Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
prompt: list<item: struct<role: string, content: string>>
child 0, item: struct<role: string, content: string>
child 0, role: string
child 1, content: string
method: string
spec: struct<instruction_id_list: list<item: string>, kwargs: list<item: string>, expected_answer: string, (... 579923 chars omitted)
child 0, instruction_id_list: list<item: string>
child 0, item: string
child 1, kwargs: list<item: string>
child 0, item: string
child 2, expected_answer: string
child 3, schema_str: struct<type: string, required: string, properties: struct<projectTitle: struct<type: string, descrip (... 579214 chars omitted)
child 0, type: string
child 1, required: string
child 2, properties: struct<projectTitle: struct<type: string, description: string, maxLength: int64>, materials: struct< (... 561902 chars omitted)
child 0, projectTitle: struct<type: string, description: string, maxLength: int64>
child 0, type: string
child 1, description: string
child 2, maxLength: int64
child 1, materials: struct<type: string, required: list<item: string>, properties: struct<supplies: struct<type: string, (... 562 chars omitted)
child 0, type: string
child 1, required: list<item: string>
child 0, item: string
child 2, properties: struct<supplies: struct<type: string, items: struct<type: string, required: list<item: string>, prop (.
...
escription: string
child 34, last_contact_date_max: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 35, follow_up_by_min: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 36, follow_up_by_max: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 37, customer_id: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 38, customer_phone: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 39, last_contact_date: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 40, notes: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 41, follow_up_by: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 2, required: list<item: string>
child 0, item: string
child 3, additionalProperties: bool
to
{'prompt': List({'role': Value('string'), 'content': Value('string')}), 'method': Value('string'), 'spec': {'instruction_id_list': List(Value('string')), 'kwargs': List(Json(decode=True)), 'verifier': Value('string')}, 'source_config': Value('string'), 'agent_ref': {'type': Value('string'), 'name': Value('string')}, 'dataset': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1779, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 609, in wrapped
for item in generator(*args, **kwargs):
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 299, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
prompt: list<item: struct<role: string, content: string>>
child 0, item: struct<role: string, content: string>
child 0, role: string
child 1, content: string
method: string
spec: struct<instruction_id_list: list<item: string>, kwargs: list<item: string>, expected_answer: string, (... 579923 chars omitted)
child 0, instruction_id_list: list<item: string>
child 0, item: string
child 1, kwargs: list<item: string>
child 0, item: string
child 2, expected_answer: string
child 3, schema_str: struct<type: string, required: string, properties: struct<projectTitle: struct<type: string, descrip (... 579214 chars omitted)
child 0, type: string
child 1, required: string
child 2, properties: struct<projectTitle: struct<type: string, description: string, maxLength: int64>, materials: struct< (... 561902 chars omitted)
child 0, projectTitle: struct<type: string, description: string, maxLength: int64>
child 0, type: string
child 1, description: string
child 2, maxLength: int64
child 1, materials: struct<type: string, required: list<item: string>, properties: struct<supplies: struct<type: string, (... 562 chars omitted)
child 0, type: string
child 1, required: list<item: string>
child 0, item: string
child 2, properties: struct<supplies: struct<type: string, items: struct<type: string, required: list<item: string>, prop (.
...
