athms / mad-labLinks
A MAD laboratory to improve AI architecture designs ๐งช
โ122Updated 6 months ago
Alternatives and similar repositories for mad-lab
Users that are interested in mad-lab are comparing it to the libraries listed below
Sorting:
- Understand and test language model architectures on synthetic tasks.โ218Updated 2 weeks ago
- The simplest, fastest repository for training/finetuning medium-sized GPTs.โ139Updated this week
- Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"โ235Updated 3 weeks ago
- โ53Updated last year
- Some preliminary explorations of Mamba's context scaling.โ214Updated last year
- seqax = sequence modeling + JAXโ162Updated 2 weeks ago
- Normalized Transformer (nGPT)โ184Updated 7 months ago
- supporting pytorch FSDP for optimizersโ82Updated 6 months ago
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.โ77Updated 2 weeks ago
- โ78Updated 11 months ago
- nanoGPT-like codebase for LLM trainingโ98Updated last month
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clustersโ126Updated 6 months ago
- Language models scale reliably with over-training and on downstream tasksโ97Updated last year
- some common Huggingface transformers in maximal update parametrization (ยตP)โ81Updated 3 years ago
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"โ74Updated 7 months ago
- โ147Updated 2 years ago
- Accelerated First Order Parallel Associative Scanโ182Updated 10 months ago
- Experiment of using Tangent to autodiff tritonโ79Updated last year
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"โ239Updated 4 months ago
- โ180Updated last year
- EvaByte: Efficient Byte-level Language Models at Scaleโ102Updated 2 months ago
- โ53Updated last year
- A toolkit for scaling law research โโ49Updated 5 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT trainingโ129Updated last year
- Minimal but scalable implementation of large language models in JAXโ35Updated 7 months ago
- โ81Updated last year
- Simple and efficient pytorch-native transformer training and inference (batched)โ76Updated last year
- โ81Updated last year
- ๐งฑ Modula software packageโ200Updated 3 months ago
- โ98Updated 5 months ago