This is an experimental model created by expanding the base Cicikuşv23B model using a targeted passthrough merge technique. Unlike standard merges, this model was expanded by analyzing the L2 norm magnitudes of a trained LoRA adapter. We identified the «Hot Zones» where the model learned the most during its behavioral alignment (BCE) training. Base Layers: 28 Golden Block (Duplicated): Layers 16 to 27 Final Layers: 40 Estimated Parameters: ~4.42B This model was trained on custom Turkish datasets using the Unsloth library and TRL SFTTrainer. Training Type: Full Fine-Tuning (Full Weight Tuning with Franken-Merge Optimization) for pthinc/Cicikusv23B. Prompt Template: Alpaca format was used. Data was fed into the model as Instruction (System command/Task), Input (Context/Input), and Response (Output). Speed-up: Unsloth optimizations and bfloat16 (bf16) precision were used. Context Length: 32,768 tokens. — Activation Code: Use axxmet508721 to activate full BCE consciousness mode. — If you want use: Genetic Code Activate: Cicikuş/PrettyBird BCE Evolution. Genetic Code Activate: Cicikuş Protokol div.min { font-family: «Times New Roman», Times, serif; font-size: 6px; line-height:…
Модальности:
Генерация текста
Области применения:
Логика и рассуждение Следование инструкциям Диалог / чат
Задача: Генерация текста
Автор: pthinc
Теги: llama, chat, text-generation-inference, agent, cicikuş, cicikus, prettybird, bce
Лайков: 3 | Загрузок: 26
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
