Alternatives and similar repositories for mup

Users that are interested in mup are comparing it to the libraries listed below

• edwardmilsom / function-space-learning-rates-paper
  Code for the paper "Function-Space Learning Rates"
  ☆23Updated 6 months ago
• cloneofsimo / min-max-gpt
  Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training
  ☆132Updated last year
• wmn-231314 / diffusion-data-constraint
  Official PyTorch implementation and models for paper "Diffusion Beats Autoregressive in Data-Constrained Settings". We find diffusion mod…
  ☆115Updated last month
• ethansmith2000 / fsdp_optimizers
  supporting pytorch FSDP for optimizers
  ☆84Updated last year
• fal-ai-community / nano-mdm
  Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun
  ☆56Updated 9 months ago
• EleutherAI / nanoGPT-mup
  The simplest, fastest repository for training/finetuning medium-sized GPTs.
  ☆180Updated 5 months ago
• cloneofsimo / scaling-guide
  WIP
  ☆93Updated last year
• katiekang1998 / reasoning_generalization
  ☆33Updated 11 months ago
• wesg52 / universal-neurons
  Universal Neurons in GPT2 Language Models
  ☆31Updated last year
• mcleish7 / arithmetic
  Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)
  ☆198Updated last year
• apoorvkh / academic-pretraining
  $100K or 100 Days: Trade-offs when Pre-Training with Academic Resources
  ☆149Updated 2 months ago
• dvruette / gidd
  Code accompanying the paper "Generalized Interpolating Discrete Diffusion"
  ☆112Updated 6 months ago
• cloneofsimo / ezmup
  Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam
  ☆84Updated last year
• cloneofsimo / minVJEPA
  ☆24Updated 7 months ago
• cloneofsimo / min-fsdp
  ☆91Updated last year
• RobertCsordas / moeut
  ☆91Updated last year
• cloneofsimo / zeroshampoo
  ☆34Updated last year
• epfml / schedules-and-scaling
  Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"
  ☆86Updated last year
• google-deepmind / alta
  ☆27Updated 3 months ago
• mlfoundations / scaling
  Language models scale reliably with over-training and on downstream tasks
  ☆100Updated last year
• JacobPfau / fillerTokens
  ☆75Updated last year
• BorealisAI / flora-opt
  This is the official repository for the paper "Flora: Low-Rank Adapters Are Secretly Gradient Compressors" in ICML 2024.
  ☆105Updated last year
• epfml / DenseFormer
  ☆82Updated last year
• berlino / seq_icl
  ☆53Updated last year
• main-horse / hnet-old
  H-Net Dynamic Hierarchical Architecture
  ☆80Updated 3 months ago
• CLAIRE-Labo / flash_attention
  A basic pure pytorch implementation of flash attention
  ☆16Updated last year
• evanatyourservice / llm-jax
  Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.
  ☆18Updated 5 months ago
• ChenWu98 / algorithmic-creativity
  [ICML 2025] Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction
  ☆80Updated 6 months ago
• thinking-machines-lab / manifolds
  Supporting code for the blog post on modular manifolds.
  ☆107Updated 2 months ago
• JoshEngels / MultiDimensionalFeatures
  Code for reproducing our paper "Not All Language Model Features Are Linear"
  ☆84Updated last year