macadeliccc/MonarchLake-7B

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Feb 22, 2024License:cc-by-nc-4.0Architecture:Transformer0.0K Open Weights Cold

macadeliccc/MonarchLake-7B is a 7 billion parameter language model, merged from mlabonne/AlphaMonarch-7B and macadeliccc/WestLake-7b-v2-laser-truthy-dpo using the SLERP method. This model is specifically equipped with enhanced emotional intelligence capabilities, building upon the AlphaMonarch-7B base. It achieves an average score of 76.10 on the Open LLM Leaderboard, demonstrating strong performance across various reasoning and language understanding benchmarks.

Loading preview...

MonarchLake-7B Overview

MonarchLake-7B is a 7 billion parameter language model developed by macadeliccc, created by merging two distinct models: mlabonne/AlphaMonarch-7B and macadeliccc/WestLake-7b-v2-laser-truthy-dpo. The merge was performed using the SLERP (Spherical Linear Interpolation) method, which combines the strengths of its constituent models.

Key Capabilities & Characteristics

  • Enhanced Emotional Intelligence: The primary focus of MonarchLake-7B is to imbue the AlphaMonarch-7B base with a strong foundation in emotional intelligence, suggesting improved performance in tasks requiring nuanced understanding of sentiment and emotional context.
  • Merged Architecture: Leverages the combined knowledge and capabilities of its two base models, aiming for a synergistic performance improvement.
  • Competitive Benchmarking: Achieves a notable average score of 76.10 on the Open LLM Leaderboard. Specific benchmark results include:
    • AI2 Reasoning Challenge (25-Shot): 74.15
    • HellaSwag (10-Shot): 89.29
    • MMLU (5-Shot): 64.44
    • TruthfulQA (0-shot): 74.97
    • Winogrande (5-shot): 85.48
    • GSM8k (5-shot): 68.31

Use Cases

This model is particularly well-suited for applications where understanding and generating emotionally intelligent responses are crucial. Its strong performance across various reasoning and language tasks also makes it a versatile choice for general-purpose language generation and comprehension, especially in scenarios benefiting from its specialized emotional intelligence.