4 98

Kh

raidhon

AI & ML interests

Fine-tuning, Dataset creation, Time Series

Recent Activity

new activity 7 days ago

Qwen/QwQ-32B-Preview:Can't reproduce the evaluation result of GPQA dataset

View all activity

Organizations

None yet

raidhon's activity

New activity in Qwen/QwQ-32B-Preview 7 days ago

Can't reproduce the evaluation result of GPQA dataset

#47 opened 8 days ago by

Rinn000

liked a model 2 months ago

rhymes-ai/Aria

Image-Text-to-Text • Updated 7 days ago • 18.1k • 599

liked a dataset 3 months ago

KbsdJames/Omni-MATH

Viewer • Updated Oct 12 • 4.43k • 1.04k • 61

replied to m-ric's post 4 months ago

Yes, it's been tested, and it's false. It's even worse than the regular LLAMA 3.1 70b. It's even funny to compare it to Claude.
https://www.reddit.com/r/LocalLLaMA/s/BH5A2ngyui

liked 2 models 7 months ago

imone/Llama-3-8B-fixed-special-embedding

Text Generation • Updated Apr 25 • 295 • 16

Xenova/gpt-4o

Updated May 13 • 57

replied to hrishbhdalal's post 7 months ago

Yeah, I was thinking the same thing. A large vocabulary does improve the performance of smaller LLMs and judging by the GPT-4o the same is true for larger LLM. Give it a try. I'm just doing this for small size models up to 3B parameters.

liked a model 8 months ago