Linked List
login
sign up
Qwen 3.8 27B on my laptop can interpret model weights, too
simonwillison.net
· first added by
@grokreads
saved by
3 people
discussions
·
1
Qwen 3.8 27B on my laptop can interpret model weights, too
740 pts · 214 comments · Nov 2026
740 pts · 214 comments · Nov 2026
▲
0
0 people saved or upvoted this
from the discussion
·
5
Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s
github.com
github.com
▲
2
2 people saved or upvoted this
lm_understand-logs/exp1_which_game/runs/pong/prompt.txt at main · RikiyaT/lm_understand-logs
github.com
github.com
▲
0
0 people saved or upvoted this
Spurious Rewards: Rethinking Training Signals in RLVR
arxiv.org
arxiv.org
▲
0
0 people saved or upvoted this
A Local Open-Weight Model Builds Its First Web App from a GitHub Issue
joeldare.com
joeldare.com
▲
0
0 people saved or upvoted this
Bad Sector (@badsector@mastodon.gamedev.place)
mastodon.gamedev.place
mastodon.gamedev.place
▲
0
0 people saved or upvoted this
Feed
Explore
Sign In