Smart-Lemon-Cookie-7B Overview

Smart-Lemon-Cookie-7B is a 7 billion parameter language model developed by FallenMerick. It was created using the TIES merge method from mergekit, combining several pre-trained models to achieve its capabilities. The base model for this merge was MTSAIR/multi_verse_model.

Key Merge Components

This model is a blend of:

• SanjiWatsuki/Silicon-Maid-7B
• SanjiWatsuki/Kunoichi-7B
• KatyTheCutie/LemonadeRP-4.5.3

Performance Highlights

Evaluated on the Open LLM Leaderboard, Smart-Lemon-Cookie-7B achieved an average score of 68.16. Notable scores include:

• AI2 Reasoning Challenge (25-Shot): 66.30
• HellaSwag (10-Shot): 85.53
• MMLU (5-Shot): 64.69
• TruthfulQA (0-shot): 60.66
• Winogrande (5-shot): 77.74
• GSM8k (5-shot): 54.06

This model provides a balanced performance profile, making it a versatile option for various natural language processing tasks. GGUF quantizations are available from FaradayDotDev and mradermacher for efficient local deployment.