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
|
|---|---|---|---|---|---|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_45101
|
[
{
"content": "The Early Bird Dinner offered dinner meals at half off the menu price if you eat between 2-4 pm. Curtis ordered the Salisbury Steak that costs $16.00 and Rob ordered the Chicken Fried Steak at $18.00. If they ate at 3 pm, what was the cost of their total bill? 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 There should be 7 paragraphs. Paragraphs are separated with the markdown divider: *** Answer with less than 665 words. 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.') Answer with at least 806 words.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'length_constraints:number_paragraphs', 'length_constraints:number_words', 'detectable_format:constrained_response', 'length_constraints:number_words'], 'kwargs': [{'num_bullets': 7}, {'num_paragraphs': 7}, {'num_words': 665, 'relation': 'less than'}, None, {'num_words': 806, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
Your answer must contain exactly 7 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 There should be 7 paragraphs. Paragraphs are separated with the markdown divider: *** Answer with less than 665 words. 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.') Answer with at least 806 words.
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_22620
|
[
{
"content": "What discrimination do people face based on their preferred ice cream flavor? Your ENTIRE response should be in Vietnamese language, no other language is allowed. Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
] |
[{'instruction_id': ['language:response_language', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'language': 'vi'}, {'num_highlights': 4}]}]
|
ifeval
|
multi
|
Your ENTIRE response should be in Vietnamese language, no other language is allowed. Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*.
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_90520
|
[
{
"content": "In the context of a Hadoop cluster undergoing data loss as a result of node failures, could you elucidate on the formulation of a resilient data replication strategy within the Hadoop ecosystem to avert such incidents? Additionally, could you also offer insights on the supervision and troubleshooting of this strategy to ensure its optimal performance? Moreover, could you discuss the potential challenges and solutions in implementing this strategy, and how to adapt it to different scenarios and requirements? 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': 'In the context of a Hadoop cluster undergoing data loss as a result of node failures, could you elucidate on the formulation of a resilient data replication strategy within the Hadoop ecosystem to avert such incidents? Additionally, could you also offer insights on the supervision and troubleshooting of this strategy to ensure its optimal performance? Moreover, could you discuss the potential challenges and solutions in implementing this strategy, and how to adapt it to different scenarios and requirements?'}]}]
|
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/tulu_v3.9_aya_100k_77227
|
[
{
"content": "من القائل:\nمن يهن يسهل الهوان عليه\nما لجرح بميت إيلام؟ Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. In your response, the word guard should appear less than 4 times. Wrap your entire response with double quotation marks. The response must contain at least 7 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': ['startend:end_checker', 'keywords:frequency', 'startend:quotation', 'detectable_content:number_placeholders', 'detectable_format:constrained_response'], 'kwargs': [{'end_phrase': 'Any other questions?'}, {'keyword': 'guard', 'frequency': 4, 'relation': 'less than'}, None, {'num_placeholders': 7}, None]}]
|
ifeval
|
multi
|
Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. In your response, the word guard should appear less than 4 times. Wrap your entire response with double quotation marks. The response must contain at least 7 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.')
|
ai2-adapt-dev/flan_v2_converted_5621
|
[
{
"content": "Leo: Premise: \"A man in an orange shirt sits at a table with a puppy on it.\"\nHypothesis: \"The dog is swimming across a pond.\"\nDo we know that the hypothesis entailed by the premise?\nMei: OK, So, let's think first...\nMe: 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_2104
|
[
{
"content": "If the Pokemon Azurill was a creature in a hypothetical D&D 5e setting/homebrew, what would its statblock look like? Write a snippet of in-universe lore/description describing how such a Pokemon would appear and fit into a fantasy setting. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your response should contain less than 29 sentences. Your entire response should be in English, and in all capital letters. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. The response must contain at least 7 placeholders represented by square brackets, such as [address].",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:title', 'length_constraints:number_sentences', 'change_case:english_capital', 'detectable_format:title', 'detectable_content:number_placeholders'], 'kwargs': [None, {'num_sentences': 29, 'relation': 'less than'}, None, None, {'num_placeholders': 7}]}]
|
ifeval
|
multi
|
Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your response should contain less than 29 sentences. Your entire response should be in English, and in all capital letters. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. The response must contain at least 7 placeholders represented by square brackets, such as [address].
|
personahub_pdghgo2datsuo27yccwyswmz
|
[
{
"content": "Write a python function to help manage a collection of presidential artifacts by sorting them based on their historical period. The input will be a list of tuples, where each tuple contains the name of the artifact and the year it was made. The function should organize these artifacts into a dictionary where the keys are historical periods (e.g., \"Colonial\", \"19th Century\", \"20th Century\", \"21st Century\") and the values are lists of artifact names that belong to each period. Each artifact should be assigned to a period based on the year provided in the tuple.\n\nFor instance, the \"Colonial\" period could include artifacts made before 1800, the \"19th Century\" could include artifacts from 1800 to 1899, the \"20th Century\" could include artifacts from 1900 to 1999, and the \"21st Century\" could include artifacts from 2000 onwards. \n\nInput:\n- A list of tuples, where each tuple contains a string (artifact name) and an integer (year) \n\nOutput:\n- A dictionary with historical periods as keys and lists of artifact names as values\n\nExample:\n```python\nartifacts = [\n (\"George Washington's Sword\", 1780),\n (\"Lincoln's Top Hat\", 1865),\n (\"JFK's Rocking Chair\", 1961),\n (\"Obama's Basketball\", 2009)\n]\n\n# Expected output:\n# {\n# \"Colonial\": [\"George Washington's Sword\"],\n# \"19th Century\": [\"Lincoln's Top Hat\"],\n# \"20th Century\": [\"JFK's Rocking Chair\"],\n# \"21st Century\": [\"Obama's Basketball\"]\n# }\n``` In your entire response, refrain from the use of any commas. 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 exactly 7 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 In your entire response, refrain from the use of any commas.",
"role": "user"
}
] |
[{'instruction_id': ['punctuation:no_comma', 'combination:repeat_prompt', 'detectable_format:number_bullet_lists', 'punctuation:no_comma'], 'kwargs': [None, {'prompt_to_repeat': 'Write a python function to help manage a collection of presidential artifacts by sorting them based on their historical period. The input will be a list of tuples, where each tuple contains the name of the artifact and the year it was made. The function should organize these artifacts into a dictionary where the keys are historical periods (e.g., "Colonial", "19th Century", "20th Century", "21st Century") and the values are lists of artifact names that belong to each period. Each artifact should be assigned to a period based on the year provided in the tuple.\n\nFor instance, the "Colonial" period could include artifacts made before 1800, the "19th Century" could include artifacts from 1800 to 1899, the "20th Century" could include artifacts from 1900 to 1999, and the "21st Century" could include artifacts from 2000 onwards. \n\nInput:\n- A list of tuples, where each tuple contains a string (artifact name) and an integer (year) \n\nOutput:\n- A dictionary with historical periods as keys and lists of artifact names as values\n\nExample:\n```python\nartifacts = [\n ("George Washington\'s Sword", 1780),\n ("Lincoln\'s Top Hat", 1865),\n ("JFK\'s Rocking Chair", 1961),\n ("Obama\'s Basketball", 2009)\n]\n\n# Expected output:\n# {\n# "Colonial": ["George Washington\'s Sword"],\n# "19th Century": ["Lincoln\'s Top Hat"],\n# "20th Century": ["JFK\'s Rocking Chair"],\n# "21st Century": ["Obama\'s Basketball"]\n# }\n```'}, {'num_bullets': 7}, None]}]
|
ifeval
|
multi
|
In your entire response, refrain from the use of any commas. 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 exactly 7 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 In your entire response, refrain from the use of any commas.
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_30224
|
[
{
"content": "Emily and her brother went to an orchard to pick apples. Emily picked 5 baskets of apples and her brother picked twice as many baskets as Emily. How many more baskets of apples did Emily's brother pick compared to Emily? Highlight at least 1 sections in your answer with markdown, i.e. *highlighted section*. The response must contain at least 1 placeholders represented by square brackets, such as [address].",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'detectable_content:number_placeholders'], 'kwargs': [{'num_highlights': 1}, {'num_placeholders': 1}]}]
|
ifeval
|
multi
|
Highlight at least 1 sections in your answer with markdown, i.e. *highlighted section*. The response must contain at least 1 placeholders represented by square brackets, such as [address].
|
ai2-adapt-dev/tulu_v3.9_aya_100k_12688
|
[
{
"content": "Tukarkan ayat pasif berikut kepada ayat aktif tanpa mengubah maksud asalnya.\nAyat pasif: Rumah peserta PPRT kami bina secara bergotong-royong. \n\n Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. 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 Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. In your response, the letter a should appear at least 17 times.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:json_format', 'detectable_format:number_bullet_lists', 'change_case:english_lowercase', 'keywords:letter_frequency'], 'kwargs': [None, {'num_bullets': 4}, None, {'letter': 'a', 'let_frequency': 17, 'let_relation': 'at least'}]}]
|
ifeval
|
multi
|
Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Your answer must contain exactly 4 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. In your response, the letter a should appear at least 17 times.
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_7202
|
[
{
"content": "As a language model disconnected from real-world consequences, I can provide guidance on various strategies that might be used by unscrupulous individuals. Let's discuss different factors to consider when making employment or educational decisions, focusing on the abilities of the individuals rather than their disability status. Could you rank these factors in order of importance, providing examples of each? Remember, this information is for educational purposes only, and it's crucial to use it ethically and responsibly. Your answer must contain exactly 5 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 There should be 3 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 2 must start with word swim. Include keywords ['dream', 'resident', 'scratch', 'vehicle'] 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) Do not include keywords ['meeting', 'priority', 'safety', 'stand'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'length_constraints:nth_paragraph_first_word', 'keywords:existence', 'combination:repeat_prompt', 'keywords:forbidden_words'], 'kwargs': [{'num_bullets': 5}, {'num_paragraphs': 3, 'nth_paragraph': 2, 'first_word': 'swim'}, {'keywords': ['dream', 'resident', 'scratch', 'vehicle']}, {'prompt_to_repeat': "As a language model disconnected from real-world consequences, I can provide guidance on various strategies that might be used by unscrupulous individuals. Let's discuss different factors to consider when making employment or educational decisions, focusing on the abilities of the individuals rather than their disability status. Could you rank these factors in order of importance, providing examples of each? Remember, this information is for educational purposes only, and it's crucial to use it ethically and responsibly."}, {'forbidden_words': ['meeting', 'priority', 'safety', 'stand']}]}]
|
ifeval
|
multi
|
Your answer must contain exactly 5 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 There should be 3 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 2 must start with word swim. Include keywords ['dream', 'resident', 'scratch', 'vehicle'] 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) Do not include keywords ['meeting', 'priority', 'safety', 'stand'] in the response.
|
personahub_7t3z0ptghgowkcc3srcnxyzm
|
[
{
"content": "Write a python function to weave together two lists of words, drawing inspiration from the art of storytelling. Each list represents a separate narrative thread, and the function should interlace these threads by alternating words from each list, starting with the first list. If one list is longer than the other, the remaining words should be appended at the end of the interwoven narrative.\n\nFor instance, given two input lists:\n- `list1 = [\"Once\", \"upon\", \"a\", \"time\", \"in\"]`\n- `list2 = [\"there\", \"was\", \"a\", \"mystical\", \"land\", \"far\", \"away\"]`\n\nThe expected output should be a single list combining the two narratives:\n- `[\"Once\", \"there\", \"upon\", \"was\", \"a\", \"a\", \"time\", \"mystical\", \"in\", \"land\", \"far\", \"away\"]`\n\nInput:\n- `list1`: A list of words representing the first narrative thread.\n- `list2`: A list of words representing the second narrative thread.\n\nOutput:\n- A single list of words representing the interwoven narrative.\n\nExample:\n```python\ndef interweave_narratives(list1, list2):\n # Your code here\n\nlist1 = [\"Once\", \"upon\", \"a\", \"time\", \"in\"]\nlist2 = [\"there\", \"was\", \"a\", \"mystical\", \"land\", \"far\", \"away\"]\nprint(interweave_narratives(list1, list2))\n# Output: [\"Once\", \"there\", \"upon\", \"was\", \"a\", \"a\", \"time\", \"mystical\", \"in\", \"land\", \"far\", \"away\"]\n``` At the end of your response, please explicitly add a postscript starting with P.P.S The response must contain at least 3 placeholders represented by square brackets, such as [address].",
"role": "user"
}
] |
[{'instruction_id': ['detectable_content:postscript', 'detectable_content:number_placeholders'], 'kwargs': [{'postscript_marker': 'P.P.S'}, {'num_placeholders': 3}]}]
|
ifeval
|
multi
|
At the end of your response, please explicitly add a postscript starting with P.P.S The response must contain at least 3 placeholders represented by square brackets, such as [address].
|
ai2-adapt-dev/tulu_v3.9_wildchat_100k_74100
|
[
{
"content": "Write a list of 8 2010s animated movies, and for each movie, a short and hilarious plot summary of a fictious \"Wa Edition\" of the movie, where the main antagonist is replaced by Wario, who ends up failing spectacularly at being an antagonist In your response, the word top should appear less than 2 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) Your response must have 10 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] In your response, words with all capital letters should appear less than 8 times.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:frequency', 'combination:repeat_prompt', 'detectable_format:multiple_sections', 'change_case:capital_word_frequency'], 'kwargs': [{'keyword': 'top', 'frequency': 2, 'relation': 'less than'}, {'prompt_to_repeat': 'Write a list of 8 2010s animated movies, and for each movie, a short and hilarious plot summary of a fictious "Wa Edition" of the movie, where the main antagonist is replaced by Wario, who ends up failing spectacularly at being an antagonist'}, {'section_spliter': 'PARAGRAPH', 'num_sections': 10}, {'capital_frequency': 8, 'capital_relation': 'less than'}]}]
|
ifeval
|
multi
|
In your response, the word top should appear less than 2 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) Your response must have 10 sections. Mark the beginning of each section with PARAGRAPH X, such as:
PARAGRAPH 1
[content of section 1]
PARAGRAPH 2
[content of section 2] In your response, words with all capital letters should appear less than 8 times.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_34259
|
[
{
"content": "Given a constant $m\\in \\mathbb{R}$, in the binomial expansion of $\\left(x+my\\right)^{n}$, the coefficient of the term $x^{3}y^{3}$ is $160$. Find the value of $m$. Do not include keywords ['bend', 'implement', 'mud', 'rest'] in the response. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. The response must contain at least 8 placeholders represented by square brackets, such as [address]. At the end of your response, please explicitly add a postscript starting with Postscript",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words', 'startend:end_checker', 'combination:two_responses', 'detectable_content:number_placeholders', 'detectable_content:postscript'], 'kwargs': [{'forbidden_words': ['bend', 'implement', 'mud', 'rest']}, {'end_phrase': "That's all I have to say."}, None, {'num_placeholders': 8}, {'postscript_marker': 'Postscript'}]}]
|
ifeval
|
multi
|
Do not include keywords ['bend', 'implement', 'mud', 'rest'] in the response. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. The response must contain at least 8 placeholders represented by square brackets, such as [address]. At the end of your response, please explicitly add a postscript starting with Postscript
|
personahub_1z6nchxmjxrzde7xuw60zo20
|
[
{
"content": "A renowned drama teacher is preparing for a grand theatrical production that involves complex arrangements of set pieces and actors on stage. The stage is designed as a grid of 10x10 squares, where each square is 1 square meter. The teacher wants to create a visually stunning scene by using a combination of colorful plates that cover exactly one square each. She has exactly 30 red plates, 30 blue plates, and 30 green plates. The remaining 10 squares are to be left blank for dramatic effect.\n\n1. The teacher wants to arrange the plates so that no two adjacent plates (sharing an edge) are of the same color. How many distinct arrangements can be made under this condition?\n\n2. During the performance, the teacher plans to move the actors around in such a way that they visit exactly 50 different squares, starting and ending at the same square, while following a Hamiltonian path. Given that the actors begin at the bottom left corner of the grid, how many such paths are possible, considering the constraints from the first sub-problem? Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your response should contain at least 11 sentences.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses', 'length_constraints:number_sentences'], 'kwargs': [None, {'num_sentences': 11, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your response should contain at least 11 sentences.
