ethansmith2000 / fsdp_optimizers
supporting pytorch FSDP for optimizers
☆79Updated 3 months ago
Alternatives and similar repositories for fsdp_optimizers:
Users that are interested in fsdp_optimizers are comparing it to the libraries listed below
- WIP☆93Updated 7 months ago
- ☆76Updated 8 months ago
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆100Updated 4 months ago
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆73Updated 7 months ago
- Focused on fast experimentation and simplicity☆69Updated 2 months ago
- ☆53Updated last year
- Fast, Modern, Memory Efficient, and Low Precision PyTorch Optimizers☆87Updated 8 months ago
- Efficient optimizers☆183Updated last week
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆122Updated 11 months ago
- A MAD laboratory to improve AI architecture designs 🧪☆107Updated 3 months ago
- ☆33Updated 6 months ago
- ☆169Updated 3 months ago
- ☆24Updated 3 months ago
- Understand and test language model architectures on synthetic tasks.☆184Updated 2 weeks ago
- ☆79Updated 11 months ago
- ☆49Updated last year
- seqax = sequence modeling + JAX☆148Updated last week
- DeMo: Decoupled Momentum Optimization☆182Updated 3 months ago
- Normalized Transformer (nGPT)☆162Updated 4 months ago
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆121Updated 6 months ago
- Implementation of PSGD optimizer in JAX☆28Updated 2 months ago
- 🧱 Modula software package☆173Updated last week
- ☆95Updated 9 months ago
- ☆89Updated 2 months ago
- Implementation of GateLoop Transformer in Pytorch and Jax☆87Updated 9 months ago
- ☆58Updated 4 months ago
- ☆51Updated 10 months ago
- Experiment of using Tangent to autodiff triton☆76Updated last year