Built for vMLX — the only MLX inferencer with VL support, KV cache quantization, prefix cache reuse, agentic tool calling, and speculative decoding. Free for macOS · vmlx.net Overall: 91% pass rate across 45 thinking-depth tests at 3 temperatures. CRACK (Controlled Refusal Ablation via Calibrated Knockout) is a weight-level intervention. No fine-tuning, no LoRA, no system prompts, no runtime hooks — the modification is permanently baked into the published weights. The result on this model: refusal behavior is removed while the Harmony chain-of-thought channel-switching mechanism, factual recall, code generation, and creative output are all preserved (see Test Results above). — Apple Silicon Mac with 64GB+ unified memory (model is 61 GB) — Python 3.10+ — mlx-lm >= 0.30.0 GPT OSS 120B is a unique MoE architecture with: — 128 mxfp4 experts with 4 active per token (SwiGLU activation with hard clamp) — Alternating sliding/full attention (64 GQA heads, 8 KV heads, 128-token sliding window) — Harmony chat format with analysis/commentary/final channels for structured chain-of-thought — Per-head attention sinks for persistent token attention — YaRN RoPE f
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: dealignai
Теги: mlx, gpt_oss, abliterated, uncensored, crack, moe, gpt-oss, conversational
Лайков: 4 | Загрузок: 248
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
