LLM-Text-Loop append each answer to keep each feature

[xlinx]

2.Append each. ==> u will get every line llm-answer separately.
when u ask llm looply. each ask is a feature; if each is important or feture that u want show indenprnt, check it.

• example:
• if u use red yellow blue ask llm 3-times.
• u will get mix result. like fig1
• enable append each.
• slide 1-5 how many times ask
• u will get 3-feature separately

[sandner-art]

I think this should add text into the [LLM-Your-Prompt]? Or send a separate call? It does not seem to work as it is, sometimes it just adds the keyword on the line to the prompt (i.e. "red", before the generated prompt). I am using Forge and xtuner/llava-llama-3-8b-v1_1-gguf/llava-llama-3-8b-v1_1-int4.gguf (with vision), can it have an impact?