### Instruction Format for User Query

Source: https://github.com/pleias/baguettotron/blob/main/README.md

This XML-like format is used for user queries during inference. It clearly delineates the user's input and the start of the assistant's response.

```xml
<|im_start|>user
Who are you?<|im_end|>
<|im_start|>assistant
<think>
```

--------------------------------

### Instruction Format for User Query with Thinking Trace

Source: https://github.com/pleias/baguettotron/blob/main/README.md

This format includes a thinking trace within the assistant's response, indicated by the <think> tag. This is recommended for multi-turn conversations to maintain context.

```xml
<|im_start|>user
Who are you?<|im_end|>
<|im_start|>assistant
</think>
```

--------------------------------

### Instruction Format for RAG with References

Source: https://github.com/pleias/baguettotron/blob/main/README.md

This format is used for Retrieval Augmented Generation (RAG) tasks. It includes special syntax to pass references (<source_X>) to the model, enabling grounded answers.

```xml
<|im_start|>user
Who are you?

<source_1>[…]</source_1>
<source_2>[…]</source_2>
<|im_end|>
<|im_start|>assistant
<think>
```

--------------------------------

### Causality/Problem Decomposition Notation

Source: https://github.com/pleias/baguettotron/blob/main/README.md

This notation is used to represent causal relationships or problem decomposition. It is a visual representation of how different features or components relate to each other.

```text
Initial query:
├─ feature1: *lorem ipsum*
├─ feature2: *lorem ipsum*
└─ feature2: *lorem ipsum*
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.