Model Overview

This model, developed by nethmid, is a fine-tuned variant of the Llama-3.2-3B-Instruct architecture, featuring 3.2 billion parameters. It was specifically trained using the Unsloth framework and Huggingface's TRL library, which enabled a significantly faster training process.

Key Characteristics

• Base Model: Fine-tuned from unsloth/Llama-3.2-3B-Instruct.
• Training Efficiency: Leverages Unsloth for 2x faster training.
• Parameter Count: 3.2 billion parameters.
• Context Length: Supports a context length of 32768 tokens.

Primary Focus

This model is specialized for applications involving cognitive distortions. Its fine-tuning aims to enhance its performance and understanding in this specific domain, making it suitable for tasks requiring nuanced interaction with such concepts.

Licensing

The model is released under the Apache-2.0 license, allowing for broad use and distribution.