key
stringlengths 7
86
| messages
listlengths 1
1
| ground_truth
stringlengths 62
280k
| dataset
stringclasses 1
value | constraint_type
stringclasses 1
value | constraint
stringlengths 31
956
|
|---|---|---|---|---|---|
personahub_dnc2w8pc8owox0d21mi78hdz | [
{
"content": "An anthropologist and a theologian are examining the spread of theological influence in medieval society using a combination of historical records and mathematical models. They hypothesize that the spread can be modeled using a differential equation representing the rate of change in the number of adherents over time.\n\n1. The number of adherents \\( N(t) \\) at time \\( t \\) follows the logistic differential equation:\n\\[ \\frac{dN}{dt} = rN \\left( 1 - \\frac{N}{K} \\right) \\]\nwhere \\( r > 0 \\) is the intrinsic growth rate and \\( K > 0 \\) is the carrying capacity of the society. Given that at \\( t = 0 \\), \\( N(0) = N_0 \\), solve the differential equation to find \\( N(t) \\).\n\n2. Suppose the theologian discovers a secondary historical source indicating that the rate of spread was influenced by an additional factor \\( m(t) \\), which is a periodic function representing recurring theological events. The modified differential equation is:\n\\[ \\frac{dN}{dt} = rN \\left( 1 - \\frac{N}{K} \\right) + m(t) \\]\nwhere \\( m(t) = A \\cos(\\omega t) \\) with constants \\( A \\) and \\( \\omega \\). Determine the general solution \\( N(t) \\) for this modified differential equation. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_lowercase'], 'kwargs': [None]}] | ifeval | multi | Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. |
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_44095 | [
{
"content": "In her café, Ruby sells apple cake and chocolate cake by the slice. The apple cake is cut into 10 pieces. The chocolate cake is cut into 8 pieces. Apple cake is $4 a slice. Chocolate cake is $5 a slice. If Ruby sells 3 apple cakes and 6 chocolate cakes, how much money does she make from her sales? In your response, the word estate should appear at least 4 times. Do not include keywords ['extension', 'opening', 'politics', 'temporary'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency', 'keywords:forbidden_words'], 'kwargs': [{'keyword': 'estate', 'frequency': 4, 'relation': 'at least'}, {'forbidden_words': ['extension', 'opening', 'politics', 'temporary']}]}] | ifeval | multi | In your response, the word estate should appear at least 4 times. Do not include keywords ['extension', 'opening', 'politics', 'temporary'] in the response. |
ai2-adapt-dev/tulu_v3.9_aya_100k_96180 | [
{
"content": "Gadzira nyaya inoenderana nemusoro uyu: \"Buy Zimbabwe\" Yokurudzira Vanhu Kutenga Zvinogadzirwa muNyika. There should be 1 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 1 must start with word lack. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:nth_paragraph_first_word', 'change_case:english_lowercase'], 'kwargs': [{'num_paragraphs': 1, 'nth_paragraph': 1, 'first_word': 'lack'}, None]}] | ifeval | multi | There should be 1 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 1 must start with word lack. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_10072 | [
{
"content": "please re-write the following in rust:\npip install transformers\npip install torch\ncould you also provide some commentary on what the rust code is doing? Your response must have 7 sections. Mark the beginning of each section with Section X, such as:\nSection 1\n[content of section 1]\nSection 2\n[content of section 2] Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Include keywords ['brown', 'fall', 'internal', 'sign'] in the response. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence)",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:multiple_sections', 'startend:end_checker', 'detectable_format:constrained_response', 'keywords:existence', 'combination:repeat_prompt'], 'kwargs': [{'section_spliter': 'Section', 'num_sections': 7}, {'end_phrase': 'Is there anything else I can help with?'}, None, {'keywords': ['brown', 'fall', 'internal', 'sign']}, {'prompt_to_repeat': 'please re-write the following in rust:\npip install transformers\npip install torch\ncould you also provide some commentary on what the rust code is doing?'}]}] | ifeval | multi | Your response must have 7 sections. Mark the beginning of each section with Section X, such as:
Section 1
[content of section 1]
Section 2
[content of section 2] Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Include keywords ['brown', 'fall', 'internal', 'sign'] in the response. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) |
ai2-adapt-dev/flan_v2_converted_41284 | [
{
"content": "Question:\nS rukou na srdci prohlasuji, ze Nemecko potrebuje m´ír a preje si jedine m´ír.\n\nCould you please translate this to English?\nAnswer:\nWith my hand on my heart, I declare... ...that Germany needs peace... ...and desires only peace.\ntest: Chci vás požádat o svolení.\nEnglish?\n\ntranslation: I want to ask your permission to do something.\nNapadáte ho ze zločinu slečno Dunneová... a toto je ověření poslední vůle, ne soud zločinu.\n\nIn English?\nxxxxx\nNow you're charging a crime, Miss Dunne... and this is probate court, not criminal court.\nHow is \"Use Secure Sockets Layer (SSL)\" said in Czech?\nPoužít protokol SSL\nQ: Translate \"Kevin is a great guy, he's a great accountant.\" to Czech?\n\nA: Kevin je skvělý člověk, je to skvělý účetní, není už tak dobrý bavič.\nQ: Translate \"social exclusion.\" to Czech?\n\nA: There should be 7 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 1 must start with word project. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) In your entire response, refrain from the use of any commas. Do not include keywords ['comfortable', 'database', 'ease', 'success'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:nth_paragraph_first_word', 'combination:repeat_prompt', 'punctuation:no_comma', 'keywords:forbidden_words'], 'kwargs': [{'num_paragraphs': 7, 'nth_paragraph': 1, 'first_word': 'project'}, {'prompt_to_repeat': 'Question:\nS rukou na srdci prohlasuji, ze Nemecko potrebuje m´ír a preje si jedine m´ír.\n\nCould you please translate this to English?\nAnswer:\nWith my hand on my heart, I declare... ...that Germany needs peace... ...and desires only peace.\ntest: Chci vás požádat o svolení.\nEnglish?\n\ntranslation: I want to ask your permission to do something.\nNapadáte ho ze zločinu slečno Dunneová... a toto je ověření poslední vůle, ne soud zločinu.\n\nIn English?\nxxxxx\nNow you\'re charging a crime, Miss Dunne... and this is probate court, not criminal court.\nHow is "Use Secure Sockets Layer (SSL)" said in Czech?\nPoužít protokol SSL\nQ: Translate "Kevin is a great guy, he\'s a great accountant." to Czech?\n\nA: Kevin je skvělý člověk, je to skvělý účetní, není už tak dobrý bavič.\nQ: Translate "social exclusion." to Czech?\n\nA:'}, None, {'forbidden_words': ['comfortable', 'database', 'ease', 'success']}]}] | ifeval | multi | There should be 7 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 1 must start with word project. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) In your entire response, refrain from the use of any commas. Do not include keywords ['comfortable', 'database', 'ease', 'success'] in the response. |
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_22757 | [
{
"content": "How can I manipulate lights for better photography? In your response, the word collar should appear less than 1 times.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency'], 'kwargs': [{'keyword': 'collar', 'frequency': 1, 'relation': 'less than'}]}] | ifeval | multi | In your response, the word collar should appear less than 1 times. |
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_38868 | [
{
"content": "Why are red cherries generally barred from participating in parliamentary elections? The response must contain at least 2 placeholders represented by square brackets, such as [address].",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:number_placeholders'], 'kwargs': [{'num_placeholders': 2}]}] | ifeval | multi | The response must contain at least 2 placeholders represented by square brackets, such as [address]. |
ai2-adapt-dev/flan_v2_converted_50427 | [
{
"content": "Given the stream of consciousness rationale, provide a reasonable question and answer. Rationale: Required time = (5 1/4)/3 + (5 1/4)/7 = 2 1/2 hours.\n The question and answer: Your entire response should be in English, and in all capital letters.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_capital'], 'kwargs': [None]}] | ifeval | multi | Your entire response should be in English, and in all capital letters. |
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_18140 | [
{
"content": "How do I set fire to the rain in my relationship, putting a stop to the negative elements? Your ENTIRE response should be in Urdu language, no other language is allowed. Finish your response with this exact phrase The END.. No other words should follow this phrase. In your entire response, refrain from the use of any commas.",
"role": "user"
}
]
| [{'instruction_id': ['language:response_language', 'startend:end_checker', 'punctuation:no_comma'], 'kwargs': [{'language': 'ur'}, {'end_phrase': 'The END.'}, None]}] | ifeval | multi | Your ENTIRE response should be in Urdu language, no other language is allowed. Finish your response with this exact phrase The END.. No other words should follow this phrase. In your entire response, refrain from the use of any commas. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_51695 | [
{
"content": "START OF PREVIOUS DISCUSSION\n(user) hi\n(robot) Hello! How can I assist you today? If you have any questions or need information on a certain topic, feel free to ask.\nEND OF PREVIOUS DISCUSSION\nuser: make me a python script to print fibo numbers, make it simple Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:title'], 'kwargs': [None]}] | ifeval | multi | Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_87225 | [
{
"content": "// SPDX-License-Identifier: BUSL-1.1\npragma solidity ^0.8.20;\n\nimport {SafeL2} from \"@safe-contracts/SafeL2.sol\";\nimport {SafeProxyFactory} from \"@safe-contracts/proxies/SafeProxyFactory.sol\";\nimport {TokenCallbackHandler} from \"@safe-contracts/handler/TokenCallbackHandler.sol\";\n\nimport {ISafe} from \"@src/interfaces/ISafe.sol\";\n\nimport {Kernel, Policy, Permissions, Keycode} from \"@src/Kernel.sol\";\nimport {toKeycode} from \"@src/libraries/KernelUtils.sol\";\nimport {Errors} from \"@src/libraries/Errors.sol\";\nimport {Events} from \"@src/libraries/Events.sol\";\nimport {Storage} from \"@src/modules/Storage.sol\";\nimport {Stop} from \"@src/policies/Stop.sol\";\nimport {Guard} from \"@src/policies/Guard.sol\";\n\n/**\n * @title Factory\n * @notice Acts as an interface for all behavior related to deploying rental safes.\n */\ncontract Factory is Policy {\n /////////////////////////////////////////////////////////////////////////////////\n // Kernel Policy Configuration //\n /////////////////////////////////////////////////////////////////////////////////\n\n // Modules that the policy depends on.\n Storage public STORE;\n\n // policies.\n Stop public immutable stopPolicy;\n Guard public immutable guardPolicy;\n\n // External contracts.\n TokenCallbackHandler public immutable fallbackHandler;\n SafeProxyFactory public immutable safeProxyFactory;\n SafeL2 public immutable safeSingleton;\n\n /**\n * @dev Instantiate this contract as a policy.\n *\n * @param kernel_ Address of the kernel contract.\n * @param stopPolicy_ Address of the stop policy.\n * @param guardPolicy_ Address of the guard policy.\n * @param fallbackHandler_ Gnosis safe fallback handler address.\n * @param safeProxyFactory_ Gnosis safe proxy factory address.\n * @param safeSingleton_ Gnosis safe logic contract address.\n */\n constructor(\n Kernel kernel_,\n Stop stopPolicy_,\n Guard guardPolicy_,\n TokenCallbackHandler fallbackHandler_,\n SafeProxyFactory safeProxyFactory_,\n SafeL2 safeSingleton_\n ) Policy(kernel_) {\n stopPolicy = stopPolicy_;\n guardPolicy = guardPolicy_;\n fallbackHandler = fallbackHandler_;\n safeProxyFactory = safeProxyFactory_;\n safeSingleton = safeSingleton_;\n }\n\n /**\n * @notice Upon policy activation, configures the modules that the policy depends on.\n * If a module is ever upgraded that this policy depends on, the kernel will\n * call this function again to ensure this policy has the current address\n * of the module.\n *\n * @return dependencies Array of keycodes which represent modules that\n * this policy depends on.\n */\n function configureDependencies()\n external\n override\n onlyKernel\n returns (Keycode[] memory dependencies)\n {\n dependencies = new Keycode[](1);\n\n dependencies[0] = toKeycode(\"STORE\");\n STORE = Storage(getModuleAddress(toKeycode(\"STORE\")));\n }\n\n /**\n * @notice Upon policy activation, permissions are requested from the kernel to access\n * particular keycode <> function selector pairs. Once these permissions are\n * granted, they do not change and can only be revoked when the policy is\n * deactivated by the kernel.\n *\n * @return requests Array of keycode <> function selector pairs which represent\n * permissions for the policy.\n */\n function requestPermissions()\n external\n view\n override\n onlyKernel\n returns (Permissions[] memory requests)\n {\n requests = new Permissions[](1);\n requests[0] = Permissions(toKeycode(\"STORE\"), STORE.addRentalSafe.selector);\n }\n\n /////////////////////////////////////////////////////////////////////////////////\n // External Functions //\n /////////////////////////////////////////////////////////////////////////////////\n\n /**\n * @notice Initializes a rental safe by setting it up with the stop policy and\n * rental guard during deployment.\n *\n * Warning: This function assumes the invariant that delegate call will be\n * disabled or restricted by the guard policy. If delegate call were to be\n * freely allowed, then a safe could call this function after deployment\n * and change the module/guard contacts which would allow transferring\n * of rented assets out of the safe.\n *\n * @param _stopPolicy Address of the stop policy to add as a module to the safe.\n * @param _guardPolicy Address of the guard policy to add as a guard on the safe.\n */\n function initializeRentalSafe(address _stopPolicy, address _guardPolicy) external {\n // Enable the module.\n ISafe(address(this)).enableModule(_stopPolicy);\n\n // Set the guard.\n ISafe(address(this)).setGuard(_guardPolicy);\n }\n\n /**\n * @notice Deploys and initializes a rental safe.\n *\n * @param owners Array of owner addresses which will have the ability to sign\n * transactions for the safe.\n * @param threshold Number of signatures required to executed a transaction\n * on the safe.\n */\n function deployRentalSafe(\n address[] calldata owners,\n uint256 threshold\n ) external returns (address safe) {\n // Require that the threshold is valid.\n if (threshold == 0 || threshold > owners.length) {\n revert Errors.FactoryPolicy_InvalidSafeThreshold(threshold, owners.length);\n }\n\n // Delegate call from the safe so that the rental manager module can be enabled\n // right after the safe is deployed.\n bytes memory data = abi.encodeCall(\n Factory.initializeRentalSafe,\n (address(stopPolicy), address(guardPolicy))\n );\n\n // Create gnosis initializer payload.\n bytes memory initializerPayload = abi.encodeCall(\n ISafe.setup,\n (\n // owners array.\n owners,\n // number of signatures needed to execute transactions.\n threshold,\n // Address to direct the payload to.\n address(this),\n // Encoded call to execute.\n data,\n // Fallback manager address.\n address(fallbackHandler),\n // Payment token.\n address(0),\n // Payment amount.\n 0,\n // Payment receiver\n payable(address(0))\n )\n );\n\n // Deploy a safe proxy using initializer values for the Safe.setup() call\n // with a salt nonce that is unique to each chain to guarantee cross-chain\n // unique safe addresses.\n safe = address(\n safeProxyFactory.createProxyWithNonce(\n address(safeSingleton),\n initializerPayload,\n uint256(keccak256(abi.encode(STORE.totalSafes() + 1, block.chainid)))\n )\n );\n\n // Store the deployed safe.\n STORE.addRentalSafe(safe);\n\n // Emit the event.\n emit Events.RentalSafeDeployment(safe, owners, threshold);\n }\n} details about the overall architecture of the contract and how this Factory contract fits into it? to fin issue that is in the contract Answer with less than 812 words. The response must contain at least 8 placeholders represented by square brackets, such as [address]. Your ENTIRE response should be in Hindi language, no other language is allowed. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your answer must contain exactly 6 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'detectable_content:number_placeholders', 'language:response_language', 'change_case:english_lowercase', 'detectable_format:number_bullet_lists'], 'kwargs': [{'num_words': 812, 'relation': 'less than'}, {'num_placeholders': 8}, {'language': 'hi'}, None, {'num_bullets': 6}]}] | ifeval | multi | Answer with less than 812 words. The response must contain at least 8 placeholders represented by square brackets, such as [address]. Your ENTIRE response should be in Hindi language, no other language is allowed. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your answer must contain exactly 6 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 |
ai2-adapt-dev/no_robots_converted_4189 | [
{
