News for hackers. Stuff that matters.
Re:LLM Powered Autonomous Agents by mirekrusin on Tuesday June 27, 2023 @03:55AM (#36490104)
Tokens are integers that map to text tokens.

Tokens are part of words, approx 4 characters or 75% of word.

It gives a list of tokens with their probabilities on output.

It's a short list with highest probabilities.

Temperature controls which tokens to pick - usually 0% = top one only (consistent results), closer to 100% means more randomness (more "creativity").

Re:LLM Powered Autonomous Agents by vorticalbox on Tuesday June 27, 2023 @05:26AM (#36490710)
so a glorified Markov chain?
Re:LLM Powered Autonomous Agents by TeMPOraL on Tuesday June 27, 2023 @05:52AM (#36490851)
Yes, in the same sense a modern digital camera is a glorified photodiode. In both cases, light comes in, voltage comes out, and we can use it to count how much light came in.
Re:LLM Powered Autonomous Agents by mirekrusin on Tuesday June 27, 2023 @06:51AM (#36491219)
Why stop there, it's just ones and zeroes.

It's "glorified markov chain" in the same sense that sqlite is just "glorified bubble sort".

Re:LLM Powered Autonomous Agents by visarga on Tuesday June 27, 2023 @07:36AM (#36491491)
Don't you know that "attention is all you need"? Attention is non-markovian. It's all-to-all with some masking, not a chain.
Re:LLM Powered Autonomous Agents by joshspankit on Tuesday June 27, 2023 @09:52AM (#36492776)
Since we’re here: Does a “resused” token count as a second token?

For example: if you limited all inout/output to the same 100 words, could you stay within the token limit permanently?