ScalingIntelligence / Archon
Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.
☆128Updated 3 weeks ago
Related projects ⓘ
Alternatives and complementary repositories for Archon
- ☆54Updated last month
- The official repo for "LLoCo: Learning Long Contexts Offline"☆113Updated 5 months ago
- Code for NeurIPS'24 paper 'Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization'☆161Updated last month
- Implementation of the Quiet-STAR paper (https://arxiv.org/pdf/2403.09629.pdf)☆42Updated 3 months ago
- Repository for the paper Stream of Search: Learning to Search in Language☆91Updated 3 months ago
- ☆112Updated last month
- A framework to study AI models in Reasoning, Alignment, and use of Memory (RAM).☆145Updated 2 weeks ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆104Updated last month
- The code for the paper ROUTERBENCH: A Benchmark for Multi-LLM Routing System☆92Updated 5 months ago
- A simple unified framework for evaluating LLMs☆145Updated last week
- Can Language Models Solve Olympiad Programming?☆100Updated 3 months ago
- ☆101Updated 3 months ago
- Functional Benchmarks and the Reasoning Gap☆78Updated last month
- ☆90Updated 4 months ago
- code for training & evaluating Contextual Document Embedding models☆117Updated this week
- ☆102Updated last month
- The official implementation of Self-Exploring Language Models (SELM)☆55Updated 5 months ago
- Benchmarking LLMs with Challenging Tasks from Real Users☆195Updated 2 weeks ago
- Evaluating LLMs with fewer examples☆134Updated 7 months ago
- A repository for transformer critique learning and generation☆86Updated 11 months ago
- Just a bunch of benchmark logs for different LLMs☆114Updated 3 months ago
- 🌍 Repository for "AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agent", ACL'24 Best Resource Pap…☆110Updated 3 weeks ago
- Parameter-Efficient Sparsity Crafting From Dense to Mixture-of-Experts for Instruction Tuning on General Tasks☆129Updated 2 months ago
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆216Updated 7 months ago
- Open Implementations of LLM Analyses☆94Updated last month
- ☆49Updated 6 months ago
- Public repository for "The Surprising Effectiveness of Test-Time Training for Abstract Reasoning"☆181Updated this week
- ☆46Updated 2 weeks ago
- Code repository for the c-BTM paper☆105Updated last year
- ☆40Updated 2 weeks ago