tomaarsen / attention_sinksLinks
Extend existing LLMs way beyond the original training length with constant memory usage, without retraining
☆729Updated last year
Alternatives and similar repositories for attention_sinks
Users that are interested in attention_sinks are comparing it to the libraries listed below
Sorting:
- Code for fine-tuning Platypus fam LLMs using LoRA☆629Updated last year
- [ICML'24 Spotlight] LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning☆660Updated last year
- batched loras☆347Updated 2 years ago
- ☆575Updated last year
- ☆415Updated 2 years ago
- [ACL2023] We introduce LLM-Blender, an innovative ensembling framework to attain consistently superior performance by leveraging the dive…☆970Updated last year
- A bagel, with everything.☆324Updated last year
- ☆547Updated 10 months ago
- Landmark Attention: Random-Access Infinite Context Length for Transformers☆426Updated last year
- YaRN: Efficient Context Window Extension of Large Language Models☆1,634Updated last year
- This repository contains code and tooling for the Abacus.AI LLM Context Expansion project. Also included are evaluation scripts and bench…☆597Updated last year
- Train Models Contrastively in Pytorch☆753Updated 7 months ago
- ☆552Updated 11 months ago
- Generate textbook-quality synthetic LLM pretraining data☆506Updated 2 years ago
- The Truth Is In There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction☆389Updated last year
- [ICML 2024] SqueezeLLM: Dense-and-Sparse Quantization☆704Updated last year
- Finetuning Large Language Models on One Consumer GPU in 2 Bits☆732Updated last year
- Tune any FALCON in 4-bit☆464Updated 2 years ago
- Official repository for LongChat and LongEval☆531Updated last year
- Official repository for ORPO☆463Updated last year
- Inference code for Persimmon-8B☆412Updated 2 years ago
- Generative Representational Instruction Tuning☆677Updated 4 months ago
- A library for easily merging multiple LLM experts, and efficiently train the merged LLM.☆494Updated last year
- This repository contains code to quantitatively evaluate instruction-tuned models such as Alpaca and Flan-T5 on held-out tasks.☆549Updated last year
- Fine-tune mistral-7B on 3090s, a100s, h100s☆714Updated 2 years ago
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,297Updated 8 months ago
- The repository for the code of the UltraFastBERT paper☆518Updated last year
- Official code for ReLoRA from the paper Stack More Layers Differently: High-Rank Training Through Low-Rank Updates☆467Updated last year
- [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning☆631Updated last year
- distributed trainer for LLMs☆583Updated last year