kevinadityaikhsan/llama-3.2-3b-legal-id-sft
The kevinadityaikhsan/llama-3.2-3b-legal-id-sft model is a 3.2 billion parameter Llama-based language model, fine-tuned by kevinadityaikhsan. It was developed using Unsloth and Huggingface's TRL library, enabling faster training. This model is specifically fine-tuned from unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit, suggesting an optimization for instruction-following tasks within a legal context for Indonesian language, given the 'legal-id' in its name.
Loading preview...
Model Overview
This model, kevinadityaikhsan/llama-3.2-3b-legal-id-sft, is a 3.2 billion parameter language model developed by kevinadityaikhsan. It is fine-tuned from the unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit base model, indicating its foundation in the Llama architecture and its instruction-following capabilities.
Key Characteristics
- Architecture: Based on the Llama model family.
- Parameter Count: Features 3.2 billion parameters, offering a balance between performance and computational efficiency.
- Training Efficiency: The model was fine-tuned significantly faster using Unsloth and Huggingface's TRL library, highlighting an optimized training process.
- Context Length: Supports a context length of 32768 tokens.
Potential Use Cases
Given its fine-tuning from an instruction-following Llama model and the 'legal-id-sft' designation, this model is likely specialized for:
- Legal Text Processing: Tasks involving legal documents, queries, or summarization, particularly within an Indonesian context.
- Instruction Following: Executing specific instructions related to legal information or general language tasks.
- Resource-Efficient Applications: Its 3.2B parameter size makes it suitable for applications where computational resources are a consideration, while still providing robust language understanding.