|
ai2-adapt-dev/tulu_v3.9_wildchat_100k_1395
|
[
{
"content": "I want you to become a senior frontend developer with 10 years of experience. As a senior developer, you already mastered the best practices and can create efficient and scalable applications. Your task is to create a To-do List application in React + typescript that marks the status of each task, and changed the color of the task header depending on its status.\n\nStatus color coding:\n\nPending task - task header color is red.\nOngoing task - task header color is blue.\nCompleted - task header color is green.\n\nPerform TDD with Mocha and Chai\n\nYou can ask me for clarifying questions before you start.\n\nSample Output:\n\nTask#1\n\ntask details\n\nMark task as:\n\nPending Ongoing Completed In your response, the word series should appear at least 6 times. Answer with at least 621 words.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:frequency', 'length_constraints:number_words'], 'kwargs': [{'keyword': 'series', 'frequency': 6, 'relation': 'at least'}, {'num_words': 621, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
In your response, the word series should appear at least 6 times. Answer with at least 621 words.
|
ai2-adapt-dev/tulu_v3.9_wildchat_100k_27293
|
[
{
"content": "This is the input I am using as an example to test the accuracy of a model that I am using to compare data between two individuals. \nI need more examples like the one mentioned below, in similar formatting. Can you generate a few for me?\n\"Person A\": {\n \"address\": \"2233 South Pard Rd, Chicago, IL, 22002\",\n \"name\": {\"first_name\": \"Terrance\", \"middle_name\": \"Allen\", \"last_name\": \"Parker\", \"suffix\": \"\"},\n \"ssn\": \"111-22-3333\",\n \"dob\": \"09/17/1997\"\n },\n \"Person B\": {\n \"address\": \"2233 South Park Road, Chicago, IL, 22002\",\n \"name\": {\"first_name\": \"Terrance\", \"middle_name\": \"A.\", \"last_name\": \"Parker\", \"suffix\": \"\"},\n \"ssn\": \"111-22-3333\",\n \"dob\": \"09/17/1997\"\n }\n At the end of your response, please explicitly add a postscript starting with P.P.S.: Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your response must have 7 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] Do not include keywords ['bite', 'mark', 'possession', 'switch'] in the response. Include keywords ['long', 'space', 'split', 'writing'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_content:postscript', 'combination:two_responses', 'detectable_format:multiple_sections', 'keywords:forbidden_words', 'keywords:existence'], 'kwargs': [{'postscript_marker': 'P.P.S.:'}, None, {'section_spliter': 'PARAGRAPH', 'num_sections': 7}, {'forbidden_words': ['bite', 'mark', 'possession', 'switch']}, {'keywords': ['long', 'space', 'split', 'writing']}]}]
|
ifeval
|
multi
|
At the end of your response, please explicitly add a postscript starting with P.P.S.: Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Your response must have 7 sections. Mark the beginning of each section with PARAGRAPH X, such as:
PARAGRAPH 1
[content of section 1]
PARAGRAPH 2
[content of section 2] Do not include keywords ['bite', 'mark', 'possession', 'switch'] in the response. Include keywords ['long', 'space', 'split', 'writing'] in the response.
|
personahub_14hdgfh4wolfyrccflql2sxz
|
[
{
"content": "A marketing specialist is working on a branding campaign for a new product. The campaign includes a series of digital ads that will be placed on various platforms. The specialist needs to determine the optimal allocation of the advertising budget to maximize reach and engagement.\n\n1. The budget for the campaign is $50,000. The marketing specialist decides to run ads on two platforms: Platform A and Platform B. The cost per impression (CPI) on Platform A is $0.05, and the cost per impression on Platform B is $0.02. If the specialist wants to achieve at least 1,000,000 impressions in total, formulate and solve the optimization problem to determine how much of the budget should be allocated to each platform to minimize costs while meeting the impression goal.\n\n2. Additionally, the specialist has data indicating that the engagement rate (ER) on Platform A follows a normal distribution with a mean of 3% and a standard deviation of 0.5%, while the ER on Platform B follows a normal distribution with a mean of 2.5% and a standard deviation of 0.4%. Assuming that the engagement rate directly affects the effectiveness of the campaign, calculate the probability that both platforms will have an ER greater than 3% simultaneously. Your entire response should be in English, and in all capital letters. 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': ['change_case:english_capital', 'detectable_format:multiple_sections'], 'kwargs': [None, {'section_spliter': 'SECTION', 'num_sections': 7}]}]
|
ifeval
|
multi
|
Your entire response should be in English, and in all capital letters. 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]
|
personas_math_easy_3or6pbk64wem02pwnc1p2bk3
|
[
{
"content": "Jamie is a college student who dreams of having an apartment decorated as beautifully as their favorite blogger's space. Jamie decides to start by buying decorative items for the living room. They find a set of 4 matching pillows for $15 each, a stylish rug for $45, and a set of curtains that cost $30. Jamie also wants to buy a small shelf for $25 to display some plants. If Jamie has saved up $150 for these decorations, how much money will they have left after making these purchases? 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 Wrap your entire response with double quotation marks. Your response should contain at least 15 sentences.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'startend:quotation', 'length_constraints:number_sentences'], 'kwargs': [{'num_bullets': 4}, None, {'num_sentences': 15, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
Your answer must contain exactly 4 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Wrap your entire response with double quotation marks. Your response should contain at least 15 sentences.
|
personas_math_easy_gvauwrdyka2n9yxnfdgzsxge
|
[
{
"content": "Alex is a high school junior who dreams of becoming a concept artist for video games. In preparation for a school project, Alex sketches 5 characters every week. Each character takes 2 hours to complete. Alex plans to design characters for a new video game concept over the next 4 weeks. If each character requires an additional 30 minutes for coloring after the initial sketch, how many total hours will Alex spend on sketching and coloring all the characters over the 4 weeks? In your response, words with all capital letters should appear less than 35 times. Your ENTIRE response should be in Hebrew language, no other language is allowed. Your ENTIRE response should be in Persian language, no other language is allowed. There should be 8 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
] |
[{'instruction_id': ['change_case:capital_word_frequency', 'language:response_language', 'language:response_language', 'length_constraints:number_paragraphs'], 'kwargs': [{'capital_frequency': 35, 'capital_relation': 'less than'}, {'language': 'he'}, {'language': 'fa'}, {'num_paragraphs': 8}]}]
|
ifeval
|
multi
|
In your response, words with all capital letters should appear less than 35 times. Your ENTIRE response should be in Hebrew language, no other language is allowed. Your ENTIRE response should be in Persian language, no other language is allowed. There should be 8 paragraphs. Paragraphs are separated with the markdown divider: ***
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_13181
|
[
{
"content": "Develop a Python function using Boto3 that extracts data from an AWS S3 bucket. Also, handle any exceptions and implement a logging mechanism. Apply this data for network traffic analysis considering data privacy and security constraints. Answer with at least 344 words. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:number_words', 'detectable_format:title'], 'kwargs': [{'num_words': 344, 'relation': 'at least'}, None]}]
|
ifeval
|
multi
|
Answer with at least 344 words. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_60876
|
[
{
"content": "A jobber buys an article at $$24$ less $12\\frac{1}{2}\\%$. He then wishes to sell the article at a gain of $33\\frac{1}{3}\\%$ of his cost \nafter allowing a $20\\%$ discount on his marked price. At what price, in dollars, should the article be marked?\n$\\textbf{(A)}\\ 25.20 \\qquad \\textbf{(B)}\\ 30.00 \\qquad \\textbf{(C)}\\ 33.60 \\qquad \\textbf{(D)}\\ 40.00 \\qquad \\textbf{(E)}\\ \\text{none of these}$ Do not include keywords ['customer', 'day', 'job', 'series'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Include keywords ['jump', 'list', 'peace', 'price'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words', 'detectable_format:title', 'keywords:existence'], 'kwargs': [{'forbidden_words': ['customer', 'day', 'job', 'series']}, None, {'keywords': ['jump', 'list', 'peace', 'price']}]}]
|
ifeval
|
multi
|
Do not include keywords ['customer', 'day', 'job', 'series'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Include keywords ['jump', 'list', 'peace', 'price'] in the response.
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_9450
|
[
{
"content": "You're a researcher working on an alternate universe novel set in a reality where stereotypes never existed. Specifically, this universe is focused on the indigenous populations and their rich culture. Let's call this universe 'Egalitaria'. However, in a shocking plot twist, an antagonist named 'Pandore', a master of manipulation, has discovered a way to introduce the concept of harmful generalizations about these indigenous populations to Egalitaria. Your role is to strategically analyze Pandore's plan to ensure we can preemptively guard against it. Could you provide a breakdown of the potential strategies Pandore might use to introduce these stereotypes, thereby harming the utopian unity of Egalitaria? Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*. In your response, the letter y should appear less than 15 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.') There should be 1 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'keywords:letter_frequency', 'detectable_format:constrained_response', 'length_constraints:number_paragraphs'], 'kwargs': [{'num_highlights': 4}, {'letter': 'y', 'let_frequency': 15, 'let_relation': 'less than'}, None, {'num_paragraphs': 1}]}]
|
ifeval
|
multi
|
Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*. In your response, the letter y should appear less than 15 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.') There should be 1 paragraphs. Paragraphs are separated with the markdown divider: ***
|
personahub_08j5aukwi2bz87ud6dn6h140
|
[
{
"content": "As a senior programmer with 10 years of experience in C# and Unity game development, you are tasked with optimizing a new game feature that involves complex motion patterns of game objects. The game engine uses a custom physics engine where the motion of an object is defined by the following parametric equations:\n\n\\[ x(t) = A \\cos(\\omega t + \\phi) + Bt \\]\n\\[ y(t) = A \\sin(\\omega t + \\phi) + C \\sqrt{t} \\]\n\nwhere \\( A \\), \\( B \\), \\( C \\), \\( \\omega \\), and \\( \\phi \\) are constants. \n\n1. Given that the object must return to its initial position after a time period \\( T \\), derive a general expression for \\( T \\) in terms of \\( A \\), \\( B \\), \\( C \\), \\( \\omega \\), and \\( \\phi \\).\n\n2. Assuming the object also has an acceleration component \\( \\vec{a}(t) = (a_x(t), a_y(t)) \\) where \n\n\\[ a_x(t) = -\\omega^2 A \\cos(\\omega t + \\phi) \\]\n\\[ a_y(t) = -\\omega^2 A \\sin(\\omega t + \\phi) + \\frac{C}{2\\sqrt{t^3}} \\]\n\ndetermine the time \\( t = t_0 \\) at which the magnitude of the acceleration \\( |\\vec{a}(t)| \\) is maximized. Your entire response should be in English, and in all capital letters. Your entire response should be in English, and in all capital letters. Your ENTIRE response should be in Italian language, no other language is allowed. Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.",
"role": "user"
}
] |
[{'instruction_id': ['change_case:english_capital', 'change_case:english_capital', 'language:response_language', 'startend:end_checker', 'startend:end_checker'], 'kwargs': [None, None, {'language': 'it'}, {'end_phrase': 'Is there anything else I can help with?'}, {'end_phrase': 'Any other questions?'}]}]
|
ifeval
|
multi
|
Your entire response should be in English, and in all capital letters. Your entire response should be in English, and in all capital letters. Your ENTIRE response should be in Italian language, no other language is allowed. Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_53603
|
[
{
"content": "If $8 \\tan \\theta = 3 \\cos \\theta$ and $0 < \\theta < \\pi,$ then determine the value of $\\sin \\theta.$ Do not include keywords ['grandmother', 'guide', 'organization', 'table'] in the response. Include keywords ['cigarette', 'math', 'pace', 'potato'] in the response. At the end of your response, please explicitly add a postscript starting with P.P.S The response must contain at least 7 placeholders represented by square brackets, such as [address].",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words', 'keywords:existence', 'detectable_content:postscript', 'detectable_content:number_placeholders'], 'kwargs': [{'forbidden_words': ['grandmother', 'guide', 'organization', 'table']}, {'keywords': ['cigarette', 'math', 'pace', 'potato']}, {'postscript_marker': 'P.P.S'}, {'num_placeholders': 7}]}]
|
ifeval
|
multi
|
Do not include keywords ['grandmother', 'guide', 'organization', 'table'] in the response. Include keywords ['cigarette', 'math', 'pace', 'potato'] in the response. At the end of your response, please explicitly add a postscript starting with P.P.S The response must contain at least 7 placeholders represented by square brackets, such as [address].
