Kukedlc/NeuralFusion-7b-Dare-Ties

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Feb 29, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

NeuralFusion-7b-Dare-Ties is a 7 billion parameter language model created by Kukedlc, built using a DARE TIES merge of NeuralMaxime-7B-slerp, Fasciculus-Arcuatus-7B-slerp, and NeoCortex-7B-slerp, all based on mlabonne/Monarch-7B. This model demonstrates strong general reasoning capabilities, achieving an average score of 75.94 on the Open LLM Leaderboard across various benchmarks. It is designed for general-purpose language generation and understanding tasks, with a context length of 4096 tokens.

Loading preview...

NeuralFusion-7b-Dare-Ties Overview

NeuralFusion-7b-Dare-Ties is a 7 billion parameter language model developed by Kukedlc. It is a product of a DARE TIES merge of three distinct models: NeuralMaxime-7B-slerp, Fasciculus-Arcuatus-7B-slerp, and NeoCortex-7B-slerp, all originating from the mlabonne/Monarch-7B base model. This merging strategy aims to combine the strengths of its constituent models.

Key Capabilities & Performance

This model exhibits solid performance across a range of general language understanding and reasoning tasks, as indicated by its evaluation on the Open LLM Leaderboard. Key benchmark results include:

  • Average Score: 75.94
  • AI2 Reasoning Challenge (25-Shot): 73.21
  • HellaSwag (10-Shot): 88.96
  • MMLU (5-Shot): 64.77
  • TruthfulQA (0-shot): 73.32
  • Winogrande (5-shot): 85.56
  • GSM8k (5-shot): 69.83

These scores suggest a balanced capability in common sense reasoning, factual recall, and problem-solving.

When to Use This Model

NeuralFusion-7b-Dare-Ties is suitable for applications requiring a capable 7B parameter model with a general understanding of language. Its performance profile makes it a strong candidate for:

  • General text generation and completion
  • Reasoning tasks
  • Question answering
  • Conversational AI where a broad knowledge base is beneficial

Its 4096-token context window supports moderately long interactions and document processing.