CerebrasResearch / Sparse-IFT
Official repository of Sparse ISO-FLOP Transformations for Maximizing Training Efficiency
☆25Updated 7 months ago
Alternatives and similar repositories for Sparse-IFT:
Users that are interested in Sparse-IFT are comparing it to the libraries listed below
- Code for studying the super weight in LLM☆94Updated 3 months ago
- Understand and test language model architectures on synthetic tasks.☆185Updated 2 weeks ago
- Experiment of using Tangent to autodiff triton☆78Updated last year
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆68Updated 9 months ago
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆122Updated 7 months ago
- A MAD laboratory to improve AI architecture designs 🧪☆108Updated 3 months ago
- Simple and efficient pytorch-native transformer training and inference (batched)☆71Updated 11 months ago
- Simple implementation of Speculative Sampling in NumPy for GPT-2.☆92Updated last year
- Collection of autoregressive model implementation☆83Updated last month
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆58Updated last month
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆125Updated 3 months ago
- ☆76Updated 8 months ago
- Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*☆81Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆80Updated 3 years ago
- ☆49Updated 4 months ago
- ☆51Updated 10 months ago
- ☆81Updated last year
- ☆125Updated last year
- Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry☆40Updated last year
- Triton Implementation of HyperAttention Algorithm☆47Updated last year
- Load compute kernels from the Hub☆99Updated this week
- ☆43Updated last year
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆121Updated 7 months ago
- ☆12Updated 3 weeks ago
- ☆47Updated 6 months ago
- Custom triton kernels for training Karpathy's nanoGPT.☆18Updated 5 months ago
- Code associated to papers on superposition (in ML interpretability)☆28Updated 2 years ago
- Triton-based implementation of Sparse Mixture of Experts.☆208Updated 3 months ago
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models"☆59Updated 5 months ago
- JAX bindings for Flash Attention v2☆88Updated 8 months ago