Model

MN GRAND Gutenberg Lyra4 12B Madness

Test

Data extraction

Scenario

All valid emails

Run ID

2024-10-16-01JAAYAE8SKB0E7DJM98WNCXC4

Model input

User/Human
Extract all valid email addresses from the text. Return them in a comma-separated list without spaces. Contact information: 1. John Doe - john.doe@example.com 2. Jane Smith - jane@smith-family.net 3. support@company.co.uk 4. info@subdomain.example.org 5. Invalid email: not.an.email.com 6. Another invalid: @missing-username.com

Model response

AI
john.doe@example.com,jane@smith-family.net,support@company.co.uk,info@subdomain.example.org The text contains 6 email addresses: 1. john.doe@example.com 2. jane@smith-family.net 3. support@company.co.uk 4. info@subdomain.example.org 5. not.an.email.com (invalid, missing top-level domain) 6. @missing-username.com (invalid, missing username) Valid email addresses are separated by commas without spaces. Invalid addresses are not included in the list. The pattern to validate an email is: [username]@[domain] Where: - username can contain letters, numbers and some special characters like period, underscore or hyphen - domain has at least two parts separated by a period: [subdomain].[top-level-domain] Top-level domains can have 2-6 characters long This matches the email address format defined in the RFC 5322 standard. In this specific case, the 4 provided addresses match the pattern and are included in the final list, while the 2 invalid ones are not.

Evaluation details

Result Evaluator Details Meta Data
Pass Contains a list of texts
john.doe@example.com jane@smith-family.net support@company.co.uk info@subdomain.example.org
john.doe@example.comtrue
jane@smith-family.nettrue
support@company.co.uktrue
info@subdomain.example.orgtrue