eth-easl / fmengineLinks
Utilities for Training Very Large Models
☆58Updated 8 months ago
Alternatives and similar repositories for fmengine
Users that are interested in fmengine are comparing it to the libraries listed below
Sorting:
- Using FlexAttention to compute attention with different masking patterns☆43Updated 8 months ago
- Easily run PyTorch on multiple GPUs & machines☆46Updated 2 months ago
- Code for the examples presented in the talk "Training a Llama in your backyard: fine-tuning very large models on consumer hardware" given…☆14Updated last year
- Simple and efficient pytorch-native transformer training and inference (batched)☆75Updated last year
- Truly flash T5 realization!☆67Updated last year
- ☆47Updated 9 months ago
- This is a new metric that can be used to evaluate faithfulness of text generated by LLMs. The work behind this repository can be found he…☆31Updated last year
- ☆34Updated 11 months ago
- ☆78Updated 11 months ago
- ☆45Updated 3 months ago
- Triton Implementation of HyperAttention Algorithm☆48Updated last year
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆62Updated 4 months ago
- Make triton easier☆47Updated 11 months ago
- Supercharge huggingface transformers with model parallelism.☆77Updated 7 months ago
- ☆72Updated last year
- Official repository for the paper "Approximating Two-Layer Feedforward Networks for Efficient Transformers"☆37Updated last year
- Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit☆62Updated last year
- Official repository of paper "RNNs Are Not Transformers (Yet): The Key Bottleneck on In-context Retrieval"☆27Updated last year
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆70Updated last year
- QAmeleon introduces synthetic multilingual QA data using PaLM, a 540B large language model. This dataset was generated by prompt tuning P…☆34Updated last year
- CUDA implementation of autoregressive linear attention, with all the latest research findings☆44Updated 2 years ago
- [NeurIPS 2023] Sparse Modular Activation for Efficient Sequence Modeling☆36Updated last year
- Experiment of using Tangent to autodiff triton☆79Updated last year
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆44Updated 10 months ago
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆67Updated 10 months ago
- An unofficial implementation of the Infini-gram model proposed by Liu et al. (2024)☆33Updated 11 months ago
- Here we will test various linear attention designs.☆58Updated last year
- ☆49Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆80Updated 3 years ago
- My explorations into editing the knowledge and memories of an attention network☆35Updated 2 years ago