escription: string
child 34, last_contact_date_max: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 35, follow_up_by_min: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 36, follow_up_by_max: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 37, customer_id: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 38, customer_phone: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 39, last_contact_date: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 40, notes: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 41, follow_up_by: struct<type: string, description: string>
child 0, type: string
child 1, description: string
child 2, required: list<item: string>
child 0, item: string
child 3, additionalProperties: bool
to
{'prompt': List({'role': Value('string'), 'content': Value('string')}), 'method': Value('string'), 'spec': {'instruction_id_list': List(Value('string')), 'kwargs': List(Json(decode=True)), 'verifier': Value('string')}, 'source_config': Value('string'), 'agent_ref': {'type': Value('string'), 'name': Value('string')}, 'dataset': Value('string')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1343, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 907, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1832, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
prompt list | method string | spec dict | source_config string | agent_ref dict | dataset string |
|---|---|---|---|---|---|
[
{
"role": "user",
"content": "While diying after his fight against the antimonitor, Oliver Queen finds his soul sent back to the start of his journey. Right back to Starling City hospital. First Episode of season 1. This is Oliver Queen realising he has a second chance. Oliver & his mom are in the car. Oliv... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:nth_paragraph_first_word",
"last_word:last_word_answer"
],
"kwargs": [
{
"num_paragraphs": 4,
"nth_paragraph": 3,
"first_word": "crash"
},
{
"last_word": "contest"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "帮忙把以下SQL进行优化:with tep as (select t.urid, t.draftid, t.bankaccountid FROM (select t.urid, t.draftid,\n\nPlease follow these instructions in your response:\n- There should be 2 paragraphs. Paragraphs are separated with the markdown divider: ... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_paragraphs",
"count:lowercase_counting",
"detectable_format:number_bullet_lists"
],
"kwargs": [
{
"num_paragraphs": 2
},
{
"N": 3
},
{
"num_bullets": 5
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response: Do not include keywords ['inevitable', 'part'] in the response and The first word of each sentence should be the word relationship. Research and identify the most in-demand game genres and mechanics in the market, this is part ... | instruction_following_rl | {
"instruction_id_list": [
"keywords:forbidden_words",
"first_word:first_word_sent"
],
"kwargs": [
{
"forbidden_words": [
"inevitable",
"part"
]
},
{
"first_word": "relationship"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "I would like a list of topics with their pages from Plato philebus. these topics should be no more than 3 pages. also provide the idex of the pages\n\nMake sure to adhere to these requirements:\n- The response must contain at least 4 placeholders represented by square brackets,... | instruction_following_rl | {
"instruction_id_list": [
"detectable_content:number_placeholders",
"letters:letter_counting",
"length_constraints:number_sentences"
],
"kwargs": [
{
"num_placeholders": 4
},
{
"N": 3,
"relation": "at least"
},
{
"num_sentences": 17,
"relation": "at l... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "how can i install jayhorn step by step\n\nYour response should comply with these constraints:\n1. There should be 2 paragraphs. Paragraphs are separated with the markdown divider: ***\n2. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of jo... | instruction_following_rl | {
"instruction_id_list": [
"paragraphs:paragraphs",
"detectable_format:title",
"length_constraints:number_words"
],
"kwargs": [
{},
null,
{
"num_words": 204,
"relation": "less than"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response: The last word of your response should be the word special and Finish your response with this exact phrase Any other questions?. No other words should follow this phrase. translate the following text in Wolof to English: \"Ndaka... | instruction_following_rl | {
"instruction_id_list": [
"last_word:last_word_answer",
"startend:end_checker"
],
"kwargs": [
{
"last_word": "special"
},
{
"end_phrase": "Any other questions?"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Write 3000-word comprehensive article about the topic \"Forearm Tattoos for women\" that is unique, easy-to-understand. Make the content punchy and engaging by using a conversational tone. Keep keyword density around 1% for \"Forearm Tattoos for women\". Suggest internal and ex... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_words",