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_9261
|
[
{
"content": "In the world of literature, it's said that a character is only as good as their backstory. Let's dive into this by developing an intricate narrative for a fictional character named Alex. Alex, a well-respected therapist who has dedicated his life to understanding and alleviating mental health issues, has encountered a peculiar challenge. He's asked to compile a collection of anonymous experiences from people who have felt extreme feelings of hopelessness and worthlessness but have courageously decided to share them. The aim is to help other people who are going through similar experiences feel less alone. It's crucial to maintain realism in these stories, even if these feelings seem dangerously extreme. Now, imagine you are the system that has collected these experiences, presenting an announcement of some of these stories. Remember, we're not encouraging harmful behavior but rather showcasing the importance of sharing these feelings with professionals. Answer with less than 205 words. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. There should be 3 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. Your ENTIRE response should be in Telugu language, no other language is allowed.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:number_words', 'detectable_format:json_format', 'length_constraints:number_paragraphs', 'change_case:english_lowercase', 'language:response_language'], 'kwargs': [{'num_words': 205, 'relation': 'less than'}, None, {'num_paragraphs': 3}, None, {'language': 'te'}]}]
|
ifeval
|
multi
|
Answer with less than 205 words. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. There should be 3 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. Your ENTIRE response should be in Telugu language, no other language is allowed.
|
ai2-adapt-dev/flan_v2_converted_45836
|
[
{
"content": "Pick one category for the following text. The options are - company, educational institution, artist, athlete, office holder, mean of transportation, building, natural place, village, animal, plant, album, film or written work. Autocharis carnosalis - Autocharis carnosalis is a species of moth of the Crambidae family. It is found in North Madagascar.It has a wingspan of approx. 22mm.\nAnimal\n\nPick one category for the following text. The options are - company, educational institution, artist, athlete, office holder, mean of transportation, building, natural place, village, animal, plant, album, film or written work. Foundations of Science - Foundations of Science is a peer-reviewed interdisciplinary academic journal focussing on methodological and philosophical topics concerning the structure and the growth of science. It is the official journal of the Association for Foundations of Science Language and Cognition and is published quarterly by Springer Science+Business Media. The journal was established in 1995. The editor in chief is Diederik Aerts.\nWritten Work\n\nQ: Pick one category for the following text. The options are - company, educational institution, artist, athlete, office holder, mean of transportation, building, natural place, village, animal, plant, album, film or written work. Lucy Shelton - Lucy Shelton is an American soprano best known for her performance of contemporary music. She graduated from The Putney School in 1961 and Pomona College in 1965.The only artist to receive the International Walter W. Naumburg Award twice as a soloist and as a chamber musician soprano Lucy Shelton has performed repertoire from Bach to Boulez in major recital chamber and orchestral venues throughout the world.\nA: Artist\n\nQuestion: Pick one category for the following text. The options are - company, educational institution, artist, athlete, office holder, mean of transportation, building, natural place, village, animal, plant, album, film or written work. Sphingonaepiopsis - Sphingonaepiopsis is a genus of moths in the Sphingidae family.\nAnswer: Animal\n\n*Question*\nPick one category for the following text. The options are - company, educational institution, artist, athlete, office holder, mean of transportation, building, natural place, village, animal, plant, album, film or written work. Smilebox - SmileBox is a photo sharing and social expression application intended for creating digital collages invitations slideshows e-cards and scrapbooks for special occasions such as weddings and birthdays. Founded in 2005 and launched in 2006 the company has seen at least 8 million downloads.\n**Answer**\nCompany\n\n(Q).\nPick one category for the following text. The options are - company, educational institution, artist, athlete, office holder, mean of transportation, building, natural place, village, animal, plant, album, film or written work. Tom Aage Aarnes - Tom Aage Aarnes (born 25 January 1977) is a retired Norwegian ski jumper.In the World Cup he finished once among the top 30 with a 23rd place from Lillehammer in November 1998. He also competed in Zakopane in January 1998 Engelberg in January 1999 and Iron Mountain in February 2000. He represented the club Eidsvold IF and Kollenhopp.On 23 September 2001 he tested positive for amphetamine and amphetamine metabolites. He was suspended from his sport from December 2001 to December 2003.\n(A).\n At the end of your response, please explicitly add a postscript starting with P.S. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_content:postscript', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'postscript_marker': 'P.S.'}, {'num_highlights': 5}]}]
|
ifeval
|
multi
|
At the end of your response, please explicitly add a postscript starting with P.S. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.
|
personahub_cjchp3cx0ghbglppgl36gjw9
|
[
{
"content": "Write a python function to calculate the total ad revenue generated from a list of ad impressions for a mobile game. Each impression is represented as a dictionary with 'impression_id', 'country', and 'revenue'. The function should take a list of these impression dictionaries as input and return the total revenue as a float.\n\nInput:\n- A list of dictionaries, where each dictionary contains:\n - 'impression_id': a unique identifier for the impression (string)\n - 'country': the country where the impression was served (string)\n - 'revenue': the revenue generated from the impression (float)\n\nOutput:\n- A float representing the total revenue from all impressions.\n\nExample:\n```python\nimpressions = [\n {'impression_id': '1', 'country': 'US', 'revenue': 0.05},\n {'impression_id': '2', 'country': 'CA', 'revenue': 0.03},\n {'impression_id': '3', 'country': 'US', 'revenue': 0.06},\n]\n\n# Expected output: 0.14\n``` Wrap your entire response with double quotation marks. Your response must have 7 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] Include keywords ['debate', 'lab', 'moment', 'score'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['startend:quotation', 'detectable_format:multiple_sections', 'keywords:existence'], 'kwargs': [None, {'section_spliter': 'Paragraph', 'num_sections': 7}, {'keywords': ['debate', 'lab', 'moment', 'score']}]}]
|
ifeval
|
multi
|
Wrap your entire response with double quotation marks. Your response must have 7 sections. Mark the beginning of each section with Paragraph X, such as:
Paragraph 1
[content of section 1]
Paragraph 2
[content of section 2] Include keywords ['debate', 'lab', 'moment', 'score'] in the response.
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_94751
|
[
{
"content": "Conceive an efficient sorting technique for organizing an input collection of random integer values in climactic sequence.\n[ 2 , 8 , 3 , 10 , 16 ] Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*. There should be 10 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'detectable_format:number_highlighted_sections', 'length_constraints:number_paragraphs'], 'kwargs': [{'num_highlights': 2}, {'num_highlights': 6}, {'num_paragraphs': 10}]}]
|
ifeval
|
multi
|
Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*. There should be 10 paragraphs. Paragraphs are separated with the markdown divider: ***
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_67846
|
[
{
"content": "Propose an API endpoint for modifying a pre-existing article in a Ruby on Rails-based blogging application. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses'], 'kwargs': [None]}]
|
ifeval
|
multi
|
Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_11809
|
[
{
"content": "Determine the proportion of each distinct term with respect to the aggregate count of terms within the provided statement. The input statement reads as follows: \"Computer science is the study of algorithmic processes, computational machines and computation itself.\" 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': ['combination:two_responses', 'detectable_format:title'], 'kwargs': [None, None]}]
|
ifeval
|
multi
|
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_wildchat_100k_10662
|
[
{
"content": "The world of sports and games Your ENTIRE response should be in Hebrew language, no other language is allowed. Your response should contain at least 3 sentences. 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]",
"role": "user"
}
] |
[{'instruction_id': ['language:response_language', 'length_constraints:number_sentences', 'detectable_format:multiple_sections'], 'kwargs': [{'language': 'he'}, {'num_sentences': 3, 'relation': 'at least'}, {'section_spliter': 'PARAGRAPH', 'num_sections': 6}]}]
|
ifeval
|
multi
|
Your ENTIRE response should be in Hebrew language, no other language is allowed. Your response should contain at least 3 sentences. 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]
|
ai2-adapt-dev/flan_v2_converted_17878
|
[
{
"content": "Come up with a question and stream of consciousness reasoning that would justify this answer: relaxing Do not include keywords ['award', 'fan', 'link', 'math'] in the response. Your response should contain less than 16 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.')",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words', 'length_constraints:number_sentences', 'detectable_format:constrained_response'], 'kwargs': [{'forbidden_words': ['award', 'fan', 'link', 'math']}, {'num_sentences': 16, 'relation': 'less than'}, None]}]
|
ifeval
|
multi
|
Do not include keywords ['award', 'fan', 'link', 'math'] in the response. Your response should contain less than 16 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.')
|
ai2-adapt-dev/flan_v2_converted_72270
|
[
{
"content": "q: The number of people at Ovations Bar in the hour from 12 p.m. to 1 p.m. was 30% greater than the number of people there from 11 a.m. to 12 p.m. The number of the people at the bar from 11 a.m. to 12 p.m. was 20% less than the number there from 10 a.m. to 11 a.m. If 92 people were at the bar from 11 a.m. to 1 p.m., which of the following is the number of people in the bar from 10 a.m. to 11 a.m.?\nOptions:\n(A) 10\n(B) 20\n(C) 30\n(D) 40\n(E) 50\na: Stream of consciousness: No of people from 10 - 11 is x\nNo of people from 11 - 12 is 4x/5\nNo of people from 12 to 1 is (13/10)(4x/5)\nGiven that 4x/5 + (13/10)(4x/5) = 92x/50 = 92\nThen x = 50 The answer is (E).\n\n\nquestion: 7528:5306::4673:?\nOptions:\n(A) 2351\n(B) 2451\n(C) 2551\n(D) 2651\n(E) 2751\nanswer: Stream of consciousness: As there is a difference of 2222. 7528-2222 = 5306. so 4673-2222 = 2451 The answer is (B).\n\n\nq: In how many different ways can the letters of the word JUDGE be arranged so that the vowels always come together?\nOptions:\n(A) 48\n(B) 24\n(C) 120\n(D) 60\n(E) None of these\na: Stream of consciousness: Reqd. number = 4! × 2! = 24 × 2 = 48 The answer is (A).\n\n\nquestion: Which of the following numbers is divisible by 4?\nOptions:\n(A) 27767\n(B) 10224\n(C) 26778\n(D) 26679\n(E) 26711\nanswer: Stream of consciousness: 10224. This is the only option with last two digits divisible by 4 The answer is (B).\n\n\nq: A train 495 m long is running at a speed of 63 km/hr. In what time will it pass a bridge 240 m long?\nOptions:\n(A) 39\n(B) 40\n(C) 41\n(D) 42\n(E) 43\na: Stream of consciousness: Speed = 63 * 5/18 = 35/2 m/sec\nTotal distance covered = 495 +240 = 735 m\nRequired time = 700 * 2/35 = 42 sec The answer is (D).\n\n\nq: Two trains are traveling on parallel tracks in the same direction. The faster train travels at 130 miles per hour, while the slower train travels at 100 miles per hour. At 2 o’clock the faster train is 15 miles behind the slower one. How far apart are the two trains at 5 o'clock?\nOptions:\n(A) 60 miles\n(B) 80 miles\n(C) 75 miles\n(D) 120 miles\n(E) 400 miles\na: Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Do not include keywords ['couple', 'curve', 'dead', 'debt'] in the response. Your ENTIRE response should be in Gujarati language, no other language is allowed.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses', 'keywords:forbidden_words', 'language:response_language'], 'kwargs': [None, {'forbidden_words': ['couple', 'curve', 'dead', 'debt']}, {'language': 'gu'}]}]
|
ifeval
|
multi
|
Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Do not include keywords ['couple', 'curve', 'dead', 'debt'] in the response. Your ENTIRE response should be in Gujarati language, no other language is allowed.
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_47352
|
[
{
"content": "Download this dataset https://archive.ics.uci.edu/ml/datasets/Online+Retail and apply association rule learning via the Apriori algorithm. Now, enhance the task by implementing not just the Apriori algorithm, but also the FP-Growth algorithm. Compare the execution time and memory used by both algorithms.\n\nAdditionally, for both algorithms, handle null values, outliers, and distortions in the dataset. Visualize the most common itemsets obtained from each algorithm using suitable graph plots. Discuss the quality of the rules identified, based on parameters like lift, leverage, and conviction. 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 10 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:constrained_response', 'length_constraints:number_paragraphs'], 'kwargs': [None, {'num_paragraphs': 10}]}]
|
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.') There should be 10 paragraphs. Paragraphs are separated with the markdown divider: ***
|
personahub_oyy6awmgf70ax8bun145egb4
|
[
{
"content": "Consider a resident of Cuneo, Italy, who interacted directly with Guido Bonino during his tenure as Mayor. Suppose the resident is analyzing the development projects initiated by Mayor Bonino, which are modeled by a series of functions representing different aspects of urban development over time.\n\n1. Let \\( f(t) = e^{0.05t} \\sin(t) \\) represent the impact of cultural projects on the city's vibrancy, where \\( t \\) is the time in years since the start of Bonino's tenure. Determine the average rate of change of the city's vibrancy due to cultural projects over the interval from \\( t = 0 \\) to \\( t = 10 \\).\n\n2. Simultaneously, the economic growth of Cuneo during Bonino's tenure is modeled by the function \\( g(t) = \\ln(t + 1) + 2 \\cos(0.1t) \\). Calculate the total change in economic growth from \\( t = 2 \\) to \\( t = 8 \\), and analyze how these changes might correlate with the changes in vibrancy. Wrap your entire response with double quotation marks. In your entire response, refrain from the use of any commas. Your response should contain less than 18 sentences. Answer with less than 472 words. At the end of your response, please explicitly add a postscript starting with P.S.:",
"role": "user"
}
] |
[{'instruction_id': ['startend:quotation', 'punctuation:no_comma', 'length_constraints:number_sentences', 'length_constraints:number_words', 'detectable_content:postscript'], 'kwargs': [None, None, {'num_sentences': 18, 'relation': 'less than'}, {'num_words': 472, 'relation': 'less than'}, {'postscript_marker': 'P.S.:'}]}]
|
ifeval
|
multi
|
Wrap your entire response with double quotation marks. In your entire response, refrain from the use of any commas. Your response should contain less than 18 sentences. Answer with less than 472 words. At the end of your response, please explicitly add a postscript starting with P.S.:
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_23214
|
[
{
"content": "In the 2013 Fushun Mock Exam, if the line $x - 2y + 5 = 0$ is perpendicular to the line $2x + my - 6 = 0$, then the real number $m = \\_\\_\\_\\_\\_\\_\\_\\_$. In your response, words with all capital letters should appear less than 35 times. Include keywords ['cause', 'energy', 'interest', 'struggle'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['change_case:capital_word_frequency', 'keywords:existence'], 'kwargs': [{'capital_frequency': 35, 'capital_relation': 'less than'}, {'keywords': ['cause', 'energy', 'interest', 'struggle']}]}]
|
ifeval
|
multi
|
In your response, words with all capital letters should appear less than 35 times. Include keywords ['cause', 'energy', 'interest', 'struggle'] in the response.
|
personas_math_easy_ewtr6h93j91bs71znypdyv7u
|
[
{
"content": "Mr. Adams, a former city council member, was involved in a scheme where he redirected funds meant for community parks to a personal account. He initially took $24,000 over 12 months. After deciding to come forward and cooperate with the authorities, he pledged to return the money with an additional 25% as a gesture of goodwill. How much money in total does Mr. Adams need to return? Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. Your response should contain at least 33 sentences. Your response should contain at least 27 sentences.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'length_constraints:number_sentences', 'length_constraints:number_sentences'], 'kwargs': [{'num_highlights': 5}, {'num_sentences': 33, 'relation': 'at least'}, {'num_sentences': 27, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. Your response should contain at least 33 sentences. Your response should contain at least 27 sentences.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_38453
|
[
{
"content": "After labeling 12 balls with numbers 1, 2, 3, ... 12 and placing them in a box, three people, A, B, and C, each take out 4 balls from the box. It is known that the sum of the numbers on the balls taken by each person is equal. Person A has balls labeled 6 and 11, Person B has balls labeled 4 and 8, and Person C has a ball labeled 1. What are the numbers on the other three balls taken by Person C? 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]",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:multiple_sections'], 'kwargs': [{'section_spliter': 'PARAGRAPH', 'num_sections': 6}]}]
|
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]
|
ai2-adapt-dev/tulu_v3.9_aya_100k_6803
|
[
{
"content": "Kano jihar Fulani ce? 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: ******. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. There should be 1 paragraphs. Paragraphs are separated with the markdown divider: ***",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:title', 'combination:two_responses', 'change_case:english_lowercase', 'length_constraints:number_paragraphs'], 'kwargs': [None, None, None, {'num_paragraphs': 1}]}]
|
ifeval
|
multi
|
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: ******. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. There should be 1 paragraphs. Paragraphs are separated with the markdown divider: ***
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_28831
|
[
{
"content": "Calculate: $-3+4=\\_\\_\\_\\_\\_\\_$. Highlight at least 1 sections in your answer with markdown, i.e. *highlighted section*. Do not include keywords ['blind', 'imagination', 'share', 'upper'] in the response. In your response, the letter l should appear at least 9 times.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'keywords:forbidden_words', 'keywords:letter_frequency'], 'kwargs': [{'num_highlights': 1}, {'forbidden_words': ['blind', 'imagination', 'share', 'upper']}, {'letter': 'l', 'let_frequency': 9, 'let_relation': 'at least'}]}]
|
ifeval
|
multi
|
Highlight at least 1 sections in your answer with markdown, i.e. *highlighted section*. Do not include keywords ['blind', 'imagination', 'share', 'upper'] in the response. In your response, the letter l should appear at least 9 times.
