athms / mad-lab
A MAD laboratory to improve AI architecture designs ๐งช
โ109Updated 3 months ago
Alternatives and similar repositories for mad-lab:
Users that are interested in mad-lab are comparing it to the libraries listed below
- Understand and test language model architectures on synthetic tasks.โ189Updated last month
- Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"โ226Updated last month
- The simplest, fastest repository for training/finetuning medium-sized GPTs.โ103Updated 4 months ago
- โ51Updated 10 months ago
- some common Huggingface transformers in maximal update parametrization (ยตP)โ80Updated 3 years ago
- nanoGPT-like codebase for LLM trainingโ91Updated this week
- โ52Updated 6 months ago
- โ81Updated last year
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clustersโ125Updated 4 months ago
- seqax = sequence modeling + JAXโ152Updated 3 weeks ago
- โ53Updated last year
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.โ59Updated 2 months ago
- โ76Updated 9 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT trainingโ123Updated 11 months ago
- Some preliminary explorations of Mamba's context scaling.โ212Updated last year
- Language models scale reliably with over-training and on downstream tasksโ96Updated last year
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"โ71Updated 5 months ago
- supporting pytorch FSDP for optimizersโ80Updated 3 months ago
- Accelerated First Order Parallel Associative Scanโ180Updated 7 months ago
- Experiment of using Tangent to autodiff tritonโ78Updated last year
- โ141Updated last year
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)โ188Updated 10 months ago
- Normalized Transformer (nGPT)โ164Updated 4 months ago
- โ182Updated this week
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"โ229Updated 2 months ago
- Single File, Single GPU, From Scratch, Efficient, Full Parameter Tuning library for "RL for LLMs"โ119Updated this week
- โ79Updated 11 months ago
- Fast and memory-efficient exact attentionโ67Updated last month
- โ74Updated 7 months ago
- โ165Updated last year