Model Overview

shadowml/BeagSake-7B is a 7 billion parameter language model developed by shadowml. It is a merged model, combining the strengths of shadowml/BeagleSempra-7B and shadowml/WestBeagle-7B using the slerp (spherical linear interpolation) merge method via LazyMergekit. This approach allows for a blend of the characteristics of its constituent models.

Key Capabilities

BeagSake-7B demonstrates solid performance across various benchmarks, indicating strong general language understanding and reasoning. Key evaluation results from the Open LLM Leaderboard include:

• Average Score: 75.38
• AI2 Reasoning Challenge (25-Shot): 72.44
• HellaSwag (10-Shot): 88.39
• MMLU (5-Shot): 65.23
• TruthfulQA (0-shot): 72.27
• Winogrande (5-shot): 82.16
• GSM8k (5-shot): 71.80

These scores suggest proficiency in common sense reasoning, factual recall, and mathematical problem-solving.

Good For

• General-purpose text generation: Creating coherent and contextually relevant text.
• Question Answering: Responding to queries based on its training data.
• Reasoning tasks: Handling tasks that require logical inference and problem-solving, as indicated by its AI2 Reasoning Challenge and GSM8k scores.