IBM / ModuleFormer
ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward experts. We released a collection of ModuleFormer-based Language Models (MoLM) ranging in scale from 4 billion to 8 billion parameters.
☆216Updated 7 months ago
Related projects ⓘ
Alternatives and complementary repositories for ModuleFormer
- Positional Skip-wise Training for Efficient Context Window Extension of LLMs to Extremely Length (ICLR 2024)☆199Updated 6 months ago
- Multipack distributed sampler for fast padding-free training of LLMs☆178Updated 3 months ago
- Code repository for the c-BTM paper☆105Updated last year
- TART: A plug-and-play Transformer module for task-agnostic reasoning☆190Updated last year
- The official repo for "LLoCo: Learning Long Contexts Offline"☆113Updated 5 months ago
- ☆93Updated last year
- Official repository for "Scaling Retrieval-Based Langauge Models with a Trillion-Token Datastore".☆129Updated this week
- Benchmarking LLMs with Challenging Tasks from Real Users☆195Updated 2 weeks ago
- LongEmbed: Extending Embedding Models for Long Context Retrieval (EMNLP 2024)☆115Updated last week
- DSIR large-scale data selection framework for language model training☆230Updated 7 months ago
- batched loras☆336Updated last year
- Experiments on speculative sampling with Llama models☆118Updated last year
- ☆184Updated last month
- This is the repo for the paper Shepherd -- A Critic for Language Model Generation☆213Updated last year
- Parameter-Efficient Sparsity Crafting From Dense to Mixture-of-Experts for Instruction Tuning on General Tasks☆129Updated 2 months ago
- ☆102Updated last month
- LOFT: A 1 Million+ Token Long-Context Benchmark☆146Updated 3 weeks ago
- Chain-of-Hindsight, A Scalable RLHF Method☆220Updated last year
- A simple unified framework for evaluating LLMs☆145Updated last week
- Spherical Merge Pytorch/HF format Language Models with minimal feature loss.☆112Updated last year
- Repo for Rho-1: Token-level Data Selection & Selective Pretraining of LLMs.☆307Updated 7 months ago
- Evaluating LLMs with CommonGen-Lite☆85Updated 8 months ago
- Manage scalable open LLM inference endpoints in Slurm clusters☆236Updated 4 months ago
- Data preparation code for Amber 7B LLM☆82Updated 6 months ago
- ☆112Updated last month
- Learning to Compress Prompts with Gist Tokens - https://arxiv.org/abs/2304.08467☆266Updated last year
- Load multiple LoRA modules simultaneously and automatically switch the appropriate combination of LoRA modules to generate the best answe…☆142Updated 9 months ago
- ☆149Updated 10 months ago
- ☆247Updated last year
- Scripts for generating synthetic finetuning data for reducing sycophancy.☆107Updated last year