alexrs / herd
Mixture of Expert (MoE) techniques for enhancing LLM performance through expert-driven prompt mapping and adapter combinations.
☆12Updated last year
Alternatives and similar repositories for herd:
Users that are interested in herd are comparing it to the libraries listed below
- Script for processing OpenAI's PRM800K process supervision dataset into an Alpaca-style instruction-response format☆27Updated last year
- ☆33Updated 10 months ago
- A public implementation of the ReLoRA pretraining method, built on Lightning-AI's Pytorch Lightning suite.☆33Updated last year
- 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
- Repository containing the SPIN experiments on the DIBT 10k ranked prompts☆24Updated last year
- Data preparation code for CrystalCoder 7B LLM☆44Updated 11 months ago
- Reference implementation for Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model☆44Updated last year
- [WIP] Transformer to embed Danbooru labelsets☆13Updated last year
- Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment☆55Updated 8 months ago
- ☆15Updated last month
- QAmeleon introduces synthetic multilingual QA data using PaLM, a 540B large language model. This dataset was generated by prompt tuning P…☆34Updated last year
- Implementation of "LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models"☆42Updated 5 months ago
- GoldFinch and other hybrid transformer components☆45Updated 9 months ago
- The repository contains code for Adaptive Data Optimization☆24Updated 4 months ago
- Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit☆63Updated last year
- Aioli: A unified optimization framework for language model data mixing☆25Updated 3 months ago
- The official repository for SkyLadder: Better and Faster Pretraining via Context Window Scheduling☆29Updated last month
- Minimum Description Length probing for neural network representations☆19Updated 3 months ago
- My Implementation of Q-Sparse: All Large Language Models can be Fully Sparsely-Activated☆32Updated 8 months ago
- Latent Large Language Models☆18Updated 8 months ago
- Repository for the Q-Filters method (https://arxiv.org/pdf/2503.02812)☆30Updated 2 months ago
- Utilities for Training Very Large Models☆58Updated 7 months ago
- Implementation of "SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models"☆27Updated 2 months ago
- Official repository for the paper "Approximating Two-Layer Feedforward Networks for Efficient Transformers"☆37Updated last year
- Code for the arXiv preprint "The Unreasonable Effectiveness of Easy Training Data"☆47Updated last year
- ☆48Updated 6 months ago
- ☆78Updated 8 months ago
- ☆44Updated 5 months ago
- NeurIPS 2023 - Cappy: Outperforming and Boosting Large Multi-Task LMs with a Small Scorer☆43Updated last year
- Lottery Ticket Adaptation☆39Updated 5 months ago