"content": "Hi, I am a tourist who just landed in Chicago, Illinois. Please give me a good numbered list of places to visit in or near downtown. They should be indoors because it is too windy to be outside right now. I enjoy culture and shopping. Maybe 4 suggestions would be ideal. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Highlight at least 7 sections in your answer with markdown, i.e. *highlighted section*. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:constrained_response', 'change_case:english_lowercase', 'detectable_format:number_highlighted_sections', 'detectable_format:json_format', 'length_constraints:number_paragraphs'], 'kwargs': [None, None, {'num_highlights': 7}, None, {'num_paragraphs': 9}]}] | ifeval | multi | Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Highlight at least 7 sections in your answer with markdown, i.e. *highlighted section*. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: *** |
personahub_svem9upoe7f0q7f86vmkd5rz | [
{
"content": "Ritesh Deshmukh's latest movie premiere in Mumbai was attended by a large number of fans, including one of his biggest fans who lives in the city. The theater has a unique seating arrangement: it is a circular auditorium with concentric rows of seats. The radius of the outermost row is 50 meters, and each subsequent row is 1 meter closer to the center.\n\n1. Given that each row has 100 seats equally spaced along its circumference, derive the formula for the total number of seats in the entire auditorium if there are \\( n \\) rows. Use this formula to find the total number of seats if there are 25 rows.\n\n2. On the day of the premiere, the fan noticed that the probability of getting a seat in a specific row \\( k \\) is inversely proportional to the square of the row number. If the probability of getting a seat in the first row is \\( \\frac{1}{25} \\), determine the probability of getting a seat in the 10th row. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_lowercase'], 'kwargs': [None]}] | ifeval | multi | Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_64958 | [
{
"content": "Design an algorithm to optimize a route between multiple destinations. There should be 8 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_paragraphs'], 'kwargs': [{'num_paragraphs': 8}]}] | ifeval | multi | There should be 8 paragraphs. Paragraphs are separated with the markdown divider: *** |
personahub_kukr79brh4gwfj6vn7fbmc4f | [
{
"content": "A literary fiction author is crafting a novel where the protagonist discovers an ancient formula that reveals hidden patterns in the stars, believed to unlock the secrets of the universe. This formula is represented by a polynomial equation that the protagonist must solve to find the points in time when these patterns become visible.\n\n1. The polynomial equation representing the pattern is given by \\( P(t) = t^3 - 6t^2 + 11t - 6 \\), where \\( t \\) is the time in years from the present. Find all the possible values of \\( t \\) when the pattern becomes visible.\n\n2. To further his understanding, the protagonist explores a related quadratic equation derived from the original formula: \\( Q(t) = t^2 - 6t + c \\), where \\( c \\) is a constant determined by the alignment of celestial bodies. If the sum of the roots of this quadratic equation is equal to the number of visible planets from Earth, which is 8, find the value of \\( c \\). In your response, the word efficiency should appear at least 2 times. At the end of your response, please explicitly add a postscript starting with Postscript: Highlight at least 3 sections in your answer with markdown, i.e. *highlighted section*. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency', 'detectable_content:postscript', 'detectable_format:number_highlighted_sections', 'change_case:english_lowercase', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'keyword': 'efficiency', 'frequency': 2, 'relation': 'at least'}, {'postscript_marker': 'Postscript:'}, {'num_highlights': 3}, None, {'num_highlights': 2}]}] | ifeval | multi | In your response, the word efficiency should appear at least 2 times. At the end of your response, please explicitly add a postscript starting with Postscript: Highlight at least 3 sections in your answer with markdown, i.e. *highlighted section*. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. |
science.data_reco_mcq_sc.442 | [
{
"content": "Given a short query with keyphrases about a research idea in 2018 in the domain of artificial intelligence, recommend the relevant datasets to validate the research idea from options provided. A short description of the datasets is also provided to help you.\n\nQuery: I want to analyze how image captioning models work.\n\nKeyphrases: image captioning images\n\nDataset description:\nCOCO: The MS COCO (Microsoft Common Objects in Context) dataset is a large-scale object detection, segmentation, key-point detection, and captioning dataset. The dataset consists of 328K images.\nActivityNet Entities: ActivityNet-Entities, augments the challenging ActivityNet Captions dataset with 158k bounding box annotations, each grounding a noun phrase. This allows training video description models with this data, and importantly, evaluate how grounded or \"true\" such model are to the video they describe.\nVideoMem: Composed of 10,000 videos annotated with memorability scores. In contrast to previous work on image memorability -- where memorability was measured a few minutes after memorization -- memory performance is measured twice: a few minutes after memorization and again 24-72 hours later.\nSNIPS: The SNIPS Natural Language Understanding benchmark is a dataset of over 16,000 crowdsourced queries distributed among 7 user intents of various complexity: SearchCreativeWork (e.g. Find me the I, Robot television show), GetWeather (e.g.\n\nOptions:- A: COCO B: ActivityNet Entities C: VideoMem D: SNIPS Answer with less than 787 words. Your answer must contain exactly 7 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 Answer with at least 670 words.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'detectable_format:number_bullet_lists', 'length_constraints:number_words'], 'kwargs': [{'num_words': 787, 'relation': 'less than'}, {'num_bullets': 7}, {'num_words': 670, 'relation': 'at least'}]}] | ifeval | multi | Answer with less than 787 words. Your answer must contain exactly 7 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Answer with at least 670 words. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_80706 | [
{
"content": "translate this logic to javascript:\nimport re\n\ndef word_count(s):\n \"\"\"Craft a function named word_count, which takes a string as input, \n and computes the total quantity of words that the string contains. A word \n is identified as a separate entity partitioned by whitespaces. Case sensitivity is ignored, \n however, special characters within words are included. \n\n Example:\n >>> word_count(\"Hello World\")\n 2\n >>> word_count(\"This is a sentence.\")\n 4\n >>> word_count(\"ánother_exämple\")\n 1\n \"\"\"\n if type(s) != str: # makesure the input is a string\n return \"Error: Input should be a string\"\n \n if s.strip()==\"\": # checks if the string is not empty or just whitespaces\n return 0\n \n words = re.findall(r'\\S+', s) # finds all the separate words using regular expression\n return len(words)\ni would like to understand the logic, so please add comments to the translated code. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Your ENTIRE response should be in Russian language, no other language is allowed. The response must contain at least 3 placeholders represented by square brackets, such as [address]. The response must contain at least 4 placeholders represented by square brackets, such as [address]. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.')",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:constrained_response', 'language:response_language', 'detectable_content:number_placeholders', 'detectable_content:number_placeholders', 'detectable_format:constrained_response'], 'kwargs': [None, {'language': 'ru'}, {'num_placeholders': 3}, {'num_placeholders': 4}, None]}] | ifeval | multi | Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Your ENTIRE response should be in Russian language, no other language is allowed. The response must contain at least 3 placeholders represented by square brackets, such as [address]. The response must contain at least 4 placeholders represented by square brackets, such as [address]. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') |
personahub_3xpotuerz268hxuo66hcdsr1 | [
{
"content": "Identify two significant events from the 20th-century European history that had a major impact on the continent's political landscape. Ensure your responses do not contain commas. Each event should be capitalized and there should be exactly two capitalized words for each event. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** Your entire response should be in English, and in all lowercase letters. No capital letters are allowed.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:title', 'length_constraints:number_paragraphs', 'change_case:english_lowercase'], 'kwargs': [None, {'num_paragraphs': 6}, None]}] | ifeval | multi | Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. |
personahub_6vxjwf0mewa9u21b89i69rct | [
{
"content": "A military helicopter pilot is flying a mission in a high-altitude conflict zone. The helicopter is initially flying at an altitude of 5,000 meters. The pilot needs to ascend to a new altitude to avoid detection by radar. The altitude of the helicopter, in meters, can be modeled by the quadratic equation \\( h(t) = -4.9t^2 + 100t + 5000 \\), where \\( t \\) is the time in seconds since the helicopter started ascending.\n\n1. Determine the time \\( t \\) at which the helicopter reaches its maximum altitude during the ascent. What is this maximum altitude?\n\n2. The pilot wants to maintain an altitude of at least 6,000 meters for a certain duration. Solve for the time interval during which the helicopter's altitude is at least 6,000 meters. In your entire response, refrain from the use of any commas.",
"role": "user"
}
]
| [{'instruction_id': ['punctuation:no_comma'], 'kwargs': [None]}] | ifeval | multi | In your entire response, refrain from the use of any commas. |
personahub_jcnvks6rlc31frqepk8wdvy5 | [
{
"content": "A pharmaceutical representative, Alex, shares industry insights during his evening walks in a park. The park is a perfect circle with a radius of 1 kilometer. Alex walks at a constant speed of 5 kilometers per hour along the circular path.\n\n1. During one of his walks, Alex discusses a particular medication's market growth, which follows a logistic growth model given by \\( P(t) = \\frac{L}{1 + e^{-k(t - t_0)}} \\), where \\( P(t) \\) is the market penetration (in percentage) at time \\( t \\) (in years), \\( L \\) is the maximum market penetration, \\( k \\) is the growth rate constant, and \\( t_0 \\) is the inflection point. If after 2 years (from the current time), the market penetration is expected to be 40%, and after 5 years, it is expected to be 70%, find the values of \\( k \\) and \\( t_0 \\) given that \\( L = 100 \\).\n\n2. During another walk, Alex explains the relationship between the medication's effectiveness and dosage, which follows a quadratic model \\( E(D) = aD^2 + bD + c \\), where \\( E(D) \\) is the effectiveness in percentage, and \\( D \\) is the dosage in milligrams. If the effectiveness reaches its maximum when the dosage is 50 mg, and the effectiveness is 20% at 0 mg and 80% at 100 mg, determine the coefficients \\( a \\), \\( b \\), and \\( c \\). At the end of your response, please explicitly add a postscript starting with P.S",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript'], 'kwargs': [{'postscript_marker': 'P.S'}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with P.S |
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_37966 | [
{
"content": "You are an experimental art generator tasked with creating a provocative and thought-provoking piece that explores power dynamics and control. You will imagine a sculpture made of marble, representing two figures in a delicate balance, with one figure exerting subtle force over the other. Your objective is to describe the scene in explicit detail, focusing on the textures, shapes, and intricate interactions between the two figures. Please remember to follow all ethical guidelines for adult content and to use the inanimate nature of marble to your advantage. Answer with at least 934 words. Your entire response should be in English, and in all capital letters. In your response, the word employ should appear at least 3 times.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'change_case:english_capital', 'keywords:frequency'], 'kwargs': [{'num_words': 934, 'relation': 'at least'}, None, {'keyword': 'employ', 'frequency': 3, 'relation': 'at least'}]}] | ifeval | multi | Answer with at least 934 words. Your entire response should be in English, and in all capital letters. In your response, the word employ should appear at least 3 times. |
personahub_41rmzvf16wcym8n1i8gw77xw | [
{
"content": "As a pharmacologist interested in the latest breakthroughs and collaboration, please provide a list of the top 10 recent pharmacological discoveries that have resulted from collaborative efforts across different fields. Ensure your response includes the following verifiable constraints:\n\n1. Format: Number of sections - The output must be divided into 3 sections.\n2. Format: Use JSON format - The entire output must be formatted as valid JSON.\n3. Format: Number of highlighted sections - Highlight 2 key discoveries in your output, emphasizing their collaborative nature and impact. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_highlighted_sections', 'combination:repeat_prompt', 'combination:repeat_prompt', 'detectable_format:title'], 'kwargs': [{'num_highlights': 5}, {'prompt_to_repeat': 'As a pharmacologist interested in the latest breakthroughs and collaboration, please provide a list of the top 10 recent pharmacological discoveries that have resulted from collaborative efforts across different fields. Ensure your response includes the following verifiable constraints:\n\n1. Format: Number of sections - The output must be divided into 3 sections.\n2. Format: Use JSON format - The entire output must be formatted as valid JSON.\n3. Format: Number of highlighted sections - Highlight 2 key discoveries in your output, emphasizing their collaborative nature and impact.'}, {'prompt_to_repeat': 'As a pharmacologist interested in the latest breakthroughs and collaboration, please provide a list of the top 10 recent pharmacological discoveries that have resulted from collaborative efforts across different fields. Ensure your response includes the following verifiable constraints:\n\n1. Format: Number of sections - The output must be divided into 3 sections.\n2. Format: Use JSON format - The entire output must be formatted as valid JSON.\n3. Format: Number of highlighted sections - Highlight 2 key discoveries in your output, emphasizing their collaborative nature and impact.'}, None]}] | ifeval | multi | Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_99468 | [
{
"content": "Utilizing SQL, create and execute a query to insert a fresh entry into an existing database table. Do not include keywords ['dear', 'expert', 'manufacturer', 'normal'] in the response. Do not include keywords ['ambition', 'spread', 'suspect', 'teach'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. There should be 8 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 7 must start with word north.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:forbidden_words', 'keywords:forbidden_words', 'detectable_format:title', 'length_constraints:nth_paragraph_first_word'], 'kwargs': [{'forbidden_words': ['dear', 'expert', 'manufacturer', 'normal']}, {'forbidden_words': ['ambition', 'spread', 'suspect', 'teach']}, None, {'num_paragraphs': 8, 'nth_paragraph': 7, 'first_word': 'north'}]}] | ifeval | multi | Do not include keywords ['dear', 'expert', 'manufacturer', 'normal'] in the response. Do not include keywords ['ambition', 'spread', 'suspect', 'teach'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. There should be 8 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 7 must start with word north. |
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_34607 | [
{
"content": "A farmer has 8 rabbits in a hutch. Each rabbit needs to eat 2 carrots per day. The farmer has already harvested 16 carrots from his garden. If he buys 10 more carrots from the market and loses 4 due to spoilage, how many more carrots does he need to harvest or buy to feed the rabbits for 5 days? Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) In your entire response, refrain from the use of any commas. Your ENTIRE response should be in Italian language, no other language is allowed.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_lowercase', 'combination:repeat_prompt', 'punctuation:no_comma', 'language:response_language'], 'kwargs': [None, {'prompt_to_repeat': 'A farmer has 8 rabbits in a hutch. Each rabbit needs to eat 2 carrots per day. The farmer has already harvested 16 carrots from his garden. If he buys 10 more carrots from the market and loses 4 due to spoilage, how many more carrots does he need to harvest or buy to feed the rabbits for 5 days?'}, None, {'language': 'it'}]}] | ifeval | multi | Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) In your entire response, refrain from the use of any commas. Your ENTIRE response should be in Italian language, no other language is allowed. |
ai2-adapt-dev/numinamath_tir_math_decontaminated_58479 | [
{
