- Designing your prompt is essentially how you “program” the model. - Adding examples of the output we’d expect for a given input helped the model provide the types of names we were looking for. - Temperature is a value between 0 and 1 that essentially lets you control how confident the model should be when making these predictions. Lowering temperature means it will take fewer risks, and completions will be more accurate and deterministic. Increasing temperature will result in more diverse completions. - When using Javascript, all API calls should be made on the server-side only - a single API request can only process up to 2,048 tokens (roughly 1,500 words) between your prompt and completion. - Price - 2 rupees for every 1000 words for the basic one. And if you want to build an advanced fine tuning one you can get it for 10 rupees - Embeddings - Embeddings is one way to provide context to an API question. The main issue with providing context is that there's only so much context that each question can take. What needs to be done is all the context is initially embedded. Following this the question is also embedded. A vector distance is calculated and the most useful strings are put in as context to the query. - You can also create a fine tuned model that allows you to upload data. You can provide a Json document with prompt and completion that the model will be additionally trained on. The more examples you give it the better. Each doubling leads to a linear increase in model quality.