Claude now has a so-called “pop-up window” of around 75,000 words, reports Anthropogenic. This pop-up is, so to speak, the chatbot’s short-term memory, an additional piece of ready-to-use knowledge that the chatbot can draw on top of the previously trained knowledge of artificial intelligence.
An example: Claudius for example, can “read” an entire book in less than a minute, then talk about every aspect of the book in simple language. This is useful when analyzing all kinds of texts, but also during conversations. Anyone who has a longer conversation with a chatbot will notice after a number of questions that the beginning is already lost. The chatbot then forgets what was discussed at the beginning of the conversation, which can lead to a circular conversation.
Memorize the meaning of words
For example, Microsoft Bing’s chatbot, which works with ChatGPT, is limited to 20 questions. After that, the chatbot may forget previous information, and strange and sometimes hurtful responses may occur. With Claude, this limit is now much higher.
In comparison: Claude can now remember around 75,000 words in his “available memory”, whereas this limit at ChatGPT is already around 3,000 words. With artificial intelligence, however, this memory is not measured in words, but in tokens: after all, not all words have the same length, moreover, it is mainly about the underlying meaning.
Not words but tokens
The AI therefore measures words in tokens, and Claude can store up to 100,000 tokens in its ready memory. With ChatGPT it’s 8,000 tokens in the paid version, or 32,000 tokens in the new, still limited GPT-4 version.
The new version of Claude is currently only available to Anthropic’s business partners. It is not public how much access to Claude costs.
Google is said to have recently invested around $400 million in Anthropic. Google recently released its own AI chatbot Bard.
“Bacon trailblazer. Certified coffee maven. Zombie lover. Tv specialist. Freelance communicator.”