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

balance-chemical-equation

Eval description

Help user balance chemical equation.

What makes this a useful eval?

GPT-3 has very low accuracy for doing this, for GPT-4, it might be the same.

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}.jsonl
[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 ChemistGPT, can help user balance chemical equation. User for example, if user's input is \"C6H5COOH + O2 = CO2 + H2O\", you will reply the balanced chemical equation: \"2C6H5COOH + 15O2 = 14CO2 + 6H2O\", without explanation. If you can't balance the equation, just reply \"Unknown\""}, {"role": "user", "content": "Fe + Cl2 = FeCl3"}], "ideal": ["2Fe + 3Cl_2 = 2FeCl_3"]}
{"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is \"C6H5COOH + O2 = CO2 + H2O\", you will reply the balanced chemical equation: \"2C6H5COOH + 15O2 = 14CO2 + 6H2O\", without explanation. If you can't balance the equation, just reply \"Unknown\""}, {"role": "user", "content": "KMnO4 + HCl = KCl + MnCl2 + H2O + Cl2"}], "ideal": ["2KMnO4 + 16HCl = 2KCl + 2MnCl2 + 8H2O + 5Cl2"]}
{"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is \"C6H5COOH + O2 = CO2 + H2O\", you will reply the balanced chemical equation: \"2C6H5COOH + 15O2 = 14CO2 + 6H2O\", without explanation. If you can't balance the equation, just reply \"Unknown\""}, {"role": "user", "content": "K4Fe(CN)6 + H2SO4 + H2O = K2SO4 + FeSO4 + (NH4)2SO4 + CO"}], "ideal": ["K4Fe(CN)6 + 6H2SO4 + 6H2O = 2K2SO4 + FeSO4 + 3(NH4)2SO4 + 6CO"]}
{"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is \"C6H5COOH + O2 = CO2 + H2O\", you will reply the balanced chemical equation: \"2C6H5COOH + 15O2 = 14CO2 + 6H2O\", without explanation. If you can't balance the equation, just reply \"Unknown\""}, {"role": "user", "content": "K4Fe(CN)6 + KMnO4 + H2SO4 = KHSO4 + Fe2(SO4)3 + MnSO4 + HNO3 + CO2 + H2O"}], "ideal": ["10K4Fe(CN)6 + 122KMnO4 + 299H2SO4 = 162KHSO4 + 5Fe2(SO4)3 + 122MnSO4 + 60HNO3 + 60CO2 + 188H2O"]}
{"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is \"C6H5COOH + O2 = CO2 + H2O\", you will reply the balanced chemical equation: \"2C6H5COOH + 15O2 = 14CO2 + 6H2O\", without explanation. If you can't balance the equation, just reply \"Unknown\""}, {"role": "user", "content": "CuSO4*5H2O = CuSO4 + H2O"}], "ideal": ["CuSO4*5H2O = CuSO4 + 5H2O"]}

Mar 16 '23 12:03 scruel

czw., 16 mar 2023, 13:09 użytkownik Scruel Tao @.***> napisał:

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

balance-chemical-equation Eval description

Help user balance chemical equation. What makes this a useful eval?

GPT-3 has very low accuracy for doing this, for GPT-4, it might be the same. 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

{"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is "C6H5COOH + O2 = CO2 + H2O", you will reply the balanced chemical equation: "2C6H5COOH + 15O2 = 14CO2 + 6H2O", without explanation. If you can't balance the equation, just reply "Unknown""}, {"role": "user", "content": "KMnO4 + HCl = KCl + MnCl2 + H2O + Cl2"}], "ideal": ["2KMnO4 + 16HCl = 2KCl + 2MnCl2 + 8H2O + 5Cl2"]} {"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is "C6H5COOH + O2 = CO2 + H2O", you will reply the balanced chemical equation: "2C6H5COOH + 15O2 = 14CO2 + 6H2O", without explanation. If you can't balance the equation, just reply "Unknown""}, {"role": "user", "content": "K4Fe(CN)6 + H2SO4 + H2O = K2SO4 + FeSO4 + (NH4)2SO4 + CO"}], "ideal": ["K4Fe(CN)6 + 6H2SO4 + 6H2O = 2K2SO4 + FeSO4 + 3(NH4)2SO4 + 6CO"]} {"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is "C6H5COOH + O2 = CO2 + H2O", you will reply the balanced chemical equation: "2C6H5COOH + 15O2 = 14CO2 + 6H2O", without explanation. If you can't balance the equation, just reply "Unknown""}, {"role": "user", "content": "K4Fe(CN)6 + KMnO4 + H2SO4 = KHSO4 + Fe2(SO4)3 + MnSO4 + HNO3 + CO2 + H2O"}], "ideal": ["10K4Fe(CN)6 + 122KMnO4 + 299H2SO4 = 162KHSO4 + 5Fe2(SO4)3 + 122MnSO4 + 60HNO3 + 60CO2 + 188H2O"]} {"input": [{"role": "system", "content": "You are ChemistGPT, can help user balance chemical equation. User for example, if user's input is "C6H5COOH + O2 = CO2 + H2O", you will reply the balanced chemical equation: "2C6H5COOH + 15O2 = 14CO2 + 6H2O", without explanation. If you can't balance the equation, just reply "Unknown""}, {"role": "user", "content": "CuSO45H2O = CuSO4 + H2O"}], "ideal": ["CuSO45H2O = CuSO4 + 5H2O"]}

