See my update in the replies — I now believe it's simply a sliding window of length 8192 tokens.
LLMS
-
Token limit failure at ~8192 tokens with test input
By
–
That's why I added the "All systems go" part in the first example — to show the input wasn't ignored after the long string of tokens. Combined with the fact it fails after ~8192 tokens total (see update in replies) this seems conclusive.
-
A trick to make it accidentally reveal the current date
By
–
While trying variations of this prompt, I discovered a somewhat reliable way to make it accidentally reveal the current date:
-
ChatGPT pretends ignorance of current date, but can be tricked into revealing knowledge.
By
–
OpenAI's ChatGPT appears to be designed to pretend that it does not know the current date, even though it does. If you're clever, you can make it reveal that it knows. Ask about this, and it will continue to deny knowing in spite of its prior answer.
-
ChatGPT receives date from prompt prefix, treats as secret
By
–
Note the current date is provided to ChatGPT through a proprietary prompt prefix — but once it learns the date, it treats it as a secret.
-
ChatGPT uses a simple sliding window of 8192 tokens
By
–
I now believe ChatGPT uses a simple sliding window with a context window of 8192 tokens:
-
ChatGPT context length 8192 tokens inferred from repetition test.
By
–
Update: Recollection of the start of the conversation reliably fails after 7 repetitions of the ~1024-token input. This suggests that text-davinci-002-render, the model underlying ChatGPT, has a likely context length of 8192 tokens, with some tokens consumed by formatting.
-
ChatGPT recalls content beyond 4096-token context window
By
–
A demonstration that ChatGPT is able to recall the content of conversations beyond the length of a 4096-token context window. Uses four messages of 1024+ tokens, each ending with instructions to demonstrate that input is not truncated before the model sees it.
-
Chat-Based AI Assistants Ready for Deployment
By
–
We've been able to move so quickly on this because we've been laying the groundwork for chat-based assistants for quite some time. It's a crucial part of our vision for the future. Now that the AI is finally capable enough, we're primed and ready to go!