stchakwdev
Follow
Lists (1)
Stars
Your own personal AI assistant. Any OS. Any Platform.
PyTorch implementation of Multifidelity Kolmogorov-Arnold Networks (MFKANs) for data-efficient learning. Train accurate models with sparse high-fidelity data by leveraging correlations with abundan…
A simple framework for humans: Server-first React with zero magic. Built to stay understandable.
Ultimate collection of Claude Code tips, tricks, hacks, and workflows that you can use to master Claude Code in minutes
Universal Slack integration for Claude Code - bidirectional communication, multi-session support, and real-time messaging
Select, weight and analyze complex sample data
Getting crystal-like representations with harmonic loss
Training small GPT-2 style models using Kolmogorov-Arnold networks.
Baantu Research: Hybrid KAN-Transformer for investigating learnable activations in LLM reasoning. Built on nanochat by Andrej Karpathy.
A rigorous 2x3 factorial comparison of neural network architectures: KAN vs MLP feedforward layers combined with Transformer vs Mamba sequence models. Investigates whether KAN advantages stem from …
bloom - evaluate any behavior immediately 🌸🌱
Empirical investigation of grokking in KAN. Key finding: KAN groks multiplication 12x faster than MLP!
AthanasiosDelis / faster-kan
Forked from ZiyaoLi/fast-kanBenchmarking and Testing FastKAN
Investigating whether language models encode anticipated social consequences in their activations. Uses a 2x2 factorial design crossing truth × social valence to show that models are more sensitive…
AI safety evaluation framework testing LLM epistemic robustness under adversarial self-history manipulation
Claude Code notifications without the context switch. A minimal, always-present session manager for macOS.
"Paper2Slides: From Paper to Presentation in One Click"
An experimental research tool for fabricating GitHub personas with AI-generated repositories
Markdown editor for scientific writing. Batteries included.
a project brain for your AI. Give LLMs instant architectural context without burning tokens
From-scratch implementation of OpenAI's GPT-OSS model in Python. No Torch, No GPUs.
Refine high-quality datasets and visual AI models
Always know what to expect from your data.
Multi-agent strategic deception evaluation framework for LLMs using Secret Hitler as a testbed. Analyzes AI reasoning, trust dynamics, and deceptive behavior patterns.