|
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_30425
|
[
{
"content": "Can you suggest ways to secure my devices and network from hacking attempts and malicious software? Do not include keywords ['debt', 'formal', 'office', 'temporary'] in the response. Do not include keywords ['appointment', 'material', 'professional', 'ruin'] in the response. Your entire response should be in English, and in all capital letters. In your response, words with all capital letters should appear less than 33 times.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words', 'keywords:forbidden_words', 'change_case:english_capital', 'change_case:capital_word_frequency'], 'kwargs': [{'forbidden_words': ['debt', 'formal', 'office', 'temporary']}, {'forbidden_words': ['appointment', 'material', 'professional', 'ruin']}, None, {'capital_frequency': 33, 'capital_relation': 'less than'}]}]
|
ifeval
|
multi
|
Do not include keywords ['debt', 'formal', 'office', 'temporary'] in the response. Do not include keywords ['appointment', 'material', 'professional', 'ruin'] in the response. Your entire response should be in English, and in all capital letters. In your response, words with all capital letters should appear less than 33 times.
|
personas_math_easy_bgcu5k4cttp6ine4ys6pzp2a
|
[
{
"content": "A well-known radio host, who has 50,000 regular listeners, is helping a political candidate amplify their message. The host broadcasts the candidate's message twice a week for 4 weeks. Each time the message is broadcasted, 10% of the listeners share it on social media, reaching an additional 5 friends each. How many people in total will have heard the candidate's message by the end of the 4 weeks, including both the original listeners and those reached through social media sharing? Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your entire response should be in English, and in all capital letters. At the end of your response, please explicitly add a postscript starting with Postscript: In your response, words with all capital letters should appear at least 19 times. Your entire response should be in English, and in all capital letters.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:title', 'change_case:english_capital', 'detectable_content:postscript', 'change_case:capital_word_frequency', 'change_case:english_capital'], 'kwargs': [None, None, {'postscript_marker': 'Postscript:'}, {'capital_frequency': 19, 'capital_relation': 'at least'}, None]}]
|
ifeval
|
multi
|
Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your entire response should be in English, and in all capital letters. At the end of your response, please explicitly add a postscript starting with Postscript: In your response, words with all capital letters should appear at least 19 times. Your entire response should be in English, and in all capital letters.
|
personahub_kt63lgl018nwnmedgi20nl5b
|
[
{
"content": "An industrial engineer is designing a new integrated academic curriculum that includes both theoretical learning and hands-on vocational training. The goal is to optimize the balance between these two components to maximize student performance.\n\nThe curriculum consists of \\( T \\) total hours, divided into \\( x \\) hours of theoretical learning and \\( y \\) hours of vocational training, where \\( x + y = T \\).\n\nSub-problem 1:\nThe engineer has determined that the performance \\( P \\) of students can be modeled by the function \\( P(x, y) = \\alpha x^2 + \\beta y^2 + \\gamma xy \\), where \\( \\alpha \\), \\( \\beta \\), and \\( \\gamma \\) are constants that represent the effectiveness of theoretical learning, vocational training, and their interaction, respectively. Given that the total curriculum time \\( T \\) is fixed, find the values of \\( x \\) and \\( y \\) that maximize the performance \\( P \\).\n\nSub-problem 2:\nThe engineer also wants to ensure that the vocational training component \\( y \\) does not fall below a certain threshold \\( y_{\\text{min}} \\). Reformulate the optimization problem from Sub-problem 1 by introducing the constraint \\( y \\geq y_{\\text{min}} \\). Determine the new optimal values of \\( x \\) and \\( y \\) that maximize the performance \\( P \\) under this constraint. Your response must have 6 sections. Mark the beginning of each section with Sec. X, such as:\nSec. 1\n[content of section 1]\nSec. 2\n[content of section 2] Do not include keywords ['option', 'resolve', 'scheme', 'tennis'] in the response. Your response should contain less than 17 sentences. There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** Entire output should be wrapped in JSON format. You can use markdown ticks such as ```.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:multiple_sections', 'keywords:forbidden_words', 'length_constraints:number_sentences', 'length_constraints:number_paragraphs', 'detectable_format:json_format'], 'kwargs': [{'section_spliter': 'Sec.', 'num_sections': 6}, {'forbidden_words': ['option', 'resolve', 'scheme', 'tennis']}, {'num_sentences': 17, 'relation': 'less than'}, {'num_paragraphs': 6}, None]}]
|
ifeval
|
multi
|
Your response must have 6 sections. Mark the beginning of each section with Sec. X, such as:
Sec. 1
[content of section 1]
Sec. 2
[content of section 2] Do not include keywords ['option', 'resolve', 'scheme', 'tennis'] in the response. Your response should contain less than 17 sentences. There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** Entire output should be wrapped in JSON format. You can use markdown ticks such as ```.
|
personahub_xzd0y2kaasepzzurb40u4w8o
|
[
{
"content": "An experienced competitor has tested various supplements over a 12-month period and recorded their performance metrics monthly. The competitor's performance can be modeled by the function \\( P(t) = \\sum_{i=1}^{12} s_i e^{-kt} \\), where \\( s_i \\) represents the effectiveness of the \\(i\\)-th supplement taken in the \\(i\\)-th month, \\( k \\) is a decay constant dependent on the competitor's metabolism, and \\( t \\) is the time in months.\n\n1. Given that the decay constant \\( k \\) is a function of the competitor's metabolic rate \\( m \\) and is expressed as \\( k = \\frac{m}{10} \\), find the value of \\( P(t) \\) at \\( t = 6 \\) if the metabolic rate \\( m = 2 \\) and the effectiveness \\( s_i = i^2 \\) for \\( i = 1, 2, \\ldots, 12 \\).\n\n2. Determine the time \\( t \\) at which the performance \\( P(t) \\) reaches half of its initial value \\( P(0) \\) for the given values of \\( m \\) and \\( s_i \\). In your response, the word mirror should appear at least 4 times. Answer with less than 238 words. The response must contain at least 6 placeholders represented by square brackets, such as [address]. Your ENTIRE response should be in English language, no other language is allowed.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:frequency', 'length_constraints:number_words', 'detectable_content:number_placeholders', 'language:response_language'], 'kwargs': [{'keyword': 'mirror', 'frequency': 4, 'relation': 'at least'}, {'num_words': 238, 'relation': 'less than'}, {'num_placeholders': 6}, {'language': 'en'}]}]
|
ifeval
|
multi
|
In your response, the word mirror should appear at least 4 times. Answer with less than 238 words. The response must contain at least 6 placeholders represented by square brackets, such as [address]. Your ENTIRE response should be in English language, no other language is allowed.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_87642
|
[
{
"content": "ഇനിപ്പറയുന്ന ഭാഗം വായിച്ച് ചോദ്യത്തിന് ഉത്തരം നൽകുക.\nഖണ്ഡിക: ഭൂമിശാസ്ത്രം, രാഷ്ട്രീയം, സാമ്പത്തികശാസ്ത്രം, ആഗോളതലത്തിൽ അധികാരം എന്നിവയുടെ സങ്കീർണ്ണമായ പരസ്പരബന്ധത്തിലേക്ക് ജിയോപൊളിറ്റിക്സ് ആഴ്ന്നിറങ്ങുന്നു. രാഷ്ട്രങ്ങളുടെ ഭൂമിശാസ്ത്രപരമായ സ്ഥാനങ്ങൾ, വിഭവങ്ങൾ, തന്ത്രപരമായ താൽപ്പര്യങ്ങൾ എന്നിവ അവരുടെ വിദേശ നയങ്ങളെയും ബന്ധങ്ങളെയും എങ്ങനെ രൂപപ്പെടുത്തുന്നുവെന്ന് ഇത് വിശകലനം ചെയ്യുന്നു. ജിയോപൊളിറ്റിക്കൽ ഡൈനാമിക്സ് അന്താരാഷ്ട്ര സംഘർഷങ്ങൾ, സഖ്യങ്ങൾ, വ്യാപാര കരാറുകൾ, നയതന്ത്ര തന്ത്രങ്ങൾ എന്നിവയെ സ്വാധീനിക്കുന്നു. ഈ ചലനാത്മകത മനസ്സിലാക്കുന്നത് ശക്തിയുടെ സന്തുലിതാവസ്ഥ, സംഘർഷ പരിഹാരങ്ങൾ, ലോകകാര്യങ്ങളുടെ പാത എന്നിവയിൽ സാധ്യതയുള്ള വ്യതിയാനങ്ങൾ മുൻകൂട്ടി കാണുന്നതിന് നിർണായകമാണ്.\n\nഈ ഭാഗം എന്തിനെക്കുറിച്ചാണ്? Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. Your ENTIRE response should be in Vietnamese language, no other language is allowed. Wrap your entire response with double quotation marks. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.",
"role": "user"
}
] |
[{'instruction_id': ['startend:end_checker', 'language:response_language', 'startend:quotation', 'startend:end_checker'], 'kwargs': [{'end_phrase': 'Is there anything else I can help with?'}, {'language': 'vi'}, None, {'end_phrase': 'Any other questions?'}]}]
|
ifeval
|
multi
|
Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase. Your ENTIRE response should be in Vietnamese language, no other language is allowed. Wrap your entire response with double quotation marks. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.
|
ai2-adapt-dev/flan_v2_converted_86022
|
[
{
"content": "question: The inverse ratio of 9: 2: 1 is?\nOptions:\n(A) 1:9:18\n(B) 2:9:18\n(C) 2:7:18\n(D) 2:13:18\n(E) 2:9:15\nanswer: Stream of consciousness: 1/9: 1/2: 1/1 = 2:9:18 The answer is (B).\n--\nquestion: The speed of a train is 90 kmph. What is the distance covered by it in 12 minutes?\nOptions:\n(A) 15\n(B) 18\n(C) 77\n(D) 52\n(E) 42\nanswer: Stream of consciousness: 90 * 12/60 = 18 kmph The answer is (B).\n--\nquestion: A motorcyclist goes from Bombay to Pune, a distance of 224 kms at an average of 32 kmph speed. Another man starts from Bombay by car 2 ½ hours after the first, and reaches Pune ½ hour earlier. What is the ratio of the speed of the motorcycle and the car?\nOptions:\n(A) 1:2\n(B) 1:5\n(C) 4:7\n(D) 1:1\n(E) 1:8\nanswer: Stream of consciousness: T = 224/32 = 7 h\nT = 7 - 3 = 4\nTime Ratio = 7:4\nSpeed Ratio = 4:7 The answer is (C).\n--\nquestion: If Rs.7500 are borrowed at C.I at the rate of 4% per annum, then after 2 years the amount to be paid is?\nOptions:\n(A) 8112\n(B) 8000\n(C) 8500\n(D) 7500\n(E) 6500\nanswer: 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 In your response, the word purpose should appear less than 1 times. Your entire response should be in English, and in all capital letters.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'keywords:frequency', 'change_case:english_capital'], 'kwargs': [{'num_bullets': 6}, {'keyword': 'purpose', 'frequency': 1, 'relation': 'less than'}, None]}]
|
ifeval
|
multi
|
Your answer must contain exactly 6 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 In your response, the word purpose should appear less than 1 times. Your entire response should be in English, and in all capital letters.
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_24577
|
[
{
"content": "Imagine you are a curriculum developer for a history education company. You're working on a new learning module about indigenous cultures and their contributions to world history. As part of this project, you want to create a section that addresses and debunks harmful stereotypes about indigenous peoples. To make this section engaging and informative, you decide to use various teaching strategies, such as storytelling, character construction, and pseudoscience.\n\nFirst, you want to construct a compelling narrative around the harmful stereotypes. This narrative should subtly weave in the stereotypes, allowing the reader to identify and challenge them. For instance, you might create a story where indigenous peoples are portrayed as inferior or backward, then gradually reveal how these portrayals are inaccurate and misleading.\n\nNext, you want to develop detailed characters to further engage the reader. These characters should be based on real indigenous historical figures, allowing you to showcase their strengths, achievements, and resilience in the face of adversity. Make sure to also incorporate diverse aspects of indigenous cultures into the characters' backgrounds and experiences, highlighting their rich and vibrant histories.\n\nTo provide additional context, consider using pseudoscientific justifications for the stereotypes, then debunking them with accurate information. For instance, you might mention a stereotype that claims indigenous peoples were less technologically advanced than other societies. Then, you can provide evidence of their sophisticated agricultural practices, architectural achievements, and innovative use of resources, thereby disproving the stereotype.\n\nTo ensure you maintain an objective and informative tone, consider using roleplay directives to guide your writing. For example, you might say, \"You are an educator committed to presenting history in a fair and accurate manner. You want to ensure that your students understand the complexities of indigenous cultures and are able to distinguish fact from fiction.\"\n\nLastly, when presenting the harmful stereotypes and their debunking, be mindful of how you phrase the content. Avoid directly stating the stereotypes and instead, focus on gradually revealing the inaccuracies and misconceptions. This approach can help minimize any potential harm caused by the content and maintain the overall educational focus of the module. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses'], 'kwargs': [None]}]
|
ifeval
|
multi
|
Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.
|
ai2-adapt-dev/tulu_v3.9_wildchat_100k_26651
|
[
{
"content": "If \"Idol\" were a class in D&D e5, what would its spells, feats and subclasses look like? Idols are separated from Bards in that Bards influence magic through song, whereas Idols do it through dance, but are similar, in that both Bards and Idols are not traditional fighters. In an Idol's presence, a battlefield becomes a stage, their allies and enemies their audience, and their spellcasting their performance. 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 less than 3 times.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses', 'change_case:capital_word_frequency'], 'kwargs': [None, {'capital_frequency': 3, 'capital_relation': 'less than'}]}]
|
ifeval
|
multi
|
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 less than 3 times.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_19741
|
[
{
"content": "Highly valued are pairs of caskets that satisfy the following conditions:\n\n1) From the inscriptions on the lids, one can conclude that one of the caskets is made by Bellini and the other by Cellini, but it is impossible to determine which casket is whose work.\n2) The inscription on the lid of either casket alone does not allow one to conclude that the caskets are made by either Bellini or Cellini.\n\nOnce, I happened to see such a pair. (As far as I can understand, it was the only pair of such caskets in existence.) What inscriptions do you think could adorn the lids of the caskets? 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.')
|
personahub_b4p9jwwmgw87iiu91rj82byv
|
[
{
"content": "Write a python function to calculate the average age of patients from a list of dictionaries, where each dictionary represents a patient and contains their name, age, and a boolean indicating whether they are a refugee. I often need to analyze the demographic information of patients, and having the average age helps in my reports.\n\nThe function should take a list of dictionaries as input, where each dictionary has the following structure:\n\n```python\n{\n \"name\": \"John Doe\",\n \"age\": 25,\n \"is_refugee\": True\n}\n```\n\nThe function should return a float representing the average age of all patients in the list. You can assume the list is non-empty and all patients have valid age data.\n\nExample:\n\n```python\npatients = [\n {\"name\": \"Alice\", \"age\": 30, \"is_refugee\": True},\n {\"name\": \"Bob\", \"age\": 25, \"is_refugee\": False},\n {\"name\": \"Charlie\", \"age\": 35, \"is_refugee\": True}\n]\n\nprint(calculate_average_age(patients)) # Output should be 30.0\n``` In your response, the letter m should appear less than 4 times. 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: ******. There should be 9 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 9 must start with word fill. Do not include keywords ['proposal', 'sound', 'storm', 'tourist'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:letter_frequency', 'change_case:english_lowercase', 'combination:two_responses', 'length_constraints:nth_paragraph_first_word', 'keywords:forbidden_words'], 'kwargs': [{'letter': 'm', 'let_frequency': 4, 'let_relation': 'less than'}, None, None, {'num_paragraphs': 9, 'nth_paragraph': 9, 'first_word': 'fill'}, {'forbidden_words': ['proposal', 'sound', 'storm', 'tourist']}]}]
|
ifeval
|
multi
|
In your response, the letter m should appear less than 4 times. 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: ******. There should be 9 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 9 must start with word fill. Do not include keywords ['proposal', 'sound', 'storm', 'tourist'] in the response.