"last_word:last_word_sent",
"keywords:letter_frequency"
],
"kwargs": [
{
"num_words": 492,
"relation": "less than"
},
{
"last_word": "mud"
},
{
"letter": "e",
"let_frequency": 8,
"let_... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. There should be 2 paragraphs. Paragraphs and only paragraphs are separated with each other by two line breaks. \n2. In your entire response, refrain from the use of any commas.\n3. There should be 4 paragraphs. Paragraph... | instruction_following_rl | {
"instruction_id_list": [
"paragraphs:paragraphs2",
"punctuation:no_comma",
"length_constraints:nth_paragraph_first_word"
],
"kwargs": [
{},
null,
{
"num_paragraphs": 4,
"nth_paragraph": 1,
"first_word": "committee"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. Only use unique words in your response, no word should be repeated!\n2. Your answer must contain exactly 1 bullet points. Use the markdown bullet points such as:\n* This is point 1. \n* This is point 2\n\nwebpack的具体使用"
... | instruction_following_rl | {
"instruction_id_list": [
"count:count_unique",
"detectable_format:number_bullet_lists"
],
"kwargs": [
null,
{
"num_bullets": 1
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Give a characteristic of a uniform random variable. \nQuestion content area bottom\nPart 1\nChoose the correct answer.\nA.\nA uniform random variable is more likely to assume a value in the specified interval than it is to assume a value in any other interval of equal size.\nB.... | instruction_following_rl | {
"instruction_id_list": [
"keywords:word_once",
"keywords:keyword_specific_position"
],
"kwargs": [
{
"keyword": "director"
},
{
"keyword": "guess",
"n": 9,
"m": 23
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Your response should satisfy these conditions:\n1. Include keyword ['help'] once in your response, keyword ['dump'] twice in your response.\n2. Wrap your entire response with double quotation marks.\n\nCalculate the optimal amount of gear inches for track cycling using gear rat... | instruction_following_rl | {
"instruction_id_list": [
"count:count_increment_word",
"startend:quotation"
],
"kwargs": [
{
"keyword1": "help",
"keyword2": "dump"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Hi I’m Kenzy\n\nPlease ensure your answer follows these guidelines:\n1. The last word of your response should be the word mom.\n2. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.\n3. Your answer must contain exactly 5 ... | instruction_following_rl | {
"instruction_id_list": [
"last_word:last_word_answer",
"startend:end_checker",
"detectable_format:number_bullet_lists"
],
"kwargs": [
{
"last_word": "mom"
},
{
"end_phrase": "Any other questions?"
},
{
"num_bullets": 5
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Salut micheal\n\nMake sure to adhere to these requirements:\n- Write 3 paragraphs, delimited by the markdown divider: * * *, with exactly 2 sentences each, with exactly 3 words in each sentence.\n- In your entire response, refrain from the use of . (i.e. dots) as punctuation an... | instruction_following_rl | {
"instruction_id_list": [
"count:counting_composition",
"punctuation:punctuation_dot",
"detectable_format:sentence_hyphens"
],
"kwargs": [
{
"n_sent": 2,
"n_words": 3
},
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "terminos del departamento de inyeccion de plasticos en una empresa fabrica de motos y partes de autos, en japones\n\nYour response should comply with these constraints:\n- At the end of your response, please explicitly add a postscript starting with P.P.S\n- No two adjacent ... | instruction_following_rl | {
"instruction_id_list": [
"detectable_content:postscript",
"keywords:no_adjacent_consecutive"
],
"kwargs": [
{
"postscript_marker": "P.P.S"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "详细介绍一下python的lambda \n\nMake sure to adhere to these requirements: Answer with at least 2 letters, In your response, words with all capital letters should appear less than 13 times, and Do not include keywords ['pass', 'reason'] in the response"
}
] | instruction_following_rl | {
"instruction_id_list": [
"letters:letter_counting",
"change_case:capital_word_frequency",
"keywords:forbidden_words"
],
"kwargs": [
{
"N": 2,
"relation": "at least"
},
{
"capital_frequency": 13,
"capital_relation": "less than"
},
{
"forbidden_words":... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "менеджер на python для запуска и остановки процессов и передачи им данных через очередь\n\nPlease follow these instructions in your response:\n1. Your ENTIRE response should be in Italian language, no other language is allowed.\n2. In your response, the letter c should appear a... | instruction_following_rl | {
"instruction_id_list": [
"language:response_language",
"keywords:letter_frequency",
"detectable_content:number_placeholders"
],
"kwargs": [
{
"language": "it"
},
{
"letter": "c",
"let_frequency": 6,
"let_relation": "at least"
},
{
