<p>Thought: we need to synchronize all communication (messengers, notes, emails, etc.) as fine-tuning of the model so that the responses become more and more like your thoughts.</p>
<p>The main question I have is how to create a question-answer link, that is, not just loading text snippets. The OpenAI API expects the format</p>
<pre><code>{"prompt": "<prompt text>", "completion": "<ideal generated text>"}</code></pre>
<p>One idea is to cut off part of the phrase and train on its ending.</p>
<p>Help me with your thoughts, and what do you think of the idea in general?</p>