Seeing in Pangram Space | Pangram Labs

Exploring the internal representations of Pangram 3.3.2

Exploring the internal representations of Pangram 3.3.2 By Elyas Masrour, Katherine Thai, and Bradley Emi June 2026 Loading visualization… Fig. 1. Each dot is a human or AI document. The animation shows the progressive separation between AI and human documents throughout Pangram's layers. 01 Introduction Since ChatGPT’s debut in 2022, AI-assisted writing has expanded at a staggering pace. Because AI-generated text now appears across so much of what we read, it has become obvious that some forms of writing lose their value when produced by a machine. In academia, essays are meant to…

saved by

• Elizabeth Qiu
• Julian H
• Ray Arayilakath

related reading

• Pangram 4 Technical Reportpangram-public.s3.us-east-1.amazonaws.com
• I don't care (anymore) if your writing got a Pangram false positiveandrewwu.substack.com
• How does Pangram work?pangram.substack.com
• How does Pangram work?substack.com
• Pangram 4 Technical Overviewpangram.com
• A Mathematical Framework for Transformer Circuitstransformer-circuits.pub
• Verbalizable Representations Form a Global Workspace in Language Modelstransformer-circuits.pub
• interpreting GPT: the logit lens — LessWronglesswrong.com
• Prism: mapping interpretable concepts and features in a latent space of language | thesephist.comthesephist.com
• Circuit Tracing: Revealing Computational Graphs in Language Modelstransformer-circuits.pub
• Why Perplexity and Burstiness Fail to Detect AI | Pangram Labspangram.com
• Does Pangram Work on Claude Fable 5? | Pangram Labspangram.com