facebookresearch / mega
Sequence modeling with Mega.
☆295Updated 2 years ago
Alternatives and similar repositories for mega:
Users that are interested in mega are comparing it to the libraries listed below
- Implementation of Mega, the Single-head Attention with Multi-headed EMA architecture that currently holds SOTA on Long Range Arena☆204Updated last year
- Implementation of the conditionally routed attention in the CoLT5 architecture, in Pytorch☆226Updated 6 months ago
- Code for the ALiBi method for transformer language models (ICLR 2022)☆520Updated last year
- Official PyTorch Implementation of Long-Short Transformer (NeurIPS 2021).☆225Updated 2 years ago
- Recurrent Memory Transformer☆149Updated last year
- FairSeq repo with Apollo optimizer☆112Updated last year
- ☆164Updated 2 years ago
- Implementation of Block Recurrent Transformer - Pytorch☆218Updated 7 months ago
- Implementation of Gated State Spaces, from the paper "Long Range Language Modeling via Gated State Spaces", in Pytorch☆99Updated 2 years ago
- Implementation of Recurrent Memory Transformer, Neurips 2022 paper, in Pytorch☆407Updated 2 months ago
- An implementation of local windowed attention for language modeling☆431Updated 2 months ago
- Long Range Arena for Benchmarking Efficient Transformers☆748Updated last year
- Randomized Positional Encodings Boost Length Generalization of Transformers☆80Updated last year
- Implementation of fused cosine similarity attention in the same style as Flash Attention☆212Updated 2 years ago
- ☆344Updated 11 months ago
- ☆141Updated last year
- Sequence Modeling with Structured State Spaces☆63Updated 2 years ago
- Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"☆374Updated last year
- An implementation of masked language modeling for Pytorch, made as concise and simple as possible☆178Updated last year
- Implementation of Memformer, a Memory-augmented Transformer, in Pytorch☆115Updated 4 years ago
- ☆165Updated last year
- Implementation of ST-Moe, the latest incarnation of MoE after years of research at Brain, in Pytorch☆323Updated 9 months ago
- Pytorch implementation of Compressive Transformers, from Deepmind☆156Updated 3 years ago
- ☆254Updated 2 years ago
- Implementation of Linformer for Pytorch☆276Updated last year
- Implementation of Rotary Embeddings, from the Roformer paper, in Pytorch☆656Updated 4 months ago
- Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT☆210Updated 7 months ago
- Code for the paper "The Impact of Positional Encoding on Length Generalization in Transformers", NeurIPS 2023☆131Updated 11 months ago
- [ICLR 2022] Official implementation of cosformer-attention in cosFormer: Rethinking Softmax in Attention☆188Updated 2 years ago
- Efficient Transformers with Dynamic Token Pooling☆60Updated last year