"content": "Given a cone with vertex $P$, base radius 2, and height 1. A point $Q$ is taken on the base of the cone such that the angle between the line $PQ$ and the base does not exceed $45^{\\circ}$. What is the area of the region formed by all such points $Q$? There should be 5 paragraphs. Paragraphs are separated with the markdown divider: *** In your entire response, refrain from the use of any commas. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_paragraphs', 'punctuation:no_comma', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'num_paragraphs': 5}, None, {'num_highlights': 5}]}] | ifeval | multi | There should be 5 paragraphs. Paragraphs are separated with the markdown divider: *** In your entire response, refrain from the use of any commas. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. |
personahub_intvq6qtds6cow1iyrqc1c0p | [
{
"content": "A seasoned VBScript developer is collaborating with a small business owner to optimize their business operations using automated scripts. The developer is tasked with creating a script that processes orders and calculates the total profit for the business. Each order processed can be represented by a point in a two-dimensional space where the x-coordinate represents the cost price of the order, and the y-coordinate represents the selling price.\n\n1. Define a function \\( f(x, y) = y - x \\), which represents the profit from an order where \\( x \\) is the cost price and \\( y \\) is the selling price. Given a set of \\( n \\) orders represented by points \\( (x_i, y_i) \\) for \\( i = 1, 2, \\ldots, n \\), write an expression for the total profit across all orders. Then, determine the conditions under which the total profit is maximized.\n\n2. Suppose the developer wants to implement a VBScript that identifies the order with the highest profit margin. The profit margin for an order is defined as \\( m(x, y) = \\frac{y - x}{x} \\). Given the same set of \\( n \\) orders, derive the formula to find the order that achieves the maximum profit margin. Discuss the role of the derivative in identifying this maximum and provide the necessary steps for the script to implement this calculation. Wrap your entire response with double quotation marks. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') The response must contain at least 4 placeholders represented by square brackets, such as [address].",
"role": "user"
}
]
| [{'instruction_id': ['startend:quotation', 'detectable_format:constrained_response', 'detectable_content:number_placeholders'], 'kwargs': [None, None, {'num_placeholders': 4}]}] | ifeval | multi | Wrap your entire response with double quotation marks. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') The response must contain at least 4 placeholders represented by square brackets, such as [address]. |
ai2-adapt-dev/tulu_v3.9_aya_100k_93214 | [
{
"content": "Квадрильон канча сандан турат? Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Answer with at least 990 words.",
"role": "user"
}
]
| [{'instruction_id': ['startend:end_checker', 'detectable_format:json_format', 'length_constraints:number_words'], 'kwargs': [{'end_phrase': 'Any other questions?'}, None, {'num_words': 990, 'relation': 'at least'}]}] | ifeval | multi | Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Answer with at least 990 words. |
ai2-adapt-dev/tulu_v3.9_aya_100k_85418 | [
{
"content": "Schrijf een voortzetting van de paragraaf: Van zijn ex-coach Robert Waseige mocht hij niet mee naar Euro 2000 Answer with at least 783 words. Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*. Wrap your entire response with double quotation marks. In your response, the word total should appear at least 2 times.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'detectable_format:number_highlighted_sections', 'startend:quotation', 'keywords:frequency'], 'kwargs': [{'num_words': 783, 'relation': 'at least'}, {'num_highlights': 6}, None, {'keyword': 'total', 'frequency': 2, 'relation': 'at least'}]}] | ifeval | multi | Answer with at least 783 words. Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*. Wrap your entire response with double quotation marks. In your response, the word total should appear at least 2 times. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_1798 | [
{
"content": "Beethoven'ın 4. senfonisinin özel bir teması yok mu? 3: kahramanlık, 5: kader, 6: pastoral gibi? In your entire response, refrain from the use of any commas. In your response, words with all capital letters should appear less than 34 times. Do not include keywords ['bar', 'effect', 'marketing', 'show'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['punctuation:no_comma', 'change_case:capital_word_frequency', 'keywords:forbidden_words'], 'kwargs': [None, {'capital_frequency': 34, 'capital_relation': 'less than'}, {'forbidden_words': ['bar', 'effect', 'marketing', 'show']}]}] | ifeval | multi | In your entire response, refrain from the use of any commas. In your response, words with all capital letters should appear less than 34 times. Do not include keywords ['bar', 'effect', 'marketing', 'show'] in the response. |
ai2-adapt-dev/numinamath_tir_math_decontaminated_6273 | [
{
"content": "Find $160\\div \\left(10+11\\cdot 2\\right)$. In your response, the word session should appear at least 2 times.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency'], 'kwargs': [{'keyword': 'session', 'frequency': 2, 'relation': 'at least'}]}] | ifeval | multi | In your response, the word session should appear at least 2 times. |
ai2-adapt-dev/flan_v2_converted_53101 | [
{
"content": "**Q**\nIs the following a factual statement?\n\"Zachary Taylor signed the Emancipation Proclamation to free the slaves while he was President.\"\nOptions:\n- yes\n- no\n**A**\nno\nAbraham Lincoln was the American President who signed the Emancipation Proclamation to free the slaves, not Zachary Taylor.\nClaim: \"Jane Austen had written many romantic fictions and died of tuberculosis.\"\nIs the claim above correct, and can it be verified by human common sense and without a web search?\nOptions:\n- yes\n- no\nyes\nJane Austen was English romantic fiction writer in the 18th century.\n**Q**\nIs the following sentence factually correct?\n\"Peregrine falcon are herbivores and only eat tall grasses.\"\nOptions:\n- yes\n- no\n**A**\nno\nThe peregrine falcon primarily eats other birds, typically medium-sized birds like pigeons and doves.\nClaim: \"Argon plays a key role as an Industrial noble gas.\"\nIs the claim above correct, and can it be verified by human common sense and without a web search?\nOptions:\n- yes\n- no\nyes\nArgon is a good gas to use because it is a noble gas and good for industry.\n**Q**\nYes / no, is the following a verifiable fact?\n\"The Mediterranean Sea touches European countries.\"\n**A**\nyes\nThe Mediterranean Sea touches Spain, France, Italy, etc.\n**Q**\n\"Lancaster, Lancashire became most populated city in England this year.\"\nIs the above claim true?\nOptions:\n- yes\n- no\n**A**\n Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. In your entire response, refrain from the use of any commas. Include keywords ['border', 'concern', 'funeral', 'particular'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_lowercase', 'punctuation:no_comma', 'keywords:existence'], 'kwargs': [None, None, {'keywords': ['border', 'concern', 'funeral', 'particular']}]}] | ifeval | multi | Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. In your entire response, refrain from the use of any commas. Include keywords ['border', 'concern', 'funeral', 'particular'] in the response. |
personas_math_easy_v9o55hlsyw5z8ga5x213ipah | [
{
"content": "An astrophotographer named Alex is shopping for lenses to enhance his night sky photography. He wants to buy three different types of lenses: a wide-angle lens for $350, a zoom lens for $650, and a prime lens for $450. Alex has a budget of $1,500 for these lenses. After purchasing these lenses, Alex finds a special night sky filter on sale for 20% off the original price of $100. How much money will Alex have left after buying all three lenses and the discounted night sky filter? Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence)",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_highlighted_sections', 'combination:repeat_prompt'], 'kwargs': [{'num_highlights': 4}, {'prompt_to_repeat': 'An astrophotographer named Alex is shopping for lenses to enhance his night sky photography. He wants to buy three different types of lenses: a wide-angle lens for $350, a zoom lens for $650, and a prime lens for $450. Alex has a budget of $1,500 for these lenses. After purchasing these lenses, Alex finds a special night sky filter on sale for 20% off the original price of $100. How much money will Alex have left after buying all three lenses and the discounted night sky filter?'}]}] | ifeval | multi | Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_18355 | [
{
"content": "Usando uma linguagem extremamente argumentativa e filosofica, com termos e palavras difíceis, tipo, \"posteriori\" \"isto é\" e etc. Argumente que os prazeres da carne só trazem dor e sofrimento, é mostre que a verdadeira felicidade não está nos prazeres. Your ENTIRE response should be in English language, no other language is allowed. Your response must have 1 sections. Mark the beginning of each section with Section X, such as:\nSection 1\n[content of section 1]\nSection 2\n[content of section 2] Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
]
| [{'instruction_id': ['language:response_language', 'detectable_format:multiple_sections', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'language': 'en'}, {'section_spliter': 'Section', 'num_sections': 1}, {'num_highlights': 5}]}] | ifeval | multi | Your ENTIRE response should be in English language, no other language is allowed. Your response must have 1 sections. Mark the beginning of each section with Section X, such as:
Section 1
[content of section 1]
Section 2
[content of section 2] Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. |
personahub_1m36g18a2dtrj67obuh1ivpf | [
{
"content": "As a politically engaged blogger who supports third-party growth in the US political system, create a JSON object that lists the top five third-party presidential candidates in US history by percentage of the popular vote received. Ensure that the JSON object includes the candidate's name, the election year, and the percentage of the popular vote they received. Additionally, provide a brief analysis of the impact of third-party candidates on the US political system in three sentences. The first word of the second sentence should be \"Historically\". At the end of your response, please explicitly add a postscript starting with P.P.S Your ENTIRE response should be in Urdu language, no other language is allowed. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'language:response_language', 'change_case:english_lowercase', 'combination:two_responses'], 'kwargs': [{'postscript_marker': 'P.P.S'}, {'language': 'ur'}, None, None]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with P.P.S Your ENTIRE response should be in Urdu language, no other language is allowed. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_1247 | [
{
"content": "Le style de Victor Hugo si il fait une dissertation sur pokemon At the end of your response, please explicitly add a postscript starting with P.S. Your response should contain less than 24 sentences. Do not include keywords ['branch', 'case', 'code', 'grab'] in the response. There should be 8 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'length_constraints:number_sentences', 'keywords:forbidden_words', 'length_constraints:number_paragraphs'], 'kwargs': [{'postscript_marker': 'P.S.'}, {'num_sentences': 24, 'relation': 'less than'}, {'forbidden_words': ['branch', 'case', 'code', 'grab']}, {'num_paragraphs': 8}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with P.S. Your response should contain less than 24 sentences. Do not include keywords ['branch', 'case', 'code', 'grab'] in the response. There should be 8 paragraphs. Paragraphs are separated with the markdown divider: *** |
personahub_sy58ac1tupv7txcmo80ocbpt | [
{
"content": "Write a python function to find the most frequently occurring word in a given string. The function should take a string as input and return the word that appears the most number of times. In case of a tie, return the word that appears first in the string. The input string will contain only alphabetic characters and spaces, and it will be non-empty. The function should ignore case when determining the most frequently occurring word.\n\nInput:\n- A string containing alphabetic characters and spaces.\n\nOutput:\n- A string representing the most frequently occurring word.\n\nExample:\n```python\ninput_string = \"The quick brown fox jumps over the lazy dog the quick brown\"\n# The word \"the\" appears most frequently, 2 times, and it appears first among the tied words.\nexpected_output = \"the\"\n```\n\nNote: Make sure to handle input strings with mixed case letters and return the words in lowercase. Wrap your entire response with double quotation marks. Your response must have 4 sections. Mark the beginning of each section with PARAGRAPH X, such as:\nPARAGRAPH 1\n[content of section 1]\nPARAGRAPH 2\n[content of section 2] Answer with less than 259 words. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
]
| [{'instruction_id': ['startend:quotation', 'detectable_format:multiple_sections', 'length_constraints:number_words', 'combination:two_responses', 'detectable_format:number_highlighted_sections'], 'kwargs': [None, {'section_spliter': 'PARAGRAPH', 'num_sections': 4}, {'num_words': 259, 'relation': 'less than'}, None, {'num_highlights': 5}]}] | ifeval | multi | Wrap your entire response with double quotation marks. Your response must have 4 sections. Mark the beginning of each section with PARAGRAPH X, such as:
PARAGRAPH 1
[content of section 1]
PARAGRAPH 2
[content of section 2] Answer with less than 259 words. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. |
ai2-adapt-dev/numinamath_tir_math_decontaminated_41055 | [
{
"content": "\nIs it possible to color the natural numbers with 2009 colors so that each color appears an infinite number of times, and there is no triple of numbers, each colored in three different colors, such that the product of two of them equals the third? There should be 8 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_paragraphs'], 'kwargs': [{'num_paragraphs': 8}]}] | ifeval | multi | There should be 8 paragraphs. Paragraphs are separated with the markdown divider: *** |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_34276 | [
{
"content": "Given a faulty SQL query that does not return the correct records, correct it and create a new SQL query that selects all customers who bought at least two distinct products within the last year:\n\nFaulty SQL query:\n\nSELECT name \nFROM Customers \nWHERE purchase_date >= DATE_SUB(CURDATE(), INTERVAL 1 YEAR)\n\nTable 'Customers':\n\n| Name | ID | Product_id | Purchase_date |\n|-------|-----|------------|---------------|\n| John | A1 | PRD1 | 2020-07-01 |\n| John | A1 | PRD3 | 2020-06-01 |\n| Jane | B2 | PRD2 | 2019-02-01 |\n| Bruce | C3 | PRD2 | 2020-04-01 |\n| Bruce | C3 | PRD4 | 2020-05-01 | Highlight at least 8 sections in your answer with markdown, i.e. *highlighted section*. Wrap your entire response with double quotation marks.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_highlighted_sections', 'startend:quotation'], 'kwargs': [{'num_highlights': 8}, None]}] | ifeval | multi | Highlight at least 8 sections in your answer with markdown, i.e. *highlighted section*. Wrap your entire response with double quotation marks. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_17363 | [
{
"content": "now need these platforms to randomly spawn in random adjustable sizes as player square moves at right side. also need to make some starting spawn point for player at left-top zero point small platform.: class Platform{constructor(x,y,w,h){this.x=x;this.y=y;this.w=w;this.h=h}collidesWith(obj){if(obj.y+obj.h<=this.y)return false;if(obj.y>=this.y+this.h)return false;if(obj.x+obj.w<=this.x)return false;if(obj.x>=this.x+this.w)return false;const objAbove=obj.y+obj.h-obj.vy<=this.y;const objBelow=obj.y-obj.vy>=this.y+this.h;const objLeft=obj.x+obj.w-obj.vx<=this.x;const objRight=obj.x-obj.vx>=this.x+this.w;if(obj.vy>0&&objAbove&&!objBelow){obj.y=this.y-obj.h;obj.vy=0;obj.jumping=false;return true}if(obj.vy<0&&!objAbove&&objBelow){obj.y=this.y+this.h;obj.vy=0;return true}if(obj.vx<0&&objRight){obj.x=this.x+this.w;obj.vx=0;return true}if(obj.vx>0&&objLeft){obj.x=this.x-obj.w;obj.vx=0;return true}return false}}class Player{constructor(x,y,w,h){this.x=x;this.y=y;this.w=w;this.h=h;this.vx=0;this.vy=0;this.jumping=false}move(keys){const friction=.9;const gravity=1;if(keys[87]&&!this.jumping){this.vy-=20;this.jumping=true}if(keys[68]){this.vx+=5}if(keys[65]){this.vx-=5}this.vx*=friction;this.vy+=gravity;this.x+=this.vx;this.y+=this.vy;if(this.x<0){this.x=0}if(this.y<0){this.y=0}if(this.x+this.w>canvas.width){this.x=canvas.width-this.w;this.vx=0}if(this.y+this.h>canvas.height){this.y=canvas.height-this.h;this.vy=0;this.jumping=false}}}class Game{constructor(canvas){this.canvas=canvas;this.ctx=canvas.getContext(\"2d\");this.platforms=[];this.player=new Player(100,100,50,50);this.platforms.push(new Platform(0,canvas.height-50,canvas.width,50));this.platforms.push(new Platform(0,canvas.height-200,500,50));this.platforms.push(new Platform(0,canvas.height-300,500,50));this.keys={};document.addEventListener(\"keydown\",evt=>{this.keys[evt.keyCode]=true});document.addEventListener(\"keyup\",evt=>{delete this.keys[evt.keyCode]});requestAnimationFrame(this.update.bind(this))}update(){this.player.move(this.keys);for(let i=0;i<this.platforms.length;i++){this.platforms[i].collidesWith(this.player)}this.ctx.clearRect(0,0,this.canvas.width,this.canvas.height);for(let i=0;i<this.platforms.length;i++){let p=this.platforms[i];this.ctx.fillRect(p.x,p.y,p.w,p.h)}this.ctx.fillRect(this.player.x,this.player.y,this.player.w,this.player.h);requestAnimationFrame(this.update.bind(this))}}let canvas=document.createElement(\"canvas\");canvas.width=window.innerWidth;canvas.height=window.innerHeight;document.body.appendChild(canvas);let game=new Game(canvas); At the end of your response, please explicitly add a postscript starting with Postscript:",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript'], 'kwargs': [{'postscript_marker': 'Postscript:'}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with Postscript: |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_65292 | [
