Tag
1 verified claim carrying this tag. Each has 2+ primary sources and an HMAC-SHA256 signature.
Grouped-Query Attention (GQA) introduced in paper: GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints (Ainslie et al., 2023).
3e9122ba60a3fe99 · 3 sources · 92% confidence