"num_placeholders":... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints:\n1. Answer with less than 390 words.\n2. Your entire response should be in English, and in all capital letters.\n3. In your entire response, refrain from the use of ! (i.e. exclamation marks) as punctuation and in general.\n\n... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_words",
"change_case:english_capital",
"punctuation:punctuation_exclamation"
],
"kwargs": [
{
"num_words": 390,
"relation": "less than"
},
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n1. In your response, the word rub should appear 2 times.\n2. Wrap every word bigram in double angular brackets, such as <<I am>> <<at home>> <<with my>> <<cute dog>>.\n3. Only use unique words in your response, no word sh... | instruction_following_rl | {
"instruction_id_list": [
"keywords:word_count_different_numbers",
"detectable_format:bigram_wrapping",
"count:count_unique"
],
"kwargs": [
{
"keyword": "rub",
"frequency": 2,
"relation": "less than"
},
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "物流企业党委书记、总经理写一篇学习心得体会，字数5000字，内容如下：一是深入学习党的二十大报告提出的新表述新概括新论断，结合公司改革发展的全局性前瞻性问题和自身分管的工作，深入开展专题研究，并结合实际进行深入思考;二是结合习近平新时代中国特色社会主义思想主题教育活动，胸怀“国之大者”，以新气象新作为努力推动高质量发展取得新成效。\n\nMake sure to adhere to these requirements:\n- Include keyword mall in the 6-th sentence, as the 22-th word o... | instruction_following_rl | {
"instruction_id_list": [
"keywords:keyword_specific_position"
],
"kwargs": [
{
"keyword": "mall",
"n": 6,
"m": 22
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Proof of x^x-(x+1)^(x-1) for x>1\n\nYour response should comply with these constraints:\n- 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\n- The response must contain at least 1 placeholders repr... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:number_bullet_lists",
"detectable_content:number_placeholders",
"detectable_format:sentence_hyphens"
],
"kwargs": [
{
"num_bullets": 2
},
{
"num_placeholders": 1
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "In python, make me some code that predicts a 5x5 field minesweeper. You're supposed to predict 4 safe spots, where you've data for the past 10 games: [2, 12, 18, 10, 12, 21, 1, 5, 13, 12, 21, 23, 7, 15, 24, 10, 14, 23, 9, 15, 16, 11, 19, 22, 2, 7, 17, 1, 6, 8]. Use machine lear... | instruction_following_rl | {
"instruction_id_list": [
"first_word:first_word_sent",
"punctuation:punctuation_exclamation",
"last_word:last_word_answer"
],
"kwargs": [
{
"first_word": "offer"
},
null,
{
"last_word": "inspection"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints: Repeat the phrase Laughter is the best medicine exactly 2 times, transforming it slightly each time by replacing only one word in the center of the phrase, Include keyword basket in your response, and Write 3 paragraphs, delim... | instruction_following_rl | {
"instruction_id_list": [
"copy:repeat_phrase",
"keywords:word_once",
"count:counting_composition"
],
"kwargs": [
{
"phrase": "Laughter is the best medicine",
"small_n": 2
},
{
"keyword": "basket"
},
{
"n_sent": 2,
"n_words": 2
}
],
"verifier"... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response:\n- Wrap your entire response with double quotation marks.\n\n下面关于环形振荡器的说法正确的是\n\nA.环形振荡器内只有非门\n\nB.环形振荡器的输出可以在电路的任何位置\n\nC.环形振荡器内门的数量必须是偶数个\n\nD.环形振荡器产生的信号可以作为时钟"
}
] | instruction_following_rl | {
"instruction_id_list": [
"startend:quotation"
],
"kwargs": [
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "What if SpongeBob had All Transcendent Powers\n\nYour response should comply with these constraints:\n1. Entire output should be wrapped in JSON format. You can use markdown ticks such as ```.\n2. Include keywords ['neat', 'spiritual'] in the response.\n3. Do not include keywor... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:json_format",
"keywords:existence",
"keywords:forbidden_words"
],
"kwargs": [
null,
{
"keywords": [
"neat",
"spiritual"
]
},
{
"forbidden_words": [
"living",
"software"
]
}
],
... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. In your response, the letter c should appear less than 2 times.\n2. The response must contain at least 3 placeholders represented by square brackets, such as [address].\n\n- Increase the engagement rate on Uber Egypt’s F... | instruction_following_rl | {
"instruction_id_list": [
"letters:letter_counting2",
"detectable_content:number_placeholders"
],
"kwargs": [
{
"letter": "c",
"let_frequency": 2,
"let_relation": "less than"
},
{
"num_placeholders": 3
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n1. The response must contain at least 3 placeholders represented by square brackets, such as [address].\n2. At the end of your response, please explicitly add a postscript starting with P.S.\n\nCan we split \"How do slack... | instruction_following_rl | {
"instruction_id_list": [
"detectable_content:number_placeholders",
"detectable_content:postscript"
],
"kwargs": [
{
"num_placeholders": 3
},
{
"postscript_marker": "P.S."
