Weyaxi/SpeechlessV1-Nova-13B

TEXT GENERATIONConcurrency Cost:1Model Size:13BQuant:FP8Ctx Length:4kArchitecture:Transformer Cold

SpeechlessV1-Nova-13B is a 13 billion parameter language model developed by Weyaxi, fine-tuned for general language understanding and generation. It demonstrates competitive performance across various benchmarks, including ARC, HellaSwag, and MMLU, making it suitable for a range of natural language processing tasks. This model is designed for applications requiring robust text comprehension and coherent response generation.

Loading preview...

Overview

Weyaxi/SpeechlessV1-Nova-13B is a 13 billion parameter language model. It is part of the SpeechlessV1 series and has been evaluated on the Open LLM Leaderboard, showcasing its capabilities across a spectrum of natural language understanding and reasoning tasks.

Key Capabilities & Performance

This model exhibits solid performance on several standard benchmarks:

  • ARC (25-shot): Achieves 61.77, indicating strong reasoning abilities.
  • HellaSwag (10-shot): Scores 82.68, demonstrating proficiency in common sense reasoning.
  • MMLU (5-shot): Reaches 57.75, reflecting its general knowledge and multi-task language understanding.
  • TruthfulQA (0-shot): Scores 51.44, suggesting a reasonable ability to generate factually correct responses.
  • Winogrande (5-shot): Achieves 77.43, showing competence in resolving pronoun ambiguity.
  • GSM8K (5-shot): Scores 5.76, indicating room for improvement in complex mathematical reasoning.
  • DROP (3-shot): Scores 9.05, suggesting moderate performance in reading comprehension with discrete answers.

Good For

  • General-purpose text generation and understanding.
  • Applications requiring common sense and multi-task language understanding.
  • Tasks benefiting from a 13B parameter model with competitive benchmark scores.