News for hackers. Stuff that matters.

A comment on Gemma 3 270M: The compact model for hyper-efficient AI. Parent comment

Re:Gemma 3 270M: The compact model for hyper-efficient AI by danielhanchen on Thursday August 14, 2025 @03:51PM (#44904841)
Oh :( Maybe the settings? Could you try

temperature = 1.0, top_k = 64, top_p = 0.95, min_p = 0.0

[ Parent | on HN ]
Re:Gemma 3 270M: The compact model for hyper-efficient AI by canyon289 on Thursday August 14, 2025 @04:41PM (#44905468)
Daniel, thanks for being here providing technical support as well. Cannot express enough how much we appreciate your all work and partnership.
[ Parent | on HN ]
Re:Gemma 3 270M: The compact model for hyper-efficient AI by danielhanchen on Thursday August 14, 2025 @05:37PM (#44905992)
Thank you and fantastic work with Gemma models!
[ Parent | on HN ]
Re:Gemma 3 270M: The compact model for hyper-efficient AI by simonw on Thursday August 14, 2025 @06:28PM (#44906449)
My topping only lets me set temperature and top_p but setting them to those values did seem to avoid the infinite loops, thanks.
[ Parent | on HN ]
Re:Gemma 3 270M: The compact model for hyper-efficient AI by danielhanchen on Thursday August 14, 2025 @07:52PM (#44907095)
Oh fantastic it worked! I was actually trying to see if we can auto set these within LM Studio (Ollama for eg has params, template) - not sure if you know how that can be done? :)
[ Parent | on HN ]