apple / ml-ademamixLinks
☆68Updated last year
Alternatives and similar repositories for ml-ademamix
Users that are interested in ml-ademamix are comparing it to the libraries listed below
Sorting:
- supporting pytorch FSDP for optimizers☆84Updated 11 months ago
- Supporting code for the blog post on modular manifolds.☆102Updated last month
- research impl of Native Sparse Attention (2502.11089)☆63Updated 9 months ago
- ☆223Updated 11 months ago
- ☆119Updated 5 months ago
- FlashRNN - Fast RNN Kernels with I/O Awareness☆163Updated last month
- DeMo: Decoupled Momentum Optimization☆197Updated 11 months ago
- Accelerated First Order Parallel Associative Scan☆192Updated last year
- ☆61Updated last year
- ☆91Updated last year
- Focused on fast experimentation and simplicity☆75Updated 10 months ago
- Explorations into the recently proposed Taylor Series Linear Attention☆99Updated last year
- Efficient optimizers☆275Updated last week
- WIP☆93Updated last year
- Tiled Flash Linear Attention library for fast and efficient mLSTM Kernels.☆74Updated last week
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆57Updated 8 months ago
- An implementation of PSGD Kron second-order optimizer for PyTorch☆97Updated 3 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆172Updated 4 months ago
- Normalized Transformer (nGPT)☆192Updated last year
- 📄Small Batch Size Training for Language Models☆63Updated last month
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"☆103Updated 10 months ago
- ☆19Updated 6 months ago
- H-Net Dynamic Hierarchical Architecture☆80Updated 2 months ago
- Experiment of using Tangent to autodiff triton☆79Updated last year
- ☆82Updated last year
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆85Updated last year
- ☆140Updated 3 weeks ago
- The AdEMAMix Optimizer: Better, Faster, Older.☆186Updated last year
- Implementation of the proposed Adam-atan2 from Google Deepmind in Pytorch☆134Updated last month