Weyaxi/Chat-AYB-Platypus2-13B
TEXT GENERATIONConcurrency Cost:1Model Size:13BQuant:FP8Ctx Length:4kArchitecture:Transformer Cold
Weyaxi/Chat-AYB-Platypus2-13B is a 13 billion parameter language model, fine-tuned for general conversational AI tasks. It demonstrates balanced performance across various benchmarks, including a 60.49% score on ARC (25-shot) and 84.03% on HellaSwag (10-shot). This model is suitable for applications requiring robust language understanding and generation capabilities, particularly in question answering and common sense reasoning.
Loading preview...
Model Overview
Weyaxi/Chat-AYB-Platypus2-13B is a 13 billion parameter language model evaluated on the Open LLM Leaderboard. It exhibits a balanced average performance of 53.39 across a suite of benchmarks, indicating its general-purpose capabilities in understanding and generating human-like text.
Key Capabilities
- Reasoning: Achieves 60.49% on ARC (25-shot), demonstrating proficiency in complex reasoning tasks.
- Common Sense: Scores 84.03% on HellaSwag (10-shot) and 75.77% on Winogrande (5-shot), indicating strong common sense reasoning abilities.
- Knowledge & Understanding: Attains 57.83% on MMLU (5-shot) and 54.52% on TruthfulQA (0-shot), reflecting its general knowledge and ability to provide factual responses.
- Reading Comprehension: Shows a 38.12% score on DROP (3-shot), suggesting moderate performance in extracting information from passages.
Good For
- General Conversational AI: Its balanced performance makes it suitable for a wide range of dialogue systems and chatbots.
- Question Answering: Effective for applications requiring answers to factual and reasoning-based questions.
- Text Generation: Can be utilized for generating coherent and contextually relevant text in various scenarios.