Lun Zima PRO

Lunzima's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

replied to their post 2 days ago

view reply

I don't know if the performance of Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v9.2 has improved or regressed because https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard/ is stuck.

posted an update 2 days ago

view post

Post

1168

I'm currently experimenting with the SFT dataset Lunzima/alpaca_like_dataset to further boost the performance of NQLSG-Qwen2.5-14B-MegaFusion-v9.x. This includes data sourced from DeepSeek-R1 or other cleaned results (excluding CoTs). Additionally, datasets that could potentially enhance the model's performance in math and programming/code, as well as those dedicated to specific uses like Swahili, are part of the mix.
@sometimesanotion @sthenno @wanlige

See translation

• 
• 1 reply

·

🚀

2

2

👍

1

1

+

Reply

posted an update 22 days ago

view post

Post

591

🚀 Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5 now excels in reasoning and coding, built on top of v4 which improved Chinese capabilities through SFT.

See translation

👍

1

1

+

Reply

posted an update 24 days ago

view post

Post

471

🚀 Created a LLM to play "NQLSG" - a quirky AMD employee known for occasional hilarious quotes! Used mergekit-multi to develop Lunzima/NQLSG-Qwen2.5-14B-MegaFusion series. Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v3 is my favorite so far!

See translation

👍

1

1

+

Reply