tech-srl / layer_norm_expressivity_roleLinks
Code for the paper "On the Expressivity Role of LayerNorm in Transformers' Attention" (Findings of ACL'2023)
☆57Updated last year
Alternatives and similar repositories for layer_norm_expressivity_role
Users that are interested in layer_norm_expressivity_role are comparing it to the libraries listed below
Sorting:
- ☆52Updated last year
- Using FlexAttention to compute attention with different masking patterns☆45Updated last year
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆84Updated 11 months ago
- Triton Implementation of HyperAttention Algorithm☆48Updated last year
- ☆38Updated last year
- ☆101Updated last year
- ☆31Updated last year
- One Initialization to Rule them All: Fine-tuning via Explained Variance Adaptation☆43Updated 11 months ago
- Official PyTorch implementation and models for paper "Diffusion Beats Autoregressive in Data-Constrained Settings". We find diffusion mod…☆101Updated last month
- ☆32Updated last year
- A basic pure pytorch implementation of flash attention☆16Updated 11 months ago
- Here we will test various linear attention designs.☆61Updated last year
- Deep Networks Grok All the Time and Here is Why☆37Updated last year
- ☆85Updated last year
- [NeurIPS 2022] Your Transformer May Not be as Powerful as You Expect (official implementation)☆32Updated 2 years ago
- Official code for the paper "Attention as a Hypernetwork"☆43Updated last year
- Official Repository of Pretraining Without Attention (BiGS), BiGS is the first model to achieve BERT-level transfer learning on the GLUE …☆114Updated last year
- Source-to-Source Debuggable Derivatives in Pure Python☆15Updated last year
- Why Do We Need Weight Decay in Modern Deep Learning? [NeurIPS 2024]☆68Updated last year
- ☆56Updated last year
- Explorations into the recently proposed Taylor Series Linear Attention☆99Updated last year
- Code for the paper "Query-Key Normalization for Transformers"☆48Updated 4 years ago
- ☆58Updated last year
- [ICML 2025] Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction☆71Updated 4 months ago
- Implementation of Infini-Transformer in Pytorch☆113Updated 9 months ago
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated last year
- HGRN2: Gated Linear RNNs with State Expansion☆54Updated last year
- The official repository for our paper "The Dual Form of Neural Networks Revisited: Connecting Test Time Predictions to Training Patterns …☆16Updated 4 months ago
- ☆81Updated last year