ChatMLFormat

class torchtune.data.ChatMLFormat

OpenAI’s Chat Markup Language used by their chat models.

It is the default chat format used by HuggingFace models.

Example

<|im_start|>system
Provide some context and/or instructions to the model.<|im_end|>
<|im_start|>user
The user’s message goes here<|im_end|>
<|im_start|>assistant
The assistant’s response goes here<|im_end|>

classmethod format(sample: List[Message]) → List[Message]

Format user and system messages with appropriate tags.

Parameters:

sample (List[Message]) – a single conversation, structured as a list of Message objects

Returns:

The formatted list of messages