aokitools/japanese-laws-egov-instruct-202508182216
The aokitools/japanese-laws-egov-instruct-202508182216 is an experimental 2 billion parameter instruction-tuned causal language model, continually pretrained from Qwen/Qwen3-1.7B. Developed by aokitools, this model is designed for tasks related to Japanese laws and e-governance, leveraging a 32768 token context length. It is intended for research and development in specialized legal and governmental text processing within the Japanese context.
Loading preview...
Model Overview
This model, aokitools/japanese-laws-egov-instruct-202508182216, is an experimental 2 billion parameter instruction-tuned language model. It is a continual pretraining of the Qwen/Qwen3-1.7B base model, utilizing the QwenTokenizer.
Key Features
- Base Model: Built upon
Qwen3-1.7Bfor its foundational language understanding. - Instruction-Tuned: Optimized for following instructions, making it suitable for various NLP tasks.
- Experimental Stage: Currently in a research phase, indicating ongoing development and refinement.
- Context Length: Supports a substantial context window of 32768 tokens, allowing for processing longer inputs.
Usage and Integration
This model can be easily integrated into projects using Ollama or directly via the Hugging Face transformers library in Python. It supports quantization with BitsAndBytesConfig for efficient memory usage.
License
Released under the Apache 2.0 + Alibaba Qianwen License, providing flexibility for various applications while adhering to specific terms.