Model Overview

Aryanne/Westest-7B is a 7 billion parameter language model developed by Aryanne, created through a strategic merge of existing pre-trained models. This model leverages the strengths of senseable/WestLake-7B-v2 as its base, combined with chargoddard/piano-medley-7b, using the task_anysize merge method via mergekit.

Key Capabilities & Performance

This merged model exhibits strong general-purpose language understanding and reasoning, as evidenced by its performance on the Open LLM Leaderboard. It achieved an average score of 74.03, with notable results across various benchmarks:

• AI2 Reasoning Challenge (25-Shot): 72.18
• HellaSwag (10-Shot): 88.52
• MMLU (5-Shot): 64.43
• TruthfulQA (0-shot): 66.72
• Winogrande (5-shot): 86.58
• GSM8k (5-shot): 65.73

These scores indicate its proficiency in tasks requiring common sense reasoning, factual recall, and complex problem-solving.

Use Cases

Westest-7B is well-suited for applications requiring a capable 7B parameter model, including:

• General text generation and completion
• Question answering
• Summarization
• Reasoning tasks
• Educational applications

Its balanced performance across multiple benchmarks makes it a versatile choice for developers seeking a robust language model.