Model Overview

The rdhawk/TR_TaskSpesificLM is a 9 billion parameter language model developed by rdhawk, derived from the google/gemma-2-9b-it base model. It has been specifically fine-tuned for the Turkish language, utilizing the unsloth framework to enhance its performance on various Turkish natural language processing tasks. The model supports a context length of 16384 tokens.

Key Capabilities

• Turkish Language Specialization: Optimized for understanding and generating text in Turkish.
• Diverse Question Answering: Trained on a mixed dataset of 50,000 rows, enabling it to handle:
  • Open-ended questions
  • Multiple-choice questions
  • Fill-in-the-blank questions

Training Details

The model underwent approximately 20 hours of training on an NVIDIA A6000 GPU. Its training data comprised a comprehensive dataset designed to cover a wide array of question formats, making it suitable for various Turkish-specific applications requiring nuanced understanding and response generation.