Tags

adamw 1 post
ai 2 posts
ai-investment 1 post
architecture 2 posts
attention 1 post
capex 1 post
colossus 1 post
common-crawl 1 post
cost 1 post
大语言模型 1 post
data-quality 1 post
datasets 1 post
dclm 1 post
deepseek 3 posts
deepseek-v3 3 posts
economics 1 post
energy 1 post
fineweb 1 post
fineweb-edu 1 post
fp8 1 post
frontier 1 post
gb200 1 post
gpt-5 1 post
gpu 1 post
gpu-cluster 1 post
h100 1 post
h200 1 post
hardware 1 post
infrastructure 2 posts
机器学习 1 post
kv-cache 1 post
llama 1 post
llm 12 posts
machine-learning 1 post
mixed-precision 1 post
mixture-of-experts 1 post
mla 1 post
muon 1 post
nemotron-cc 1 post
optimization 1 post
optimizer 1 post
power 1 post
pre-training 3 posts
project-rainier 1 post
quantization 1 post
qwen 3 posts
shampoo 1 post
soap 1 post
stargate 2 posts
tpu 1 post
training 1 post
training-budget 1 post
trainium 1 post
预训练 1 post