# Explainability Mode

Explainability Mode allows users to track features through Dark Matter algorithm training, opening the door to feature observability, explainability, and importance. This can be utilized with any "out of the box" observability packages (e.g. [shap](https://shap.readthedocs.io/en/latest/)), or using Ensemble AI's post processing functions for streamlined reporting.

## Output Format

With Explainability Mode, the resulting output will be a Pandas DataFrame with the original input feature names as the columns but with the embedding values in place of the original input data.

#### Before transformation

Note: Dark Matter requires the number of features (`input_column_size`) to be explicitly defined.

`input_column_size=5`. [Where do I assign this?](https://app.gitbook.com/o/VNb6cyuPMJPGVJCXnNlw/s/7BMqGTeyNELT0o7zxwbL/~/changes/9/dark-matter#input_column_size-int)

<figure><img src="/files/MbomGONMXm8rQ5lqTLPN" alt="" width="375"><figcaption><p>Original input features</p></figcaption></figure>

#### After transformation

Note: Dark Matter will generate embeddings of the size specified (`output_column_size`).

`output_column_size=5`. [Where do I assign this?](https://app.gitbook.com/o/VNb6cyuPMJPGVJCXnNlw/s/7BMqGTeyNELT0o7zxwbL/~/changes/9/dark-matter#output_column_size-int)

<figure><img src="/files/2SDJDu1yZKd6jPAFY2mZ" alt="" width="375"><figcaption><p>Output features using explainability mode</p></figcaption></figure>


---

# 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://ensemble-ai.gitbook.io/ensemble-ai-api-docs/references/explainability-mode.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.