lucidrains / memory-compressed-attentionLinks
Implementation of Memory-Compressed Attention, from the paper "Generating Wikipedia By Summarizing Long Sequences"
☆69Updated 2 years ago
Alternatives and similar repositories for memory-compressed-attention
Users that are interested in memory-compressed-attention are comparing it to the libraries listed below
Sorting:
- Axial Positional Embedding for Pytorch☆83Updated 7 months ago
- Implementation of Token Shift GPT - An autoregressive model that solely relies on shifting the sequence space for mixing☆50Updated 3 years ago
- My explorations into editing the knowledge and memories of an attention network☆35Updated 2 years ago
- Standalone Product Key Memory module in Pytorch - for augmenting Transformer models☆83Updated last year
- A variant of Transformer-XL where the memory is updated not with a queue, but with attention☆49Updated 5 years ago
- ResiDual: Transformer with Dual Residual Connections, https://arxiv.org/abs/2304.14802☆95Updated 2 years ago
- Implementation of a Transformer using ReLA (Rectified Linear Attention) from https://arxiv.org/abs/2104.07012☆49Updated 3 years ago
- Implementation of Memformer, a Memory-augmented Transformer, in Pytorch☆123Updated 4 years ago
- Implementation of N-Grammer, augmenting Transformers with latent n-grams, in Pytorch☆76Updated 2 years ago
- Implementation of Long-Short Transformer, combining local and global inductive biases for attention over long sequences, in Pytorch☆120Updated 4 years ago
- Implementation of Multistream Transformers in Pytorch☆54Updated 4 years ago
- Another attempt at a long-context / efficient transformer by me☆38Updated 3 years ago
- Implementation of Gated State Spaces, from the paper "Long Range Language Modeling via Gated State Spaces", in Pytorch☆101Updated 2 years ago
- Large Scale Distributed Model Training strategy with Colossal AI and Lightning AI☆56Updated 2 years ago
- Some personal experiments around routing tokens to different autoregressive attention, akin to mixture-of-experts☆122Updated 11 months ago
- Skyformer: Remodel Self-Attention with Gaussian Kernel and Nystr\"om Method (NeurIPS 2021)☆63Updated 3 years ago
- Implementation of a Transformer that Ponders, using the scheme from the PonderNet paper☆81Updated 3 years ago
- Implementation of the conditionally routed attention in the CoLT5 architecture, in Pytorch☆230Updated last year
- Implementation of Mega, the Single-head Attention with Multi-headed EMA architecture that currently holds SOTA on Long Range Arena☆207Updated 2 years ago
- Implementation of TableFormer, Robust Transformer Modeling for Table-Text Encoding, in Pytorch☆39Updated 3 years ago
- GPT, but made only out of MLPs☆89Updated 4 years ago
- 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.☆81Updated 3 years ago
- An implementation of Transformer with Expire-Span, a circuit for learning which memories to retain☆34Updated 4 years ago
- ☆29Updated 3 years ago
- ☆67Updated last year
- Unofficial PyTorch implementation of Fastformer based on paper "Fastformer: Additive Attention Can Be All You Need"."☆133Updated 4 years ago
- Sequence modeling with Mega.☆300Updated 2 years ago
- Implementation of Fast Transformer in Pytorch☆177Updated 4 years ago
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated 2 years ago
- Transformers at any scale☆41Updated last year