Alternatives and similar repositories for unit-scaling

Users that are interested in unit-scaling are comparing it to the libraries listed below

• meta-pytorch / float8_experimental
  This repository contains the experimental PyTorch native float8 training UX
  ☆227Updated last year
• proger / accelerated-scan
  Accelerated First Order Parallel Associative Scan
  ☆193Updated last year
• srush / triton-autodiff
  Experiment of using Tangent to autodiff triton
  ☆81Updated last year
• mgmalek / efficient_cross_entropy
  ☆121Updated last year
• cloneofsimo / min-fsdp
  ☆91Updated last year
• nshepperd / flash_attn_jax
  JAX bindings for Flash Attention v2
  ☆101Updated last month
• ethansmith2000 / fsdp_optimizers
  supporting pytorch FSDP for optimizers
  ☆84Updated last year
• HazyResearch / based
  Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"
  ☆243Updated 6 months ago
• MatX-inc / seqax
  seqax = sequence modeling + JAX
  ☆169Updated 5 months ago
• young-geng / scalax
  A simple library for scaling up JAX programs
  ☆144Updated last month
• athms / mad-lab
  A MAD laboratory to improve AI architecture designs 🧪
  ☆135Updated last year
• davisyoshida / lorax
  LoRA for arbitrary JAX models and functions
  ☆143Updated last year
• lucidrains / flash-attention-jax
  Implementation of Flash Attention in Jax
  ☆222Updated last year
• HomebrewML / HeavyBall
  Efficient optimizers
  ☆277Updated this week
• epfml / dynamic-sparse-flash-attention
  ☆150Updated 2 years ago
• google-research / jaxpruner
  ☆234Updated 10 months ago
• HazyResearch / zoology
  Understand and test language model architectures on synthetic tasks.
  ☆246Updated 2 months ago
• graphcore-research / out-of-the-box-fp8-training
  Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.
  ☆46Updated last year
• shikaiqiu / compute-better-spent
  ☆62Updated last year
• lucidrains / triton-transformer
  Implementation of a Transformer, but completely in Triton
  ☆277Updated 3 years ago
• shawntan / scattermoe
  Triton-based implementation of Sparse Mixture of Experts.
  ☆257Updated 2 months ago
• modula-systems / modula
  🧱 Modula software package
  ☆316Updated 4 months ago
• google / aqt
  ☆340Updated 2 weeks ago
• JonasGeiping / linear_cross_entropy_loss
  A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.
  ☆73Updated last year
• cloneofsimo / min-max-gpt
  Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training
  ☆132Updated last year
• EleutherAI / nanoGPT-mup
  The simplest, fastest repository for training/finetuning medium-sized GPTs.
  ☆180Updated 5 months ago
• google / praxis
  ☆190Updated this week
• warner-benjamin / optimi
  Fast, Modern, and Low Precision PyTorch Optimizers
  ☆118Updated 3 months ago
• insuhan / hyper-attn
  ☆83Updated 2 years ago
• apple / ml-ademamix
  ☆68Updated last year