huggingface/lighteval

[BUG] Optimize tokenization

Aperta

#732 aperta il 15 mag 2025

 (1 commento) (0 reazioni) (1 assegnatario)Python (514 fork)auto 404
buggood first issuehelp wantedscience-team

Metriche repository

Star
 (2496 stelle)
Metriche merge PR
 (Metriche PR in attesa)

Descrizione

Describe the bug

For now tokenization is bing made in a for loop, making the whole process very expensive for large benchmarks.

To Reproduce

run any big benchmarks

Expected behavior

Use batch tokenize to speed it up.

Version info

Please provide your operating system, lighteval version or commit if you installed from main, and pip/conda environment if your problem concerns dependencies.

Guida contributor