SeongryongJung/Qwen3-4B-Biology-RLSD-TR
TEXT GENERATIONConcurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 2, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
SeongryongJung/Qwen3-4B-Biology-RLSD-TR is a 4 billion parameter Qwen3-based language model fine-tuned using the RLSD_TR method specifically for biology-related tasks. This model, with a context length of 32768 tokens, is optimized for performance on the SciKnowEval biology dataset, achieving a best validation mean@16 of 48.50%. It is designed to provide specialized knowledge and reasoning capabilities within the biological domain.
Loading preview...
Model Overview
SeongryongJung/Qwen3-4B-Biology-RLSD-TR is a specialized 4 billion parameter language model built upon the Qwen3 architecture. It has been fine-tuned using the RLSD_TR (Reinforcement Learning from Scientific Data with Trust-Region) method, specifically targeting the biology domain.
Key Capabilities
- Biology-specific knowledge: Optimized for understanding and generating content related to biological sciences.
- Performance on SciKnowEval: Achieved a best validation
mean@16score of 48.50% on the SciKnowEval biology dataset, indicating strong performance in this specialized area. - Extended Context Window: Supports a maximum prompt length of 2048 tokens and a maximum response length of 8192 tokens, with a total model length of 10240 tokens, allowing for processing of substantial biological texts.
- Training Methodology: Utilizes a sophisticated RLSD_TR training approach with a constant learning rate of 1e-6 and a batch size of 32, focusing on robust generalization within the biology dataset.
Good For
- Biological text analysis: Tasks requiring deep understanding or generation of biological concepts, research papers, or scientific data.
- Specialized question answering: Answering queries within the field of biology where domain-specific knowledge is crucial.
- Research and academic applications: Supporting tools or systems that require a strong foundation in biological information.