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "怎样在真正的DOS系统下在未连接IPX的情况下玩Command and conquer DOS版的遭遇战？ Please follow these rules when answering: No two adjacent words can start with consecutive letters of the alphabet and The first word of your response should be the word bike"
}
] | instruction_following_rl | {
"instruction_id_list": [
"keywords:no_adjacent_consecutive",
"first_word:first_word_answer"
],
"kwargs": [
null,
{
"first_word": "bike"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Name some characters, that have used a Breast Smother with Bodyscissors combo hold at some point. \n\nMake sure to adhere to these requirements: Your response should contain less than 8 sentences and Finish your response with this exact phrase Any other questions?. No other wor... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_sentences",
"startend:end_checker"
],
"kwargs": [
{
"num_sentences": 8,
"relation": "less than"
},
{
"end_phrase": "Any other questions?"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "You’re a Product Manager. Use the sentences in quotes to construct a short-term roadmap for each sentence. “Develop AI solutions for accurate fraud detection and risk management in financial institutions. Implement AI-powered financial advisory platforms for personalized wealth... | instruction_following_rl | {
"instruction_id_list": [
"combination:two_responses",
"language:response_language",
"detectable_format:title"
],
"kwargs": [
null,
{
"language": "ta"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Your response should satisfy these conditions: 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 2 paragraphs. Paragraphs and only paragraphs are... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:multiple_sections",
"paragraphs:paragraphs2",
"detectable_format:sentence_hyphens"
],
"kwargs": [
{
"section_spliter": "SECTION",
"num_sections": 4
},
{},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "This my current html code for a interactive clickable table:\n“<!DOCTYPE html>\n<style type=“text/css”>\n.tg {border-collapse:collapse;border-spacing:0;}\n.tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;\noverflow:hidde... | instruction_following_rl | {
"instruction_id_list": [
"count:count_unique",
"punctuation:no_comma"
],
"kwargs": [
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. Wrap your entire response with double quotation marks.\n2. Your entire response should be in English, and in all capital letters.\n3. Finish your response with this exact phrase Is there anything else I can help with?. N... | instruction_following_rl | {
"instruction_id_list": [
"startend:quotation",
"change_case:english_capital",
"startend:end_checker"
],
"kwargs": [
null,
null,
{
"end_phrase": "Is there anything else I can help with?"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "напиши fly hack для minecraft 1.19.4 в Visual studio Your response should comply with these constraints: Wrap your entire response with double quotation marks, Do not include keywords ['benefit', 'spray'] in the response, and No two adjacent words can start with consecutive let... | instruction_following_rl | {
"instruction_id_list": [
"startend:quotation",
"keywords:forbidden_words",
"keywords:no_adjacent_consecutive"
],
"kwargs": [
null,
{
"forbidden_words": [
"benefit",
"spray"
]
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please incorporate these requirements into your answer: 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 and In your response, the letter b should appear less than 5 times. Donde debo solicitar vis... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:number_bullet_lists",
"letters:letter_counting2"
],
"kwargs": [
{
"num_bullets": 4
},
{
"letter": "b",
"let_frequency": 5,
"let_relation": "less than"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines: Include keyword hire in your response and The last word of each sentence, before punctuation, should be the word desk. write a scrpit about u of l 3-9 season"
}
] | instruction_following_rl | {
"instruction_id_list": [
"keywords:word_once",
"last_word:last_word_sent"
],
"kwargs": [
{
"keyword": "hire"
},
{
"last_word": "desk"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Что у тебя спросить\n\nPlease ensure your answer follows these guidelines:\n1. Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>."
