MaxBelitsky / cache-steeringLinks
KV Cache Steering for Inducing Reasoning in Small Language Models
☆41Updated 3 months ago
Alternatives and similar repositories for cache-steering
Users that are interested in cache-steering are comparing it to the libraries listed below
Sorting:
- Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment☆60Updated last year
- Lottery Ticket Adaptation☆40Updated 11 months ago
- Code and pretrained models for the paper: "MatMamba: A Matryoshka State Space Model"☆61Updated 11 months ago
- ☆55Updated 11 months ago
- ReBase: Training Task Experts through Retrieval Based Distillation☆29Updated 8 months ago
- A repository for research on medium sized language models.☆78Updated last year
- ☆78Updated 2 months ago
- Co-LLM: Learning to Decode Collaboratively with Multiple Language Models☆122Updated last year
- ☆57Updated 3 weeks ago
- Repo hosting codes and materials related to speeding LLMs' inference using token merging.☆37Updated 3 weeks ago
- Codebase for Instruction Following without Instruction Tuning☆36Updated last year
- Aioli: A unified optimization framework for language model data mixing☆28Updated 9 months ago
- [NeurIPS 2024] Goldfish Loss: Mitigating Memorization in Generative LLMs☆92Updated 11 months ago
- Official repository for "BLEUBERI: BLEU is a surprisingly effective reward for instruction following"☆28Updated 4 months ago
- ☆86Updated last year
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS …☆60Updated last year
- Resa: Transparent Reasoning Models via SAEs☆44Updated last month
- Reference implementation for Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model☆43Updated last month
- Code for PHATGOOSE introduced in "Learning to Route Among Specialized Experts for Zero-Shot Generalization"☆90Updated last year
- Codebase accompanying the Summary of a Haystack paper.☆79Updated last year
- Matrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data genera…☆99Updated this week
- Code for RATIONALYST: Pre-training Process-Supervision for Improving Reasoning https://arxiv.org/pdf/2410.01044☆35Updated last year
- ☆69Updated last year
- Source code for the collaborative reasoner research project at Meta FAIR.☆103Updated 6 months ago
- Implementation of the paper: "Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention" from Google in pyTO…☆56Updated last week
- The first dense retrieval model that can be prompted like an LM☆89Updated 5 months ago
- ☆26Updated last year
- Understanding the correlation between different LLM benchmarks☆29Updated last year
- ☆52Updated last year
- MEXMA: Token-level objectives improve sentence representations☆42Updated 9 months ago