davisyoshida / abnormal-floatsLinks
Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)
☆21Updated 2 years ago
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 NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆84Updated 10 months ago
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆53Updated last year
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆73Updated last year
- ☆89Updated last year
- ☆53Updated last year
- Simple and efficient pytorch-native transformer training and inference (batched)☆78Updated last year
- ☆49Updated last year
- ☆32Updated last year
- Experiment of using Tangent to autodiff triton☆81Updated last year
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated last year
- A toolkit for scaling law research ⚖☆51Updated 8 months ago
- ☆13Updated 4 months ago
- Language models scale reliably with over-training and on downstream tasks☆100Updated last year
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆70Updated last year
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆46Updated last year
- sigma-MoE layer☆20Updated last year
- LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence☆59Updated 3 years ago
- Using FlexAttention to compute attention with different masking patterns☆44Updated last year
- Triton Implementation of HyperAttention Algorithm☆48Updated last year
- Here we will test various linear attention designs.☆62Updated last year
- [NeurIPS 2023] Sparse Modular Activation for Efficient Sequence Modeling☆40Updated last year
- Awesome Triton Resources☆33Updated 5 months ago
- train with kittens!☆62Updated 11 months ago
- Utilities for Training Very Large Models☆58Updated last year
- Official repository for the paper "Approximating Two-Layer Feedforward Networks for Efficient Transformers"☆38Updated 3 months ago
- ☆39Updated 3 weeks ago
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year
- Minimum Description Length probing for neural network representations☆18Updated 8 months ago
- Revisiting Efficient Training Algorithms For Transformer-based Language Models (NeurIPS 2023)☆81Updated 2 years ago
- Official Repository of Pretraining Without Attention (BiGS), BiGS is the first model to achieve BERT-level transfer learning on the GLUE …☆115Updated last year