Model Overview

abdulrahman-nuzha/belal-finetuned-llama2-v1.0 is a 7 billion parameter language model, fine-tuned from the Llama 2 architecture. While specific training details and datasets are not provided in the model card, its performance on the Open LLM Leaderboard indicates its general language understanding and reasoning capabilities.

Key Capabilities

• Reasoning: Achieved 52.82 on the AI2 Reasoning Challenge (25-Shot).
• Common Sense: Scored 77.75 on HellaSwag (10-Shot) and 74.35 on Winogrande (5-shot).
• General Knowledge: Demonstrated 43.51 on MMLU (5-Shot).
• Factuality: Scored 39.09 on TruthfulQA (0-shot).
• Mathematical Reasoning: Achieved 10.69 on GSM8k (5-shot).

Performance Highlights

The model's average score on the Open LLM Leaderboard is 49.70. Detailed evaluation results are available on the Hugging Face Open LLM Leaderboard.

Intended Use

This model is suitable for direct use in various natural language processing tasks where a Llama 2-based model with general reasoning and language understanding is beneficial. Users should be aware of the inherent biases and limitations common to large language models.