dhmeltzer/Llama-2-13b-hf-ds_eli5_1024_r_64_alpha_16_merged
dhmeltzer/Llama-2-13b-hf-ds_eli5_1024_r_64_alpha_16_merged is a 13 billion parameter Llama-2 based language model developed by dhmeltzer. This model is fine-tuned with a 4096 token context length and achieves an average score of 47.8 on the Open LLM Leaderboard benchmarks. It demonstrates capabilities across various tasks including common sense reasoning, language understanding, and question answering, making it suitable for general-purpose text generation and comprehension tasks.
Loading preview...
Model Overview
dhmeltzer/Llama-2-13b-hf-ds_eli5_1024_r_64_alpha_16_merged is a 13 billion parameter language model built upon the Llama-2 architecture. It features a 4096 token context window, making it capable of processing moderately long inputs for various natural language processing tasks. This model has been fine-tuned to enhance its performance across a range of benchmarks.
Key Capabilities & Performance
This model's performance is evaluated on the Open LLM Leaderboard, where it achieves an overall average score of 47.8. Specific benchmark results include:
- ARC (25-shot): 59.13
- HellaSwag (10-shot): 82.13
- MMLU (5-shot): 54.98
- TruthfulQA (0-shot): 44.23
- Winogrande (5-shot): 76.4
- GSM8K (5-shot): 8.11
- DROP (3-shot): 9.6
These scores indicate a solid foundation in common sense reasoning, reading comprehension, and general knowledge, with particular strength in HellaSwag and Winogrande.
When to Use This Model
This model is a good candidate for use cases requiring:
- General text generation: Creating coherent and contextually relevant text.
- Question answering: Responding to queries based on provided context or general knowledge.
- Language understanding tasks: Analyzing and interpreting text for various applications.
- Prototyping and experimentation: Its 13B parameter size offers a balance between performance and computational requirements for many applications.
Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.