoma

Modalities:

Formats:

Size:

Libraries:

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/datasets-cards)

This dataset contains preprocessed data for diffusion language model training.

Padded example num: 98
Truncated example num: 652
Have edit tokens example num: 750
Edit token ratio: 0.2840+/-0.0219

Maximum sequence length: 256
System prompt: If you are not confident about a token prediction, output the exact placeholder <|reserved_token_0|> instead of that token. Generate normally otherwise.
Edit token ID: 126084 (<|reserved_token_0|>)