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
尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
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。
一分鐘完成驗證一個用 Rust 編寫的推理引擎,支援 GGUF、AWQ 與 FP8(W8A8)檢查點,採用手寫 CUDA 與 Metal 核心——不使用 PyTorch、不使用 libtorch、也不使用 ggml
AI 翻譯
An inference engine written in Rust for GGUF, AWQ, and FP8 (W8A8) checkpoints, with hand-written CUDA and Metal kernels — no PyTorch, no libtorch, no ggml
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
An inference engine written in Rust for GGUF, AWQ, native FP8 (W8A8), and
native NVFP4 (W4A4) checkpoints, with hand-written kernels for both NVIDIA
(CUDA) and Apple Silicon (Metal) GPUs — no PyTorch, no libtorch, no ggml.
CUDA is the primary, most complete backend; Metal covers dense decode, GQA
attention and GatedDeltaNet today, with MoE, the vision tower, and the
INT4/FP8/NVFP4 tensor-core GEMM paths still CUDA-only (see
Metal below). The whole path from a model
checkpoint on disk to an OpenAI-compatible HTTP response is in this
repository.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。
An inference engine written in Rust for GGUF, AWQ, native FP8 (W8A8), and
native NVFP4 (W4A4) checkpoints, with hand-written kernels for both NVIDIA
(CUDA) and Apple Silicon (Metal) GPUs — no PyTorch, no libtorch, no ggml.
CUDA is the primary, most complete backend; Metal covers dense decode, GQA
attention and GatedDeltaNet today, with MoE, the vision tower, and the
INT4/FP8/NVFP4 tensor-core GEMM paths still CUDA-only (see
Metal below). The whole path from a model
checkpoint on disk to an OpenAI-compatible HTTP response is in this
repository.檢視完整 README ↗