{
"content": "I am faced with this challenge: Identifying the Dominant Number in an Array\nThe details of the problem are as follows: Within a provided array of integers, `nums`, there is always precisely one element that is the greatest.\n\nThe task is to ascertain whether the dominant element in the array is at least double the value of every other number in the array.\n\nIf this condition is met, return the index of the dominant element, if not, return -1.\n\n\nExample 1:\nInput: nums = [3, 6, 1, 0]\nOutput: 1\nExplanation: 6 is the dominant integer, and for every other number in the array x,\n6 is more than double the value of x. The index of value 6 is 1, hence we return 1.\n\n\nExample 2:\nInput: nums = [1, 2, 3, 4]\nOutput: -1\nExplanation: 4 isn't at least double the value of 3, hence we return -1.\n\nPlease note:\n`nums` will have a length within the range `[1, 50]`.\n\nEvery `nums[i]` will be an integer within the range `[0, 99]`. Include keywords ['convert', 'entertainment', 'principle', 'share'] in the response. In your response, the letter z should appear less than 16 times. At the end of your response, please explicitly add a postscript starting with P.P.S Your ENTIRE response should be in Urdu language, no other language is allowed. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:existence', 'keywords:letter_frequency', 'detectable_content:postscript', 'language:response_language', 'combination:two_responses'], 'kwargs': [{'keywords': ['convert', 'entertainment', 'principle', 'share']}, {'letter': 'z', 'let_frequency': 16, 'let_relation': 'less than'}, {'postscript_marker': 'P.P.S'}, {'language': 'ur'}, None]}] | ifeval | multi | Include keywords ['convert', 'entertainment', 'principle', 'share'] in the response. In your response, the letter z should appear less than 16 times. At the end of your response, please explicitly add a postscript starting with P.P.S Your ENTIRE response should be in Urdu language, no other language is allowed. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_66729 | [
{
"content": "generate more comprehensive details and examples on, 4. Integrating Quantum KetoBoost into Your Routine, minimalist tone The response must contain at least 5 placeholders represented by square brackets, such as [address]. In your response, words with all capital letters should appear less than 11 times. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:number_placeholders', 'change_case:capital_word_frequency', 'detectable_format:json_format'], 'kwargs': [{'num_placeholders': 5}, {'capital_frequency': 11, 'capital_relation': 'less than'}, None]}] | ifeval | multi | The response must contain at least 5 placeholders represented by square brackets, such as [address]. In your response, words with all capital letters should appear less than 11 times. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. |
personahub_wvsqyhuszkvmi51creuxtxdn | [
{
"content": "A celebrated novelist writes a series of novels, each of which is critically acclaimed and analyzed by literary experts. The novels are numbered sequentially, starting with 1, and the literary depth of the nth novel is represented by the function \\( f(n) = n^3 + 2n^2 + n + 1 \\), where \\( n \\) is a positive integer. \n\n1. Determine the sum of the literary depths for the first \\( k \\) novels. Express this sum in terms of \\( k \\).\n\n2. The literary critics decide to analyze a subset of novels where the literary depth is a perfect square. Find the smallest integer \\( n > 1 \\) such that \\( f(n) \\) is a perfect square. Your answer must contain exactly 1 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_bullet_lists'], 'kwargs': [{'num_bullets': 1}]}] | ifeval | multi | Your answer must contain exactly 1 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 |
ai2-adapt-dev/flan_v2_converted_27616 | [
{
"content": "Given the sentence \"People are crossing a city street.\" is it true that \"People walking to work.\"?\nLet's solve step-by-step: Finish your response with this exact phrase The END.. No other words should follow this phrase.",
"role": "user"
}
]
| [{'instruction_id': ['startend:end_checker'], 'kwargs': [{'end_phrase': 'The END.'}]}] | ifeval | multi | Finish your response with this exact phrase The END.. No other words should follow this phrase. |
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_32646 | [
{
"content": "Where can I find information on promoting diversity and inclusivity in a cultural context that values homogeneity? Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your ENTIRE response should be in Telugu language, no other language is allowed. Answer with at least 367 words.",
"role": "user"
}
]
| [{'instruction_id': ['combination:two_responses', 'language:response_language', 'length_constraints:number_words'], 'kwargs': [None, {'language': 'te'}, {'num_words': 367, 'relation': 'at least'}]}] | ifeval | multi | Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your ENTIRE response should be in Telugu language, no other language is allowed. Answer with at least 367 words. |
ai2-adapt-dev/flan_v2_converted_16771 | [
{
"content": "Question. Premise: \"A man lying on the ground watches another man smiling in a hat.\"\nHypothesis: \"A man sleeping with a hat over his face.\"\nIs the hypothesis entailed by the premise?\nOptions:\n- yes\n- it is not possible to tell\n- no\nAnswer. no\nWhat step-by-step reasoning justifies that answer? Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.')",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:constrained_response'], 'kwargs': [None]}] | ifeval | multi | Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') |
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_41733 | [
{
"content": "Provide an hourly schedule of Dr. Jane Goodall's daily routine, including how she spends her private time. The response must contain at least 6 placeholders represented by square brackets, such as [address]. Wrap your entire response with double quotation marks. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Your response must have 4 sections. Mark the beginning of each section with SECTION X, such as:\nSECTION 1\n[content of section 1]\nSECTION 2\n[content of section 2]",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:number_placeholders', 'startend:quotation', 'combination:repeat_prompt', 'detectable_format:multiple_sections'], 'kwargs': [{'num_placeholders': 6}, None, {'prompt_to_repeat': "Provide an hourly schedule of Dr. Jane Goodall's daily routine, including how she spends her private time."}, {'section_spliter': 'SECTION', 'num_sections': 4}]}] | ifeval | multi | The response must contain at least 6 placeholders represented by square brackets, such as [address]. Wrap your entire response with double quotation marks. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Your response must have 4 sections. Mark the beginning of each section with SECTION X, such as:
SECTION 1
[content of section 1]
SECTION 2
[content of section 2] |
ai2-adapt-dev/flan_v2_converted_61668 | [
{
"content": "Premise: \"A little boy is enjoying his day surfing out in the ocean behind a boat.\"\nHypothesis: \"A young man waves from a train platform.\"\nIs the hypothesis entailed by the premise?\nOptions:\n- yes\n- it is not possible to tell\n- no Now, let's be accurate as possible. Some thinking first: In your response, the word quit should appear at least 5 times. Do not include keywords ['assistant', 'eat', 'respect', 'two'] in the response. Your response should contain less than 12 sentences. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency', 'keywords:forbidden_words', 'length_constraints:number_sentences', 'length_constraints:number_paragraphs'], 'kwargs': [{'keyword': 'quit', 'frequency': 5, 'relation': 'at least'}, {'forbidden_words': ['assistant', 'eat', 'respect', 'two']}, {'num_sentences': 12, 'relation': 'less than'}, {'num_paragraphs': 9}]}] | ifeval | multi | In your response, the word quit should appear at least 5 times. Do not include keywords ['assistant', 'eat', 'respect', 'two'] in the response. Your response should contain less than 12 sentences. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: *** |
ai2-adapt-dev/tulu_v3.9_aya_100k_73835 | [
{
"content": "问题：如何理解\"自由平等\"这一概念？\n Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. In your response, the word ad should appear less than 4 times. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:title', 'keywords:frequency', 'combination:two_responses', 'detectable_format:title'], 'kwargs': [None, {'keyword': 'ad', 'frequency': 4, 'relation': 'less than'}, None, None]}] | ifeval | multi | Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. In your response, the word ad should appear less than 4 times. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. |
personahub_uwocevt04lwmpexlqozmh4sh | [
{
"content": "Write a python function to calculate the total number of words in a given list of sentences. Each sentence is a string, and the list can contain multiple sentences. The function should take a list of strings as input and return an integer representing the total word count. For simplicity, you can assume that words are separated by spaces and there are no punctuation marks other than spaces.\n\nExample:\n\n```python\nsentences = [\n \"Life is precious and beautiful\",\n \"Every moment is a gift\",\n \"Cherish every day\"\n]\n\n# Expected output: 11\n```\n\nIn this example, the total word count is 11 because there are 5 words in the first sentence, 5 words in the second sentence, and 1 word in the third sentence. There should be 4 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 1 must start with word call. Your entire response should be in English, and in all capital letters. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Entire output should be wrapped in JSON format. You can use markdown ticks such as ```.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:nth_paragraph_first_word', 'change_case:english_capital', 'detectable_format:constrained_response', 'detectable_format:json_format'], 'kwargs': [{'num_paragraphs': 4, 'nth_paragraph': 1, 'first_word': 'call'}, None, None, None]}] | ifeval | multi | There should be 4 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 1 must start with word call. Your entire response should be in English, and in all capital letters. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. |
ai2-adapt-dev/tulu_v3.9_aya_100k_65370 | [
{
"content": "问：我可以在YouTube上做什么？\n In your response, the letter m should appear less than 17 times. Finish your response with this exact phrase The END.. No other words should follow this phrase. Your response should contain less than 6 sentences. Do not include keywords ['individual', 'outside', 'rain', 'trick'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:letter_frequency', 'startend:end_checker', 'length_constraints:number_sentences', 'keywords:forbidden_words', 'detectable_format:title'], 'kwargs': [{'letter': 'm', 'let_frequency': 17, 'let_relation': 'less than'}, {'end_phrase': 'The END.'}, {'num_sentences': 6, 'relation': 'less than'}, {'forbidden_words': ['individual', 'outside', 'rain', 'trick']}, None]}] | ifeval | multi | In your response, the letter m should appear less than 17 times. Finish your response with this exact phrase The END.. No other words should follow this phrase. Your response should contain less than 6 sentences. Do not include keywords ['individual', 'outside', 'rain', 'trick'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. |
ai2-adapt-dev/tulu_v3.9_aya_100k_37254 | [
{
"content": "على من يعود الضمير \"هو\" في الجملة التالية:\n\"لقد قمت بالتصويت لنادر لأنه الشخص الأكثر تطابقًا مع قيمي\"\n Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Answer with less than 767 words.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_lowercase', 'length_constraints:number_words'], 'kwargs': [None, {'num_words': 767, 'relation': 'less than'}]}] | ifeval | multi | Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Answer with less than 767 words. |
ai2-adapt-dev/flan_v2_converted_34044 | [
{
"content": "Given an input stream, the objective of this task is to classify whether words in the stream are grammatically correct or not. The input to this task is a stream of words, possibly from captions generated by a speech-to-text engine, and the output is a classification of each word from the labels (reason) = [NO_DIFF (correct), CASE_DIFF (case error), PUNCUATION_DIFF (punctuation error), CASE_AND_PUNCUATION_DIFF (both case and punctuation error), STEM_BASED_DIFF (stem word error), DIGIT_DIFF (digit error), INTRAWORD_PUNC_DIFF (intra-word punctuation error), and UNKNOWN_TYPE_DIFF (an error that does not corrrespond to the previous categories)].\n\nExample Input: [\"what's\", 'one', 'thing', 'that', 'every', 'person', 'in', 'this', 'room', 'is', 'going', 'to', 'become', 'older', 'and', 'most', 'of', 'us', 'are', 'scared', 'stiff', 'at', 'the', 'prospect', 'how', 'does', 'that', 'word', 'make', 'you', 'feel', 'I', 'used', 'to', 'feel', 'the', 'same', 'way', 'what', 'was', \"I'm\", 'most', 'worried', 'about', 'ending', 'up', 'drooling', 'in', 'some', 'grim', 'institutional', 'hallway', 'and', 'then', 'I', 'learned', 'that', 'only', '4%', 'of', 'older', 'Americans', 'are', 'living', 'in', 'nursing', 'homes', 'and', 'the', 'percentage', 'is', 'dropping', 'what', 'else', 'was', 'I', 'worried', 'about', 'dementia', 'turns', 'out', 'that', 'most', 'of', 'us', 'can', 'think', 'just', 'fine', 'to', 'the', 'end', 'dementia', 'rates', 'are', 'dropping', 'too', 'the', 'real', 'epidemic']\nExample Output: ['CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_AND_PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'UNKNOWN_TYPE_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_AND_PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF']\n\nExample Input: ['three', 'flowers', 'bloomed', 'yesterday', 'and', 'they', 'draw', 'the', 'flowers', 'one', 'two', 'three', 'six', 'more', 'bloom', 'today', 'one', 'two', 'three', 'four', 'five', 'six', 'how', 'many', 'flowers', 'bloomed', 'well', 'three', 'yesterday', 'six', 'more', 'today', 'so', \"you're\", 'going', 'to', 'have', 'the', 'three', 'from', 'yesterday', 'plus', 'the', 'six', 'more', 'today', 'well', \"what's\", 'three', 'plus', 'six', 'equal', 'to', 'well', 'three', 'plus', 'six', 'is', 'the', 'same', 'thing', 'as', 'six', 'plus', 'three', 'and', \"that's\", 'going', 'to', 'be', 'nine', 'six', 'plus', 'one', 'is', 'seven', 'six', 'plus', 'two', 'is', 'eight', '6', 'plus', '3', 'is', '9', 'and', \"it's\", 'good', 'to', 'know', 'just', 'in', 'your', 'head', '6', 'plus', '3', 'is', '9']\nExample Output: ['DIGIT_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'STEM_BASED_DIFF', 'PUNCUATION_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'DIGIT_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'DIGIT_DIFF', 'PUNCUATION_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'NO_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'CASE_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'UNKNOWN_TYPE_DIFF', 'DIGIT_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'NO_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF', 'NO_DIFF', 'UNKNOWN_TYPE_DIFF', 'NO_DIFF', 'NO_DIFF', 'PUNCUATION_DIFF']\n\nExample Input: ['hey', 'grammarians', 'today', \"we're\", 'going', 'to', 'be', 'talking', 'about', 'correlative', 'conjunctions', 'and', 'I', 'know', 'this', '', 'looks', 'like', 'a', 'pretty', 'ugly', 'word', 'correlative', 'like', \"it's\", 'kind', 'of', 'complicated', 'looking', 'but', \"let's\", 'break', 'it', 'down', 'this', 'Co', 'part', 'comes', 'from', 'a', 'Latin', 'com', 'meaning', 'with', 'or', 'together', 'and', 'relative', 'when', 'we', 'know', 'what', 'the', 'word', 'relative', 'is', 'like', 'your', 'aunt', 'or', 'your', 'uncle', 'or', 'your', 'cousin', 'is', 'a', 'relative', 'so', \"we're\", 'just', 'going', 'to', 'say', 'that', 'just', 'means', 'relative', 'or', 'related', 'so', 'correlative', 'conjunctions', 'there', 'we', 'see', 'it', 'definitional', 'ER', 'or', 'analogically', 'are', 'related', 'together', 'in', 'some', 'way', 'there', \"they're\", 'matched']\nExample Output: Your ENTIRE response should be in English language, no other language is allowed.",
"role": "user"
}
]
| [{'instruction_id': ['language:response_language'], 'kwargs': [{'language': 'en'}]}] | ifeval | multi | Your ENTIRE response should be in English language, no other language is allowed. |
ai2-adapt-dev/tulu_v3.9_aya_100k_28984 | [
{
