KnutJaegersberg/Deita-32b
Deita-32b is a 32.5 billion parameter language model developed by KnutJaegersberg, designed for general-purpose text generation and understanding. It features a 32768-token context window, enabling it to process and generate extensive text sequences. The model demonstrates strong performance across various reasoning and language understanding benchmarks, making it suitable for a wide range of conversational and analytical AI applications.
Loading preview...
Overview
Deita-32b is a large language model with 32.5 billion parameters, developed by KnutJaegersberg. It is designed to handle a wide array of natural language processing tasks, leveraging a substantial context window of 32768 tokens for comprehensive understanding and generation of long texts.
Key Capabilities
- General-purpose text generation: Capable of producing coherent and contextually relevant text for various prompts.
- Reasoning and understanding: Demonstrates strong performance on reasoning and common sense benchmarks.
- Extended context processing: Utilizes a 32768-token context window, allowing for detailed analysis and generation based on large inputs.
Performance Highlights
Deita-32b has been evaluated on the Open LLM Leaderboard, showcasing competitive results across several key metrics:
- Average Score: 72.16
- AI2 Reasoning Challenge (25-Shot): 63.05
- HellaSwag (10-Shot): 84.67
- MMLU (5-Shot): 73.95
- TruthfulQA (0-shot): 58.11
- Winogrande (5-shot): 80.82
- GSM8k (5-shot): 72.33
Good for
- Applications requiring robust general-purpose language understanding and generation.
- Tasks benefiting from a large context window, such as summarization of long documents or complex conversational agents.
- Use cases where strong performance on reasoning and common sense benchmarks is critical.
Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.