davisyoshida / abnormal-floatsLinks
Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)
☆19Updated last year
Alternatives and similar repositories for abnormal-floats
Users that are interested in abnormal-floats are comparing it to the libraries listed below
Sorting:
- Code for the paper "Function-Space Learning Rates"☆20Updated last month
- Implementation of Hyena Hierarchy in JAX☆10Updated 2 years ago
- ☆32Updated last year
- Experiment of using Tangent to autodiff triton☆79Updated last year
- ☆46Updated last year
- Parallel Associative Scan for Language Models☆18Updated last year
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆54Updated last year
- Using FlexAttention to compute attention with different masking patterns☆43Updated 8 months ago
- Awesome Triton Resources☆28Updated last month
- ☆19Updated 2 weeks ago
- [Oral; Neurips OPT2024 ] μLO: Compute-Efficient Meta-Generalization of Learned Optimizers☆12Updated 2 months ago
- ☆53Updated 8 months ago
- JAX Scalify: end-to-end scaled arithmetics☆16Updated 7 months ago
- ☆29Updated 2 years ago
- ☆78Updated 11 months ago
- Here we will test various linear attention designs.☆58Updated last year
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated last year
- Triton Implementation of HyperAttention Algorithm☆48Updated last year
- LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence☆60Updated 3 years ago
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆44Updated 10 months ago
- ☆29Updated 6 months ago
- ☆20Updated last year
- ☆52Updated last year
- My explorations into editing the knowledge and memories of an attention network☆35Updated 2 years ago
- ☆32Updated last year
- Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.☆30Updated last week
- An implementation of the Llama architecture, to instruct and delight☆21Updated this week
- ☆33Updated 8 months ago
- See https://github.com/cuda-mode/triton-index/ instead!☆10Updated last year
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year