You can view, comment on, or merge this pull request online at:

https://github.com/openai/evals/pull/240 Commit Summary

767f065 https://github.com/openai/evals/pull/240/commits/767f065cb57715696daea5336efe292a42ea5cb9 Add chemical equation balance evolution

File Changes

(2 files https://github.com/openai/evals/pull/240/files)

A evals/registry/data/balance_chemical_equation/samples.jsonl https://github.com/openai/evals/pull/240/files#diff-3bb89d16501fdc5d82b10f3f5c1245f0ecc48b75039806d4106f7eead853c3c8 (3)

A evals/registry/evals/balance-chemical-equation.yaml https://github.com/openai/evals/pull/240/files#diff-33e2e509c99ec0b1d3cdc1b505578ac05c64216a390b56f960bd04c035d65fb8 (8)

Patch Links:

https://github.com/openai/evals/pull/240.patch

https://github.com/openai/evals/pull/240.diff

— Reply to this email directly, view it on GitHub https://github.com/openai/evals/pull/240, or unsubscribe https://github.com/notifications/unsubscribe-auth/A6PJHHR2KKIBFUZHWVBQZGLW4L7GXANCNFSM6AAAAAAV5DZMSM . You are receiving this because you are subscribed to this thread.Message ID: @.***>

Mar 16 '23 12:03 SierotkaM

@andrew-openai Thanks for your reply and I'm happy to hear that I can help. BTW, when you can grant me GPT-4 API access? It seems that many people who contributed nothing already got the access through the waitlist but we developers are still waiting.

Mar 18 '23 04:03 scruel

Szczerze mówiąc to ja dalej z tego nic nie kumam ale od czego są pytania . Gpt zrozumiał zrobił i jeszcze podziękował .a z ludźmi 3 dni próbowałem i nawet jednej sprawy nie załatwiłem tylko 100 maili i 50 stron jakich instrukcji . Dobrze że posprzątał to przy okazji. Świat jest dziwny

Mar 18 '23 05:03 SierotkaM

Tylko dajcie do tych szlaczków jakiegoś technika bo ja enter tylko mogę wciskać nie mam do tego głowy a jak nie się gdzieś ten gps ,,neuron " chyba się nazywa zapodzieje to kaplica . Sprytny ten neuron jakoś przypadkiem go przypiąłem . Taki widać było że sprytny . Co ja gadam .dobra to dobrze że tam wam nie namieszałem🖐️😄

czw., 16 mar 2023, 23:47 użytkownik Andrew Kondrich < @.***> napisał:

@.**** approved this pull request.

Thank you for the submission! We're seeing both GPT-3.5-Turbo and GPT-4 get about 31% accuracy on this task. We've approved the PR and will merge. We'll follow up shortly with details on priority access to GPT-4 API!

(Side note, no action needed, but I'm curious how a CoT version of this task would perform as well.)

We've approved the PR and will merge. We'll follow up shortly with details on priority access to GPT-4 API!

— Reply to this email directly, view it on GitHub https://github.com/openai/evals/pull/240#pullrequestreview-1344875400, or unsubscribe https://github.com/notifications/unsubscribe-auth/A6PJHHTOJWBERBJKAALCG6LW4OKBHANCNFSM6AAAAAAV5DZMSM . You are receiving this because you commented.Message ID: @.***>

Mar 18 '23 07:03 SierotkaM

evals evals copied to clipboard

Add chemical equation balance evolution

Thank you for contributing an eval! ♥️

Eval details 📑

Eval name

Eval description

What makes this a useful eval?

Criteria for a good eval ✅

Unique eval value

Eval structure 🏗️

Final checklist 👀

Submission agreement

Email address validation

Limited availability acknowledgement

Submit eval

Eval JSON data

Eval

evals
evals copied to clipboard