jlamprou / Infini-Attention

Efficient Infinite Context Transformers with Infini-attention Pytorch Implementation + QwenMoE Implementation + Training Script + 1M context keypass retrieval
https://arxiv.org/abs/2404.07143
64 stars 5 forks source link

missing import of apply_rotary_pos_emb #2

Open winglian opened 6 months ago