A library for unit scaling in PyTorch
☆137Jul 11, 2025Updated last year
Alternatives and similar repositories for unit-scaling
Users that are interested in unit-scaling are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆46Jul 17, 2024Updated 2 years ago
- JAX for Graphcore IPU (experimental)☆21Mar 12, 2024Updated 2 years ago
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆89Jul 28, 2024Updated 2 years ago
- Flexibly track outputs and grad-outputs of torch.nn.Module.☆13Oct 6, 2023Updated 3 years ago
- JAX Scalify: end-to-end scaled arithmetics☆18Oct 30, 2024Updated 2 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- WIP☆97Aug 13, 2024Updated 2 years ago
- ☆22Nov 9, 2024Updated last year
- ☆30Dec 2, 2024Updated last year
- 🧱 Modula software package☆347Aug 18, 2025Updated last year
- Official implementation of 'A Large-Scale Exploration of mu-Transfer' (CoRR 2024)☆32Jun 5, 2025Updated last year
- These papers will provide unique insightful concepts that will broaden your perspective on neural networks and deep learning☆48Sep 3, 2023Updated 3 years ago
- Code for the paper "Function-Space Learning Rates"☆23Jun 3, 2025Updated last year
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆206Jan 19, 2026Updated 9 months ago
- Generic floating-point types in Python☆20Oct 18, 2026Updated last week
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Apr 17, 2024Updated 2 years ago
- maximal update parametrization (µP)☆1,756Jul 17, 2024Updated 2 years ago
- Scaling is a distributed training library and installable dependency designed to scale up neural networks, with a dedicated module for tr…☆66Nov 18, 2025Updated 11 months ago
- An implementation of the Llama architecture, to instruct and delight☆21May 31, 2025Updated last year
- Interactive version of the CuTe layout paper☆56Apr 14, 2026Updated 6 months ago
- ☆307Jul 15, 2024Updated 2 years ago
- ☆72Sep 11, 2026Updated last month
- Understand and test language model architectures on synthetic tasks.☆291Mar 22, 2026Updated 7 months ago
- ☆126May 28, 2024Updated 2 years ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆605Aug 14, 2026Updated 2 months ago
- Reviving Any-Order Autoregressive Models via Principled Parallel Sampling and Speculative Decoding☆16Nov 16, 2025Updated 11 months ago
- Latent Large Language Models☆19Aug 24, 2024Updated 2 years ago
- PySCF on IPU☆44Mar 12, 2024Updated 2 years ago
- Code to reproduce key results accompanying "SAEs (usually) Transfer Between Base and Chat Models"☆13Jul 18, 2024Updated 2 years ago
- Repository for Skill Set Optimization☆14Jul 26, 2024Updated 2 years ago
- seqax = sequence modeling + JAX☆199Jul 23, 2025Updated last year
- Efficient optimizers☆340Sep 9, 2026Updated last month
- Schedule-Free Optimization in PyTorch☆2,333Jul 28, 2026Updated 3 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Implementation of papers in 101 lines of code.☆18Nov 12, 2023Updated 2 years ago
- ☆19Dec 4, 2025Updated 10 months ago
- Accelerated First Order Parallel Associative Scan☆197Jan 7, 2026Updated 9 months ago
- ☆48Nov 1, 2025Updated last year
- Microsoft Automatic Mixed Precision Library☆634Dec 1, 2025Updated 10 months ago
- Maximal Update Parametrization (μP) with Flax & Optax.☆16Dec 27, 2023Updated 2 years ago
- ☆51Jan 18, 2024Updated 2 years ago