Menu

#Parameters

17 posts

Feed·
16 of 17 posts
Dense vs. MoE Models: Active Parameters, Throughput, and When to Choose Each
🖼️
0Read More

Dense vs. MoE Models: Active Parameters, Throughput, and When to Choose Each

NVIDIA Technical Blog·Elizabeth Goodman·9 days ago
#nSbSdGrd
#developer#model#token#dense#memory

How can a 30B-parameter model activate only 3B parameters per token, and still use the capacity of the larger model? Nemotron 3.5…

15s
PersonymAI Topic Discovery: AI That Studies Your Channel Before Commenting
🖼️
0Read More

PersonymAI Topic Discovery: AI That Studies Your Channel Before Commenting

DEV Community: sass·PersonymAi·5 months ago
#FWUJWAWC
#dev#user#topic#parameters#global

If you're building an AI product for content channels, you'll hit one question sooner or later: how...

15s
Params Vs Compute
📰
0Read More

Params Vs Compute

Hacker News·6 months ago
#AcXrdO94
#parl#computation#models#model#parameters

When we talk about the power of a deep learning model, often the only metric we pay attention to is its size, which is measured by the number parameters in that model.…

15s