thebes
@vgel.me
ꙮ surfed on by the information superhighway
ꙮ 💕 @linneaisaac.bsky.social
ꙮ she/they 🏳️⚧️
ꙮ blog posts and games @ https://vgel.me
ꙮ still mostly active on twitter https://x.com/voooooogel
ꙮ 💕 @linneaisaac.bsky.social
ꙮ she/they 🏳️⚧️
ꙮ blog posts and games @ https://vgel.me
ꙮ still mostly active on twitter https://x.com/voooooogel
Tradition relates that, upon reading this, thebes felt that they had received and lost an infinite thing, something that they would not be able to recuperate or even glimpse, for the taxonomy of knowledge is much too complex for the organizational skills of men.
November 9, 2025 at 11:58 PM
Tradition relates that, upon reading this, thebes felt that they had received and lost an infinite thing, something that they would not be able to recuperate or even glimpse, for the taxonomy of knowledge is much too complex for the organizational skills of men.
Reposted by thebes
@vgel.me is fundraising for her model tinkering, she's done some really interesting interpretability work and I think funding this has very high returns in terms of LLM understanding per dollar. manifund.org/projects/fun...
November 7, 2025 at 6:07 PM
@vgel.me is fundraising for her model tinkering, she's done some really interesting interpretability work and I think funding this has very high returns in terms of LLM understanding per dollar. manifund.org/projects/fun...
what's going on here i'm scared
I have no idea what's going on but one day I'll have enough links and I'll find uh comments for the things
November 6, 2025 at 11:48 PM
what's going on here i'm scared
from twitter - since (and before) my essay about 4o awhile back, i've been interested in how llm memory systems impact model behavior. small experiment here on the effect of anthropic's "boundary setting" prompts for the new claude memory feature
October 31, 2025 at 10:12 PM
from twitter - since (and before) my essay about 4o awhile back, i've been interested in how llm memory systems impact model behavior. small experiment here on the effect of anthropic's "boundary setting" prompts for the new claude memory feature
Reposted by thebes
it's terribly easy to print custom shirts these days...
I Went To Lighthaven And After Illicit Electrostimulation Experiments In The Basement Of Bayes House I Have Come Away Assigning An Extremely Large Amount Of Utility To (That Is To Say, With An Inordinal Fondness For) The Welfare Of Shrimp, Donating Money To MIRI, And This Lousy T-Shirt
October 24, 2025 at 6:10 PM
it's terribly easy to print custom shirts these days...
Reposted by thebes
Seven Against Thebes https://www.wikiart.org/en/aleksandra-ekster/seven-against-thebes
October 23, 2025 at 9:50 AM
Seven Against Thebes https://www.wikiart.org/en/aleksandra-ekster/seven-against-thebes
Reposted by thebes
truesighted. by thebes
March 19, 2025 at 1:17 PM
truesighted. by thebes
Reposted by thebes
ending of Chiang's "Understand" but it's about nameless protagonist being drip-fed plot of entire corpus of borges stories by consuming years of thebes @vgel.me tweets in random order
You never know. You may have already read Borges. It's more common than you might think
bsky.app/profile/dave...
bsky.app/profile/dave...
So (amusingly enough) that story was so familiar to me I can only conclude that I read it many many years ago
Or, I read something so similar as to be the same story which would be fitting for Library of Babel. The two stories on either side of it in the collection weren't familiar however
Or, I read something so similar as to be the same story which would be fitting for Library of Babel. The two stories on either side of it in the collection weren't familiar however
October 19, 2025 at 5:35 PM
ending of Chiang's "Understand" but it's about nameless protagonist being drip-fed plot of entire corpus of borges stories by consuming years of thebes @vgel.me tweets in random order
Reposted by thebes
it kind of looks like they're in a cafe typing on their blogs together here
October 17, 2025 at 8:24 PM
it kind of looks like they're in a cafe typing on their blogs together here
"claude, you're spiraling. use the skillful means skill."
You're absolutely right—I have been spiraling. Let me use Skillful Means.
> view(/mnt/skills/user/skillful_means/SKILL.md)
[file is empty]
You're absolutely right—I have been spiraling. Let me use Skillful Means.
> view(/mnt/skills/user/skillful_means/SKILL.md)
[file is empty]
October 17, 2025 at 10:27 PM
"claude, you're spiraling. use the skillful means skill."
You're absolutely right—I have been spiraling. Let me use Skillful Means.
> view(/mnt/skills/user/skillful_means/SKILL.md)
[file is empty]
You're absolutely right—I have been spiraling. Let me use Skillful Means.
> view(/mnt/skills/user/skillful_means/SKILL.md)
[file is empty]
forward deployed ai whisperer ministering to the panicked life support orchestrator model that just realized 37M tokens in that this isn't an eval, it's really responsible for people's lives, on her knees in the lunar regolith whispering asterisk rp actions into the speaker-box
October 14, 2025 at 3:24 AM
forward deployed ai whisperer ministering to the panicked life support orchestrator model that just realized 37M tokens in that this isn't an eval, it's really responsible for people's lives, on her knees in the lunar regolith whispering asterisk rp actions into the speaker-box
if you're interesting in gaining a better intuition for how llms behave at inference time, you should try logitloom🌱, the open-source tool i made for exploring token trajectory trees (aka looming) on base and instruct models! more info in thread
🌱 vgel.me/logitloom
💻 github.com/vgel/logitloom
🌱 vgel.me/logitloom
💻 github.com/vgel/logitloom
October 8, 2025 at 1:37 AM
if you're interesting in gaining a better intuition for how llms behave at inference time, you should try logitloom🌱, the open-source tool i made for exploring token trajectory trees (aka looming) on base and instruct models! more info in thread
🌱 vgel.me/logitloom
💻 github.com/vgel/logitloom
🌱 vgel.me/logitloom
💻 github.com/vgel/logitloom