neuron-explorer
Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results
翻译准备中
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
在项目文件夹中运行 /nomorevibe verify,即可验证所有权并自行管理介绍。
一分钟完成验证在 harness 工具選擇迴圈中嵌入的決策層
AI 翻译
A decision layer embedded into the harness tool-selection loop
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
Choose the next tool inside an existing agent loop.
Harness Router takes a goal, the current observation, and a set of available tools,
then returns a tool choice, confidence, or a fallback to the host planner. It uses
Jev through OpenRouter for discrete decisions, caches reusable choices, and supports
optional Monte Carlo Tree Search (MCTS) over simulated outcomes.
The host harness generates arguments, checks permissions, executes the tool, and
processes its result. Router supplies the next-tool decision at the lifecycle
boundary the host exposes.查看完整 README ↗目前无法加载完整 README。请通过原文链接阅读。 以下为已保存的摘录。
Choose the next tool inside an existing agent loop.
Harness Router takes a goal, the current observation, and a set of available tools,
then returns a tool choice, confidence, or a fallback to the host planner. It uses
Jev through OpenRouter for discrete decisions, caches reusable choices, and supports
optional Monte Carlo Tree Search (MCTS) over simulated outcomes.
The host harness generates arguments, checks permissions, executes the tool, and
processes its result. Router supplies the next-tool decision at the lifecycle
boundary the host exposes.查看完整 README ↗