warner-benjamin / optimiLinks
Fast, Modern, and Low Precision PyTorch Optimizers
☆116Updated 2 months ago
Alternatives and similar repositories for optimi
Users that are interested in optimi are comparing it to the libraries listed below
Sorting:
- ☆121Updated last year
- supporting pytorch FSDP for optimizers☆83Updated 11 months ago
- ☆20Updated 2 years ago
- ☆91Updated last year
- Various transformers for FSDP research☆38Updated 3 years ago
- ☆50Updated last year
- ☆21Updated last year
- HomebrewNLP in JAX flavour for maintable TPU-Training☆51Updated last year
- A library for unit scaling in PyTorch☆132Updated 4 months ago
- Griffin MQA + Hawk Linear RNN Hybrid☆89Updated last year
- A fast implementation of T5/UL2 in PyTorch using Flash Attention☆110Updated 2 weeks ago
- Easily run PyTorch on multiple GPUs & machines☆47Updated 2 weeks ago
- A byte-level decoder architecture that matches the performance of tokenized Transformers.☆66Updated last year
- An implementation of the Llama architecture, to instruct and delight☆21Updated 5 months ago
- LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence☆60Updated 3 years ago
- A place to store reusable transformer components of my own creation or found on the interwebs☆59Updated 3 weeks ago
- ☆81Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆86Updated 3 years ago
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"☆103Updated 10 months ago
- Automatically take good care of your preemptible TPUs☆37Updated 2 years ago
- Efficient optimizers☆276Updated 3 weeks ago
- research impl of Native Sparse Attention (2502.11089)☆63Updated 8 months ago
- ☆34Updated last year
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated 2 years ago
- ☆53Updated last year
- Experiment of using Tangent to autodiff triton☆79Updated last year
- Fast modular code to create and train cutting edge LLMs☆68Updated last year
- Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.☆18Updated 3 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- Load compute kernels from the Hub☆326Updated this week