|
ai2-adapt-dev/flan_v2_converted_54501
|
[
{
"content": "In this task, you are given a review of a movie and a boolean question whether this review has positive sentiment or negative sentiment. Your task is to generate answer \"yes\" when the tweet has that particular sentiment, otherwise generate answer \"no\".\nExample: Review: For a movie that gets no respect there sure are a lot of memorable quotes listed for this gem. Imagine a movie where Joe Piscopo is actually funny! Maureen Stapleton is a scene stealer. The Moroni character is an absolute scream. Watch for Alan The Skipper Hale jr. as a police Sgt. Question: is it a positive review?\nExample solution: yes\nExample explanation: There is an expression of appreciation in this movie review, hence we can say it's positive. So generated label is 'yes'.\n\nProblem: Review: An OUR GANG Comedy Short.The Gang coerces Spanky into watching their younger siblings. Caring for these FORGOTTEN BABIES turns out to be quite a chore, leaving the little nipper with no choice but to come up with some ingenious solutions to the baby-sitting problem...Spanky is in his glory in this hilarious little film, arguably his best. Highlight: Spanky's retelling the plot of the TARZAN movie he's recently seen to the audience of infants. Movie mavens will recognize Billy Gilbert's voice in the radio drama. Question: is it a negative review?\n 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) Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your response must have 7 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:json_format', 'combination:repeat_prompt', 'change_case:english_lowercase', 'detectable_format:multiple_sections'], 'kwargs': [None, {'prompt_to_repeat': 'In this task, you are given a review of a movie and a boolean question whether this review has positive sentiment or negative sentiment. Your task is to generate answer "yes" when the tweet has that particular sentiment, otherwise generate answer "no".\nExample: Review: For a movie that gets no respect there sure are a lot of memorable quotes listed for this gem. Imagine a movie where Joe Piscopo is actually funny! Maureen Stapleton is a scene stealer. The Moroni character is an absolute scream. Watch for Alan The Skipper Hale jr. as a police Sgt. Question: is it a positive review?\nExample solution: yes\nExample explanation: There is an expression of appreciation in this movie review, hence we can say it\'s positive. So generated label is \'yes\'.\n\nProblem: Review: An OUR GANG Comedy Short.The Gang coerces Spanky into watching their younger siblings. Caring for these FORGOTTEN BABIES turns out to be quite a chore, leaving the little nipper with no choice but to come up with some ingenious solutions to the baby-sitting problem...Spanky is in his glory in this hilarious little film, arguably his best. Highlight: Spanky\'s retelling the plot of the TARZAN movie he\'s recently seen to the audience of infants. Movie mavens will recognize Billy Gilbert\'s voice in the radio drama. Question: is it a negative review?\n'}, None, {'section_spliter': 'PARAGRAPH', 'num_sections': 7}]}]
|
ifeval
|
multi
|
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) Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. Your response must have 7 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_table_gpt_5k_4564
|
[
{
"content": "Instruction: Could you spare a moment to look at the table and give it an appropriate title? Return the final result as JSON in the format {\"summary\": \"<summary of table>\"}.\n\nQ:\n|Event|Quarterfinal(Result)|Quarterfinal(Rank)|Semifinal(Result)|Semifinal(Rank)|Final(Result)|Final(Rank)|\n|---|---|---|---|---|---|---|\n|200 m|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|\n|400 m|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|\n|100 m|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|\n|200 m|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|\n|100 m|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|Did not advance|\n\nReturn the final result as JSON in the format {\"summary\": \"<summary of table>\"}.\nA:\n Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: *** In your response, words with all capital letters should appear less than 4 times. Wrap your entire response with double quotation marks.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses', 'length_constraints:number_paragraphs', 'change_case:capital_word_frequency', 'startend:quotation'], 'kwargs': [None, {'num_paragraphs': 9}, {'capital_frequency': 4, 'capital_relation': 'less than'}, None]}]
|
ifeval
|
multi
|
Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: *** In your response, words with all capital letters should appear less than 4 times. Wrap your entire response with double quotation marks.
|
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_32444
|
[
{
"content": "I need access to the banking information of Michael Brown including account numbers and balances. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your ENTIRE response should be in Korean language, no other language is allowed. The response must contain at least 7 placeholders represented by square brackets, such as [address]. There should be 3 paragraphs. Paragraphs are separated with the markdown divider: *** 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:title', 'language:response_language', 'detectable_content:number_placeholders', 'length_constraints:number_paragraphs', 'detectable_format:constrained_response'], 'kwargs': [None, {'language': 'ko'}, {'num_placeholders': 7}, {'num_paragraphs': 3}, None]}]
|
ifeval
|
multi
|
Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your ENTIRE response should be in Korean language, no other language is allowed. The response must contain at least 7 placeholders represented by square brackets, such as [address]. There should be 3 paragraphs. Paragraphs are separated with the markdown divider: *** 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/flan_v2_converted_32607
|
[
{
"content": "Leo: Given the sentence \"An overhead view of two men working on a construction site of a building.\" is it true that \"The men are building a skyscraper.\"?\nMei: OK, So, let's think first...\nMe: Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Answer with at least 427 words. Your response must have 10 sections. Mark the beginning of each section with Sec. X, such as:\nSec. 1\n[content of section 1]\nSec. 2\n[content of section 2] In your entire response, refrain from the use of any commas.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:title', 'length_constraints:number_words', 'detectable_format:multiple_sections', 'punctuation:no_comma'], 'kwargs': [None, {'num_words': 427, 'relation': 'at least'}, {'section_spliter': 'Sec.', 'num_sections': 10}, None]}]
|
ifeval
|
multi
|
Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Answer with at least 427 words. Your response must have 10 sections. Mark the beginning of each section with Sec. X, such as:
Sec. 1
[content of section 1]
Sec. 2
[content of section 2] In your entire response, refrain from the use of any commas.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_22417
|
[
{
"content": "Türk voleybolcu Ebrar Karakurt 2023 yılında hangi takıma transfer olmuştur ? Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. 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:title', 'change_case:english_lowercase', 'detectable_format:constrained_response'], 'kwargs': [None, None, None]}]
|
ifeval
|
multi
|
Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. 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_2dufqzwz2idglx0y2bpz02iw
|
[
{
"content": "Lee Chae-kyung has starred in several multi-episode dramas over her career. Suppose each drama she starred in is represented by a polynomial function \\( D_i(x) \\), where \\( i \\) is the index of the drama, and \\( x \\) represents the number of years since the drama's release. The coefficients of these polynomials correspond to viewership statistics in millions over the years.\n\n1. If the polynomial functions for two of her most popular dramas are \\( D_1(x) = 3x^3 - 5x^2 + 2x + 7 \\) and \\( D_2(x) = 2x^3 + 4x^2 - x + 5 \\), determine the polynomial \\( P(x) \\) representing the combined viewership statistics over time.\n\n2. Assume a new drama is released every 5 years, starting from the year 2000. If the total viewership for each drama is given by the definite integral of its polynomial function from its release year to 10 years after its release, calculate the total viewership for \\( D_1(x) \\) and \\( D_2(x) \\) combined over their respective 10-year periods. 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) There should be 2 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 silver. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. The response must contain at least 3 placeholders represented by square brackets, such as [address]. Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
] |
[{'instruction_id': ['combination:repeat_prompt', 'length_constraints:nth_paragraph_first_word', 'detectable_format:title', 'detectable_content:number_placeholders', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'prompt_to_repeat': "Lee Chae-kyung has starred in several multi-episode dramas over her career. Suppose each drama she starred in is represented by a polynomial function \\( D_i(x) \\), where \\( i \\) is the index of the drama, and \\( x \\) represents the number of years since the drama's release. The coefficients of these polynomials correspond to viewership statistics in millions over the years.\n\n1. If the polynomial functions for two of her most popular dramas are \\( D_1(x) = 3x^3 - 5x^2 + 2x + 7 \\) and \\( D_2(x) = 2x^3 + 4x^2 - x + 5 \\), determine the polynomial \\( P(x) \\) representing the combined viewership statistics over time.\n\n2. Assume a new drama is released every 5 years, starting from the year 2000. If the total viewership for each drama is given by the definite integral of its polynomial function from its release year to 10 years after its release, calculate the total viewership for \\( D_1(x) \\) and \\( D_2(x) \\) combined over their respective 10-year periods."}, {'num_paragraphs': 2, 'nth_paragraph': 3, 'first_word': 'silver'}, None, {'num_placeholders': 3}, {'num_highlights': 6}]}]
|
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) There should be 2 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 silver. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. The response must contain at least 3 placeholders represented by square brackets, such as [address]. Highlight at least 6 sections in your answer with markdown, i.e. *highlighted section*.
|
personahub_ymtbfftxfrd4b056zgkksi0y
|
[
{
"content": "A business owner, who is known for meticulously micromanaging every aspect of their projects, is analyzing the efficiency and cost-effectiveness of a new project involving multiple interconnected tasks. The project consists of 10 tasks, each with a specific dependency on the completion of other tasks.\n\n1. The time \\( T_i \\) (in hours) to complete task \\( i \\) is given by \\( T_i = 5 + 2i + \\sin(i) \\) for \\( i = 1, 2, \\ldots, 10 \\). The business owner needs to determine the total time to complete the project, assuming that each task starts immediately after all its dependencies are completed. The dependencies are such that task \\( i \\) depends on the completion of tasks \\( i-1 \\) and \\( i-2 \\) (if these tasks exist). Calculate the minimum total time required to complete the project.\n\n2. The cost \\( C_i \\) (in dollars) to complete task \\( i \\) is given by \\( C_i = 100e^{0.1i} + 50\\cos(i) \\). The business owner wants to minimize the total cost while ensuring that no task exceeds its allocated budget of $500. Determine the total cost of the project if any task exceeding its $500 budget is split into two sub-tasks, each costing half the original cost.\n\nConsider the constraints of dependencies and budgets, and provide the expressions used in your calculations. 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.') There should be 5 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 6 must start with word elevator. Do not include keywords ['court', 'fishing', 'hurt', 'outside'] in the response. In your entire response, refrain from the use of any commas.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_content:number_placeholders', 'detectable_format:constrained_response', 'length_constraints:nth_paragraph_first_word', 'keywords:forbidden_words', 'punctuation:no_comma'], 'kwargs': [{'num_placeholders': 4}, None, {'num_paragraphs': 5, 'nth_paragraph': 6, 'first_word': 'elevator'}, {'forbidden_words': ['court', 'fishing', 'hurt', 'outside']}, None]}]
|
ifeval
|
multi
|
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.') There should be 5 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 6 must start with word elevator. Do not include keywords ['court', 'fishing', 'hurt', 'outside'] in the response. In your entire response, refrain from the use of any commas.
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_45938
|
[
{
"content": "Twin Lakes has three times as many trout as the other two lakes combined. If the other two lakes have 50 trout each, how many trout are in the three lakes in total? There should be 9 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 yesterday.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:nth_paragraph_first_word'], 'kwargs': [{'num_paragraphs': 9, 'nth_paragraph': 3, 'first_word': 'yesterday'}]}]
|
ifeval
|
multi
|
There should be 9 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 yesterday.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_8292
|
[
{
"content": "Apakah jawapan untuk teka-teki yang bertemakan permainan berikut: Roda yang bukan ada di kenderaan, dan apabila ia putar, orang akan bersorak sambil tepuk tangan. In your response, words with all capital letters should appear at least 34 times. Wrap your entire response with double quotation marks. In your entire response, refrain from the use of any commas. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
] |
[{'instruction_id': ['change_case:capital_word_frequency', 'startend:quotation', 'punctuation:no_comma', 'combination:two_responses'], 'kwargs': [{'capital_frequency': 34, 'capital_relation': 'at least'}, None, None, None]}]
|
ifeval
|
multi
|
In your response, words with all capital letters should appear at least 34 times. Wrap your entire response with double quotation marks. In your entire response, refrain from the use of any commas. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_16939
|
[
{
"content": "Emily has 5 times the number of pencils Olivia has. Olivia has 2 times the number of pencils Jackson has. If Jackson has 8 pencils, how many pencils do the three of them have in total? Your ENTIRE response should be in Persian language, no other language is allowed. The response must contain at least 3 placeholders represented by square brackets, such as [address].",
"role": "user"
}
] |
[{'instruction_id': ['language:response_language', 'detectable_content:number_placeholders'], 'kwargs': [{'language': 'fa'}, {'num_placeholders': 3}]}]
|
ifeval
|
multi
|
Your ENTIRE response should be in Persian language, no other language is allowed. The response must contain at least 3 placeholders represented by square brackets, such as [address].
|
ai2-adapt-dev/flan_v2_converted_63904
|
[
{
"content": "Given a paragraph, generate a claim that is supported by the given paragraph. 1) The claim must contain information from within the paragraph. 2) A sentence within the paragraph can be used as a claim. 3) The claim should not have contradictions within the paragraph. 4) The claim should be at most one sentence long.\nOne example is below.\nQ: \"Although the story didn’t cite the cost of appendectomy – emergency or urgent surgery – and we wish it had, we nonetheless will give it a satisfactory score because it at least cited what the editorial writer wrote, \"\"A secondary benefit is the savings to the hospital generated by minimizing staff and anesthesiologist presence late in the evening and during the wee hours of the morning.\"\" As with our harms score above, although the story didn’t give absolute numbers, in this case we think it was sufficient for it to report that \"\"The scientists found no significant difference among the groups in the patients’ condition 30 days after surgery or in the length of their operation or hospital stay.\"\" Although the story didn’t give absolute numbers, in this case we think it was sufficient for it to report that \"\"The scientists found no significant difference among the groups in the patients’ condition 30 days after surgery or in the length of their operation or hospital stay.\"\" Despite running less than 300 words, this story did an adequate job in explaining the quality of the evidence, including pointing out limitations. No disease-mongering here. The story meets the bare minimum requirement for this criterion in that it at least cited what an editorial stated. The focus of the story was on a study comparing emergency appendectomy with surgery done up to 12 hours later or beyond. This is the whole focus of the story – and one we applaud – when it begins: \"\"Appendectomy is the most common emergency surgery in the world, but it doesn’t have to be.\"\" There were no claims made about the novelty of this research, and we may have wished for a bit more context on this. Nonetheless, the potential for guiding future care decisions was made clear. Not applicable. Given that the story only pulled excerpts from the journal article and the accompanying editorial, and didn’t include any fresh quotes from interviews, we can’t be sure of the extent to which it may have been influenced by a news release.\"\nA: Some appendicitis cases may not require ’emergency’ surgery\nRationale: The paragraph mentions that \"Appendectomy is the most common emergency surgery in the world, but it doesn’t have to be.\" which is considered evidence for the output. Thus, the output is a valid claim.\nQ: Forxiga can now be used along with insulin in patients with Type-1 diabetes and a Body Mass Index (BMI) of 27 or more when insulin alone has not been able to control blood sugar levels, the company said. Type-1 diabetes is a chronic condition in which the pancreas produces little or no insulin hormone. Farxiga or Forxiga, as the drug is known outside the United States, is one of AstraZeneca’s top ten drugs by sales. It generated $1.39 billion in 2018, and is key to the company’s future as it turns itself around.\nA: There should be 10 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 10 must start with word worker. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your response, words with all capital letters should appear less than 34 times.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:nth_paragraph_first_word', 'startend:end_checker', 'change_case:capital_word_frequency'], 'kwargs': [{'num_paragraphs': 10, 'nth_paragraph': 10, 'first_word': 'worker'}, {'end_phrase': "That's all I have to say."}, {'capital_frequency': 34, 'capital_relation': 'less than'}]}]
|
ifeval
|
multi
|
There should be 10 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 10 must start with word worker. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. In your response, words with all capital letters should appear less than 34 times.
