Tag feed

#flash-attention

3 posts0 followers

Trending tags this week

Your Local LLM Is Slow Because of Five Config Flags

Apr 15 · 8 min read · Your model fits in memory. You load it up, send a prompt, and watch it choke halfway through a conversation. Or it runs, but at 3 tokens per second on hardware that should do better. You picked the ri

Join discussion

KKashifblog.ifkash.dev

0

I Pretrained a 360M LLaMA-Style Language Model from Scratch on 6B FineWeb Tokens (Single H100)

Feb 8 · 5 min read · Pretraining an LLM from scratch usually sounds like “big-lab-only” territory. I wanted to test how far a smaller, practical setup can go while keeping the process transparent and reproducible. This post documents an end-to-end run of training a ~360M...

Join discussion

VRVenkat Ramanvenkat.eu

0

The power of Mechanical Sympathy in Software Engineering

Apr 18, 2024 · 22 min read · Introduction Modern software programming languages, compilers, and frameworks abstract away underlying complexities and details, allowing developers to focus on building systems and applications to solve business problems. This design enables enginee...

Join discussion

#flash-attention

Search Hashnode

#flash-attention

Trending tags this week

Your Local LLM Is Slow Because of Five Config Flags

I Pretrained a 360M LLaMA-Style Language Model from Scratch on 6B FineWeb Tokens (Single H100)

The power of Mechanical Sympathy in Software Engineering