HendrikStrobelt/detecting-fake-text
Giant Language Model Test Room
Provides real-time token-level probability analysis to detect LLM-generated text by visualizing model rank distributions and likelihood scores across input sequences. Implements a pluggable backend architecture supporting multiple models (GPT-2, BERT) via a Python server with REST API, paired with an interactive web client for linguistic analysis and inspection.
495 stars. No commits in the last 6 months.
Stars
495
Forks
116
Language
TypeScript
License
Apache-2.0
Category
Last pushed
Jan 18, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/HendrikStrobelt/detecting-fake-text"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
ssciwr/AMMICO
AI-based Media and Misinformation Content Analysis Tool: Analyze text and images
ryuryukke/MINT
A unified evaluation suite for membership inference and machine text detection.
AFLucas-UOM/AI-Early-Detection-System-for-Dyslexia-ADHD
Proof-of-concept behavioural analysis system for early detection of dyslexia and ADHD using...
CyranoB/slop-detector
Detect AI-like writing patterns in creative prose. Returns a 0-100 score where higher = more AI-like.
superannotateai/generated_text_detector
SuperAnnotate HTTP service for Generated Text Detection