Overview

Gille/StrangeMerges_8-7B-slerp is a 7 billion parameter language model developed by Gille. It is a product of merging two existing models, Gille/StrangeMerges_7-7B-slerp and Gille/StrangeMerges_5-7B-ties, utilizing the slerp (spherical linear interpolation) merge method via LazyMergekit. This merging technique allows for combining the strengths of its constituent models.

Key Capabilities & Performance

This model has been evaluated on the Open LLM Leaderboard, achieving an overall average score of 73.39. Specific benchmark results include:

• AI2 Reasoning Challenge (25-Shot): 71.08
• HellaSwag (10-Shot): 87.75
• MMLU (5-Shot): 65.26
• TruthfulQA (0-shot): 64.52
• Winogrande (5-shot): 84.45
• GSM8k (5-shot): 67.25

These scores indicate a strong performance across various reasoning, common sense, and language understanding tasks. The model operates with a context length of 4096 tokens.

Good For

• General text generation and conversational AI.
• Applications requiring robust performance in reasoning and common sense tasks.
• Developers looking for a merged model that balances different capabilities from its base components.