neuron-explorer
Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results
Simple Python functions become specialized agents that coordinate through a message-driven Reef.
Translation pending
Developer toolsPython · ShellGitHub @aiexplorationsOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
Multi-agent framework for building large scale Agentic AI. Simple agents, powerful results
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
Build model-backed agents and message-driven agent teams in Python Praval is a focused Python framework for agent systems. One agent can use a provider-neutral model runtime, tools, streaming, structured output, memory, multimodal input, and request-based voice. Teams of agents can collaborate through a native layer called Reef. You can start small with one Agent, then add Reef and Spores when work needs to move between specialists. These are two parts of the same framework, not competing APIs. [README 설치·사용 부분 발췌] InstallRead the full README ↗
The full README is unavailable right now. Read it at the original link. The saved excerpt is shown below.
Build model-backed agents and message-driven agent teams in Python Praval is a focused Python framework for agent systems. One agent can use a provider-neutral model runtime, tools, streaming, structured output, memory, multimodal input, and request-based voice. Teams of agents can collaborate through a native layer called Reef. You can start small with one Agent, then add Reef and Spores when work needs to move between specialists. These are two parts of the same framework, not competing APIs. [README 설치·사용 부분 발췌] InstallRead the full README ↗