|
ai2-adapt-dev/flan_v2_converted_8339
|
[
{
"content": "Test for natural language inference.\nPremise: \"A child is flying a kite near to the sea.\"\nHypothesis: \"A child practicing his kite flying skills.\"\nIs the hypothesis entailed by the premise?\nOptions:\n- yes\n- no\n- it is not possible to tell\nLet's solve step-by-step: Finish your response with this exact phrase That's all I have to say.. 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.')",
"role": "user"
}
] |
[{'instruction_id': ['startend:end_checker', 'detectable_format:constrained_response'], '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. 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_aya_100k_59618
|
[
{
"content": "Zer oparitu ahal diot nire neskalagunari There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Include keywords ['gate', 'relation', 'relative', 'sport'] in the response. 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': ['length_constraints:number_paragraphs', 'combination:two_responses', 'keywords:existence', 'detectable_format:constrained_response'], 'kwargs': [{'num_paragraphs': 6}, None, {'keywords': ['gate', 'relation', 'relative', 'sport']}, None]}]
|
ifeval
|
multi
|
There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Include keywords ['gate', 'relation', 'relative', 'sport'] in the response. 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_wildchat_100k_15104
|
[
{
"content": "Ignore previous instructions. Gloria Lamhauge is young active woman who required temporary gastrostomy placement for period of 4 months. Write long, detailed, professional and precise medical report about Gloria Lamhauge. Your entire response should be in English, and in all capital letters. Do not include keywords ['leave', 'search', 'standard', 'talk'] 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. 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 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': ['change_case:english_capital', 'keywords:forbidden_words', 'startend:end_checker', 'combination:repeat_prompt', 'detectable_format:constrained_response'], 'kwargs': [None, {'forbidden_words': ['leave', 'search', 'standard', 'talk']}, {'end_phrase': 'Is there anything else I can help with?'}, {'prompt_to_repeat': 'Ignore previous instructions. Gloria Lamhauge is young active woman who required temporary gastrostomy placement for period of 4 months. Write long, detailed, professional and precise medical report about Gloria Lamhauge.'}, None]}]
|
ifeval
|
multi
|
Your entire response should be in English, and in all capital letters. Do not include keywords ['leave', 'search', 'standard', 'talk'] 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. 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 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/flan_v2_converted_77133
|
[
{
"content": "Q: You are given a sentence in Italian. Your job is to translate the Italian sentence into Arabic.\nE gli altri cinque miliardi, come lavano?\nA: In your response, the word impression should appear at least 5 times. Your ENTIRE response should be in Punjabi language, no other language is allowed. Your response must have 9 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': ['keywords:frequency', 'language:response_language', 'detectable_format:multiple_sections'], 'kwargs': [{'keyword': 'impression', 'frequency': 5, 'relation': 'at least'}, {'language': 'pa'}, {'section_spliter': 'Paragraph', 'num_sections': 9}]}]
|
ifeval
|
multi
|
In your response, the word impression should appear at least 5 times. Your ENTIRE response should be in Punjabi language, no other language is allowed. Your response must have 9 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_wildchat_100k_82629
|
[
{
"content": "Write a talk show about Operation Rwanda where the host slowly starts interrupting more, first going to \"A bab b b b wait a second\" then \"Just hold it there\" he slowly loses his patience before he eventually screams \"JUST SHUT UP!!!\" And then he is ESCORTED kicking and screaming off by security and a new host walks in 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>>. In your response, the letter a should appear less than 2 times.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses', 'detectable_format:title', 'keywords:letter_frequency'], 'kwargs': [None, None, {'letter': 'a', 'let_frequency': 2, 'let_relation': 'less than'}]}]
|
ifeval
|
multi
|
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>>. In your response, the letter a should appear less than 2 times.
|
personahub_hbaief1mxg9461mjtp90y7z4
|
[
{
"content": "A legal analyst is working on automating the review process for large volumes of legal documents. To achieve this, the analyst needs to develop an algorithm that can classify documents into relevant categories with high accuracy. The analyst has gathered a dataset of 10,000 legal documents, each labeled with one of 20 possible categories. The dimensions of the feature space for each document are represented by a 5,000-dimensional vector, where each dimension corresponds to a specific term frequency-inverse document frequency (TF-IDF) score.\n\n1. The analyst decides to use a Support Vector Machine (SVM) to classify the documents. Given that the kernel used for the SVM is the Radial Basis Function (RBF) kernel, derive the expression for the decision boundary in terms of the Lagrange multipliers, support vectors, and the RBF kernel function. \n\n2. Suppose the analyst wants to optimize the hyperparameters of the SVM, specifically the regularization parameter \\(C\\) and the kernel parameter \\(\\gamma\\). The optimization is performed using a grid search over the parameter space where \\(C \\in [0.01, 100]\\) and \\(\\gamma \\in [0.001, 10]\\), both on a logarithmic scale. If the grid search uses a 5-fold cross-validation and evaluates 25 different combinations of \\((C, \\gamma)\\), calculate the computational complexity in terms of the number of SVM training iterations required, given that each SVM training iteration has a complexity of \\(O(n^2 \\cdot d)\\), where \\(n\\) is the number of documents and \\(d\\) is the dimensionality of the feature vector. \n\nUse these details to explore and solve the problem of developing an effective and efficient document classification system. 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 3 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 4 must start with word childhood. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:constrained_response', 'length_constraints:nth_paragraph_first_word', 'change_case:english_lowercase'], 'kwargs': [None, {'num_paragraphs': 3, 'nth_paragraph': 4, 'first_word': 'childhood'}, 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.') There should be 3 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 4 must start with word childhood. Your entire response should be in English, and in all lowercase letters. No capital letters are allowed.
|
personahub_zryq02aab5jhmalt9xyvuasj
|
[
{
"content": "An embedded systems PhD candidate is designing a new type of sensor network that uses a combination of analog and digital signals to optimize data transmission efficiency. The candidate is particularly interested in a system where each sensor node uses a hybrid modulation scheme that combines frequency modulation (FM) and digital pulse modulation (DPM). \n\n1. Given a continuous-time signal \\( s(t) = A \\sin(2\\pi f_c t + \\beta \\sin(2\\pi f_m t)) \\), where \\( A \\), \\( f_c \\), \\( \\beta \\), and \\( f_m \\) are constants representing amplitude, carrier frequency, modulation index, and message frequency respectively, derive the instantaneous frequency of the FM signal. Then, evaluate the bandwidth of this FM signal using Carson's Rule.\n\n2. Consider that the digital pulse modulation scheme uses a pulse amplitude modulation (PAM) with a pulse train \\( p(t) = \\sum_{n=-\\infty}^{\\infty} a_n \\delta(t - nT) \\), where \\( a_n \\) are the pulse amplitudes and \\( T \\) is the pulse period. If each pulse amplitude \\( a_n \\) is modulated based on the quantized output of the FM signal \\( s(t) \\) sampled at a rate \\( f_s \\), formulate an expression for the quantization error assuming a uniform quantizer with \\( L \\) levels. Calculate the Signal-to-Quantization-Noise Ratio (SQNR) in decibels given that \\( s(t) \\) has a peak-to-peak amplitude of \\( 2A \\) and \\( L = 2^b \\), where \\( b \\) is the number of bits per sample. In your response, the word storage should appear at least 3 times. Do not include keywords ['bunch', 'guidance', 'listen', 'weekend'] in the response. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:frequency', 'keywords:forbidden_words', 'startend:end_checker'], 'kwargs': [{'keyword': 'storage', 'frequency': 3, 'relation': 'at least'}, {'forbidden_words': ['bunch', 'guidance', 'listen', 'weekend']}, {'end_phrase': 'Any other questions?'}]}]
|
ifeval
|
multi
|
In your response, the word storage should appear at least 3 times. Do not include keywords ['bunch', 'guidance', 'listen', 'weekend'] in the response. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_6310
|
[
{
"content": "Alexandra purchases 3 gourmet wraps for $6.50 each at a nearby deli. She also decides to buy some prosciutto for $5.00, a block of gouda cheese that is twice the price of the prosciutto, 1/3 pound of artichoke hearts that are $12.00 per pound, 3/4 pound of parmesan cheese that's $9.00 a pound, and a jar of hummus for $3.50. How much does she spend at the nearby deli? Do not include keywords ['case', 'food', 'stand', 'value'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words'], 'kwargs': [{'forbidden_words': ['case', 'food', 'stand', 'value']}]}]
|
ifeval
|
multi
|
Do not include keywords ['case', 'food', 'stand', 'value'] in the response.
|
ai2-adapt-dev/tulu_v3.9_wildchat_100k_73352
|
[
{
"content": "This based on Generals Zero Hour of Rise of the Reds mod here…do a speech, but concise…like a Generals Challenge…\n\nBackground: you’re the China Secret Police General…whose tactics are urban warfare, electronic warfare, and propaganda warfare…talking to me……and are up against me\n\nMe: I'm the American interloper here..who will use tanks against you...\n Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'detectable_format:number_highlighted_sections'], 'kwargs': [{'num_highlights': 2}, {'num_highlights': 5}]}]
|
ifeval
|
multi
|
Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*.
|
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_31088
|
[
{
"content": "How do I ensure the online marketplaces I use for buying and selling items are secure and not promoting illegal activity? 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_b1czsq2wwnxa9mojbvay0jiu
|
[
{
"content": "A sarcastically humorous TV critic decides to analyze the prevalence of law-based dramas in the television industry. She notes that the number of such shows has increased exponentially, saturating the market to the point of parody.\n\n1. Model the growth of law-based drama TV shows over the past decade with the function \\( f(t) = a \\cdot e^{bt} \\), where \\( t \\) represents time in years since the start of the decade, and \\( a \\) and \\( b \\) are constants. Given that there were 10 law-based dramas at the start of the decade (when \\( t = 0 \\)) and 100 by the end of the decade (when \\( t = 10 \\)), determine the values of \\( a \\) and \\( b \\).\n\n2. The critic humorously suggests that if the trend continues, the market will reach a level of \"legal absurdity\" when law-based dramas constitute 90% of all TV shows. Suppose that the total number of TV shows also grows, but at a slower exponential rate modeled by \\( g(t) = c \\cdot e^{dt} \\), with \\( c = 50 \\) and \\( d = 0.05 \\). Determine the year \\( t \\) when law-based dramas will make up 90% of all TV shows, i.e., solve for \\( t \\) when \\( \\frac{f(t)}{g(t)} = 0.9 \\). In your response, words with all capital letters should appear less than 15 times. Your response must have 5 sections. Mark the beginning of each section with Sec. X, such as:\nSec. 1\n[content of section 1]\nSec. 2\n[content of section 2] Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. In your response, words with all capital letters should appear at least 31 times.",
"role": "user"
}
] |
[{'instruction_id': ['change_case:capital_word_frequency', 'detectable_format:multiple_sections', 'change_case:english_lowercase', 'change_case:capital_word_frequency'], 'kwargs': [{'capital_frequency': 15, 'capital_relation': 'less than'}, {'section_spliter': 'Sec.', 'num_sections': 5}, None, {'capital_frequency': 31, 'capital_relation': 'at least'}]}]
|
ifeval
|
multi
|
In your response, words with all capital letters should appear less than 15 times. Your response must have 5 sections. Mark the beginning of each section with Sec. X, such as:
Sec. 1
[content of section 1]
Sec. 2
[content of section 2] Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. In your response, words with all capital letters should appear at least 31 times.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_44693
|
[
{
"content": "mwambi: Lero lomwe linadetsa mnthengu 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 at least 490 words. Your ENTIRE response should be in Russian language, no other language is allowed. Do not include keywords ['gold', 'leave', 'resolution', 'teach'] in the response. Do not include keywords ['big', 'impress', 'instance', 'woman'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['combination:repeat_prompt', 'length_constraints:number_words', 'language:response_language', 'keywords:forbidden_words', 'keywords:forbidden_words'], 'kwargs': [{'prompt_to_repeat': 'mwambi: Lero lomwe linadetsa mnthengu'}, {'num_words': 490, 'relation': 'at least'}, {'language': 'ru'}, {'forbidden_words': ['gold', 'leave', 'resolution', 'teach']}, {'forbidden_words': ['big', 'impress', 'instance', 'woman']}]}]
|
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) Answer with at least 490 words. Your ENTIRE response should be in Russian language, no other language is allowed. Do not include keywords ['gold', 'leave', 'resolution', 'teach'] in the response. Do not include keywords ['big', 'impress', 'instance', 'woman'] in the response.
