learning-at-home / lean_transformer
Memory-efficient transformer. Work in progress.
☆19Updated 2 years ago
Related projects: ⓘ
- Implementation of Token Shift GPT - An autoregressive model that solely relies on shifting the sequence space for mixing☆47Updated 2 years ago
- ☆29Updated last year
- Implementation of some personal helper functions for Einops, my most favorite tensor manipulation library ❤️☆52Updated last year
- Another attempt at a long-context / efficient transformer by me☆37Updated 2 years ago
- A GPT, made only of MLPs, in Jax☆55Updated 3 years ago
- Using FlexAttention to compute attention with different masking patterns☆28Updated last week
- High performance pytorch modules☆18Updated last year
- An open source implementation of CLIP.☆32Updated last year
- Implementation of Kronecker Attention in Pytorch☆17Updated 4 years ago
- An attempt to merge ESBN with Transformers, to endow Transformers with the ability to emergently bind symbols☆14Updated 3 years ago
- A python library for highly configurable transformers - easing model architecture search and experimentation.☆50Updated 2 years ago
- A collection of optimizers, some arcane others well known, for Flax.☆29Updated 3 years ago
- "Towards Crowdsourced Training of Large Neural Networks using Decentralized Mixture-of-Experts" (NeurIPS 2020), original PyTorch implemen…☆53Updated 3 years ago
- ☆27Updated 3 years ago
- A case study of efficient training of large language models using commodity hardware.☆68Updated 2 years ago
- My explorations into editing the knowledge and memories of an attention network☆34Updated last year
- ☆15Updated this week
- ☆40Updated this week
- Triton Implementation of HyperAttention Algorithm☆46Updated 9 months ago
- JAX/Flax implementation of the Hyena Hierarchy☆29Updated last year
- ☆21Updated this week
- An implementation of Transformer with Expire-Span, a circuit for learning which memories to retain☆33Updated 3 years ago
- Blog post☆16Updated 7 months ago
- A collection of Models, Datasets, DataModules, Callbacks, Metrics, Losses and Loggers to better integrate pytorch-lightning with transfor…☆47Updated last year
- Hacks for PyTorch☆17Updated last year
- Standalone Product Key Memory module in Pytorch - for augmenting Transformer models☆72Updated last month
- A minimal TPU compatible Jax implementation of NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis.☆13Updated 2 years ago
- Contains my experiments with the `big_vision` repo to train ViTs on ImageNet-1k.☆22Updated last year
- CUDA implementation of autoregressive linear attention, with all the latest research findings☆43Updated last year