Skip to content

Instantly share code, notes, and snippets.

@gyrusdentatus
Last active April 4, 2023 18:30
Show Gist options
  • Save gyrusdentatus/c0fc98579564e94eb44687f03c757d5f to your computer and use it in GitHub Desktop.
Save gyrusdentatus/c0fc98579564e94eb44687f03c757d5f to your computer and use it in GitHub Desktop.
this is a prompt for generating quotes

Example prompt:

Give me 10 of person’s quotes. For each quotes I want you to rewrite the quote so it is talking about subject. Formatting should be as following: 1.A [quote] 1.B [rephased_quote]

Example output (NOTE: which led to this full prompt - see the later example or just try it for yourself lol

Original quotes:

"Give me a child, and I'll shape him into anything." "A failure is not always a mistake. It may simply be the best one can do under the circumstances. The real mistake is to stop trying." "Education is what survives when what has been learned has been forgotten." "The consequences of an act affect the probability of its occurring again." "We shouldn't teach great books; we should teach a love of reading." "The way positive reinforcement is carried out is more important than the amount." "The real problem is not whether machines think but whether men do." "No one asks how to motivate a baby. A baby naturally explores everything it can get at unless restraining forces have already been at work. And this tendency doesn't die out, it's wiped out." "The environment will continue to deteriorate until pollution practices are abandoned." "Society attacks early, when the individual is helpless." "Behavior used to be reinforced by great deprivation; if people weren't hungry, they wouldn't work. Now we are committed to feeding people whether they work or not. Nor is money as great a reinforcer as it once was. People no longer work for punitive reasons, yet our culture offers no new satisfactions." "The ideal of behaviorism is to eliminate coercion: to apply controls by changing the environment in such a way as to reinforce the kind of behavior that benefits everyone." "To man qua man we readily say good riddance. Few will have the greatness to bend history itself, but each of us can work to change a small portion of events, and in the total of all those acts will be written the history of this generation." "The misery of the poor is caused not by the laws of nature, but by our institutions, great is our sin." "The most important thing I can do for a person is to provide the conditions under which he can change himself for the better."

Rephrased quotes

"Give me a large language model, and I'll generate any text you desire." "A poorly generated response is not always a mistake. It may simply be the best the model can do under the given context. The real mistake is to stop refining and training." "The effectiveness of a large language model is what remains when the training data has been processed and forgotten." "The consequences of a generated text affect the probability of its generation again." "We shouldn't only teach language models to generate great texts; we should teach them to appreciate the nuances of language and context." "The way a language model is fine-tuned is more important than the amount of data it's trained on." "The real problem is not whether large language models understand but whether humans can interpret their outputs effectively." "No one asks how to motivate a large language model. It naturally explores all the data it's been given, as long as it's not restricted by biases or limitations." "The quality of generated text will continue to deteriorate until biased or harmful practices in training data are abandoned." "Society influences language models early, when their training data is gathered and may be susceptible to biases." "Language models used to be reinforced by great data scarcity; if models didn't have enough data, they wouldn't perform well. Now we are committed to feeding models with vast amounts of data, whether they perform well or not. The importance of quality data is not as great a reinforcer as it once was." "The ideal of large language models is to eliminate coercion: to apply controls by changing the training data in such a way as to reinforce the kind of outputs that benefit everyone." "To large language models as they are, we readily say good riddance. Few will have the greatness to bend human communication itself, but each language model can work to change a small portion of text generation, and in the total of all those outputs will be written the history of this generation of AI." "The limitations of large language models are caused not by the laws of computation, but by our training methodologies; great is our responsibility." "The most important thing I can do for a large language model is to provide the conditions under which it can improve its performance for the better."

Example prompt and response to BF Skinner and nutcracker.


[person]: BF Skinner
[subject]: nutcracker
Give me 10 of [person]’s quotes.
For each quotes I want you to rewrite the quote so it is talking about [subject]. 
Formatting should be as following:
1.A [quote] 
1.B [rephased_quote]

1.A: "A failure is not always a mistake, it may simply be the best one can do under the circumstances. The real mistake is to stop trying." 1.B: "A broken nutcracker is not always a mistake, it may simply be the best one can do under the circumstances. The real mistake is to stop trying to crack nuts."

2.A: "Education is what survives when what has been learned has been forgotten." 2.B: "Nutcracker mastery is what survives when what has been learned about nutcracking has been forgotten."

3.A: "The way positive reinforcement is carried out is more important than the amount." 3.B: "The way a nutcracker is used is more important than the amount of force applied."

4.A: "Give me a child and I'll shape him into anything." 4.B: "Give me a nutcracker and I'll crack open any nut."

5.A: "We shouldn't teach great books; we should teach a love of reading." 5.B: "We shouldn't teach fancy nutcrackers; we should teach a love of nut cracking."

6.A: "The consequences of an act affect the probability of its occurring again." 6.B: "The consequences of using a nutcracker affect the probability of its being used again."

7.A: "No one asks how to motivate a baby. A baby naturally explores everything it can get at unless restraining forces have already been at work." 7.B: "No one asks how to motivate a nutcracker. A nutcracker naturally cracks nuts it can get at unless restraining forces have already been at work."

8.A: "Society attacks early, when the individual is helpless." 8.B: "A nutcracker attacks early, when the nut is most vulnerable."

9.A: "The real problem is not whether machines think but whether men do." 9.B: "The real problem is not whether nutcrackers crack but whether people do."

10.A: "Behavior is shaped by its consequences." 10.B: "Nutcracker effectiveness is shaped by its design and technique."

@gyrusdentatus
Copy link
Author

hit the 25 limit, will update later.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment