Model Overview

Mayonnaise-4in1-02 is a 7 billion parameter causal language model developed by The Kaitchup. It is an ensemble model created using the TIES-merging technique with MergeKit, based on the mistralai/Mistral-7B-v0.1 architecture. This model combines the strengths of several fine-tuned models, including mncai/mistral-7b-dpo-v5, flemmingmiguel/MBX-7B, and BarryFutureman/NeuralTurdusVariant1-7B, to enhance its overall performance.

Key Capabilities & Performance

The model demonstrates strong general-purpose language understanding and generation capabilities, as evidenced by its performance on the Open LLM Leaderboard. Key evaluation results include:

• Avg. Score: 75.21
• AI2 Reasoning Challenge (25-Shot): 73.38
• HellaSwag (10-Shot): 88.51
• MMLU (5-Shot): 64.89
• TruthfulQA (0-shot): 69.04
• Winogrande (5-shot): 84.37
• GSM8k (5-shot): 71.04

These scores indicate proficiency in reasoning, common sense, factual recall, and mathematical problem-solving. The model is licensed under Apache 2.0.

When to Use This Model

This model is suitable for a variety of English NLP tasks where a balanced performance across different benchmarks is desired. Its TIES-merging approach aims to leverage the specialized knowledge of its constituent models, making it a robust option for applications requiring:

• General text generation and understanding.
• Reasoning and logical inference.
• Question answering and factual recall.
• Common sense understanding.