zjunlp/knowlm-13b-base-v1.0
TEXT GENERATIONPricing:Input $1.5 / Output $2.1Concurrent Unit Cost:1Model Size:13BQuant:FP8Context Size:4kLicense:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold
The zjunlp/knowlm-13b-base-v1.0 is a 13 billion parameter language model built upon the LLaMA-13b architecture. It features a 4096-token context length and has undergone a secondary full-scale pretraining phase using both Chinese and English bilingual data. This additional training significantly enhances the model's comprehension of Chinese language, making it suitable for bilingual applications.
Loading preview...
KnowLM-13b-Base-v1.0 Overview
KnowLM-13b-Base-v1.0 is a 13 billion parameter language model developed by zjunlp, leveraging the foundational LLaMA-13b architecture. This model distinguishes itself through a specialized secondary pretraining phase. During this phase, it was exposed to a substantial volume of both Chinese and English bilingual data.
Key Capabilities
- Enhanced Chinese Comprehension: The bilingual pretraining significantly improves the model's understanding and processing of the Chinese language.
- LLaMA-13b Foundation: Benefits from the robust architecture and general language understanding capabilities of the LLaMA-13b base model.
- Bilingual Data Integration: Incorporates knowledge from both English and Chinese datasets, making it versatile for cross-lingual tasks.
Good For
- Applications requiring strong Chinese language understanding.
- Bilingual natural language processing tasks involving both Chinese and English.
- Research and development building upon the LLaMA architecture with an emphasis on East Asian languages.