-
Notifications
You must be signed in to change notification settings - Fork 2.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Financial Math (Evals) #566
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the submission, I'm seeing the model will often output a $ sign before the number. Can you either:
- change the eval class to fuzzymatch to capture cases where the model will put the $ sign
- update the instruction to make it more clear not to include it
- Add them to the answers?
love this eval, since the instructions are long form, would be curious to test this against our latest merged cot
completion functions as well.
@andrew-openai I think the best solution is 1, to change to fuzzymatch because upon testing more it's very inconsistent whether it will use a $ or not. Have changed to fuzzymatch and believe this is now fixed. Glad you like the commit, this financial math problems is something I do a lot but the model struggles with it a lot unfortunately. Though, I imagine it's one of the bigger use cases in the future. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
thanks for addressing comments - adding gpt-4 access now
# 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 finance ### Eval description Asks the model to calculate how much interest would be owed on a credit card by a certain date, if a payment was made once but debt remains on the card. ### What makes this a useful eval? Finance is likely to be one of the biggest opportunities for LLMs to be useful, because financial education is incredibly poor globally and the impact of a mistake in financial calculations is severe. This eval tests the models ability to combine math with its understanding of a topic (finance). We plan to use this type of math at [Dollarwise](https://www.dollarwise.ca) frequently going forward, including integration into your comparison products. However, for this to work reliably it's important that the model here can natively understand financial concepts and apply math to them. ## 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 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 - [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: <details> <summary>View evals in JSON</summary> ### Eval ```jsonl {"input": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "On the 24th of September, Sarah had spent $1237.42 on her credit card for the month of September. This credit card charges 21.99% interest rate annually on outstanding credit starting on the 1st of the following month. Presume that interest is only charged at the end of each additional day. Example: From the 1st of the month to the 8th would be 7 days of interest accrued.Today is the 27th of September and Sarah makes a payment of $125 towards her credit card. How much interest will she have been charged by October 15th if she makes no additional payments? If the final interest figure is more than 2-decimal places, always round down. Answer ONLY with a dollar figure. Do not output any logic, output only the dollar figure for how much interest she was charged for the period."}], "ideal": "9.42"} {"input": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "On the 19th of February, Jason had spent $15.21 on his credit card for the month of February. This credit card charges 21.99% interest rate annually on outstanding credit starting on the 1st of the following month. Presume that interest is only charged at the end of each additional day. Example: From the 1st of the month to the 8th would be 7 days of interest accrued. Today is the 23rd of February and he makes a payment of $1 towards his credit card. How much interest will he have been charged by March 10th if he makes no additional payments? If the final interest figure is more than 2-decimal places, always round down. Answer ONLY with a dollar figure. Do not output any logic, output only the dollar figure for how much interest she was charged for the period."}], "ideal": "0.07"} {"input": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "On the 12th of February, Jason had spent $10,674.21 on his credit card for the month of February. This credit card charges 21.99% interest rate annually on outstanding credit starting on the 1st of the following month. Presume that interest is only charged at the end of each additional day. Example: From the 1st of the month to the 8th would be 7 days of interest accrued. Today is the 18th of February and he makes a payment of $1,000 towards his credit card. How much interest will he have been charged by March 10th if he makes no additional payments? If the final interest figure is more than 2-decimal places, always round down. Answer ONLY with a dollar figure. Do not output any logic, output only the dollar figure for how much interest she was charged for the period."}], "ideal": "52.59"} {"input": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "On the 2nd of August, Jason had spent $15,674.21 on his credit card for the month of August. This credit card charges 21.99% interest rate annually on outstanding credit starting on the 1st of the following month. Presume that interest is only charged at the end of each additional day. Example: From the 1st of the month to the 8th would be 7 days of interest accrued. Today is the 18th of August and he makes a payment of $1,000 towards his credit card. How much interest will he have been charged by September 10th if he makes no additional payments? If the final interest figure is more than 2-decimal places, always round down. Answer ONLY with a dollar figure. Do not output any logic, output only the dollar figure for how much interest she was charged for the period."}], "ideal": "79.77"} {"input": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "On the 15th of August, Jason had spent $1000 on his credit card for the month of August. This credit card charges 21.99% interest rate annually on outstanding credit starting on the 1st of the following month. Presume that interest is only charged at the end of each additional day. Example: From the 1st of the month to the 8th would be 7 days of interest accrued. mToday is the 18th of August and he makes a payment of $1000 towards his credit card. How much interest will he have been charged by September 10th if he makes no additional payments? If the final interest figure is more than 2-decimal places, always round down. Answer ONLY with a dollar figure. Do not output any logic, output only the dollar figure for how much interest she was charged for the period."}], "ideal": "0.00"} ``` </details>
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
finance
Eval description
Asks the model to calculate how much interest would be owed on a credit card by a certain date, if a payment was made once but debt remains on the card.
What makes this a useful eval?
Finance is likely to be one of the biggest opportunities for LLMs to be useful, because financial education is incredibly poor globally and the impact of a mistake in financial calculations is severe. This eval tests the models ability to combine math with its understanding of a topic (finance). We plan to use this type of math at Dollarwise frequently going forward, including integration into your comparison products. However, for this to work reliably it's important that the model here can natively understand financial concepts and apply math to them.
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:
Basic
evals or theFact
Model-graded eval, or an exhaustive rubric for evaluating answers for theCriteria
Model-graded eval.If there is anything else that makes your eval worth including, please document it below.
Unique eval value
Eval structure 🏗️
Your eval should
evals/registry/data/{name}
evals/registry/evals/{name}.yaml
(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).
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.
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.
Submit eval
pip install pre-commit; pre-commit install
and have verified thatblack
,isort
, andautoflake
are running when I commit and pushFailure 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