davisyoshida / haiku-mupLinks
A port of muP to JAX/Haiku
☆25Updated 2 years ago
Alternatives and similar repositories for haiku-mup
Users that are interested in haiku-mup are comparing it to the libraries listed below
Sorting:
- ☆17Updated 11 months ago
- LoRA for arbitrary JAX models and functions☆140Updated last year
- JMP is a Mixed Precision library for JAX.☆207Updated 6 months ago
- If it quacks like a tensor...☆58Updated 8 months ago
- A JAX implementation of stochastic addition.☆14Updated 2 years ago
- A functional training loops library for JAX☆88Updated last year
- minGPT in JAX☆48Updated 3 years ago
- [NeurIPS'19] Deep Equilibrium Models Jax Implementation☆40Updated 4 years ago
- Silly twitter torch implementations.☆46Updated 2 years ago
- JAX Synergistic Memory Inspector☆177Updated last year
- ☆24Updated 6 years ago
- ☆104Updated last year
- A simple library for scaling up JAX programs☆140Updated 9 months ago
- Pytorch implementation of preconditioned stochastic gradient descent (Kron and affine preconditioner, low-rank approximation precondition…☆180Updated last week
- ☆61Updated 3 years ago
- Pytorch-like dataloaders for JAX.☆94Updated 2 months ago
- JAX Arrays for human consumption☆106Updated last month
- ☆158Updated last year
- Neural Networks for JAX☆84Updated 10 months ago
- Maximal Update Parametrization (μP) with Flax & Optax.☆16Updated last year
- Fast Discounted Cumulative Sums in PyTorch☆96Updated 3 years ago
- ☆40Updated last year
- A Python package of computer vision models for the Equinox ecosystem.☆107Updated last year
- Stochastic Automatic Differentiation library for PyTorch.☆205Updated 11 months ago
- ☆80Updated 3 years ago
- 🧱 Modula software package☆210Updated last week
- ☆115Updated this week
- Visualize, create, and operate on pytrees in the most intuitive way possible.☆45Updated 6 months ago
- Running Jax in PyTorch Lightning☆109Updated 7 months ago
- Code accompanying our paper "Feature Learning in Infinite-Width Neural Networks" (https://arxiv.org/abs/2011.14522)☆62Updated 4 years ago