Hugging Face's logo
Mike Ravkine's picture
🤝 Open to Collab

Mike Ravkine PRO

mike-ravkine
49 27 678
Frangils's profile picture dipankarsarkar's profile picture branikita's profile picture
·

AI & ML interests

LLM Research / Development / Evaluation

Recent Activity

liked a model 1 day ago
amd/Instella-MoE-16B-A3B-Think
repliedto their post 2 days ago
I went into this expecting to find a ~Q2 garbage dumpster but https://huggingface.co/prism-ml/Ternary-Bonsai-27B-gguf is a slick feat of QAT engineering. It is weaker then FP16 on a handful of tasks where quants usually degrade, as per their own paper the loss is "concentrated on sustained chains of reasoning / agentic" and in ReasonScape this bites on Sort, Shuffle and Dates, but counter-acting this are some noticeable improvements to thinking length without accuracy loss on several other tasks (Shapes, Cars). I haven't had a chance to run the Binary yet, but PQ2 + Bonsai QAT are confirmed to be pretty darn impressive.
repliedto their post 2 days ago
I went into this expecting to find a ~Q2 garbage dumpster but https://huggingface.co/prism-ml/Ternary-Bonsai-27B-gguf is a slick feat of QAT engineering. It is weaker then FP16 on a handful of tasks where quants usually degrade, as per their own paper the loss is "concentrated on sustained chains of reasoning / agentic" and in ReasonScape this bites on Sort, Shuffle and Dates, but counter-acting this are some noticeable improvements to thinking length without accuracy loss on several other tasks (Shapes, Cars). I haven't had a chance to run the Binary yet, but PQ2 + Bonsai QAT are confirmed to be pretty darn impressive.
View all activity

Organizations

Blog-explorers's profile picture