HomebrewML / HeavyBallLinks
Efficient optimizers
☆254Updated 3 weeks ago
Alternatives and similar repositories for HeavyBall
Users that are interested in HeavyBall are comparing it to the libraries listed below
Sorting:
- supporting pytorch FSDP for optimizers☆84Updated 8 months ago
- ☆208Updated 8 months ago
- 🧱 Modula software package☆222Updated 3 weeks ago
- CIFAR-10 speedruns: 94% in 2.6 seconds and 96% in 27 seconds☆283Updated last month
- Accelerated First Order Parallel Associative Scan☆187Updated last year
- ☆275Updated last year
- The AdEMAMix Optimizer: Better, Faster, Older.☆184Updated 11 months ago
- An implementation of PSGD Kron second-order optimizer for PyTorch☆96Updated 3 weeks ago
- A library for unit scaling in PyTorch☆129Updated last month
- ☆307Updated last year
- WIP☆94Updated last year
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆152Updated last month
- A repository for log-time feedforward networks☆223Updated last year
- For optimization algorithm research and development.☆530Updated this week
- Quick implementation of nGPT, learning entirely on the hypersphere, from NvidiaAI☆289Updated 2 months ago
- Normalized Transformer (nGPT)☆186Updated 9 months ago
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆85Updated last year
- ☆65Updated 9 months ago
- Annotated version of the Mamba paper☆487Updated last year
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆130Updated last year
- DeMo: Decoupled Momentum Optimization☆190Updated 8 months ago
- Pytorch implementation of preconditioned stochastic gradient descent (Kron and affine preconditioner, low-rank approximation precondition…☆179Updated this week
- ☆87Updated last year
- Universal Tensor Operations in Einstein-Inspired Notation for Python.☆399Updated 4 months ago
- Fast, Modern, and Low Precision PyTorch Optimizers☆108Updated 3 weeks ago
- Scalable and Performant Data Loading☆291Updated last week
- seqax = sequence modeling + JAX☆166Updated last month
- Official repository for the paper "Grokfast: Accelerated Grokking by Amplifying Slow Gradients"☆560Updated last year
- Understand and test language model architectures on synthetic tasks.☆221Updated last month
- ☆115Updated 2 months ago