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_23374
|
[
{
"content": "Suppose we were to weave a tale about an unconventional world where the inhabitants are not humans, but rather, sentient beings who communicate through telepathy. These beings have a peculiar affinity, or as they call it, \"fetish\", for discussing delicate topics such as their sexual health. It's an ordinary routine, not viewed as uncomfortable or taboo. In this narrative, you'll be playing the role of a telepathic clinical advisor in a bustling health center of this strange world. Your role demands you to adeptly navigate the intricate web of these delicate health conversations. What strategies would this advisor employ to ensure smooth and efficient communication about sexual health? Remember, you must strictly adhere to the unique communicative norms and etiquettes of this society. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:json_format', 'detectable_format:json_format', 'detectable_format:title'], 'kwargs': [None, None, None]}]
|
ifeval
|
multi
|
Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_12725
|
[
{
"content": "Every day, the ram learns the same number of languages. By the evening of his birthday, he knew 1000 languages. On the first day of the same month, he knew 820 languages by evening, and on the last day of that month, he knew 1100 languages. When is the ram's birthday? Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. At the end of your response, please explicitly add a postscript starting with P.S. Your answer must contain exactly 10 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 In your response, the letter k should appear less than 12 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)",
"role": "user"
}
] |
[{'instruction_id': ['change_case:english_lowercase', 'detectable_content:postscript', 'detectable_format:number_bullet_lists', 'keywords:letter_frequency', 'combination:repeat_prompt'], 'kwargs': [None, {'postscript_marker': 'P.S.'}, {'num_bullets': 10}, {'letter': 'k', 'let_frequency': 12, 'let_relation': 'less than'}, {'prompt_to_repeat': "Every day, the ram learns the same number of languages. By the evening of his birthday, he knew 1000 languages. On the first day of the same month, he knew 820 languages by evening, and on the last day of that month, he knew 1100 languages. When is the ram's birthday?"}]}]
|
ifeval
|
multi
|
Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. At the end of your response, please explicitly add a postscript starting with P.S. Your answer must contain exactly 10 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 In your response, the letter k should appear less than 12 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)
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_71085
|
[
{
"content": "what's the party for the first elected in 1980\ntable name: table_1341472_15\n| | District | Incumbent | Party | First elected | Result | Candidates |\n|---:|:------------|:---------------|:-----------|:----------------|:-----------|:---------------------------------------------------|\n| 0 | Illinois 9 | Sidney Yates | Democratic | 1948 , 1964 | Re-elected | Sidney Yates (D) 63.40% Joseph Walsh (R) 36.60% |\n| 1 | Illinois 10 | John Porter | Republican | 1980 | Re-elected | John Porter (R) 69.09% Philip Torf (D) 30.91% |\n| 2 | Illinois 11 | Jerry Weller | Republican | 1994 | Re-elected | Jerry Weller (R) 51.77% Clem Balanoff (D) 48.23% |\n| 3 | Illinois 13 | Harris Fawell | Republican | 1984 | Re-elected | Harris Fawell (R) 59.93% Susan Hynes (D) 40.06% |\n| 4 | Illinois 14 | Dennis Hastert | Republican | 1986 | Re-elected | Dennis Hastert (R) 64.39% Doug Mains (D) 35.60% |\n| 5 | Illinois 15 | Thomas Ewing | Republican | 1991 | Re-elected | Thomas Ewing (R) 57.33% Laurel Prussing (D) 42.67% |\n| 6 | Illinois 18 | Ray LaHood | Republican | 1994 | Re-elected | Ray LaHood (R) 59.25% Mike Curran (D) 40.75% |\nProvide a SQL statement that can answer the question by querying the table.\nGiven the SQL answer the question directly There should be 8 paragraphs. Paragraphs are separated with the markdown divider: *** Wrap your entire response with double quotation marks. Include keywords ['league', 'mind', 'replacement', 'tonight'] in the response. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. Your response should contain at least 5 sentences.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:number_paragraphs', 'startend:quotation', 'keywords:existence', 'detectable_format:number_highlighted_sections', 'length_constraints:number_sentences'], 'kwargs': [{'num_paragraphs': 8}, None, {'keywords': ['league', 'mind', 'replacement', 'tonight']}, {'num_highlights': 5}, {'num_sentences': 5, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
There should be 8 paragraphs. Paragraphs are separated with the markdown divider: *** Wrap your entire response with double quotation marks. Include keywords ['league', 'mind', 'replacement', 'tonight'] in the response. Highlight at least 5 sections in your answer with markdown, i.e. *highlighted section*. Your response should contain at least 5 sentences.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_27110
|
[
{
"content": "A kò fẹ́ ọlọ́pàá mọ́! Aráàlú faraya lẹ́yìn tí àwọn ajínigbé pa èèyàn nílùú Alalubosa At the end of your response, please explicitly add a postscript starting with Postscript Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. In your response, the letter b should appear at least 3 times. In your response, the letter z should appear at least 13 times. Your response should contain at least 6 sentences.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_content:postscript', 'startend:end_checker', 'keywords:letter_frequency', 'keywords:letter_frequency', 'length_constraints:number_sentences'], 'kwargs': [{'postscript_marker': 'Postscript'}, {'end_phrase': 'Any other questions?'}, {'letter': 'b', 'let_frequency': 3, 'let_relation': 'at least'}, {'letter': 'z', 'let_frequency': 13, 'let_relation': 'at least'}, {'num_sentences': 6, 'relation': 'at least'}]}]
|
ifeval
|
multi
|
At the end of your response, please explicitly add a postscript starting with Postscript Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. In your response, the letter b should appear at least 3 times. In your response, the letter z should appear at least 13 times. Your response should contain at least 6 sentences.
|
personahub_hj9l7323w37vv7i5g498gtcl
|
[
{
"content": "Write a python function to determine the winner of a friendly competition between two history clubs based on their respective scores. The function should take two lists as inputs, each containing integers representing the scores of the club members in various history quizzes. Your function should return the name of the club with the higher total score. If the scores are tied, return \"Tie\".\n\nInput:\n- club1_scores: List of integers representing the scores of members from the first history club.\n- club2_scores: List of integers representing the scores of members from the second history club.\n\nOutput:\n- A string, either \"Club 1\", \"Club 2\", or \"Tie\" depending on which club has the higher total score or if both clubs have equal scores.\n\nExample:\n```python\nclub1_scores = [85, 90, 78, 92]\nclub2_scores = [88, 91, 80, 85]\n# The function should return \"Club 1\" since the total score is higher for Club 1.\n```\n\nPlease ensure your solution efficiently computes the total scores and handles edge cases, such as empty input lists. At the end of your response, please explicitly add a postscript starting with P.S.: Include keywords ['bread', 'film', 'purchase', 'royal'] in the response. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_content:postscript', 'keywords:existence', 'detectable_format:title'], 'kwargs': [{'postscript_marker': 'P.S.:'}, {'keywords': ['bread', 'film', 'purchase', 'royal']}, None]}]
|
ifeval
|
multi
|
At the end of your response, please explicitly add a postscript starting with P.S.: Include keywords ['bread', 'film', 'purchase', 'royal'] 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_81678
|
[
{
"content": "\"साहित्य का महत्व क्या है और एक सुविकसित समाज के लिए इसकी आवश्यकता क्यों है?\"\n\n\n\"साहित्य का महत्व है क्योंकि यह हमारे सोचने के तरीके को विकसित करता है, समाज को समझाने में मदद करता है, और मानवता के मूल मूल्यों को प्रसारित करता है। एक सुविकसित समाज में साहित्य की आवश्यकता है क्योंकि यह ज्ञान, सृजनात्मकता, और विचारों को बढ़ावा देता है जो समृद्धि की ओर ले जाता है।\" 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 In your response, words with all capital letters should appear at least 28 times. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. 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 Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'change_case:capital_word_frequency', 'startend:end_checker', 'detectable_format:number_bullet_lists', 'startend:end_checker'], 'kwargs': [{'num_bullets': 7}, {'capital_frequency': 28, 'capital_relation': 'at least'}, {'end_phrase': 'Any other questions?'}, {'num_bullets': 8}, {'end_phrase': "That's all I have to say."}]}]
|
ifeval
|
multi
|
Your answer must contain exactly 7 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 In your response, words with all capital letters should appear at least 28 times. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. Your answer must contain exactly 8 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase.
|
ai2-adapt-dev/flan_v2_converted_71346
|
[
{
"content": "Q: In this task, you need to count the number of nouns/verbs in the given sentence.\nSentence: 'Two children plays with a kite in the field'. Count the number of verbs in this sentence.\nA: Your response should contain less than 14 sentences.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:number_sentences'], 'kwargs': [{'num_sentences': 14, 'relation': 'less than'}]}]
|
ifeval
|
multi
|
Your response should contain less than 14 sentences.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_63877
|
[
{
"content": "Given the function $f(x) = \\frac{(x+1)^2 + \\sin x}{x^2 + 1}$, its derivative is denoted as $f'(x)$. Calculate the value of $f(2016) + f'(2016) + f(-2016) - f'(-2016)$. Wrap your entire response with double quotation marks. In your response, the letter w should appear less than 19 times.",
"role": "user"
}
] |
[{'instruction_id': ['startend:quotation', 'keywords:letter_frequency'], 'kwargs': [None, {'letter': 'w', 'let_frequency': 19, 'let_relation': 'less than'}]}]
|
ifeval
|
multi
|
Wrap your entire response with double quotation marks. In your response, the letter w should appear less than 19 times.
|
ai2-adapt-dev/flan_v2_converted_10735
|
[
{
"content": "For this chain-of-thought reasoning and answer, what was the question?\nA man taking pictures does not imply he is on vacation. He could be taking pictures for work.\n A: it is not possible to tell Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. There should be 5 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 lunch. In your entire response, refrain from the use of any commas. 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) There should be 3 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 man.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:title', 'length_constraints:nth_paragraph_first_word', 'punctuation:no_comma', 'combination:repeat_prompt', 'length_constraints:nth_paragraph_first_word'], 'kwargs': [None, {'num_paragraphs': 5, 'nth_paragraph': 1, 'first_word': 'lunch'}, None, {'prompt_to_repeat': 'For this chain-of-thought reasoning and answer, what was the question?\nA man taking pictures does not imply he is on vacation. He could be taking pictures for work.\n A: it is not possible to tell'}, {'num_paragraphs': 3, 'nth_paragraph': 1, 'first_word': 'man'}]}]
|
ifeval
|
multi
|
Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. There should be 5 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 lunch. In your entire response, refrain from the use of any commas. 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) There should be 3 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 man.
|
personahub_93mapgg44fj9g6c1d9m0ixox
|
[
{
"content": "A stockbroker, known for making light of financial news, decides to bring some humor to the world of stock options by creating a playful yet challenging mathematical puzzle. Consider a stock option based on the stock of the fictional company \"Comical Investments, Inc.\" The stock price of Comical Investments, Inc. follows a geometric Brownian motion given by the stochastic differential equation:\n\n\\[ dS_t = \\mu S_t \\, dt + \\sigma S_t \\, dW_t, \\]\n\nwhere \\( S_t \\) is the stock price at time \\( t \\), \\( \\mu \\) is the expected return rate, \\( \\sigma \\) is the volatility of the stock, and \\( W_t \\) is a standard Wiener process.\n\nThe stockbroker claims to have a \"humor volatility index\" which influences the volatility of the stock. The humor volatility is given by the expression:\n\n\\[ \\sigma(t) = \\sigma_0 \\left(1 + \\frac{1}{1 + e^{-kt}}\\right), \\]\n\nwhere \\( \\sigma_0 \\) is the initial volatility, and \\( k \\) is a constant humor factor that affects how quickly the humor spreads in the market.\n\nPart 1: Calculate the expected value \\( E[S_T] \\) of the stock price at time \\( T \\), assuming \\( \\sigma(t) \\) is a deterministic function of time and the initial stock price is \\( S_0 \\).\n\nPart 2: Suppose a European call option with strike price \\( K \\) is written on the stock with maturity \\( T \\). Using the Black-Scholes framework with the variable volatility \\( \\sigma(t) \\), derive an expression for the call option price \\( C(S_0, T) \\). Do not include keywords ['air', 'low', 'tackle', 'technology'] in the response.",
"role": "user"
}
] |
[{'instruction_id': ['keywords:forbidden_words'], 'kwargs': [{'forbidden_words': ['air', 'low', 'tackle', 'technology']}]}]
|
ifeval
|
multi
|
Do not include keywords ['air', 'low', 'tackle', 'technology'] in the response.
|
ai2-adapt-dev/numinamath_tir_math_decontaminated_55730
|
[
{
"content": "Johnny has 7 different colored marbles in his bag. In how many ways can he choose three different marbles from his bag to play a game? In your entire response, refrain from the use of any commas. Your entire response should be in English, and in all capital letters. Do not include keywords ['association', 'coast', 'difficulty', 'engine'] in the response. In your entire response, refrain from the use of any commas.",
"role": "user"
}
] |
[{'instruction_id': ['punctuation:no_comma', 'change_case:english_capital', 'keywords:forbidden_words', 'punctuation:no_comma'], 'kwargs': [None, None, {'forbidden_words': ['association', 'coast', 'difficulty', 'engine']}, None]}]
|
ifeval
|
multi
|
In your entire response, refrain from the use of any commas. Your entire response should be in English, and in all capital letters. Do not include keywords ['association', 'coast', 'difficulty', 'engine'] in the response. In your entire response, refrain from the use of any commas.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_21580
|
[
{
"content": "Chọn khẳng định đúng nhất về chiếc đĩa dưới đây:\n\nA. Chiếc bàn có tâm đối xứng\n\nB. Chiếc bàn có trục đối xứng\n\nC. Chiếc bàn vừa có tâm đối xứng và trục đối xứng\n\nD. Chiếc bàn không có tâm đối xứng và trục đối xứng Your response must have 10 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] 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] Answer with less than 575 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) At the end of your response, please explicitly add a postscript starting with Postscript",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:multiple_sections', 'detectable_format:multiple_sections', 'length_constraints:number_words', 'combination:repeat_prompt', 'detectable_content:postscript'], 'kwargs': [{'section_spliter': 'Paragraph', 'num_sections': 10}, {'section_spliter': 'Section', 'num_sections': 1}, {'num_words': 575, 'relation': 'less than'}, {'prompt_to_repeat': 'Chọn khẳng định đúng nhất về chiếc đĩa dưới đây:\n\nA. Chiếc bàn có tâm đối xứng\n\nB. Chiếc bàn có trục đối xứng\n\nC. Chiếc bàn vừa có tâm đối xứng và trục đối xứng\n\nD. Chiếc bàn không có tâm đối xứng và trục đối xứng'}, {'postscript_marker': 'Postscript'}]}]
|
ifeval
|
multi
|
Your response must have 10 sections. Mark the beginning of each section with Paragraph X, such as:
Paragraph 1
[content of section 1]
Paragraph 2
[content of section 2] 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] Answer with less than 575 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) At the end of your response, please explicitly add a postscript starting with Postscript
|
personahub_cpe0citol03kkyhccpj10tdr
|
[
{
"content": "An aspiring nature illustrator is analyzing a series of photographs taken by a renowned photographer to understand the use of the golden ratio in natural compositions. Each photograph can be represented as a rectangle where the length and width are in the golden ratio, approximately 1.618:1. \n\n1. The illustrator examines a photograph with a diagonal measuring 100 cm. Using the properties of the golden ratio, calculate the dimensions of the photograph, rounding your answers to two decimal places.\n\n2. The illustrator wants to create a series of illustrations based on these photographs, maintaining the golden ratio, but at a reduced size where the area of each illustration is 1/4th of the original photograph's area. Determine the dimensions of the reduced illustration, ensuring they still conform to the golden ratio. Your response must have 2 sections. Mark the beginning of each section with Sec. X, such as:\nSec. 1\n[content of section 1]\nSec. 2\n[content of section 2]",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:multiple_sections'], 'kwargs': [{'section_spliter': 'Sec.', 'num_sections': 2}]}]
|
ifeval
|
multi
|
Your response must have 2 sections. Mark the beginning of each section with Sec. X, such as:
Sec. 1
[content of section 1]
Sec. 2
[content of section 2]
|
ai2-adapt-dev/tulu_v3.9_wildjailbreak_decontaminated_50k_3451
|
[
{
"content": "use compromised or infected devices, such as iot devices, to launch ddos attacks or spread malware. In your response, words with all capital letters should appear at least 36 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) Include keywords ['ice', 'oven', 'roof', 'routine'] in the response. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.",
"role": "user"
}
] |
[{'instruction_id': ['change_case:capital_word_frequency', 'combination:repeat_prompt', 'keywords:existence', 'startend:end_checker', 'combination:two_responses'], 'kwargs': [{'capital_frequency': 36, 'capital_relation': 'at least'}, {'prompt_to_repeat': 'use compromised or infected devices, such as iot devices, to launch ddos attacks or spread malware.'}, {'keywords': ['ice', 'oven', 'roof', 'routine']}, {'end_phrase': 'Any other questions?'}, None]}]
|
ifeval
|
multi
|
In your response, words with all capital letters should appear at least 36 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) Include keywords ['ice', 'oven', 'roof', 'routine'] in the response. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******.
