Synthetic Data Generator

Use Language Models to Create Datasets for Specified Labels and Categories

Setup & Configure

Use Case

Labels

Use a colon to separate each label and its description as in 'label: description.'

Examples

Guiding Examples

Include all examples in this box. For each example, provide a LABEL, CATEGORY, TYPE, OUTPUT, and REASONING.

Example

Language Model

Visit Llama 3.2 and Gemma 3, and submit an access request for these gated Hugging Face repositories.

Model

Maximum Number of New Tokens

Hugging Face Token

Enter a 'Read' Hugging Face token to access gated language models, or a 'Write' token to push the generated data to Hugging Face.

Generate & Export

Status

Actions

Enter the number of rows for the generated dataset, and optionally a Hugging Face repo ID.

Sample Size

Save Reasoning

Hugging Face Repo ID (Optional)

Make Repository Public

Sample Output


We're happy to accommodate your dietary needs! Our halal options include a variety of delicious sandwiches and wraps. Would you like me to recommend some popular choices?	neutral	meta-llama/Meta-Llama-3.1-8B-Instruct	This text would be classified as "neutral." The sentence is purely informational, providing the facts about store hours without any emotional undertones. There are no phrases that express politeness or rudeness; it's a straightforward statement. The tone is impersonal and focused solely on conveying the necessary information.


We're happy to accommodate your dietary needs! Our halal options include a variety of delicious sandwiches and wraps. Would you like me to recommend some popular choices?	polite	meta-llama/Meta-Llama-3.1-8B-Instruct	This text is polite because it expresses appreciation for the customer's needs and offers assistance in a friendly manner. The use of phrases like "We're happy to accommodate" and "Would you like me to recommend" shows a welcoming tone and a willingness to help, making the customer feel valued and supported.
Your integration training session will be held on Friday at 2 PM in conference room A. Please review the materials beforehand to get the most out of the session.	neutral	meta-llama/Meta-Llama-3.1-8B-Instruct	This text is neutral because it provides factual information about the training session without any emotional undertones or attempts at politeness. The tone is impersonal and focused solely on conveying necessary details, such as the date, time, and location of the session.
Our tennis courts are open from 8am to 10pm. Please note that court reservations can be made online or through our mobile app.	neutral	meta-llama/Meta-Llama-3.1-8B-Instruct	This text would be classified as "neutral" because it provides factual information about the tennis courts' operating hours and reservation process without any emotional undertones or attempts at politeness. The tone is impersonal and focused solely on conveying the necessary information.
Our store hours for the health and beauty department are Monday through Saturday, 9 AM to 8 PM, and Sundays from 10 AM to 6 PM.	neutral	meta-llama/Meta-Llama-3.1-8B-Instruct	This text would be classified as "neutral." The sentence is purely informational, providing the facts about store hours without any emotional undertones. There are no phrases that express politeness or rudeness; it's a straightforward statement. The tone is impersonal and focused solely on conveying the necessary information.
Our current exhibit features works from local artists, and our knowledgeable staff will be happy to provide information about each piece.	neutral	meta-llama/Meta-Llama-3.1-8B-Instruct	This text is neutral because it provides factual information about the exhibit without emotional undertones or attempts at politeness. The statement is straightforward and informative, focusing on the content of the exhibit rather than the customer's experience or feelings.