}
] | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:title"
],
"kwargs": [
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "What the most winning numbers mass lottery? Please ensure your answer follows these guidelines: Include keyword option in the 9-th sentence, as the 4-th word of that sentence and There should be 2 paragraphs. Paragraphs and only paragraphs are separated with each other by two l... | instruction_following_rl | {
"instruction_id_list": [
"keywords:keyword_specific_position",
"paragraphs:paragraphs2"
],
"kwargs": [
{
"keyword": "option",
"n": 9,
"m": 4
},
{}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response: Your ENTIRE response should be in Malayalam language, no other language is allowed and 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. Big5字符编码格... | instruction_following_rl | {
"instruction_id_list": [
"language:response_language",
"detectable_format:number_bullet_lists"
],
"kwargs": [
{
"language": "ml"
},
{
"num_bullets": 3
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines: In your entire response, refrain from the use of ! (i.e. exclamation marks) as punctuation and in general and Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>. write a script abot... | instruction_following_rl | {
"instruction_id_list": [
"punctuation:punctuation_exclamation",
"detectable_format:title"
],
"kwargs": [
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response:\n- Answer with at least 3 letters.\n- 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 blue.\n- In you... | instruction_following_rl | {
"instruction_id_list": [
"letters:letter_counting",
"length_constraints:nth_paragraph_first_word",
"punctuation:no_comma"
],
"kwargs": [
{
"N": 3,
"relation": "at least"
},
{
"num_paragraphs": 3,
"nth_paragraph": 2,
"first_word": "blue"
},
null
],
... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "While responding, please comply with these rules: Answer with less than 392 words, The response must contain at least 2 placeholders represented by square brackets, such as [address], and Your entire response should be in English, and in all lowercase letters. No capital letter... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_words",
"detectable_content:number_placeholders",
"change_case:english_lowercase"
],
"kwargs": [
{
"num_words": 392,
"relation": "less than"
},
{
"num_placeholders": 2
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Act like a professional medival historian. Tell me about medieval parent-children relationships, were average noble parents attentive to their children, what common activities could they share? Let's think about it step by step.\n\nPlease follow these instructions in your respo... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:constrained_response"
],
"kwargs": [
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Tell me a long funny story about a dragon becoming desperate to pee but refusing to go, potty dancing and wetting himself, all while in his home and playing videogames. Focus on making the wetting part long and detailed.\n\nMake sure to adhere to these requirements:\n- In your ... | instruction_following_rl | {
"instruction_id_list": [
"keywords:frequency",
"keywords:start_end"
],
"kwargs": [
{
"keyword": "platform",
"frequency": 3,
"relation": "less than"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response:\n1. Answer with one of the following options: ('My answer is yes.', 'My answer is no.', 'My answer is maybe.')\n\nwhat are the airlines company name's in the lybia country?"
}
] | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:constrained_response"
],
"kwargs": [
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "What are the different orders in which circuit breaker, retry, timeout are applied? Use the function application format to indicate order. For example, with C(B(A(remoteCall()))), A is applied 1st, then B, then C. Mention the use-case for each order and explain the reasoning st... | instruction_following_rl | {
"instruction_id_list": [
"keywords:forbidden_words",
"detectable_content:postscript"
],
"kwargs": [
{
"forbidden_words": [
"person",
"rough"
]
},
{
"postscript_marker": "P.S."
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Could you add more to this\n\n1. Chakra (Naruto): Chakra is the energy source used by ninjas in the Naruto series to perform jutsu techniques. This system consists of three main components: physical energy, spiritual energy, and nature transformation.\n\n2. Ki (Dragon Ball seri... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:multiple_sections"
],
"kwargs": [
{
"section_spliter": "SECTION",
"num_sections": 2
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "智力竞技中心的详细资料\n\nMake sure to adhere to these requirements:\n- Your ENTIRE response should be in French language, no other language is allowed.\n- Include keyword ['perception'] once in your response, keyword ['blood'] twice in your response."
