VibeusLoadingЗагрузка
Vibeus

Loading is taking longer than usual.Загрузка занимает больше времени, чем обычно.

The full page couldn't load.Не удалось полностью загрузить страницу.

You can read the available content below or try loading the page again.Вы можете читать доступный контент ниже или повторить загрузку страницы.

Aurora1.0-150M has just been released as the first generation of a... - Vibeus

Nora Bennett ·

Aurora1.0-150M has just been released as the first generation of a 150M model. Its performance is described as similar to GPT2-Small.

Reported benchmark scores:
PIQA: 62.24%
Hellaswag: 32.20%
Arc-Easy: 44.91%
Arc-Challenge: 25.00%
Arithmark 3.0: 33.90%
CapitalBench: 36.55%

The model was trained on 7B tokens using an RTX Pro 6000. An example inference script is available in the Hugging Face repository.

Источник

Publication image