# Examples

In the last section we learned the difference between a [field](/key-concepts/fields.md), its reference, and its value. 

Now let's talk about "examples."

Intuitively, examples are what you use to evaluate prompts. They are also what you use to fine-tune a model.

If you picture a typical spreadsheet where you have multiple columns and rows of data, each row of data is one example. Each cell in the row is a field value in Entry Point terminology.

An example includes values for both inputs and outputs.

Templates are used to format examples. That makes it very easy to test different formatting on your data *without modifying the actual data itself.* This is important because formatting can have an outsize impact on model performance. At Entry Point, we think of formatting as important as any other [fine-tuning hyperparameter](/guides/fine-tune-a-model/fine-tuning-hyperparameters.md).

Now let's put templates, fields, and examples together.

### Opposite Generator

Let's say we need a model that will generate a word that is the opposite of whatever word we enter.

We would create two fields of type `text`:

1. Input Word (reference: `input_word`)
2. Output Word (reference: `output_word`)

Then, we would write a template:

| Template Part    | Content                                                                                                        |
| ---------------- | -------------------------------------------------------------------------------------------------------------- |
| System Prompt    | Based on the following word, come up with a word that is most opposite to it. Only respond with a single word. |
| User Prompt      | {input\_word}                                                                                                  |
| Assistant Output | {output\_word}                                                                                                 |

Finally, we need examples, which contain the concrete datapoints to replace the references above.

We would write a few examples:

| Input Word | Output Word |
| ---------- | ----------- |
| hot        | cold        |
| near       | far         |
| wet        | dry         |
| high       | low         |

Each row above is one example.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.entrypointai.com/key-concepts/examples.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.