When using wildcards in your prompts, the AI generates responses based on the provided guidelines, but it has some degree of randomness in selecting which elements to include. While wildcards provide general instructions, the AI has the flexibility to decide how to interpret and incorporate them. It also depends heavily on the checkpoints you are using.

If you find that certain descriptors or elements are consistently not rendering in the generated image, you can experiment with different approaches. One option is to rephrase or restructure your prompt, emphasizing the specific parts you want to see. For example, if you want a "jukebox" to appear, you can mention it as a single word with higher weight, like (jukebox:1.5). Increasing the weight assigns greater importance to that particular element, increasing the chances of it being included.

Additionally, you can try removing or reducing the emphasis on other words or descriptors in the prompt that might be overshadowing the desired elements. By simplifying and narrowing down the instructions, you can focus the AI's attention on the specific details you want to see in the generated image.

Remember that experimentation and iteration are key when working with wildcards and fine-tuning your prompts. By tweaking and refining your instructions, you can increase the likelihood of achieving the desired results.

no loras used