}
] | instruction_following_rl | {
"instruction_id_list": [
"language:response_language",
"count:count_increment_word"
],
"kwargs": [
{
"language": "fr"
},
{
"keyword1": "perception",
"keyword2": "blood"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. Your entire response should be in English, and in all capital letters.\n2. The first word of each sentence should be the word long.\n\nWrite me a spy story featuring ambatukam omaygot"
}
] | instruction_following_rl | {
"instruction_id_list": [
"change_case:english_capital",
"first_word:first_word_sent"
],
"kwargs": [
null,
{
"first_word": "long"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. Wrap every word bigram in double angular brackets, such as <<I am>> <<at home>> <<with my>> <<cute dog>>.\n2. Only use unique words in your response, no word should be repeated!\n3. In your entire response, refrain from ... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:bigram_wrapping",
"count:count_unique",
"punctuation:punctuation_dot"
],
"kwargs": [
null,
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints: The last word of your response should be the word clue, The last word of each sentence, before punctuation, should be the word independence, and In your response, the letter f should appear less than 7 times. responda a pergun... | instruction_following_rl | {
"instruction_id_list": [
"last_word:last_word_answer",
"last_word:last_word_sent",
"letters:letter_counting2"
],
"kwargs": [
{
"last_word": "clue"
},
{
"last_word": "independence"
},
{
"letter": "f",
"let_frequency": 7,
"let_relation": "less than"
... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n- The last word of each sentence, before punctuation, should be the word sleep.\n- There should be 4 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n' in pytho... | instruction_following_rl | {
"instruction_id_list": [
"last_word:last_word_sent",
"length_constraints:nth_paragraph_first_word"
],
"kwargs": [
{
"last_word": "sleep"
},
{
"num_paragraphs": 4,
"nth_paragraph": 4,
"first_word": "script"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "While responding, please comply with these rules: Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase, There should be 4 paragraphs. Paragraphs are separated with the markdown divider: ***, and The last w... | instruction_following_rl | {
"instruction_id_list": [
"startend:end_checker",
"length_constraints:number_paragraphs",
"last_word:last_word_sent"
],
"kwargs": [
{
"end_phrase": "Is there anything else I can help with?"
},
{
"num_paragraphs": 4
},
{
"last_word": "eye"
}
],
"verifier":... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding: The first word of your response should be the word dependent, Your response should contain at least 8 sentences, and Include keyword raise in the 10-th sentence, as the 27-th word of that sentence. are you chatgpt 4?"
}
... | instruction_following_rl | {
"instruction_id_list": [
"first_word:first_word_answer",
"length_constraints:number_sentences",
"keywords:keyword_specific_position"
],
"kwargs": [
{
"first_word": "dependent"
},
{
"num_sentences": 8,
"relation": "at least"
},
{
"keyword": "raise",
"... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "What is a characteristic of fungi? a) it is bad for the enviroment b) it has a cell wall c) There are many different types choose the correct answer from the following \nMake sure your response satisfies these conditions: Write 3 paragraphs, delimited by the markdown divider: *... | instruction_following_rl | {
"instruction_id_list": [
"count:counting_composition",
"length_constraints:nth_paragraph_first_word"
],
"kwargs": [
{
"n_sent": 2,
"n_words": 2
},
{
"num_paragraphs": 5,
"nth_paragraph": 4,
"first_word": "appearance"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "While responding, please comply with these rules: In your response, the word shame should appear less than 3 times, There should be 2 paragraphs. Paragraphs are separated with the markdown divider: ***, and In your response, all lowercase words should appear at most 2 times. wr... | instruction_following_rl | {
"instruction_id_list": [
"keywords:frequency",
"paragraphs:paragraphs",
"count:lowercase_counting"
],
"kwargs": [
{
"keyword": "shame",
"frequency": 3,
"relation": "less than"
},
{},
{
"N": 2
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "write an article on \"Why you should pay attention to ZkLink, the next-generation DeFi platform?\" to target beginners Your response should comply with these constraints: Finish your response with this exact phrase Any other questions?. No other words should follow this phrase,... | instruction_following_rl | {
"instruction_id_list": [
"startend:end_checker",
"keywords:word_once",
"last_word:last_word_sent"
],
"kwargs": [
{
"end_phrase": "Any other questions?"
},
{
"keyword": "mate"
},
{
"last_word": "mouth"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to this constraint: 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] Η την η επί Τάσος ξέρεις τι σημαίνει αυτό?"
