mcleish7 / arithmeticLinks
Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)
☆193Updated last year
Alternatives and similar repositories for arithmetic
Users that are interested in arithmetic are comparing it to the libraries listed below
Sorting:
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆166Updated 3 months ago
- Understand and test language model architectures on synthetic tasks.☆233Updated last month
- ☆142Updated last month
- Code for NeurIPS'24 paper 'Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization'☆233Updated 3 months ago
- Functional Benchmarks and the Reasoning Gap☆89Updated last year
- A MAD laboratory to improve AI architecture designs 🧪☆131Updated 10 months ago
- ☆200Updated last month
- Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"☆241Updated 4 months ago
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆82Updated 10 months ago
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆172Updated 9 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Updated last year
- Repository for the paper Stream of Search: Learning to Search in Language☆151Updated 8 months ago
- $100K or 100 Days: Trade-offs when Pre-Training with Academic Resources☆147Updated 3 weeks ago
- ☆185Updated last year
- RuLES: a benchmark for evaluating rule-following in language models☆238Updated 8 months ago
- ☆73Updated last year
- PyTorch library for Active Fine-Tuning☆93Updated 3 weeks ago
- nanoGPT-like codebase for LLM training☆107Updated 5 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆222Updated 10 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆130Updated 10 months ago
- Open source interpretability artefacts for R1.☆163Updated 6 months ago
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆218Updated last week
- Implementation of the Llama architecture with RLHF + Q-learning☆167Updated 8 months ago
- ☆86Updated last year
- ☆102Updated 3 months ago
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆224Updated last month
- ☆81Updated last year
- Extract full next-token probabilities via language model APIs☆247Updated last year
- A curated reading list of research in Adaptive Computation, Inference-Time Computation & Mixture of Experts (MoE).☆155Updated 9 months ago
- ☆124Updated 10 months ago