LFM2-12B-A1B-GLM-4.7-Thinking-Quantum-IQ1C-P-TR1-S2 [Stage 2] Fine tune of «LFM2-8B-A1B-GLM-4.7-Flash-Thinking-Quantum-IQ1C-P» using Unsloth using custom dataset(s), 128k context in 16 bit precision. Then expanded to 12B (via custom python script), and training on an extensive dataset [18k]. This model is a sparse mixture of experts model (32) with 4 experts activated. Speed exceeds 50-100 t/s on CPU // 200 t/s on most cards // 400 t/s + on 5090 at QUANT Q6K [4 experts]. LFM2-12B-A1B-Instruct-Quantum-IQ1C-P-TR1-S2 q8-hi 0.520,0.759,0.736,0.659,0.414,0.756,0.605 LFM2-8B-A1B-GLM-4.7-Flash-Thinking-Quantum-IQ1C-P q8-hi 0.529,0.744,0.745,0.658,0.412,0.760,0.597 Explain 10 ways (in extensive detail) to use the «night» time cooling of radiant energy into space to reduce global temperatures. Include charts, and tables. The user wants a detailed explanation of 10 ways to use nighttime radiant cooling to reduce global temperatures. 1. Understand the Core Concept: Nighttime Radiant Cooling What is it? The Earth emits thermal infrared radiation (heat) into the cold vacuum of space at night. This is a natural heat sink. Current context: The planet is accumulating heat (global…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: DavidAU
Теги: lfm2_moe, finetune, unsloth, mixture of experts, sparse moe, moe, heretic, uncensored
Лайков: 4 | Загрузок: 61
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
