Anterior C.C.
@philasophy.bsky.social
2nd year PhD student | @Metacogmission on Twitter
Reposted by Anterior C.C.
🚨New Working Paper!🚨
How do Americans mentally map political coalitions?
We often talk as if U.S. politics is just two sorted camps. But when people think about which groups go with which, the picture is more complicated.
🧵👇
How do Americans mentally map political coalitions?
We often talk as if U.S. politics is just two sorted camps. But when people think about which groups go with which, the picture is more complicated.
🧵👇
April 19, 2026 at 9:30 PM
🚨New Working Paper!🚨
How do Americans mentally map political coalitions?
We often talk as if U.S. politics is just two sorted camps. But when people think about which groups go with which, the picture is more complicated.
🧵👇
How do Americans mentally map political coalitions?
We often talk as if U.S. politics is just two sorted camps. But when people think about which groups go with which, the picture is more complicated.
🧵👇
Meta's TRIBE v2 is cool.
I took some pilot data I had from a Turing Test style study. I fed it to the model to see if human messages had distinctively different neural activity than messages generated by Gemini 3 Flash after controlling for word length.
I took some pilot data I had from a Turing Test style study. I fed it to the model to see if human messages had distinctively different neural activity than messages generated by Gemini 3 Flash after controlling for word length.
March 29, 2026 at 7:48 AM
Meta's TRIBE v2 is cool.
I took some pilot data I had from a Turing Test style study. I fed it to the model to see if human messages had distinctively different neural activity than messages generated by Gemini 3 Flash after controlling for word length.
I took some pilot data I had from a Turing Test style study. I fed it to the model to see if human messages had distinctively different neural activity than messages generated by Gemini 3 Flash after controlling for word length.
Need synthetic data but don't want to use the API? I made an MCP specifically for small dataset curation with cli agents.
Tell the model what columns you want, the data you need generated, and how many rows. It includes a validation layer to make sure output is unique.
github.com/IMNMV/Genguard
Tell the model what columns you want, the data you need generated, and how many rows. It includes a validation layer to make sure output is unique.
github.com/IMNMV/Genguard
GitHub - IMNMV/Genguard: A data generation framework for cli agents
A data generation framework for cli agents. Contribute to IMNMV/Genguard development by creating an account on GitHub.
github.com
January 8, 2026 at 3:22 AM
Need synthetic data but don't want to use the API? I made an MCP specifically for small dataset curation with cli agents.
Tell the model what columns you want, the data you need generated, and how many rows. It includes a validation layer to make sure output is unique.
github.com/IMNMV/Genguard
Tell the model what columns you want, the data you need generated, and how many rows. It includes a validation layer to make sure output is unique.
github.com/IMNMV/Genguard
I've been wanting to explore how people talk to Grok since asking Grok for information became common place. To test out a new scraping method, I took a look at tweets with @grok today since there have been an uptick in requesting sexual modifications to images and analyzed them w/ Opus 4.5 in R.
January 1, 2026 at 3:31 AM
I've been wanting to explore how people talk to Grok since asking Grok for information became common place. To test out a new scraping method, I took a look at tweets with @grok today since there have been an uptick in requesting sexual modifications to images and analyzed them w/ Opus 4.5 in R.
I was curious to check out how frontier LLM companies each handled terms of service between consumers and their enterprise agreements.
I was primarily interested in: arbitration, indemnification, and user data training. Using Opus 4.5 it scraped the consumer and business TOS for major AI companies.
I was primarily interested in: arbitration, indemnification, and user data training. Using Opus 4.5 it scraped the consumer and business TOS for major AI companies.
December 30, 2025 at 8:31 AM
I was curious to check out how frontier LLM companies each handled terms of service between consumers and their enterprise agreements.
I was primarily interested in: arbitration, indemnification, and user data training. Using Opus 4.5 it scraped the consumer and business TOS for major AI companies.
I was primarily interested in: arbitration, indemnification, and user data training. Using Opus 4.5 it scraped the consumer and business TOS for major AI companies.
Given the recent executive order by Trump to federally regulate AI, I was curious to explore what standalone bills Congress has managed to pass - and who has led the charge in passing them.
With Opus 4.5 in R, I pulled bills with AI related terms then made topics from those will full text to analyze
With Opus 4.5 in R, I pulled bills with AI related terms then made topics from those will full text to analyze
December 30, 2025 at 8:23 AM
Given the recent executive order by Trump to federally regulate AI, I was curious to explore what standalone bills Congress has managed to pass - and who has led the charge in passing them.
With Opus 4.5 in R, I pulled bills with AI related terms then made topics from those will full text to analyze
With Opus 4.5 in R, I pulled bills with AI related terms then made topics from those will full text to analyze