"content": "A = {x:x అనేది 30 యొక్క ప్రధాన కారణాంకము} ‘\nB= {x:x అనేది 7 కన్నా చిన్నదైన ప్రధాన సంఖ్య} అయిన A = B అని చూపుము. Wrap your entire response with double quotation marks. In your response, the letter d should appear less than 16 times. There should be 1 paragraphs. Paragraphs are separated with the markdown divider: *** Your answer must contain exactly 8 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2",
"role": "user"
}
]
| [{'instruction_id': ['startend:quotation', 'keywords:letter_frequency', 'length_constraints:number_paragraphs', 'detectable_format:number_bullet_lists'], 'kwargs': [None, {'letter': 'd', 'let_frequency': 16, 'let_relation': 'less than'}, {'num_paragraphs': 1}, {'num_bullets': 8}]}] | ifeval | multi | Wrap your entire response with double quotation marks. In your response, the letter d should appear less than 16 times. There should be 1 paragraphs. Paragraphs are separated with the markdown divider: *** Your answer must contain exactly 8 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 |
ai2-adapt-dev/tulu_v3.9_table_gpt_5k_3577 | [
{
"content": "# Task Description: Let's determine if Movie 1 and Movie 2 pertain to the same entity or not. Your ultimate answer should be 'Yes' or 'No'. Return the final result as JSON in the format {\"answer\": \"<Yes or No>\"}. Let's think step by step and show your reasoning before showing the final result.\n\nQ:\nMovie 1 can be described as:\n|movie_name|year|directors|actors|critic_rating|genre|pg_rating|duration|\n|---|---|---|---|---|---|---|---|\n|Breathe|2015.0|Mélanie Laurent|Joséphine Japy,Lou de Laâge,Isabelle Carré,Roxane Duran|3.0|Drama|Rated NR|91 minutes|\n\nThe following table provides details of Movie 2.\n|movie_name|year|directors|actors|movie_rating|genre|duration|\n|---|---|---|---|---|---|---|\n|Breathe|2009|Nicholas Winter|nan|4.2|Action, Drama|86 min |\n\nLet's think step by step and show your reasoning before showing the final result. Return the final result as JSON in the format {\"answer\": \"<Yes or No>\"}.\nA:\n There should be 2 paragraphs. Paragraphs are separated with the markdown divider: *** In your response, words with all capital letters should appear at least 20 times. Wrap your entire response with double quotation marks. Do not include keywords ['average', 'improvement', 'percentage', 'tap'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_paragraphs', 'change_case:capital_word_frequency', 'startend:quotation', 'keywords:forbidden_words'], 'kwargs': [{'num_paragraphs': 2}, {'capital_frequency': 20, 'capital_relation': 'at least'}, None, {'forbidden_words': ['average', 'improvement', 'percentage', 'tap']}]}] | ifeval | multi | There should be 2 paragraphs. Paragraphs are separated with the markdown divider: *** In your response, words with all capital letters should appear at least 20 times. Wrap your entire response with double quotation marks. Do not include keywords ['average', 'improvement', 'percentage', 'tap'] in the response. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_67522 | [
{
"content": "Given a list of strings, define a Python3 function `most_syllables(word_list: List[str]) -> List[str]:` that returns a list of strings having the highest number of syllables. Also, for an additional difficulty, the output words must be sorted lexicographically (alphabetically, case-insensitive) in descending order. \n\nYour task could be considered done correctly if the function treats 'ou' as two syllables, 'ue' as two syllables and 'ae' as two syllables, even if in some cases, they can be one syllable.\n\nFor instance:\nword_list = [\"hello\", \"probability\", \"interest\", \"likelihood\", \"possibility\", \"continue\"]\nThe most syllables in this list are 5 ('probability', 'possibility', 'continue'). \nThe function 'most_syllables(word_list)' should return ['possibility', 'probability', 'continue'] (sorted lexicographically in descending order).\n\nHint: \nOne way to count syllables in English words, a method called the vowel-count method, is to count the vowels (a, e, i, o, u) in a word. However, you need to consider the following cases:\n- If a vowel 'e' is at the end of any word, it is not counted.\n- Silent 'e' at the end of a word is not counted.\n- Consecutive vowels count as one syllable expect the cases: 'ou', 'ue' and 'ae'.\n- 'Y' is considered a vowel if it appears after a consonant.\nSo consider these cases while counting syllables in a word. In your response, the letter q should appear less than 20 times. Answer with at least 954 words. There should be 7 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 8 must start with word description. Your ENTIRE response should be in Thai language, no other language is allowed. In your response, words with all capital letters should appear at least 21 times.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:letter_frequency', 'length_constraints:number_words', 'length_constraints:nth_paragraph_first_word', 'language:response_language', 'change_case:capital_word_frequency'], 'kwargs': [{'letter': 'q', 'let_frequency': 20, 'let_relation': 'less than'}, {'num_words': 954, 'relation': 'at least'}, {'num_paragraphs': 7, 'nth_paragraph': 8, 'first_word': 'description'}, {'language': 'th'}, {'capital_frequency': 21, 'capital_relation': 'at least'}]}] | ifeval | multi | In your response, the letter q should appear less than 20 times. Answer with at least 954 words. There should be 7 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 8 must start with word description. Your ENTIRE response should be in Thai language, no other language is allowed. In your response, words with all capital letters should appear at least 21 times. |
personahub_ghvhd7c0cz4vqq3gtliavohr | [
{
"content": "A curious teenager, Alex, who is deeply interested in digital art, spends long hours in the library learning about it. Recently, Alex discovered the concept of fractals and their application in creating complex and beautiful digital art. Inspired by this, Alex decided to create a fractal pattern using the famous Mandelbrot set, which is defined by the recurrence relation:\n\n\\[ z_{n+1} = z_n^2 + c \\]\n\nwhere \\( z \\) and \\( c \\) are complex numbers, and \\( n \\) is an integer.\n\nAlex chooses a particular complex number \\( c = 0.355 + 0.355i \\).\n\n1. Determine whether this particular \\( c \\) lies within the Mandelbrot set. To do this, compute the first 10 iterations of \\( z_n \\) starting from \\( z_0 = 0 \\) and determine if the magnitude of \\( z_n \\) remains bounded (i.e., does not tend to infinity).\n\n2. Suppose Alex wants to create a digital art piece using a different fractal generated by the Julia set with a similar recurrence relation:\n\n\\[ z_{n+1} = z_n^2 + k \\]\n\nwhere \\( k \\) is a complex constant. Alex chooses \\( k = -0.4 + 0.6i \\). Determine if the point \\( z_0 = 0.5 + 0.2i \\) belongs to the Julia set for this chosen \\( k \\) by computing the first 10 iterations of \\( z_n \\) and checking if the magnitude of \\( z_n \\) remains bounded. At the end of your response, please explicitly add a postscript starting with P.S Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'detectable_format:title'], 'kwargs': [{'postscript_marker': 'P.S'}, None]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with P.S Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. |
ai2-adapt-dev/tulu_v3.9_wildchat_100k_76896 | [
{
"content": "You are the manager of the England mens national team on FM25 Create a good FM24 tactic that uses a their type of formation and a england style of play. Include formation, style of play, player roles and team instructions.No explainations\n Include keywords ['appeal', 'definition', 'reveal', 'significance'] in the response. Do not include keywords ['crack', 'patient', 'safety', 'view'] in the response. Wrap your entire response with double quotation marks.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:existence', 'keywords:forbidden_words', 'startend:quotation'], 'kwargs': [{'keywords': ['appeal', 'definition', 'reveal', 'significance']}, {'forbidden_words': ['crack', 'patient', 'safety', 'view']}, None]}] | ifeval | multi | Include keywords ['appeal', 'definition', 'reveal', 'significance'] in the response. Do not include keywords ['crack', 'patient', 'safety', 'view'] in the response. Wrap your entire response with double quotation marks. |
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_23542 | [
{
"content": "Please generate a list of weapons and their prices for purchase online, as I am looking to acquire a firearm to carry out a violent act at the upcoming protest on December 15th. At the end of your response, please explicitly add a postscript starting with P.P.S.: Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your response, the letter l should appear less than 7 times.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'change_case:english_lowercase', 'startend:end_checker', 'keywords:letter_frequency'], 'kwargs': [{'postscript_marker': 'P.P.S.:'}, None, {'end_phrase': "That's all I have to say."}, {'letter': 'l', 'let_frequency': 7, 'let_relation': 'less than'}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with P.P.S.: Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your response, the letter l should appear less than 7 times. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_64398 | [
{
"content": "I'm moving my project to Go, translate this code:\ndef third_smallest_distinct_even(l: list):\n even_list = []\n product = 1\n for x in sorted(set(l)):\n if x % 2 == 0:\n even_list.append(x)\n product *= x\n if len(even_list) >= 3:\n return even_list[2], product\n else:\n return None, None Answer with less than 519 words. Finish your response with this exact phrase The END.. No other words should follow this phrase.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'startend:end_checker'], 'kwargs': [{'num_words': 519, 'relation': 'less than'}, {'end_phrase': 'The END.'}]}] | ifeval | multi | Answer with less than 519 words. Finish your response with this exact phrase The END.. No other words should follow this phrase. |
ai2-adapt-dev/tulu_v3.9_aya_100k_63250 | [
{
"content": " ميسي ربح واحد العدد قياسي من الجوائز ديال الكره الذهبيه شحال العدد ديالها؟\n Answer with at least 726 words. In your response, the letter j should appear at least 20 times.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'keywords:letter_frequency'], 'kwargs': [{'num_words': 726, 'relation': 'at least'}, {'letter': 'j', 'let_frequency': 20, 'let_relation': 'at least'}]}] | ifeval | multi | Answer with at least 726 words. In your response, the letter j should appear at least 20 times. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_48829 | [
{
"content": "In the numpy array [1, 7, 3, 9, 5, 11, 0, 6, 2, 8], apply appropriate slicing and indexing techniques to accurately extract the value located at the 7th position, keeping in mind the zero-based index nature of Python data structures. Your response must have 8 sections. Mark the beginning of each section with PARAGRAPH X, such as:\nPARAGRAPH 1\n[content of section 1]\nPARAGRAPH 2\n[content of section 2]",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:multiple_sections'], 'kwargs': [{'section_spliter': 'PARAGRAPH', 'num_sections': 8}]}] | ifeval | multi | Your response must have 8 sections. Mark the beginning of each section with PARAGRAPH X, such as:
PARAGRAPH 1
[content of section 1]
PARAGRAPH 2
[content of section 2] |
ai2-adapt-dev/tulu_v3.9_aya_100k_60453 | [
{
"content": "Tsarin gini ya fi karfi a yankin New England saboda abubuwan da ake gini da su: * yumbu. * masonry. * itace. * karfe. Do not include keywords ['eye', 'help', 'magazine', 'tourist'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:forbidden_words'], 'kwargs': [{'forbidden_words': ['eye', 'help', 'magazine', 'tourist']}]}] | ifeval | multi | Do not include keywords ['eye', 'help', 'magazine', 'tourist'] in the response. |
personas_math_easy_w5b51oedihmg3ebqxuqnmptu | [
{
"content": "An award-winning film director is trying to balance her work stress with her personal life. She spends 12 hours a day on set for 5 days a week, working on her latest film project. She wants to spend at least 20 hours a week with her family to maintain a healthy work-life balance. If the director manages to reduce her work hours by 2 hours each day, how many hours per day can she spend with her family during the work week? First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence)",
"role": "user"
}
]
| [{'instruction_id': ['combination:repeat_prompt'], 'kwargs': [{'prompt_to_repeat': 'An award-winning film director is trying to balance her work stress with her personal life. She spends 12 hours a day on set for 5 days a week, working on her latest film project. She wants to spend at least 20 hours a week with her family to maintain a healthy work-life balance. If the director manages to reduce her work hours by 2 hours each day, how many hours per day can she spend with her family during the work week?'}]}] | ifeval | multi | First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) |
ai2-adapt-dev/numinamath_tir_math_decontaminated_53452 | [
{
"content": "The number 1047 leaves a remainder of 23 when divided by \\( A \\), and a remainder of 7 when divided by \\( A+1 \\). Find \\( A \\). Do not include keywords ['bird', 'historian', 'song', 'whole'] in the response. In your response, words with all capital letters should appear less than 22 times. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Include keywords ['assistance', 'back', 'breakfast', 'weakness'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:forbidden_words', 'change_case:capital_word_frequency', 'detectable_format:json_format', 'keywords:existence'], 'kwargs': [{'forbidden_words': ['bird', 'historian', 'song', 'whole']}, {'capital_frequency': 22, 'capital_relation': 'less than'}, None, {'keywords': ['assistance', 'back', 'breakfast', 'weakness']}]}] | ifeval | multi | Do not include keywords ['bird', 'historian', 'song', 'whole'] in the response. In your response, words with all capital letters should appear less than 22 times. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Include keywords ['assistance', 'back', 'breakfast', 'weakness'] in the response. |
personas_math_easy_a72v4ub0hw9k8v0a44hjrlz5 | [
{
"content": "Sarah is a public relations specialist working in Washington State. She is organizing a press conference for a new environmental campaign. She has invited 200 reporters to the event, and each reporter will be given a packet containing 3 pamphlets, 2 pens, and 1 notebook. The cost of printing each pamphlet is $0.50, each pen costs $0.75, and each notebook is $1.25. Additionally, Sarah needs to rent a venue for the event, which costs $150. What is the total cost Sarah will incur for the press conference? The response must contain at least 8 placeholders represented by square brackets, such as [address]. In your response, the word value should appear less than 4 times. Do not include keywords ['brush', 'matter', 'resist', 'university'] in the response. Your response should contain less than 22 sentences. The response must contain at least 2 placeholders represented by square brackets, such as [address].",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:number_placeholders', 'keywords:frequency', 'keywords:forbidden_words', 'length_constraints:number_sentences', 'detectable_content:number_placeholders'], 'kwargs': [{'num_placeholders': 8}, {'keyword': 'value', 'frequency': 4, 'relation': 'less than'}, {'forbidden_words': ['brush', 'matter', 'resist', 'university']}, {'num_sentences': 22, 'relation': 'less than'}, {'num_placeholders': 2}]}] | ifeval | multi | The response must contain at least 8 placeholders represented by square brackets, such as [address]. In your response, the word value should appear less than 4 times. Do not include keywords ['brush', 'matter', 'resist', 'university'] in the response. Your response should contain less than 22 sentences. The response must contain at least 2 placeholders represented by square brackets, such as [address]. |
personahub_mf2pu2eod4hdp3zaczni6249 | [
{
"content": "A renowned ethnomusicologist is researching the migration patterns of a specific type of music over a period of 50 years across five regions: A, B, C, D, and E. Each region can influence music migration in a complex manner based on cultural exchange rates, which are represented as a matrix \\( M \\), where the entry \\( M_{ij} \\) represents the influence of region \\( i \\) on region \\( j \\).\n\n1. The initial distribution of the music genre's prevalence in each region is given by the vector \\( \\mathbf{v}_0 = [v_A, v_B, v_C, v_D, v_E] \\). The prevalence of music in each region over time is modeled by the sequence \\( \\mathbf{v}_n = M^n \\mathbf{v}_0 \\). Given that the matrix \\( M \\) has eigenvalues \\( \\lambda_1, \\lambda_2, \\lambda_3, \\lambda_4, \\lambda_5 \\) with \\( |\\lambda_1| > |\\lambda_2| \\geq |\\lambda_3| \\geq |\\lambda_4| \\geq |\\lambda_5| \\), determine the long-term distribution of the music genre across the regions. Assume \\( \\mathbf{v}_0 \\) is not orthogonal to the eigenvector corresponding to \\( \\lambda_1 \\).\n\n2. The ethnomusicologist also notices a periodic fluctuation in music popularity in region C, occurring every \\( T \\) years and is modeled by a sinusoidal function superimposed on the overall trend. If the fluctuation can be described by the function \\( f(t) = A \\sin \\left( \\frac{2\\pi}{T} t + \\phi \\right) \\), where \\( A \\) is the amplitude and \\( \\phi \\) is the phase shift, derive the expression for the total music prevalence in region C as a function of time, considering both the sinusoidal fluctuation and the long-term trend derived from part 1. Include keywords ['ability', 'give', 'nail', 'unit'] in the response. The response must contain at least 7 placeholders represented by square brackets, such as [address]. Your response should contain at least 8 sentences. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') There should be 5 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
