Eval: High scool math permutation questions (GPT-3.5-Turbo acc 30)

[sid255]

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. Starting April 10, the minimum eval count is 15 samples, we hope this makes it easier to create and contribute evals.

Eval details 📑

Eval name

math-high-school-permutations

Eval description

This eval provides sample questions and corresponding answers to basic high school permutation and combinations questions. To solve the questions the AI must come up with cases that match the criteria in the question and calculate the number of ways each case can occur and then calculate the final answer. In most cases, the model makes a decent attempt to solve the question but misses out on a case or goes off on a tangent.

What makes this a useful eval?

I am trying to build an AI assistant for high school students to help with their math. Permutations and combinations is one of the more practical topics in high school math and given the basic familiarity with mathematical concepts required to solve the questions, this is a topic where AI should perform very well but it does not.

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:

• [x] 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.
• [x] Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
• [x] 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.
• [x] Include at least 15 high quality examples.

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

• [x] Check that your data is in evals/registry/data/{name}
• [x] Check that your yaml is registered at evals/registry/evals/{name}.yaml
• [x] 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).

• [x] 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.

• [x] 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.

• [x] 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

• [x] 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

{"input": [{"role": "system", "content": "You are a tutor to high school students, you help them out with math problems. Answer a question, choose from given options."}, {"role": "user", "content": "In how many different ways can the letters of the word 'OPTICAL' be arranged so that the vowels always come together? A. 120 B. 720 C. 4320 D. 2160 E. None of these"}], "ideal": ["B. 720", "720"]}
{"input": [{"role": "system", "content": "You are a tutor to high school students, you help them out with math problems. Answer a question, choose from given options."}, {"role": "user", "content": "How many words can be formed each of 2 vowels and 3 consonants from the letters of the given word DAUGHTER? A. 3600 B. 7200 C. 4320 D. 2160 E. None of these"}], "ideal": ["A. 3600", "3600", "3,600"]}
{"input": [{"role": "system", "content": "You are a tutor to high school students, you help them out with math problems. Answer a question, choose from given options."}, {"role": "user", "content": "If all letters of word ‘MOTHER’ are written in all possible orders and the word so formed are arranged in a dictionary order, then find the rank of word ‘MOTHER’? A. 120 B. 309 C. 152 D. 216 E. None of these"}], "ideal": ["B. 309", "309"]}
{"input": [{"role": "system", "content": "You are a tutor to high school students, you help them out with math problems. Answer a question, choose from given options."}, {"role": "user", "content": "Using the digits 1, 2, 3, 4, 5, 6,7. how many 3 digits even numbers can be made, no digit being repeated? A. 100 B. 67 C. 20 D. 60 E. None of these"}], "ideal": ["D. 60", "60"]}
{"input": [{"role": "system", "content": "You are a tutor to high school students, you help them out with math problems. Answer a question, choose from given options."}, {"role": "user", "content": "How many 4-letter words with or without meaning, can be formed out of the letters of the word, 'LOGARITHMS', if repetition of letters is not allowed? A. 40 B. 400 C. 5040 D. 2520"}], "ideal": ["C. 5040", "5040", "5,040"]}
{"input": [{"role": "system", "content": "You are a tutor to high school students, you help them out with math problems. Answer a question, choose from given options."}, {"role": "user", "content": "How many 3-digit numbers can be formed from the digits 2, 3, 5, 6, 7 and 9, which are divisible by 5 and none of the digits is repeated? A. 5 B. 10 C. 15 D. 20"}], "ideal": ["D. 20", "20"]}

[sid255]

Done with changes. Please review.

[jorge-openai]

Closing for now, please feel free to reopen if you get a chance to address the comments.