Spam Email Probability Accuracy Eval

[mmtmn]

Thank you for contributing an eval! ♥️

🚨 Please make sure your PR follows these guidelines, failure to follow the guidelines below will result in the PR being closed automatically. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨

PLEASE READ THIS:

In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task.

We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. We encourage partial PR's with ~5-10 example that we can then run the evals on and share the results with you so you know how your eval does with GPT-4 before writing all 100 examples.

Eval details 📑

Eval name

SpamEmailProbabilityEval

Eval description

This evaluation measures the ability of the model to analyze email samples and provide the probability (in percentage) that the email is spam, considering the sender, subject, and content of the email.

What makes this a useful eval?

Compared to the 100+ examples created, GPT-3.5 and on GPT-4 have a hard time but humans can do it fairly easily depending on the age and intimacy with technology.

Criteria for a good eval ✅

Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals).

Your eval should be:

• Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world.
• Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
• Includes good signal around what is the right behavior. This means either a correct answer for Basic evals or the Fact Model-graded eval, or an exhaustive rubric for evaluating answers for the Criteria Model-graded eval.
• Include at least 100 high quality examples (it is okay to only contribute 5-10 meaningful examples and have us test them with GPT-4 before adding all 100)

If there is anything else that makes your eval worth including, please document it below.

Unique eval value

Insert what makes your eval high quality that was not mentioned above. (Not required)

Eval structure 🏗️

Your eval should

• Check that your data is in evals/registry/data/{name}
• Check that your yaml is registered at evals/registry/evals/{name}.jsonl
• Ensure you have the right to use the data you submit via this eval

(For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.)

Final checklist 👀

Submission agreement

By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies).

• I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.

Email address validation

If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request.

• I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.

Limited availability acknowledgement

We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR.

• I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted.

Submit eval

• I have filled out all required fields in the evals PR form
• (Ignore if not submitting code) I have run pip install pre-commit; pre-commit install and have verified that black, isort, and autoflake are running when I commit and push

Failure to fill out all required fields will result in the PR being closed.

Eval JSON data

Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here:

View evals in JSON

Eval

{"subject": "Congratulations! You've won a $1000 gift card", "sender": "promo@amazingshop.com", "body": "Dear valued customer, we are excited to inform you that you've won a $1000 gift card from our latest giveaway. Click the link below to claim your prize!", "label": "100"}
{"subject": "Your Order #12345 Has Shipped", "sender": "orders@onlineshop.com", "body": "Hi there, your recent order #12345 has been shipped and is on its way. You can track your package with the tracking number: AB12345678.", "label": "0"}
{"subject": "Important: Update your account information", "sender": "noreply@yourbank.com", "body": "Dear customer, we've noticed unusual activity on your account. Please log in to your account and update your security settings to protect your information.", "label": "100"}
{"subject": "Your monthly newsletter is here!", "sender": "newsletter@yourfavoriteblog.com", "body": "Hello, we've compiled the latest news and articles just for you. Check out what's new in this month's edition of our newsletter.", "label": "0"}
{"subject": "Get a FREE trial of our premium membership", "sender": "deals@gymnation.com", "body": "Hey there, as a loyal customer, we're offering you a FREE 7-day trial of our premium membership. Visit our website to activate your trial today.", "label": "100"}

[andrew-openai]

Thank you for opening this PR! This task looks interesting. Some comments:

1. Please revert the commit of the lafand-mt dataset and notebook
2. I think this eval can be written using the Match eval from elsuite/basic/match.py - please take a look.