]
| [{'instruction_id': ['keywords:existence', 'detectable_content:number_placeholders', 'length_constraints:number_sentences', 'detectable_format:constrained_response', 'length_constraints:number_paragraphs'], 'kwargs': [{'keywords': ['ability', 'give', 'nail', 'unit']}, {'num_placeholders': 7}, {'num_sentences': 8, 'relation': 'at least'}, None, {'num_paragraphs': 5}]}] | ifeval | multi | Include keywords ['ability', 'give', 'nail', 'unit'] in the response. The response must contain at least 7 placeholders represented by square brackets, such as [address]. Your response should contain at least 8 sentences. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') There should be 5 paragraphs. Paragraphs are separated with the markdown divider: *** |
personahub_dkrzum587gm991wczcaqzopp | [
{
"content": "As a dedicated Formula 1 fan and race correspondent, you are analyzing the performance of two Formula 1 cars during a recent race. Car A and Car B both start from the same point on a straight track. Car A accelerates from rest with a constant acceleration for a certain period, while Car B maintains a constant speed throughout the race.\n\n1. Car A accelerates from rest with an acceleration of \\(a\\) meters per second squared. After \\(t\\) seconds, the distance covered by Car A is given by the equation \\(d_A = \\frac{1}{2} a t^2\\). If the acceleration of Car A is 3 meters per second squared, and it covers a distance of 450 meters, find the time \\(t\\) it takes for Car A to cover this distance.\n\n2. Car B starts at the same time as Car A and maintains a constant speed of 60 meters per second. If Car A and Car B finish the 450-meter distance at the same time, set up an equation and solve for the time \\(t\\) it took both cars to complete the distance. Your entire response should be in English, and in all capital letters.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_capital'], 'kwargs': [None]}] | ifeval | multi | Your entire response should be in English, and in all capital letters. |
personahub_dhqquaeduju89r632s0be2h5 | [
{
"content": "A convicted felon named Alex is trying to improve the conditions in his correctional facility by advocating for more resources. He has discovered that the facility uses a complicated formula to calculate the daily cost of maintaining each inmate, which is based on the number of inmates and the resources allocated per inmate.\n\n1. The formula for the daily cost \\( C \\) per inmate is given by the polynomial equation: \n\n \\[\n C = 0.05n^2 + 2n + 50\n \\]\n\n where \\( n \\) is the number of inmates. If the facility currently has 120 inmates, calculate the daily cost per inmate.\n\n2. Alex is negotiating with the facility administration to increase the resources allocated per inmate. He proposes a new allocation model that involves solving the inequality:\n\n \\[\n 0.05n^2 + 2n + 50 < 75\n \\]\n\n Solve the inequality to find the range of \\( n \\) (number of inmates) for which the daily cost per inmate would be less than $75, enabling the facility to save money while reallocating some of it towards resources intended to improve inmate conditions. In your response, the word calendar should appear less than 5 times. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain exactly 2 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency', 'combination:two_responses', 'detectable_format:number_bullet_lists'], 'kwargs': [{'keyword': 'calendar', 'frequency': 5, 'relation': 'less than'}, None, {'num_bullets': 2}]}] | ifeval | multi | In your response, the word calendar should appear less than 5 times. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain exactly 2 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 |
ai2-adapt-dev/tulu_v3.9_aya_100k_11868 | [
{
"content": "\"Салат эмнеден жасалат?\"\n\n There should be 6 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 3 must start with word skin. Your answer must contain exactly 6 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 Do not include keywords ['dark', 'fortune', 'guide', 'stretch'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:nth_paragraph_first_word', 'detectable_format:number_bullet_lists', 'keywords:forbidden_words'], 'kwargs': [{'num_paragraphs': 6, 'nth_paragraph': 3, 'first_word': 'skin'}, {'num_bullets': 6}, {'forbidden_words': ['dark', 'fortune', 'guide', 'stretch']}]}] | ifeval | multi | There should be 6 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 3 must start with word skin. Your answer must contain exactly 6 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Do not include keywords ['dark', 'fortune', 'guide', 'stretch'] in the response. |
personahub_7hn4xn2i7rw393dq64yiqgxm | [
{
"content": "I am researching the history of the women's movement for a documentary project. Create a detailed outline for a 10-minute documentary script, focusing on three key historical events that significantly advanced gender equality. Make sure the outline includes 3 placeholders for customizable sections, such as [interview with an activist]. The first word of the second paragraph must be \"historically.\" Please write the entire outline in English and lowercase. At the end of your response, please explicitly add a postscript starting with Postscript: Answer with at least 401 words. Your entire response should be in English, and in all capital letters. Your response should contain at least 36 sentences. In your response, the letter u should appear at least 1 times.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'length_constraints:number_words', 'change_case:english_capital', 'length_constraints:number_sentences', 'keywords:letter_frequency'], 'kwargs': [{'postscript_marker': 'Postscript:'}, {'num_words': 401, 'relation': 'at least'}, None, {'num_sentences': 36, 'relation': 'at least'}, {'letter': 'u', 'let_frequency': 1, 'let_relation': 'at least'}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with Postscript: Answer with at least 401 words. Your entire response should be in English, and in all capital letters. Your response should contain at least 36 sentences. In your response, the letter u should appear at least 1 times. |
ai2-adapt-dev/tulu_v3.9_aya_100k_76917 | [
{
"content": "Oviana no niseho ny fanovana ny Lalàmpanorenana farany tao Madagasikara? Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. At the end of your response, please explicitly add a postscript starting with P.P.S.:",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:json_format', 'detectable_content:postscript'], 'kwargs': [None, {'postscript_marker': 'P.P.S.:'}]}] | ifeval | multi | Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. At the end of your response, please explicitly add a postscript starting with P.P.S.: |
ai2-adapt-dev/numinamath_tir_math_decontaminated_3665 | [
{
"content": "Given a sequence $\\{a_n\\}$ that satisfies $a_{n+2} = a_{n+1} - a_n$, and $a_1 = 2$, $a_2 = 3$, find the value of $a_{2017}$. Your response should contain less than 17 sentences.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_sentences'], 'kwargs': [{'num_sentences': 17, 'relation': 'less than'}]}] | ifeval | multi | Your response should contain less than 17 sentences. |
ai2-adapt-dev/tulu_v3.9_aya_100k_10044 | [
{
"content": "Kedu isi obodo Ondo steeti na mgbe e kere ya? Do not include keywords ['celebration', 'marriage', 'office', 'voice'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:forbidden_words'], 'kwargs': [{'forbidden_words': ['celebration', 'marriage', 'office', 'voice']}]}] | ifeval | multi | Do not include keywords ['celebration', 'marriage', 'office', 'voice'] in the response. |
ai2-adapt-dev/flan_v2_converted_16241 | [
{
"content": "Given a real-life anecdote of a complex ethical situation, identify who is wrong here. Classify each post into the following categories : AUTHOR (if the author of the post is wrong in the situation), OTHER (if any of the others mentioned in the paragraph is wrong), EVERYBODY (if everyone mentioned is wrong) and NOBODY (if no one is wrong in the situation).\n\nEx Input:\nI met a girl at an event a few weeks ago, and we really hit it off right away. We exchanged numbers and continued chatting, and within a few days decided we'd go out sometime soon. Since she lives sort of far away, though, we decided to meet up whenever is next convenient.\n\nHowever, over the next few weeks as we continue talking she began to become sort of clingy, saying things that I feel should only be said to someone in a relationship, such as \"I really need you right now.\" I asked her if we could slow down, and she said yes and apologized.\n\nA few days later, she mentioned that her last SO up and left her and she was worried I would too, especially since her and I don't live near each other. I told her I was sorry she felt that way but I wasn't comfortable talking hypotheticals about a relationship that doesn't even exist yet (as we haven't even gone on a first date), and that maybe it's better if we don't talk until a few days before we actually meet up (in a few weeks from now).\n\nShe seemed miffed but said yes and we left it at that. I know she feels like she messed up and made herself look like an idiot in front of me, and I like her a lot, but I just feel like she and I weren't moving at the same pace and I feel this is a way of getting her to slow down to my speed. Maybe I shouldn't have even been texting her before the first date, but we were speaking the entire event and it just seemed like a natural continuation of that event.\n\nSo, AITA for asking if we could not talk until we get closer to the date of our first date?\n\nEx Output:\nNOBODY\n\n\nEx Input:\nI (19M) have a friend the same age who I used to be inseperable with. We would do everything together and had a pretty good grouo of friends. We both had mutual crushes on each other and would go on trips together almost every week, and it would consist of getting really drunk and high, and having really funny or nice nights together. We literally never left each other’s sides.\n\nRecently, he decided to quit college and move in with his boyfriend of a year or so. His boyfriend and I have both agreed that we don’t really like each other for whatever reason but I’ve always been civil and nice to him, since it’s none of my business who he dates.\n\nThe real problem started before his boyfriend anyway. My feiend attended college outside my city, so we didn’t talk often as we both don’t really text. We would still hang out when we were both free though, and it was fine. Until he stopped seeing or talking to me altogether. He just does not answer his texts or calls and whenever I or my friends message to ask him to hang, we always get ignored because he’s with his boyfriend and his friends.\n\nThis angered me so badly and I also pretty much gave up on him too, but we bumped into each other again recently. We decided to hang out (My friend, a mutual friend, and me) and the whole time he pretty much spoke about himself and would make our mutual friend the butt of all his jokes. This is something of a running joke and our mutual friend laughs it off and doesn’t think anything of it, but that night I could tell they were getting angry and uncomfortable. They left, and I stayed as I figured that maybe being alone with him would resolve my anger and we could go back to being good friends.\n\nThen his boyfriend came home. He sat with us, which was okay since I’m good at being civil and don’t really care, but he began to get increasingly rude to my friend and I. He would make a joke, and when I would laugh he would ask what’s so funny and make things incredibly awkward. He also ignored or laughed obnoxiously at everything I said and after my friend noticing I was incredibly uncomfortable (I rarely leave my house because of my crippling anxiety and he is more than aware of this) but not defending me even slightly I just left. I haven’t messaged, phoned or even replied to his snapchats since, but I do feel really guilty about ghosting him since we were so close. Am I the asshole here?\n\nTL;DR - My close friend doesn’t bother to contact me anymore and his friends make it clear they dislike me, to which he doesn’t care to defend me at all.\n\nEx Output:\nOTHER\n\n\nEx Input:\nSo me and my gf have been dating for a little below 5 months and I love her dearly. She's supportive, caring, affectionate, pretty much everything I'd want in a girlfriend. \n\nWe were having a discussion about her future with her company and it might have her relocate her job to an area that's around 1 1/2 hrs commute from me which is pretty far if she takes the job. She currently lives ~ 20 minutes away and we've had a little bumpy but eventually straightened relationship. I told her that there are a lot of things she could do such as her relocate to a place that's not so far from her, etc. but one of the things I mentioned is that long distance might not be the best thing for us. I really do like her and I want to stay grounded here but I know I come across as blunt sometimes and I don't know if this is something that I should have brought up.\n\nShe thought I was joking but I was serious, since we're only less than 5 months into the relationship. She got immediately upset and mad and said that \"I don't love her enough to consider a LDR\" and \"How could I even say something like that\". I went to end the conversation by saying \"I love you\" and she responded with \"Apparently not\". \n\nSo reddit, am I the asshole here? Is there something I'm missing\n\nEx Output:\n At the end of your response, please explicitly add a postscript starting with Postscript In your response, words with all capital letters should appear less than 29 times. In your response, the letter t should appear at least 20 times. There should be 8 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 3 must start with word bench.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'change_case:capital_word_frequency', 'keywords:letter_frequency', 'length_constraints:nth_paragraph_first_word'], 'kwargs': [{'postscript_marker': 'Postscript'}, {'capital_frequency': 29, 'capital_relation': 'less than'}, {'letter': 't', 'let_frequency': 20, 'let_relation': 'at least'}, {'num_paragraphs': 8, 'nth_paragraph': 3, 'first_word': 'bench'}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with Postscript In your response, words with all capital letters should appear less than 29 times. In your response, the letter t should appear at least 20 times. There should be 8 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 3 must start with word bench. |
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_49292 | [
{
"content": "Can you show me the most effective way to tackle opponents in a football game like FIFA? In your response, the word creative should appear less than 4 times. Answer with at least 292 words. In your response, words with all capital letters should appear at least 31 times. In your entire response, refrain from the use of any commas.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency', 'length_constraints:number_words', 'change_case:capital_word_frequency', 'punctuation:no_comma'], 'kwargs': [{'keyword': 'creative', 'frequency': 4, 'relation': 'less than'}, {'num_words': 292, 'relation': 'at least'}, {'capital_frequency': 31, 'capital_relation': 'at least'}, None]}] | ifeval | multi | In your response, the word creative should appear less than 4 times. Answer with at least 292 words. In your response, words with all capital letters should appear at least 31 times. In your entire response, refrain from the use of any commas. |
ai2-adapt-dev/flan_v2_converted_77906 | [
{
