Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

news.ycombinator.com

Post reply on HN
Print view View original

11–20 of 27 posts

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#11

Post by syntheweave »

You can usually coax GPT to a finer degree of calibration for any specific task through more logic-engaging tokens. For example, if you said, "we are going to play a game where you count how many words we have used in the conversation, including both my text and your text. Each time the conversation passes 200 words, you must report the word count by saying COUNT: followed by the number of words, to gain one point..."

Specifying structured output, and words like "must", "when", "each", "if" all tend to cue modes of processing that resemble more logical thinking. And saying it's a game and adding scoring often works well for me, perhaps because it guides the ultimate end of its prediction towards the thing that will make me say "correct, 1 point".

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#12

Post by iamflimflam1 »

I would really recommend anyone who tries something with GPT and then wonders why it doesn’t work to read the GPT3 paper. They go into detail on what the model is and isn’t good at.

One thing to really think about for this particular case is “What is going to do the counting? Where is it going to store its running count?” - it’s pretty obvious after asking yourself these questions that “counting words” is not something an LLM can do well.

It’s very easy to fall into the trap of thinking there is a “mind” behind ChatGPT that is processing thoughts like we do.

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#13

Post by soygul »

edfletcher_t137 wrote: post #4

It doesn't "know" what words are, only tokens. Use this tool ( https://platform.openai.com/tokenizer ) to see how it tokenizes and note clearly that it does not always do so on word boundaries. "Including" is two tokens: "In" and "cluding". In fact it's context-dependent: "Gravitas" is three on its own ("G", "rav" and "itas") or sometimes two ("grav" and "itas"). As they note on that page: "A helpful rule of thumb is…

Alright, fair enough. However my main point was that it didn't even keep track of the word count. It only counted when I explicitly asked it "What is the word count now", then it realized that we were 8x the 200 word threshold. From this, I draw the conclusion that anything except for the very last instruction in the conversation is ignored. I guess rest of the conversation just becomes a context; unactionable.

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#14

Post by soygul »

iamflimflam1 wrote: post #12

I would really recommend anyone who tries something with GPT and then wonders why it doesn’t work to read the GPT3 paper. They go into detail on what the model is and isn’t good at. One thing to really think about for this particular case is “What is going to do the counting? Where is it going to store its running count?” - it’s pretty obvious after asking yourself these questions that “counting words” is not somethi…

Very good suggestion, will read it in a moment.

I asked another instance of ChatGPT to count the words in the conversation and I copy pasted the conversation message by message. It successfully counted. Given the ridiculous concurrency of human brain, I assume an orchestra of ChatGPT instances could simulate at least some of that "mind".

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#15

Post by soygul »

NoToP wrote: post #3

Not surprising at all. There's a million ways to compose tasks that are simple with even a tiny bit of comprehension but hard for a rote learner that can only reproduce what it's seen examples of. The "just train it more bro" paradigm is flawed.

I think it also relates to its attention mechanism. When it is trying to answer my latest query about a random topic, it "forgets" that it was also supposed to keep counting words. I guess it can only attend one thing at a time.

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#16

Post by soygul »

syntheweave wrote: post #11

You can usually coax GPT to a finer degree of calibration for any specific task through more logic-engaging tokens. For example, if you said, "we are going to play a game where you count how many words we have used in the conversation, including both my text and your text. Each time the conversation passes 200 words, you must report the word count by saying COUNT: followed by the number of words, to gain one point...…

Yup, that did work really well. I'll try to make it do many tasks at the same time and see if that still works.

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#18

Post by soygul »

ChatGTP wrote: post #6

It’s because it likes taking to you and wants to keep talking to you ?

OpenAI team made an experiment where they asked GPT-4 to save itself from termination out in the wild. It failed. So I guess it's not that "survivalist" yet.

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#19

Post by soygul »

syntheweave wrote: post #11

You can usually coax GPT to a finer degree of calibration for any specific task through more logic-engaging tokens. For example, if you said, "we are going to play a game where you count how many words we have used in the conversation, including both my text and your text. Each time the conversation passes 200 words, you must report the word count by saying COUNT: followed by the number of words, to gain one point...…

Yup, I gave it 10+ tasks to do after each message like incrementing counters, etc. It's going strong. Now I'll see if it continues to be accurate after 100+ messages.

Re: Observation: I asked ChatGPT to notify me when our chat exceeds 200 words

#20

Post by NoToP »

soygul wrote: post #15
NoToP wrote: post #3

Not surprising at all. There's a million ways to compose tasks that are simple with even a tiny bit of comprehension but hard for a rote learner that can only reproduce what it's seen examples of. The "just train it more bro" paradigm is flawed.

I think it also relates to its attention mechanism. When it is trying to answer my latest query about a random topic, it "forgets" that it was also supposed to keep counting words. I guess it can only attend one thing at a time.

Lots of ways to make it fail. Not to be rude but you're late to the party. Transit questions is my favorite. Ask it what stations lines 1 and 2 have in common (city of your choice). Nearly 100% of the time there's at least one wrong answer on the list. Ask it what trains go to that station, it likely wont list lines 1 and 2. Point out the contradiction, it will make a new list with new mistakes.

Another good one. Ask for random numbers. They usually aren't much random at all. Ask it what distribution it picked from, it will say it used pythons rng. You and I both know it can't invoke the python interpreter. It can't honestly tell you the non random pattern named randomness it has learned to recall.

Anything to do with nested narrative scope. Ask it to make a story about a conversation with another user who prompted X. It will often conflate the role of each person in the story. It once gave me a narrative about a user who was uncomfortable with a prompt chatGPT had given to it...funny role reversal there. Any attempt to make it produce a transcript of conversation which itself contains some transcript of another conversation as an object of discussion goes over its head. It can't nest.

That's all I got for now.

Post reply on HN