> [!Important] > Model trained on Claude Sonnet 4.6 Adaptive Thinking (The claude.ai Claude, not the API Claude). The training data is PRIVATE. LFM2.5-1.2B-Distilled-Claude-4.6 (Liquid Claude) is a distillation of Claude into LFM2.5-1.2B-Thinking via LoRA. THINK BLOCK PATTERNS Agentic think blocks (Action/Observation): 8473 Pure reasoning think blocks: 3005 CONSECUTIVE USER MESSAGES Total consecutive user-user pairs: 319 Assistant messages total: 11738 With agentic tool calls in think: 2151 (18.3%) Total chars in dataset: 60,487,999 Approx tokens (~4 chars/token): 15,121,999 Conversations with 5 think blocks in a single assistant msg: 319 !image (Ignore the fact that it took 1min to reason, i got a i3-6006u / 12GB as hardwa
Модальности:
Генерация текста
Области применения:
Логика и рассуждение Диалог / чат
Задача: Генерация текста
Автор: FlameF0X
Теги: lfm2, liquid, reasoning, cot, distillate, conversational, en, eval-results
Лайков: 4 | Загрузок: 178
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
