lucaslingle / mu_transformer
Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.
☆30Updated 3 months ago
Alternatives and similar repositories for mu_transformer:
Users that are interested in mu_transformer are comparing it to the libraries listed below
- Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.☆17Updated 2 weeks ago
- Minimal but scalable implementation of large language models in JAX☆34Updated 5 months ago
- Code for the paper "Function-Space Learning Rates"☆17Updated last month
- Using FlexAttention to compute attention with different masking patterns☆42Updated 6 months ago
- ☆19Updated last week
- Machine Learning eXperiment Utilities☆46Updated 9 months ago
- ☆33Updated 6 months ago
- JAX implementation of the Mistral 7b v0.2 model☆35Updated 8 months ago
- An implementation of the Llama architecture, to instruct and delight☆21Updated 2 months ago
- Automatically take good care of your preemptible TPUs☆36Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆80Updated 3 years ago
- Combining SOAP and MUON☆13Updated last month
- [Oral; Neurips OPT2024 ] μLO: Compute-Efficient Meta-Generalization of Learned Optimizers☆12Updated 2 weeks ago
- ☆47Updated last week
- ☆20Updated last year
- ☆53Updated last year
- ☆30Updated 4 months ago
- JAX Scalify: end-to-end scaled arithmetics☆16Updated 5 months ago
- My explorations into editing the knowledge and memories of an attention network☆34Updated 2 years ago
- ☆76Updated 8 months ago
- Implementation of Gradient Agreement Filtering, from Chaubard et al. of Stanford, but for single machine microbatches, in Pytorch☆23Updated 2 months ago
- ☆87Updated 2 weeks ago
- Triton Implementation of HyperAttention Algorithm☆47Updated last year
- ☆31Updated 11 months ago
- Jax like function transformation engine but micro, microjax☆30Updated 5 months ago
- ☆51Updated 10 months ago
- Here we will test various linear attention designs.☆60Updated 11 months ago
- Make triton easier☆47Updated 9 months ago
- Custom triton kernels for training Karpathy's nanoGPT.☆18Updated 5 months ago
- Explorations into adversarial losses on top of autoregressive loss for language modeling☆35Updated last month