instruction clarity

Instruction clarity is the simple, high-payoff habit of telling the model exactly what you want, in unambiguous and specific terms. A model cannot read your mind; it can only act on the words in front of it, and where those words are vague it will fill the gap with a reasonable guess that may not match your intent. Asking write about dogs invites anything; asking write a 100-word friendly paragraph for children explaining why dogs wag their tails gets you the thing you pictured.

Clear instructions usually name the task, the audience, the desired length and format, the tone, and any constraints or things to avoid. It helps to state the goal positively, to put the most important requirement up front, and to give an example of the output shape when words alone are slippery. Much of what looks like a model failing is really an under-specified request, and tightening the wording fixes it faster than any clever trick. The model takes your words fairly literally, so say precisely what you mean and do not assume it shares context you only have in your head.