vhab10/Llama-3.2-3B-Instruct-Nepali-merged-16bit
vhab10/Llama-3.2-3B-Instruct-Nepali-merged-16bit is a 3.2 billion parameter instruction-tuned Llama-3.2 model developed by vhab10. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is designed for instruction-following tasks, particularly in Nepali, leveraging its efficient training methodology.
Loading preview...
Model Overview
vhab10/Llama-3.2-3B-Instruct-Nepali-merged-16bit is an instruction-tuned language model based on the Llama-3.2 architecture, featuring 3.2 billion parameters. Developed by vhab10, this model was fine-tuned using the Unsloth library in conjunction with Huggingface's TRL library, which facilitated a 2x speedup in the training process.
Key Characteristics
- Architecture: Llama-3.2-3B-Instruct
- Parameter Count: 3.2 billion
- Training Efficiency: Utilizes Unsloth for accelerated fine-tuning.
- Context Length: Supports a context length of 32768 tokens.
Use Cases
This model is particularly suited for:
- Instruction-following tasks.
- Applications requiring a compact yet capable Llama-3.2 variant.
- Scenarios where efficient fine-tuning is a priority.
Licensing
The model is released under the Apache 2.0 license, allowing for broad usage and distribution.