Analyzing LLM Contamination in the Wild
Back to Blog
November 28, 2024

Analyzing LLM Contamination in the Wild

Some LLMs are scoring suspiciously high on benchmarks. A data-driven analysis of which models likely saw test data during training and how to spot it.

Jonathan Chavez
Jonathan Chavez
Software Engineer @ Datadog

Disclaimer: The views and opinions expressed in this blog are my own and do not necessarily reflect the official position of my employer.

Analyzing LLM Contamination in the Wild

Continue Reading

01

Which LLM Benchmarks Don't Matter in 2024?

02

Are We Measuring LLMs Wrong? A Deep Dive into Evaluation Metrics

03

Are Long Context Windows Worth It? A Context vs RAG Analysis

04

Can Open Source LLMs Really Compete with Closed Models?

05

How to calculate the cost of serving LLMs in production