Alternatives and similar repositories for SOAP

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

• cloneofsimo / zeroshampoo
  ☆34Updated last year
• main-horse / hnet-old
  H-Net Dynamic Hierarchical Architecture
  ☆81Updated 4 months ago
• ethansmith2000 / fsdp_optimizers
  supporting pytorch FSDP for optimizers
  ☆84Updated last year
• euclaise / supertrainer2000
  ☆50Updated last year
• cloneofsimo / min-max-gpt
  Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training
  ☆132Updated last year
• fal-ai-community / NativeSparseAttention
  research impl of Native Sparse Attention (2502.11089)
  ☆63Updated 11 months ago
• lucidrains / grokfast-pytorch
  Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"
  ☆103Updated last year
• warner-benjamin / optimi
  Fast, Modern, and Low Precision PyTorch Optimizers
  ☆124Updated last month
• amirzandieh / HyperAttention
  Triton Implementation of HyperAttention Algorithm
  ☆48Updated 2 years ago
• ethansmith2000 / TransformerExperiments
  ☆19Updated 2 months ago
• dvruette / barrel-rec-pytorch
  ☆53Updated 2 years ago
• epfml / DenseFormer
  ☆82Updated last year
• joey00072 / ohara
  Collection of autoregressive model implementation
  ☆85Updated 3 weeks ago
• evanatyourservice / llm-jax
  Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.
  ☆18Updated 6 months ago
• BlinkDL / SmallInitEmb
  LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence
  ☆61Updated 3 years ago
• zaydzuhri / flame
  Fork of Flame repo for training of some new stuff in development
  ☆19Updated last month
• bloc97 / DeMo
  DeMo: Decoupled Momentum Optimization
  ☆198Updated last year
• kaiokendev / cutoff-len-is-context-len
  Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit
  ☆63Updated 2 years ago
• HomebrewML / Olmax
  HomebrewNLP in JAX flavour for maintable TPU-Training
  ☆51Updated 2 years ago
• lucidrains / transformer-lm-gan
  Explorations into adversarial losses on top of autoregressive loss for language modeling
  ☆41Updated last month
• zaydzuhri / softpick-attention
  Landing repository for the paper "Softpick: No Attention Sink, No Massive Activations with Rectified Softmax"
  ☆86Updated 4 months ago
• lucidrains / GAF-microbatch-pytorch
  Implementation of Gradient Agreement Filtering, from Chaubard et al. of Stanford, but for single machine microbatches, in Pytorch
  ☆25Updated last year
• neilrathi / token-filtering
  Shaping capabilities with token-level pretraining data filtering
  ☆75Updated last week
• cloneofsimo / min-fsdp
  ☆92Updated last year
• lucidrains / pause-transformer
  Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…
  ☆53Updated 2 years ago
• lucidrains / taylor-series-linear-attention
  Explorations into the recently proposed Taylor Series Linear Attention
  ☆100Updated last year
• recursal / GoldFinch-paper
  GoldFinch and other hybrid transformer components
  ☆45Updated last year
• cat-state / tinypar
  ☆20Updated 2 years ago
• crowsonkb / LDLM
  Latent Diffusion Language Models
  ☆70Updated 2 years ago
• crowsonkb / torch-dist-utils
  Utilities for PyTorch distributed
  ☆25Updated 11 months ago