Overview

NeuralNovel/Tiger-7B-v0.1 is a 7 billion parameter language model developed by NeuralNovel, resulting from a strategic merge of two distinct models: NeuralNovel/Mistral-7B-Instruct-v0.2-Neural-Story and NeuralNovel/Gecko-7B-v0.1-DPO. This merge was performed using the SLERP (Spherical Linear Interpolation) method, aiming to combine and enhance the capabilities of its base models.

Key Capabilities & Performance

This model demonstrates solid performance across various benchmarks, as evaluated on the Open LLM Leaderboard:

• Average Score: 65.02
• AI2 Reasoning Challenge (25-Shot): 59.98
• HellaSwag (10-Shot): 83.21
• MMLU (5-Shot): 61.42
• TruthfulQA (0-shot): 61.03
• Winogrande (5-shot): 77.66
• GSM8k (5-shot): 46.78

Use Cases

Given its balanced performance across reasoning, common sense, and language understanding tasks, Tiger-7B-v0.1 is suitable for a range of general-purpose applications. Its foundation on instruction-tuned models suggests proficiency in following directives and generating coherent text. Developers can consider this model for tasks requiring a capable 7B parameter model with a 4096-token context window.