lucaslingle / mu_transformerLinks
Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.
☆32Updated 4 months ago
Alternatives and similar repositories for mu_transformer
Users that are interested in mu_transformer are comparing it to the libraries listed below
Sorting:
- An implementation of the Llama architecture, to instruct and delight☆21Updated 4 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- JAX implementation of the Mistral 7b v0.2 model☆36Updated last year
- Code for the paper "Function-Space Learning Rates"☆23Updated 4 months ago
- Automatically take good care of your preemptible TPUs☆36Updated 2 years ago
- Jax like function transformation engine but micro, microjax☆32Updated 11 months ago
- ☆91Updated last year
- ☆64Updated 6 months ago
- Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.☆19Updated 2 months ago
- some common Huggingface transformers in maximal update parametrization (µP)☆82Updated 3 years ago
- Make triton easier☆47Updated last year
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆164Updated 3 months ago
- Machine Learning eXperiment Utilities☆47Updated 2 months ago
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆46Updated last year
- Triton Implementation of HyperAttention Algorithm☆48Updated last year
- ☆34Updated last year
- ☆58Updated last year
- Minimal but scalable implementation of large language models in JAX☆35Updated last month
- ☆53Updated last year
- Simple repository for training small reasoning models☆40Updated 8 months ago
- Experiment of using Tangent to autodiff triton☆81Updated last year
- Unofficial but Efficient Implementation of "Mamba: Linear-Time Sequence Modeling with Selective State Spaces" in JAX☆89Updated last year
- A place to store reusable transformer components of my own creation or found on the interwebs☆60Updated 2 weeks ago
- ☆15Updated last year
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"☆102Updated 9 months ago
- ☆19Updated 4 months ago
- Exploration into the proposed "Self Reasoning Tokens" by Felipe Bonetto☆57Updated last year
- supporting pytorch FSDP for optimizers☆84Updated 9 months ago
- ☆46Updated last year
- GoldFinch and other hybrid transformer components☆45Updated last year