Following the Long Echo philosophy and the Long Echo Toolkit, I used CTK (Conversation Toolkit) to export and preserve my ChatGPT conversations.

The Archive

The live archive is offline at the moment while I think through what to publish and how. The numbers below describe what was built; if you want a copy of the source corpus, the sanitized data is published as the chatgpt-conversation-corpus repo.

Stats

1,956 conversations (sanitized from 2,219, removed 263 containing sensitive data)