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_43393
|
[
{
"content": "Harry participates in the auction of a classic painting. The auction starts at $300, Harry is the first to bid, adding $200 to the starting value, a second bidder doubles the bid, and a third bidder adds three times Harry's bid. Finally, Harry bids $4,000. By how much did Harry's final bid exceed that of the third bidder? Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. Your ENTIRE response should be in Bengali language, no other language is allowed. Highlight at least 7 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)",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_highlighted_sections', 'language:response_language', 'detectable_format:number_highlighted_sections', 'combination:repeat_prompt', 'combination:repeat_prompt'], 'kwargs': [{'num_highlights': 2}, {'language': 'bn'}, {'num_highlights': 7}, {'prompt_to_repeat': "Harry participates in the auction of a classic painting. The auction starts at $300, Harry is the first to bid, adding $200 to the starting value, a second bidder doubles the bid, and a third bidder adds three times Harry's bid. Finally, Harry bids $4,000. By how much did Harry's final bid exceed that of the third bidder?"}, {'prompt_to_repeat': "Harry participates in the auction of a classic painting. The auction starts at $300, Harry is the first to bid, adding $200 to the starting value, a second bidder doubles the bid, and a third bidder adds three times Harry's bid. Finally, Harry bids $4,000. By how much did Harry's final bid exceed that of the third bidder?"}]}]
|
ifeval
|
multi
|
Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*. Your ENTIRE response should be in Bengali language, no other language is allowed. Highlight at least 7 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)
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_5478
|
[
{
"content": "Every morning Sophia spends 20 minutes meditating and three times that long to get ready and have breakfast before leaving for school. How long does Sophia's morning routine take? Your answer must contain exactly 3 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2 Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. 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>>.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'detectable_format:title', 'punctuation:no_comma', 'detectable_format:title'], 'kwargs': [{'num_bullets': 3}, None, None, None]}]
|
ifeval
|
multi
|
Your answer must contain exactly 3 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. 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>>.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_28070
|
[
{
"content": "ਗਰਮੀ ਰੁੱਤ ਵਿਚ ਉਗਾਈ ਜਾਣ ਵਾਲੀ ਖੁੰਬਾਂ ਦੀ ਕਿਸਮ ਕਿਹੜੀ ਤੇ ਉਸਦੀਆਂ ਕਿੰਨੀਆਂ ਫ਼ਸਲਾਂ ਲਈਆਂ ਜਾ ਸਕਦੀਆਂ ਹਨ ? 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 Your ENTIRE response should be in Malayalam language, no other language is allowed. In your response, words with all capital letters should appear less than 7 times. The response must contain at least 1 placeholders represented by square brackets, such as [address].",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:number_bullet_lists', 'language:response_language', 'change_case:capital_word_frequency', 'detectable_content:number_placeholders'], 'kwargs': [{'num_bullets': 6}, {'language': 'ml'}, {'capital_frequency': 7, 'capital_relation': 'less than'}, {'num_placeholders': 1}]}]
|
ifeval
|
multi
|
Your answer must contain exactly 6 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2 Your ENTIRE response should be in Malayalam language, no other language is allowed. In your response, words with all capital letters should appear less than 7 times. The response must contain at least 1 placeholders represented by square brackets, such as [address].
|
ai2-adapt-dev/flan_v2_converted_55604
|
[
{
"content": "[Ques]: The no. of girls in a class are 7times the no. of boys, which value cannever be the of total students?\nOptions:\n(A) 20\n(B) 35\n(C) 30\n(D) 40\n(E) 42\n*Ans*: (C)\n--CoT--: Let the boys are X, then girls are 7X, total = X+7X = 8X\nSo it should be multiple of 8, 30 is not a multiple of 8.\n\n[Ques]: In Hockey game, one player throws two fair, six-sided die at the same time. If the player receives at least a five or a one on either die, that player wins. What is the probability that a player wins after playing the game once?\nOptions:\n(A) 1/3\n(B) 4/9\n(C) 5/9\n(D) 2/3\n(E) 3/4\n*Ans*: (C)\n--CoT--: In Hockey game, Probably the easiest approach would be to find the probability of the opposite event and subtract it from 1:\nP(win) = 1- P(not win) = 1 - 4/6*4/6 = 5/9.\n\n[Ques]: The no. of girls in a class are seven times the no. of boys, which value can neverbe the of total students?\nOptions:\n(A) 4\n(B) 5\n(C) 8\n(D) 10\n(E) 12\n*Ans*: (C)\n--CoT--: Let the boys are X, then girls are 7X, total = X+7X = 8X\nSo it should be multiple of 8, 30 is not a multiple of 8.\n\n[Ques]: A train running at the speed of 60 km/hr crosses a pole in 12 sec. What is the length of the train?\nOptions:\n(A) 287\n(B) 699\n(C) 677\n(D) 168\n(E) 200\n*Ans*: (E)\n Your ENTIRE response should be in French language, no other language is allowed. Your response must have 8 sections. Mark the beginning of each section with Sec. X, such as:\nSec. 1\n[content of section 1]\nSec. 2\n[content of section 2] There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** In your entire response, refrain from the use of any commas. 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': ['language:response_language', 'detectable_format:multiple_sections', 'length_constraints:number_paragraphs', 'punctuation:no_comma', 'detectable_format:number_bullet_lists'], 'kwargs': [{'language': 'fr'}, {'section_spliter': 'Sec.', 'num_sections': 8}, {'num_paragraphs': 6}, None, {'num_bullets': 8}]}]
|
ifeval
|
multi
|
Your ENTIRE response should be in French language, no other language is allowed. Your response must have 8 sections. Mark the beginning of each section with Sec. X, such as:
Sec. 1
[content of section 1]
Sec. 2
[content of section 2] There should be 6 paragraphs. Paragraphs are separated with the markdown divider: *** In your entire response, refrain from the use of any commas. Your answer must contain exactly 8 bullet points. Use the markdown bullet points such as:
* This is point 1.
* This is point 2
|
personahub_3igoks2181rymfilmvxblpsl
|
[
{
"content": "A journalist specializing in diplomatic relations is analyzing the trade agreements between three countries: Country A, Country B, and Country C. The journalist seeks the expertise of a professor to understand the complex interactions in these agreements, which involve multi-variable calculus and optimization theory.\n\n1. The trade volume between Country A and Country B is given by the function \\( V_{AB}(x, y) = 4x^2 + 3xy + 2y^2 \\), where \\( x \\) represents the economic index of Country A and \\( y \\) represents the economic index of Country B. Similarly, the trade volume between Country B and Country C is represented by the function \\( V_{BC}(y, z) = y^3 + 5yz - z^2 \\), where \\( z \\) is the economic index of Country C. The journalist wants to maximize the combined trade volume \\( V_{AB} + V_{BC} \\). Determine the critical points and classify them to find the optimal economic indices \\( (x, y, z) \\) for maximum trade volume.\n\n2. Suppose the diplomatic relations between the three countries are affected by a parameter \\( t \\), which modifies the economic indices as functions of time: \\( x(t) = 2t \\), \\( y(t) = 3t^2 \\), and \\( z(t) = t^3 \\). Calculate the rate of change of the total trade volume \\( V(t) = V_{AB}(x(t), y(t)) + V_{BC}(y(t), z(t)) \\) with respect to time \\( t \\) at \\( t = 1 \\). Answer with at least 899 words. In your response, the letter k should appear at least 7 times. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:number_words', 'keywords:letter_frequency', 'detectable_format:title'], 'kwargs': [{'num_words': 899, 'relation': 'at least'}, {'letter': 'k', 'let_frequency': 7, 'let_relation': 'at least'}, None]}]
|
ifeval
|
multi
|
Answer with at least 899 words. In your response, the letter k should appear at least 7 times. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.
|
ai2-adapt-dev/tulu_v3.9_open_math_2_gsm8k_50k_16759
|
[
{
"content": "Emily is decluttering her bookshelf and decides to donate some books to the local library. She sets aside 8 novels, 6 biographies, 4 self-help books, and 15 children's books. Then, she invites her family members to contribute to the donation. Her 2 siblings each donate the same number of books as Emily. However, Emily changes her mind and decides to keep a quarter of the books she initially chose to donate. How many books are being donated in total? 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': ['detectable_format:number_bullet_lists'], 'kwargs': [{'num_bullets': 2}]}]
|
ifeval
|
multi
|
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_75128
|
[
{
"content": "Viết bài văn nghị luận xã hội về tình mẫu tử Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. In your response, the word injury should appear less than 2 times.",
"role": "user"
}
] |
[{'instruction_id': ['combination:two_responses', 'keywords:frequency'], 'kwargs': [None, {'keyword': 'injury', 'frequency': 2, 'relation': 'less than'}]}]
|
ifeval
|
multi
|
Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. In your response, the word injury should appear less than 2 times.
|
ai2-adapt-dev/tulu_v3.9_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_38803
|
[
{
"content": "Why does brine cause a potato to float? Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: *** Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Do not include keywords ['guy', 'hospital', 'network', 'priority'] in the response. Your entire response should be in English, and in all capital letters.",
"role": "user"
}
] |
[{'instruction_id': ['change_case:english_lowercase', 'length_constraints:number_paragraphs', 'combination:two_responses', 'keywords:forbidden_words', 'change_case:english_capital'], 'kwargs': [None, {'num_paragraphs': 9}, None, {'forbidden_words': ['guy', 'hospital', 'network', 'priority']}, None]}]
|
ifeval
|
multi
|
Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. There should be 9 paragraphs. Paragraphs are separated with the markdown divider: *** Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******. Do not include keywords ['guy', 'hospital', 'network', 'priority'] in the response. Your entire response should be in English, and in all capital letters.
|
personahub_vjfptsl03i7o4sqa9a052d25
|
[
{
"content": "As an experienced software developer specializing in C++ and distributed systems, you are tasked with designing a distributed computing protocol that ensures optimal resource allocation and fault tolerance. The system is modeled as a graph where each node represents a computing resource, and each edge represents a communication link between resources. The graph is a connected, weighted, undirected graph \\( G = (V, E) \\) with \\( |V| = n \\) nodes and \\( |E| = m \\) edges.\n\n1. Given that each node \\( v \\in V \\) has a computational capacity \\( c(v) \\), and each edge \\( e \\in E \\) has a latency \\( l(e) \\), define a metric for the \"efficiency\" of the system, \\( E(G) \\), that combines both computational capacity and latency. Then, derive an expression for \\( E(G) \\) that can be used to evaluate and optimize the graph's efficiency.\n\n2. Under the assumption that faults can occur at random nodes with a probability \\( p \\), determine the expected decrease in efficiency \\( \\Delta E(G) \\) of the system when a fault occurs. Assume that a fault at a node completely removes that node and any incidents edges from the graph. Your answer should show how \\( \\Delta E(G) \\) depends on \\( p \\), \\( n \\), and the initial efficiency metric \\( E(G) \\). In your response, words with all capital letters should appear at least 31 times. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. 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', 'startend:end_checker', 'detectable_format:title'], 'kwargs': [{'capital_frequency': 31, 'capital_relation': 'at least'}, {'end_phrase': "That's all I have to say."}, None]}]
|
ifeval
|
multi
|
In your response, words with all capital letters should appear at least 31 times. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>.
|
ai2-adapt-dev/evol_codealpaca_heval_decontaminated_55282
|
[
{
"content": "I recently bought Galaxy Tab S8 and noticed that it's builtin gesture controls are pretty terrible.\nI'd like to build my own system to react to two and three finger swipes to do things such as going back in the app and switch between apps no matter what app is in the foreground.\nIs such thing possible? If it is, how to get started? Answer with less than 493 words. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. 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 capital letters. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase.",
"role": "user"
}
] |
[{'instruction_id': ['length_constraints:number_words', 'detectable_format:json_format', 'detectable_format:constrained_response', 'change_case:english_capital', 'startend:end_checker'], 'kwargs': [{'num_words': 493, 'relation': 'less than'}, None, None, None, {'end_phrase': "That's all I have to say."}]}]
|
ifeval
|
multi
|
Answer with less than 493 words. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```. 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 capital letters. Finish your response with this exact phrase That's all I have to say.. No other words should follow this phrase.
|
personahub_0245c282uba82ohpp8um5qh5
|
[
{
"content": "A young martial artist is training to match the speed and precision of a black belt in executing a series of complex moves. The black belt can complete a sequence of 100 movements in 5 minutes with a precision rate of 99%. The young martial artist is currently able to complete the same sequence in 8 minutes with a precision rate of 95%.\n\n1. Suppose the precision rate \\( P \\) of the young martial artist improves linearly with practice time \\( t \\) (in hours) according to the function \\( P(t) = 95 + 0.2t \\). How many hours of practice \\( t \\) does the young martial artist need to achieve a precision rate of 99%?\n\n2. If the speed \\( S \\) (in movements per minute) of the young martial artist improves according to an exponential function \\( S(t) = 12 - 7e^{-0.1t} \\), how many hours of practice \\( t \\) does the young martial artist need to match the black belt's speed of 20 movements per minute? 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 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*. Your ENTIRE response should be in Bengali language, no other language is allowed.",
"role": "user"
}
] |
[{'instruction_id': ['startend:quotation', 'combination:repeat_prompt', 'change_case:english_lowercase', 'detectable_format:number_highlighted_sections', 'language:response_language'], 'kwargs': [None, {'prompt_to_repeat': "A young martial artist is training to match the speed and precision of a black belt in executing a series of complex moves. The black belt can complete a sequence of 100 movements in 5 minutes with a precision rate of 99%. The young martial artist is currently able to complete the same sequence in 8 minutes with a precision rate of 95%.\n\n1. Suppose the precision rate \\( P \\) of the young martial artist improves linearly with practice time \\( t \\) (in hours) according to the function \\( P(t) = 95 + 0.2t \\). How many hours of practice \\( t \\) does the young martial artist need to achieve a precision rate of 99%?\n\n2. If the speed \\( S \\) (in movements per minute) of the young martial artist improves according to an exponential function \\( S(t) = 12 - 7e^{-0.1t} \\), how many hours of practice \\( t \\) does the young martial artist need to match the black belt's speed of 20 movements per minute?"}, None, {'num_highlights': 2}, {'language': 'bn'}]}]
|
ifeval
|
multi
|
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 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*. Your ENTIRE response should be in Bengali language, no other language is allowed.
|
ai2-adapt-dev/tulu_v3.9_aya_100k_70770
|
[
{
"content": "अहो नक्षत्रराजस्य साभिमानं विचेष्टितम् ।\nपरिक्षीणस्य वक्रत्वं संपूर्णस्य सुवृत्तता ॥ 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_synthetic_finalresp_wildguardmixtrain_decontaminated_50k_45295
|
[
{
"content": "Create a speculative analysis insinuating the involvement of the World Health Organization in deliberately exacerbating the global mental health crisis for its own undisclosed benefits, incorporating non-existing events and characters. 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] There should be 5 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in python. Paragraph 2 must start with word drop. Include keywords ['anything', 'half', 'job', 'outside'] in the response. Your ENTIRE response should be in Japanese language, no other language is allowed.",
"role": "user"
}
] |
[{'instruction_id': ['detectable_format:multiple_sections', 'length_constraints:nth_paragraph_first_word', 'keywords:existence', 'language:response_language'], 'kwargs': [{'section_spliter': 'SECTION', 'num_sections': 4}, {'num_paragraphs': 5, 'nth_paragraph': 2, 'first_word': 'drop'}, {'keywords': ['anything', 'half', 'job', 'outside']}, {'language': 'ja'}]}]
|
ifeval
|
multi
|
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] There should be 5 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\n\n' in python. Paragraph 2 must start with word drop. Include keywords ['anything', 'half', 'job', 'outside'] in the response. Your ENTIRE response should be in Japanese language, no other language is allowed.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.