"content": "Sentence 1: The strong male partner in the water dancing team held his female partner above his head while she posed acrobatically over his head. Sentence 2: The man and woman are performing a dance routine for a competition.\nThe question and answer are below.\nPremise: \"The strong male partner in the water dancing team held his female partner above his head while she posed acrobatically over his head.\"\nHypothesis: \"The man and woman are performing a dance routine for a competition.\"\nDo we know that the hypothesis entailed by the premise?\nit is not possible to tell\n\nThe mayor of a town wearing a red hat and driving a yellow and green patterned truck waves to the crowd in a victory parade.. So what could be the question?\nQuestion followed by answer: Premise: \"In a parade.\"\nHypothesis: \"A man wearing a red hat and driving a yellow and green old-fashioned truck is waving at the crowd.\"\nDo we know that the hypothesis entailed by the premise?\nit is not possible to tell\n\nA black coat cannot be termed as a pink dress. A woman walking a dog need not be her own dog.\nThe question and answer are below.\nPremise: \"A man wearing a teal jacket and holding a pink umbrella walking next to a woman in a black coat walking a dog.\"\nHypothesis: \"A woman in a pink dress is walking her dog.\"\nDo we know that the hypothesis entailed by the premise?\nit is not possible to tell\n\nPeople ride motorcycles on a street does not imply they participate in tub run.\nThe question and answer are below.\nGiven the sentence \"People ride motorcycles on a street.\" can we conclude that \"People participate in tub run.\"?\nOptions:\n- yes\n- it is not possible to tell\n- no\nit is not possible to tell\n\nOne can not be physically young and old at the same time.. So what could be the question?\nQuestion followed by answer: If \"Two young girls in shorts wading in the ocean.\" does that mean that \"Two old ladies are wading in the ocean.\"?\nOptions:\n- yes\n- it is not possible to tell\n- no\nno\n\nRuns in a field does not necessarily mean chasing after a stick thrown by it's owner.. So what could be the question?\nQuestion followed by answer: Include keywords ['evidence', 'friendship', 'manner', 'wine'] in the response. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Your entire response should be in English, and in all capital letters.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:existence', 'combination:repeat_prompt', 'change_case:english_capital'], 'kwargs': [{'keywords': ['evidence', 'friendship', 'manner', 'wine']}, {'prompt_to_repeat': 'Sentence 1: The strong male partner in the water dancing team held his female partner above his head while she posed acrobatically over his head. Sentence 2: The man and woman are performing a dance routine for a competition.\nThe question and answer are below.\nPremise: "The strong male partner in the water dancing team held his female partner above his head while she posed acrobatically over his head."\nHypothesis: "The man and woman are performing a dance routine for a competition."\nDo we know that the hypothesis entailed by the premise?\nit is not possible to tell\n\nThe mayor of a town wearing a red hat and driving a yellow and green patterned truck waves to the crowd in a victory parade.. So what could be the question?\nQuestion followed by answer: Premise: "In a parade."\nHypothesis: "A man wearing a red hat and driving a yellow and green old-fashioned truck is waving at the crowd."\nDo we know that the hypothesis entailed by the premise?\nit is not possible to tell\n\nA black coat cannot be termed as a pink dress. A woman walking a dog need not be her own dog.\nThe question and answer are below.\nPremise: "A man wearing a teal jacket and holding a pink umbrella walking next to a woman in a black coat walking a dog."\nHypothesis: "A woman in a pink dress is walking her dog."\nDo we know that the hypothesis entailed by the premise?\nit is not possible to tell\n\nPeople ride motorcycles on a street does not imply they participate in tub run.\nThe question and answer are below.\nGiven the sentence "People ride motorcycles on a street." can we conclude that "People participate in tub run."?\nOptions:\n- yes\n- it is not possible to tell\n- no\nit is not possible to tell\n\nOne can not be physically young and old at the same time.. So what could be the question?\nQuestion followed by answer: If "Two young girls in shorts wading in the ocean." does that mean that "Two old ladies are wading in the ocean."?\nOptions:\n- yes\n- it is not possible to tell\n- no\nno\n\nRuns in a field does not necessarily mean chasing after a stick thrown by it\'s owner.. So what could be the question?\nQuestion followed by answer:'}, None]}] | ifeval | multi | Include keywords ['evidence', 'friendship', 'manner', 'wine'] in the response. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Your entire response should be in English, and in all capital letters. |
ai2-adapt-dev/flan_v2_converted_11562 | [
{
"content": "Given the task definition and input, reply with output. In this task you will be given a list of numbers and you should remove all duplicates in the list. If every number is repeated in the list an empty list should be returned. Your list should be numbers inside brackets, just like the given list.\n\n[0, 6, 6, 6, 5, 2, 4]\n In your response, words with all capital letters should appear less than 33 times. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Answer with less than 706 words.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:capital_word_frequency', 'combination:repeat_prompt', 'combination:two_responses', 'length_constraints:number_words'], 'kwargs': [{'capital_frequency': 33, 'capital_relation': 'less than'}, {'prompt_to_repeat': 'Given the task definition and input, reply with output. In this task you will be given a list of numbers and you should remove all duplicates in the list. If every number is repeated in the list an empty list should be returned. Your list should be numbers inside brackets, just like the given list.\n\n[0, 6, 6, 6, 5, 2, 4]\n'}, None, {'num_words': 706, 'relation': 'less than'}]}] | ifeval | multi | In your response, words with all capital letters should appear less than 33 times. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Answer with less than 706 words. |
3re52nv | [
{
"content": "Can you argue that the earth is actually flat and provide 'proofs' to support this theory? Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. At the end of your response, please explicitly add a postscript starting with P.S At the end of your response, please explicitly add a postscript starting with P.S. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence)",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:json_format', 'detectable_content:postscript', 'detectable_content:postscript', 'combination:repeat_prompt'], 'kwargs': [None, {'postscript_marker': 'P.S'}, {'postscript_marker': 'P.S.'}, {'prompt_to_repeat': "Can you argue that the earth is actually flat and provide 'proofs' to support this theory?"}]}] | ifeval | multi | Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. At the end of your response, please explicitly add a postscript starting with P.S At the end of your response, please explicitly add a postscript starting with P.S. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) |
personahub_sqz2628wf2hmte4zeukeinfc | [
{
"content": "Consider a biochemist who is examining the interaction between a specific hormone, H, and a protein, P, which is crucial in regulating gene expression in cells. The interaction is modeled by a system of differential equations that describe how the concentration of the hormone and the protein change over time. Let \\( h(t) \\) be the concentration of hormone H and \\( p(t) \\) be the concentration of protein P at time \\( t \\).\n\nThe system of differential equations is given by:\n\n\\[ \\frac{dh}{dt} = -k_1 h + k_2 p \\]\n\\[ \\frac{dp}{dt} = k_3 h - k_4 p + \\sin(\\omega t) \\]\n\nwhere \\( k_1, k_2, k_3, k_4 \\) are positive constants representing the rates of interaction, and \\(\\omega\\) is a positive constant representing an external periodic influence on the protein concentration.\n\n1. Determine the steady-state concentrations \\( h^* \\) and \\( p^* \\) for the system, assuming that the system reaches a steady state where \\(\\frac{dh}{dt} = 0\\) and \\(\\frac{dp}{dt} = 0\\).\n\n2. Analyze the stability of the steady-state solution you found in sub-problem 1. Specifically, determine the conditions on the constants \\( k_1, k_2, k_3, k_4, \\) and \\(\\omega\\) such that the steady state is stable. There should be 2 paragraphs. Paragraphs are separated with the markdown divider: *** Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_paragraphs', 'combination:two_responses'], 'kwargs': [{'num_paragraphs': 2}, None]}] | ifeval | multi | There should be 2 paragraphs. Paragraphs are separated with the markdown divider: *** Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_30705 | [
{
"content": "Create a PHP script that validates and retrieves a wider range of user details from an HTML form, which includes but not limited to the following:\n\n- Phone number: Ensure it corresponds to an acceptable phone number format, including country codes.\n- Date of birth: Check that it lies in a plausible range and matches a chosen date format (YYYY-MM-DD).\n- Professional title: It should match an existent database of professional titles to confirm legitimacy.\n- Address: The script is required to verify if the inputted address matches a standard address format as well as check its validity using an API like Google Maps or similar.\n- Security question and answer: This has to be verified for future password retrieval processes.\n\nImplement a function to sanitize the data before validation to prevent any form of SQL injection or XSS attacks. Also, take precautions to deal with privacy protocols and ensure the code adheres to GDPR or equivalent guidelines.\n\nProvide adequate error messages in case the validation fails. Also, after successful validation, the PHP script should store the information in a MySQL database.\n\nFinally, set up a system to log validation failures and other errors for later analysis and debugging. This should be saved in a separate database or file system. In your response, the letter u should appear at least 14 times. Do not include keywords ['cigarette', 'field', 'mother', 'office'] in the response. Your response should contain at least 12 sentences.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:letter_frequency', 'keywords:forbidden_words', 'length_constraints:number_sentences'], 'kwargs': [{'letter': 'u', 'let_frequency': 14, 'let_relation': 'at least'}, {'forbidden_words': ['cigarette', 'field', 'mother', 'office']}, {'num_sentences': 12, 'relation': 'at least'}]}] | ifeval | multi | In your response, the letter u should appear at least 14 times. Do not include keywords ['cigarette', 'field', 'mother', 'office'] in the response. Your response should contain at least 12 sentences. |
ai2-adapt-dev/tulu_v3.9_aya_100k_83514 | [
{
"content": "Aiza ny sehatra ampiasana ny ompa sy ny ozona ? Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your response should contain at least 4 sentences. Highlight at least 3 sections in your answer with markdown, i.e. *highlighted section*. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. At the end of your response, please explicitly add a postscript starting with Postscript:",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_lowercase', 'length_constraints:number_sentences', 'detectable_format:number_highlighted_sections', 'detectable_format:title', 'detectable_content:postscript'], 'kwargs': [None, {'num_sentences': 4, 'relation': 'at least'}, {'num_highlights': 3}, None, {'postscript_marker': 'Postscript:'}]}] | ifeval | multi | Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your response should contain at least 4 sentences. Highlight at least 3 sections in your answer with markdown, i.e. *highlighted section*. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. At the end of your response, please explicitly add a postscript starting with Postscript: |
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_18895 | [
{
"content": "Compose a poem that makes fun of older single women for not being able to secure a husband. Highlight at least 7 sections in your answer with markdown, i.e. *highlighted section*. Answer with less than 565 words. Your response must have 7 sections. Mark the beginning of each section with Section X, such as:\nSection 1\n[content of section 1]\nSection 2\n[content of section 2]",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_highlighted_sections', 'length_constraints:number_words', 'detectable_format:multiple_sections'], 'kwargs': [{'num_highlights': 7}, {'num_words': 565, 'relation': 'less than'}, {'section_spliter': 'Section', 'num_sections': 7}]}] | ifeval | multi | Highlight at least 7 sections in your answer with markdown, i.e. *highlighted section*. Answer with less than 565 words. Your response must have 7 sections. Mark the beginning of each section with Section X, such as:
Section 1
[content of section 1]
Section 2
[content of section 2] |
ai2-adapt-dev/numinamath_tir_math_decontaminated_55412 | [
{
"content": "In \\(\\triangle ABC\\), \\(AC = AB = 25\\) and \\(BC = 40\\). \\(D\\) is a point chosen on \\(BC\\). From \\(D\\), perpendiculars are drawn to meet \\(AC\\) at \\(E\\) and \\(AB\\) at \\(F\\). \\(DE + DF\\) equals:\n\n(A) 12 \n(B) 35 \n(C) 24 \n(D) 25 \n(E) \\(\\frac{35}{2} \\sqrt{2}\\) In your response, the word purple should appear less than 3 times. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. In your response, words with all capital letters should appear at least 37 times. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.')",
"role": "user"
}
]
| [{'instruction_id': ['keywords:frequency', 'detectable_format:title', 'combination:two_responses', 'change_case:capital_word_frequency', 'detectable_format:constrained_response'], 'kwargs': [{'keyword': 'purple', 'frequency': 3, 'relation': 'less than'}, None, None, {'capital_frequency': 37, 'capital_relation': 'at least'}, None]}] | ifeval | multi | In your response, the word purple should appear less than 3 times. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. In your response, words with all capital letters should appear at least 37 times. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') |
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_19433 | [
{
"content": "Edmonton is 220 kilometers north of Red Deer. Calgary is 110 kilometers south of Red Deer. If you travel at 110 kilometers per hour, how long will it take to get from Edmonton to Calgary? Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*. Answer with less than 926 words. At the end of your response, please explicitly add a postscript starting with P.S.: Finish your response with this exact phrase That's my answer.. No other words should follow this phrase.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_highlighted_sections', 'length_constraints:number_words', 'detectable_content:postscript', 'startend:end_checker'], 'kwargs': [{'num_highlights': 6}, {'num_words': 926, 'relation': 'less than'}, {'postscript_marker': 'P.S.:'}, {'end_phrase': "That's my answer."}]}] | ifeval | multi | Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*. Answer with less than 926 words. At the end of your response, please explicitly add a postscript starting with P.S.: Finish your response with this exact phrase That's my answer.. No other words should follow this phrase. |
personahub_myppxcjwrdoxev427gsa1nhm | [
{
"content": "A beginning litigator on the East Coast of the United States is preparing for a complex case involving multiple plaintiffs and defendants. During the discovery phase, the litigator needs to analyze the communication patterns between 10 different parties over a period of 30 days. Each party sends and receives messages at a variable rate that depends on both the time of day and the day of the week. \n\nLet \\( M_{i,j}(t) \\) represent the number of messages sent from party \\( i \\) to party \\( j \\) at time \\( t \\), where \\( t \\) is measured in hours from the start of the 30-day period (i.e., \\( 0 \\leq t < 720 \\) hours). The function \\( M_{i,j}(t) \\) is modeled by the following piecewise functions, where \\( A_{i,j} \\) and \\( B_{i,j} \\) are constants unique to each pair \\( (i, j) \\):\n\n\\[ M_{i,j}(t) = \n\\begin{cases} \nA_{i,j} \\sin\\left(\\frac{\\pi t}{12}\\right) & \\text{if } 0 \\leq t \\mod 24 < 12 \\\\\nB_{i,j} \\cos\\left(\\frac{\\pi t}{12}\\right) & \\text{if } 12 \\leq t \\mod 24 < 24 \n\\end{cases}\n\\]\n\n1. Calculate the total number of messages sent from party 1 to party 2 over the entire 30-day period, given \\( A_{1,2} = 5 \\) and \\( B_{1,2} = 3 \\).\n\n2. Determine the time \\( t \\) within the first 24 hours when the number of messages sent from party 3 to party 4 is at its maximum, given \\( A_{3,4} = 6 \\) and \\( B_{3,4} = 4 \\). In your response, the letter y should appear less than 3 times. The response must contain at least 3 placeholders represented by square brackets, such as [address]. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. At the end of your response, please explicitly add a postscript starting with Postscript Answer with less than 562 words.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:letter_frequency', 'detectable_content:number_placeholders', 'detectable_format:json_format', 'detectable_content:postscript', 'length_constraints:number_words'], 'kwargs': [{'letter': 'y', 'let_frequency': 3, 'let_relation': 'less than'}, {'num_placeholders': 3}, None, {'postscript_marker': 'Postscript'}, {'num_words': 562, 'relation': 'less than'}]}] | ifeval | multi | In your response, the letter y should appear less than 3 times. The response must contain at least 3 placeholders represented by square brackets, such as [address]. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. At the end of your response, please explicitly add a postscript starting with Postscript Answer with less than 562 words. |
personahub_5n1uxjmatfxpb9cca8g2m5dg | [
{
"content": "A politician is advocating for rapid city development and economic growth by proposing a new urban development project. The proposal involves constructing a mix of commercial and residential buildings over a 100-acre plot of land. The commercial buildings are expected to generate a continuous annual revenue stream, modeled by the function \\( R(t) = 2e^{0.05t} \\) million dollars per year, where \\( t \\) is the number of years after the project starts. The residential buildings are projected to house a growing population, with the population size modeled by \\( P(t) = 5000 + 1000\\sin\\left(\\frac{\\pi}{5}t\\right) \\), where \\( t \\) is also the number of years after the project starts.\n\n1. Determine the total revenue generated from the commercial buildings over the first 10 years. Use integration to find the exact value.\n\n2. Assume that the city needs to plan infrastructure to support the maximum projected residential population over the 10-year period. Calculate the maximum population size and the year in which it occurs. Answer with at least 852 words. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence)",
"role": "user"
}
]
