neuron-explorer
Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results
翻譯準備中
個人 GitHub 帳號下公開儲存庫的擁有者可以登入並驗證所有權,自行管理介紹,無需管理員核准。
登入並產生驗證檔案,在 30 分鐘內將其提交至此儲存庫的預設分支。其他分支或分叉儲存庫中的檔案無效。然後在下方檢查檔案,即可無需管理員核准直接管理專案。請保留檔案以維持管理權限。
如需登入,將前往登入頁面。登入後請回到此處,再次點擊所選的驗證按鈕。
在專案資料夾中執行 /nomorevibe verify,即可驗證所有權並自行管理介紹。
一分鐘完成驗證用於程式碼代理的防護欄桿:破壞性指令分流、一層委派、卡住迴圈偵測、允許清單提交驗證。
AI 翻譯
Guardrails for coding agents: destructive-command triage, one-level delegation, stuck-loop detection, allowlisted commit verification.
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
Unofficial community project by Scalably. Not affiliated with or endorsed by Anthropic or OpenAI. Claude, Claude Code and Codex are trademarks of their respective owners. Eight guardrails that run in front of a coding agent. They are the command-hook form of guards we run in our own agent platform, where agents do real work on real systems all day. They are boring on purpose. Each one is a shell script that reads JSON on standard input and answers with an exit code. No daemon, no network calls, no telemetry, nothing to sign up for. jq is the only dependency. [README 설치·사용 부분 발췌] Install檢視完整 README ↗
目前無法載入完整 README。請透過原文連結閱讀。 以下為已儲存的摘錄。
Unofficial community project by Scalably. Not affiliated with or endorsed by Anthropic or OpenAI. Claude, Claude Code and Codex are trademarks of their respective owners. Eight guardrails that run in front of a coding agent. They are the command-hook form of guards we run in our own agent platform, where agents do real work on real systems all day. They are boring on purpose. Each one is a shell script that reads JSON on standard input and answers with an exit code. No daemon, no network calls, no telemetry, nothing to sign up for. jq is the only dependency. [README 설치·사용 부분 발췌] Install檢視完整 README ↗