A 1-billion-parameter instruction fine-tuning model based on the Meta Llama 3.2 architecture, specifically optimized for text generation tasks. This model uses the PEFT (Parameter-Efficient Fine-Tuning) technique to enhance instruction following and dialogue capabilities while maintaining the capabilities of the base model.

Natural Language ProcessingPeft

A fine-tuned version of the google/flan-t5-base model for dialogue summarization tasks, using PEFT for parameter-efficient fine-tuning

Natural Language ProcessingTensorboardEnglish

This is a LoRA adapter extracted from DeepSeek-R1-Distill-Qwen-32B, based on the Qwen2.5-32B base model, suitable for parameter-efficient fine-tuning.

Natural Language ProcessingTransformers

Athena-70B-L3 is a large language model with 70B parameters, supporting English text generation tasks and utilizing parameter-efficient fine-tuning techniques.

Natural Language ProcessingTransformersEnglish

Camelidae-8x34B is a Mixture of Experts (MoE) model trained using parameter-efficient sparse construction technology. It uses 8 experts with a parameter scale of 34B and performs excellently in general instruction fine-tuning tasks, especially in code and mathematical domain tasks.

Natural Language ProcessingTransformersEnglish

Camelidae-8x13B is a Mixture of Experts model based on parameter-efficient sparse construction technology. It helps dense models learn multi-domain knowledge such as code and math through instruction fine-tuning, improving model performance while maintaining efficient parameter utilization.

Natural Language ProcessingTransformersEnglish

KULLM-Polyglot-5.8B-v2 is a large language model version that performs parameter-efficient fine-tuning on the KULLM v2 dataset based on EleutherAI/polyglot-ko-5.8b. It is specifically optimized for Korean natural language processing tasks and has a scale of 5.8 billion parameters.

Natural Language ProcessingTransformersKorean