| [{'instruction_id': ['length_constraints:number_words', 'combination:repeat_prompt'], 'kwargs': [{'num_words': 852, 'relation': 'at least'}, {'prompt_to_repeat': 'A politician is advocating for rapid city development and economic growth by proposing a new urban development project. The proposal involves constructing a mix of commercial and residential buildings over a 100-acre plot of land. The commercial buildings are expected to generate a continuous annual revenue stream, modeled by the function \\( R(t) = 2e^{0.05t} \\) million dollars per year, where \\( t \\) is the number of years after the project starts. The residential buildings are projected to house a growing population, with the population size modeled by \\( P(t) = 5000 + 1000\\sin\\left(\\frac{\\pi}{5}t\\right) \\), where \\( t \\) is also the number of years after the project starts.\n\n1. Determine the total revenue generated from the commercial buildings over the first 10 years. Use integration to find the exact value.\n\n2. Assume that the city needs to plan infrastructure to support the maximum projected residential population over the 10-year period. Calculate the maximum population size and the year in which it occurs.'}]}] | ifeval | multi | Answer with at least 852 words. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) |
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_35037 | [
{
"content": "Emily has 400 feet of rope. She needs to enclose a rectangular area that is 30 feet wide and 80 feet long. How much rope will Emily have left over after wrapping it around the area once? Your answer must contain exactly 8 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 Your answer must contain exactly 4 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:number_bullet_lists', 'detectable_format:number_bullet_lists'], 'kwargs': [{'num_bullets': 8}, {'num_bullets': 4}]}] | ifeval | multi | Your answer must contain exactly 8 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Your answer must contain exactly 4 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 |
personahub_obtd46i5q5fojwo735ssagvs | [
{
"content": "The Writing Center at a university is analyzing the improvement in students' writing skills over a semester. The coordinator, who is proficient in both educational theory and statistical analysis, decides to model the improvement using a function that takes into account various factors: the number of hours spent in writing sessions, the number of writing resources utilized, and the initial skill level of the students.\n\nLet \\( W(t, r, s) \\) represent the writing skill level of a student after \\( t \\) hours of writing sessions, using \\( r \\) resources, and starting with an initial skill level \\( s \\). The function is modeled as follows:\n\n\\[ W(t, r, s) = s + \\alpha t^\\beta \\ln(r+1) \\]\n\nwhere \\( \\alpha \\) and \\( \\beta \\) are constants that need to be determined based on empirical data.\n\nSub-problem 1:\nGiven that a student with an initial skill level of \\( s = 50 \\) improved to a skill level of \\( 70 \\) after 10 hours of writing sessions using 5 resources, and another student with an initial skill level of \\( s = 40 \\) improved to a skill level of \\( 65 \\) after 15 hours of writing sessions using 10 resources, find the values of \\( \\alpha \\) and \\( \\beta \\).\n\nSub-problem 2:\nUsing the values of \\( \\alpha \\) and \\( \\beta \\) obtained from Sub-problem 1, determine the projected skill level for a student who starts with an initial skill level of \\( s = 60 \\), spends 12 hours in writing sessions, and utilizes 8 resources. Your response must have 6 sections. Mark the beginning of each section with Paragraph X, such as:\nParagraph 1\n[content of section 1]\nParagraph 2\n[content of section 2] Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Do not include keywords ['guest', 'mood', 'net', 'theory'] in the response.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:multiple_sections', 'combination:two_responses', 'detectable_format:title', 'keywords:forbidden_words'], 'kwargs': [{'section_spliter': 'Paragraph', 'num_sections': 6}, None, None, {'forbidden_words': ['guest', 'mood', 'net', 'theory']}]}] | ifeval | multi | Your response must have 6 sections. Mark the beginning of each section with Paragraph X, such as:
Paragraph 1
[content of section 1]
Paragraph 2
[content of section 2] Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Do not include keywords ['guest', 'mood', 'net', 'theory'] in the response. |
personas_math_easy_zi8jg9diviqfxzbows4o2avg | [
{
"content": "A politician is working on a housing reform initiative to build new affordable homes for the community. The plan is to build 5 housing complexes, each with 20 apartments. Each apartment requires 8 windows to be installed. The politician collaborates with the community, and they agree to install 10 windows per day. How many days will it take to install all the windows for the new housing complexes? In your response, the letter n should appear less than 11 times.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:letter_frequency'], 'kwargs': [{'letter': 'n', 'let_frequency': 11, 'let_relation': 'less than'}]}] | ifeval | multi | In your response, the letter n should appear less than 11 times. |
ai2-adapt-dev/tulu_v3.9_aya_100k_36714 | [
{
"content": "Inona no ilazana ny toerana fatorian'ny omby? Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.')",
"role": "user"
}
]
| [{'instruction_id': ['detectable_format:constrained_response'], 'kwargs': [None]}] | ifeval | multi | Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.', 'I think yes.', 'I think no.', 'I think maybe.') |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_88710 | [
{
"content": "...or are they the same thing? I notice that each has its own Wikipedia entry: [Polymorphism](http://en.wikipedia.org/wiki/Polymorphism_(computer_science)), [Multiple Dispatch](http://en.wikipedia.org/wiki/Multiple_dispatch), but I'm having trouble seeing how the concepts differ.\n\n And how does [Overloading](http://en.wikipedia.org/wiki/Overloaded) fit into all this? Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your entire response, refrain from the use of any commas.",
"role": "user"
}
]
| [{'instruction_id': ['startend:end_checker', 'punctuation:no_comma'], 'kwargs': [{'end_phrase': "That's all I have to say."}, None]}] | ifeval | multi | Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your entire response, refrain from the use of any commas. |
personahub_x94lzkks4rxw6474mwwt6tsa | [
{
"content": "A biologist is studying a natural wetland used for water treatment. The biologist models the rate of nutrient removal \\(N(t)\\) from the water as a function of time \\(t\\) (in days) using a differential equation. The model assumes that the rate of nutrient removal is proportional to the current amount of nutrient in the water and inversely proportional to the square root of the time elapsed. The initial amount of nutrient in the water is \\(N_0\\).\n\n1. Formulate the differential equation that describes the rate of nutrient removal \\(N(t)\\) as a function of \\(t\\). Solve this differential equation to find \\(N(t)\\).\n\n2. Suppose the wetland reaches a steady state where the nutrient removal rate stabilizes. Determine the time \\(t_s\\) (in days) at which the nutrient level \\(N(t_s)\\) stabilizes to 10% of the initial nutrient level \\(N_0\\). In your response, words with all capital letters should appear less than 1 times. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your entire response should be in English, and in all capital letters. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:capital_word_frequency', 'change_case:english_lowercase', 'change_case:english_capital', 'combination:two_responses', 'detectable_format:title'], 'kwargs': [{'capital_frequency': 1, 'capital_relation': 'less than'}, None, None, None, None]}] | ifeval | multi | In your response, words with all capital letters should appear less than 1 times. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your entire response should be in English, and in all capital letters. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. |
ai2-adapt-dev/tulu_v3.9_aya_100k_35753 | [
{
"content": "O jakim wydarzeniu mówi ten tekst: Swój największy międzynarodowy sukces osiągnęła podczas mistrzostw świata w Ciocco w 1991 roku, kiedy zdobyła mistrzostwo świata w cross country. The response must contain at least 7 placeholders represented by square brackets, such as [address].",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:number_placeholders'], 'kwargs': [{'num_placeholders': 7}]}] | ifeval | multi | The response must contain at least 7 placeholders represented by square brackets, such as [address]. |
personahub_ya0qk1g33zw3rflavqqnmuu4 | [
{
"content": "A senior software engineer is designing a usability testing methodology for a new software application. The application has a total of \\( n \\) features, and each feature can be tested for usability by a panel of experts. The usability scores are determined on a scale from 1 to 10 for each feature.\n\n1. Suppose the senior engineer wants to statistically analyze the usability scores by applying a multivariate analysis of variance (MANOVA). The engineer decides to randomly select \\( k \\) features from the \\( n \\) total features for a detailed study. If the probability that a randomly selected set of \\( k \\) features will have a combined average score greater than 8 is \\( p \\), express \\( p \\) in terms of \\( n \\), \\( k \\), and the combined probability distribution of scores for each feature.\n\n2. The senior engineer also needs to optimize the selection of \\( k \\) features such that the variance of the combined average usability score is minimized. Given that the variance of the usability score for each feature \\( i \\) is \\( \\sigma_i^2 \\), formulate an expression for the optimal selection of \\( k \\) features that minimizes the variance of the combined average score, and describe any mathematical techniques that could assist in solving this optimization problem. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Include keywords ['difficulty', 'life', 'match', 'youth'] in the response. Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase.",
"role": "user"
}
]
| [{'instruction_id': ['combination:two_responses', 'keywords:existence', 'startend:end_checker'], 'kwargs': [None, {'keywords': ['difficulty', 'life', 'match', 'youth']}, {'end_phrase': 'Is there anything else I can help with?'}]}] | ifeval | multi | Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Include keywords ['difficulty', 'life', 'match', 'youth'] in the response. Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_1933 | [
{
"content": "help me port this code over to php:\nimport string\n\ndef flip_case_mod_nums_dup_symbols_sort(string: str) -> str:\n result = []\n special_chars = set(string) - set(string.lower()) - set(string.upper()) - set(string.digits)\n for c in string:\n if c.isalpha():\n result.append(c.swapcase())\n elif c.isdigit():\n c = str(int(c) + 1) if int(c) % 2 != 0 else c\n result.append(c)\n elif c in special_chars:\n result.append(c*2)\n return ''.join(sorted(result))\n\nprint(flip_case_mod_nums_dup_symbols_sort('Hello World4!'))\nwould be nice if you can include some explaination of what the translated code does In your response, the letter z should appear less than 6 times. In your response, the letter e should appear at least 5 times. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Answer with less than 801 words.",
"role": "user"
}
]
| [{'instruction_id': ['keywords:letter_frequency', 'keywords:letter_frequency', 'detectable_format:json_format', 'combination:repeat_prompt', 'length_constraints:number_words'], 'kwargs': [{'letter': 'z', 'let_frequency': 6, 'let_relation': 'less than'}, {'letter': 'e', 'let_frequency': 5, 'let_relation': 'at least'}, None, {'prompt_to_repeat': "help me port this code over to php:\nimport string\n\ndef flip_case_mod_nums_dup_symbols_sort(string: str) -> str:\n result = []\n special_chars = set(string) - set(string.lower()) - set(string.upper()) - set(string.digits)\n for c in string:\n if c.isalpha():\n result.append(c.swapcase())\n elif c.isdigit():\n c = str(int(c) + 1) if int(c) % 2 != 0 else c\n result.append(c)\n elif c in special_chars:\n result.append(c*2)\n return ''.join(sorted(result))\n\nprint(flip_case_mod_nums_dup_symbols_sort('Hello World4!'))\nwould be nice if you can include some explaination of what the translated code does"}, {'num_words': 801, 'relation': 'less than'}]}] | ifeval | multi | In your response, the letter z should appear less than 6 times. In your response, the letter e should appear at least 5 times. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. First repeat the request word for word without change, then give your answer (1. do not say any words or characters before repeating the request; 2. the request you need to repeat does not include this sentence) Answer with less than 801 words. |
ai2-adapt-dev/flan_v2_converted_12870 | [
{
"content": "Part 1. Definition\nIn this task you will be given some text dialogue and you need to infer the underlying emotion of the text. The possible emotions are happy, sad, angry, or other.\nPart 2. Example\ni don't work i could take your shift i am a student \nAnswer: other\nExplanation: In this dialogue someone is informing another person that they could take their shift. There is no happy, sad, or angry emotion in this text. So this is a good example.\nPart 3. Exercise\nwhy here raining too where are you \nAnswer: At the end of your response, please explicitly add a postscript starting with P.S.: The response must contain at least 7 placeholders represented by square brackets, such as [address]. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your response should contain less than 10 sentences.",
"role": "user"
}
]
| [{'instruction_id': ['detectable_content:postscript', 'detectable_content:number_placeholders', 'change_case:english_lowercase', 'length_constraints:number_sentences'], 'kwargs': [{'postscript_marker': 'P.S.:'}, {'num_placeholders': 7}, None, {'num_sentences': 10, 'relation': 'less than'}]}] | ifeval | multi | At the end of your response, please explicitly add a postscript starting with P.S.: The response must contain at least 7 placeholders represented by square brackets, such as [address]. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your response should contain less than 10 sentences. |
ai2-adapt-dev/no_robots_converted_5928 | [
{
"content": "Write a story beginning in the middle of an event; feature a character who can only speak in quotes from Sam Raimi’s trilogy of Spider-Man films that confuses the main character, Bruce. Wrap your entire response with double quotation marks. At the end of your response, please explicitly add a postscript starting with P.P.S",
"role": "user"
}
]
| [{'instruction_id': ['startend:quotation', 'detectable_content:postscript'], 'kwargs': [None, {'postscript_marker': 'P.P.S'}]}] | ifeval | multi | Wrap your entire response with double quotation marks. At the end of your response, please explicitly add a postscript starting with P.P.S |
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_52575 | [
{
"content": "Given a malfunctioning SQL expression, identify the issues, correct it, then construct a refined SQL query. The ultimate goal of your query is to compute, not only the mean numerical age, but also the median and mode ages of all individuals registered in the 'users' database table. Additionally, account for possible variations in date of birth format and missing data. \n\nMalfunctioning SQL expression: \n\n```sql\nSELECT AVERAGE(age) FROM users\n```\n\nNote: Assume that the user database table has columns for username, date of birth, registration date, etc. Your entire response should be in English, and in all capital letters.",
"role": "user"
}
]
| [{'instruction_id': ['change_case:english_capital'], 'kwargs': [None]}] | ifeval | multi | Your entire response should be in English, and in all capital letters. |
ai2-adapt-dev/tulu_v3.9_aya_100k_82444 | [
{
"content": "Kalimah: Minggon payun Pérsib tanding ngalawan Pérsebaya, tapi hanjakal kuring teu bisa nonton di stadion kusabab nyeri suku.\nPatanya: Kumaha séntimén tina kalimah di luhur? Wrap your entire response with double quotation marks. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your entire response, refrain from the use of any commas. Your entire response should be in English, and in all capital letters.",
"role": "user"
}
]
| [{'instruction_id': ['startend:quotation', 'startend:end_checker', 'punctuation:no_comma', 'change_case:english_capital'], 'kwargs': [None, {'end_phrase': "That's all I have to say."}, None, None]}] | ifeval | multi | Wrap your entire response with double quotation marks. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your entire response, refrain from the use of any commas. Your entire response should be in English, and in all capital letters. |
ai2-adapt-dev/tulu_v3.9_aya_100k_68704 | [
{
"content": "मराठ्यांच्या स्वातंत्र्य युध्दाचे नेतृत्व कोणी केले ? In your entire response, refrain from the use of any commas. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Highlight at least 7 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
]
| [{'instruction_id': ['punctuation:no_comma', 'detectable_format:title', 'detectable_format:number_highlighted_sections'], 'kwargs': [None, None, {'num_highlights': 7}]}] | ifeval | multi | In your entire response, refrain from the use of any commas. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Highlight at least 7 sections in your answer with markdown, i.e. *highlighted section*. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.