Edit model card

Malaysian Qwen2.5-0.5B Instruct

Continue finetuning https://huggingface.co/Qwen/Qwen2.5-0.5B on highly curated 1.5B tokens Malaysian instruction dataset.

Improvement

  1. Support respond in Manglish, Mandarin, Tamil, Jawi, Johor, Kedah, Kelantan, Pahang, Perak, Sabah, Sarawak, Selangor, Negeri Sembilan and Terengganu.
  2. Able to code in Manglish, Mandarin, Tamil, Jawi, Johor, Kedah, Kelantan, Pahang, Perak, Sabah, Sarawak, Selangor, Negeri Sembilan and Terengganu.
  3. Multi-turn Malaysian context such as related to Malaysian Legislation, politics, religions and languages.
  4. Malaysian role-playing.
  5. Standard RAG.

WanDB at https://wandb.ai/huseinzol05/lora-embedding-256-Qwen2.5-0.5B-multipack

Downloads last month
4
Safetensors
Model size
630M params
Tensor type
BF16
·
Inference API
Unable to determine this model's library. Check the docs .