CarperAI/trlx
Ver no GitHubAdd tensor padding options to improve hardware utilization
Open
#284 aberto em 6 de fev. de 2023
contributions welcomefeature requestgood first issue
Métricas do repositório
- Stars
- (4.184 stars)
- Métricas de merge de PR
- (Nenhuma PRs mesclada em 30d)
Description
🚀 The feature, motivation, and pitch
Recent discussion on Twitter highlighted the importance of tensor padding to improve hardware util. NeMo seems to already support GPU-friendly vocab size padding from discussions with @cat-state but we should also consider adding an optional pad_to_multiple_of arg to our tokenizer calls with the accelerate/transformers backend to satisfy Tensor Core requirements.

The implementer should also provide system plots to display any improvements/findings.
Alternatives
No response
Additional context
No response