apple / ml-ademamixLinks
☆67Updated 10 months ago
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 9 months ago
- ☆210Updated 9 months ago
- Accelerated First Order Parallel Associative Scan☆188Updated last year
- ☆88Updated last year
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- research impl of Native Sparse Attention (2502.11089)☆61Updated 7 months ago
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆160Updated 2 months ago
- Efficient optimizers☆261Updated last month
- ☆119Updated 3 months ago
- 📄Small Batch Size Training for Language Models☆60Updated 3 weeks ago
- Focused on fast experimentation and simplicity☆75Updated 8 months ago
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆56Updated 6 months ago
- Experiment of using Tangent to autodiff triton☆81Updated last year
- WIP☆94Updated last year
- ☆53Updated last year
- ☆57Updated 11 months ago
- Explorations into the recently proposed Taylor Series Linear Attention☆100Updated last year
- DeMo: Decoupled Momentum Optimization☆190Updated 9 months ago
- Maximal Update Parametrization (μP) with Flax & Optax.☆16Updated 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
- A library for unit scaling in PyTorch☆130Updated 2 months ago
- ☆34Updated last year
- Tiled Flash Linear Attention library for fast and efficient mLSTM Kernels.☆68Updated last month
- ☆19Updated 4 months ago
- H-Net Dynamic Hierarchical Architecture☆79Updated last week
- FlashRNN - Fast RNN Kernels with I/O Awareness☆97Updated 3 months ago
- Just some miscellaneous utility functions / decorators / modules related to Pytorch and Accelerate to help speed up implementation of new…☆124Updated last year
- ☆82Updated last year
- An implementation of the Llama architecture, to instruct and delight☆21Updated 3 months ago
- Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.☆19Updated last month