This is a uncensored, full deep thinking Ernie 21B-A3B (MOE, 64 experts) fine tune using Claude 4.5 High Opus reasoning dataset via Unsloth via local hardware, Linux (for windows). Note this model is mostly uncensored right from the «factory» so to speak. Model excels in creative (brainstorming, creative prose) as well as general usage. Reasoning is compact, but detailed (very detailed) and right to the «point» so to speak. CRITICAL SETTINGS: — for creative suggest rep pen of 1.01 to 1.1 — for general work rep pen of 1 (off), 1.05 or 1.1 — Lower quants MAY LOOP in some cases. Reasoning affects: — General model operation. — Output generation — Benchmarks. Model Features: — 128k context — Temp range .1 to 2.5. — Reasoning is temp stable. You may want to visit Baidu’s repo for this model for root/core benchmarks and settings. Settings: CHAT / ROLEPLAY and/or SMOOTHER operation of this model: In «KoboldCpp» or «oobabooga/text-generation-webui» or «Silly Tavern» ; -> if using GGUFs you need to use «llama_HF» (which involves downloading some config files from the SOURCE version of this model)…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: DavidAU
Теги: ernie4_5_moe, mixture of experts, moe, 64 experts, uncensored, unsloth, finetune, All use cases
Лайков: 3 | Загрузок: 49
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