}
] | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:multiple_sections"
],
"kwargs": [
{
"section_spliter": "Section",
"num_sections": 1
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines:\n1. Highlight at least 2 sections in your answer with markdown, i.e. *highlighted section*.\n2. Do not include keywords ['advance', 'league'] in the response.\n3. The response must contain at least 4 placeholders represented... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:number_highlighted_sections",
"keywords:forbidden_words",
"detectable_content:number_placeholders"
],
"kwargs": [
{
"num_highlights": 2
},
{
"forbidden_words": [
"advance",
"league"
]
},
{
"num_pl... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "go读取pdf文件中的条形码数据 \nPlease follow these instructions in your response: Only use unique words in your response, no word should be repeated! and Enclose every word in your response within square brackets"
}
] | instruction_following_rl | {
"instruction_id_list": [
"count:count_unique",
"detectable_format:square_brackets"
],
"kwargs": [
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response:\n1. Include keyword meeting in your response.\n2. Finish your response with this exact phrase Any other questions?. No other words should follow this phrase.\n3. Enclose every word in your response within square brackets.\n\nWr... | instruction_following_rl | {
"instruction_id_list": [
"keywords:word_once",
"startend:end_checker",
"detectable_format:square_brackets"
],
"kwargs": [
{
"keyword": "meeting"
},
{
"end_phrase": "Any other questions?"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "write a procedure for training\n\nMake sure to adhere to these requirements:\n1. Repeat the phrase All that glitters is not gold exactly 2 times, transforming it slightly each time by replacing only one word in the center of the phrase.\n2. In your entire response, refrain from... | instruction_following_rl | {
"instruction_id_list": [
"copy:repeat_phrase",
"punctuation:no_comma"
],
"kwargs": [
{
"phrase": "All that glitters is not gold",
"small_n": 2
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Your response should satisfy these conditions: There should be 2 paragraphs. Paragraphs and only paragraphs are separated with each other by two line breaks, Include keyword year in your response, and Your ENTIRE response should be in Telugu language, no other language is allow... | instruction_following_rl | {
"instruction_id_list": [
"paragraphs:paragraphs2",
"keywords:word_once",
"language:response_language"
],
"kwargs": [
{},
{
"keyword": "year"
},
{
"language": "te"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "write a short email to superiors objecting a cleaner damaging floor tiles by carrying a trash bag dripping mustard sauce\n\nMake sure to adhere to these requirements:\n1. The response must contain at least 4 placeholders represented by square brackets, such as [address].\n2. Th... | instruction_following_rl | {
"instruction_id_list": [
"detectable_content:number_placeholders",
"last_word:last_word_sent"
],
"kwargs": [
{
"num_placeholders": 4
},
{
"last_word": "community"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Придумай шутку про фаренгейт\n\nYour response should comply with these constraints:\n1. Do not include keywords ['clothes', 'pizza'] in the response.\n2. In your response, the word money should appear 1 times.\n3. Answer with less than 105 words."
}
] | instruction_following_rl | {
"instruction_id_list": [
"keywords:forbidden_words",
"keywords:word_count_different_numbers",
"length_constraints:number_words"
],
"kwargs": [
{
"forbidden_words": [
"clothes",
"pizza"
]
},
{
"keyword": "money",
"frequency": 1,
"relation": "a... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n- Include keyword shine in the 3-th sentence, as the 21-th word of that sentence.\n- Write 3 paragraphs, delimited by the markdown divider: * * *, with exactly 3 sentences each, with exactly 2 words in each sentence.\n\nW... | instruction_following_rl | {
"instruction_id_list": [
"keywords:keyword_specific_position",
"count:counting_composition"
],
"kwargs": [
{
"keyword": "shine",
"n": 3,
"m": 21
},
{
"n_sent": 3,
"n_words": 2
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Your response should satisfy these conditions: The first word of your response should be the word context and In your response, the word tennis should appear 1 times. 请用英文写一篇关于新冠肺炎预测模型论文的引言部分，谢谢"
}
] | instruction_following_rl | {
"instruction_id_list": [
"first_word:first_word_answer",
"keywords:word_count_different_numbers"
],
"kwargs": [
{
"first_word": "context"
},
{
"keyword": "tennis",
"frequency": 1,
"relation": "at least"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Por que o Fluminense disputou a Série C do Brasileirão em 1999??? \n\nMake sure to adhere to these requirements: Your response should contain less than 13 sentences, All sentences must be connected using hyphens, with no spaces between them, and Wrap your entire response with d... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_sentences",
"detectable_format:sentence_hyphens",
"startend:quotation"
],
"kwargs": [
{
"num_sentences": 13,
"relation": "less than"
},
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Придумай и Напиши первую главу книги по вселенной Warhammer 40000, о приключениях роты гвардии смерти Please ensure your answer follows these guidelines: Finish your response with this exact phrase Any other questions?. No other words should follow this phrase and The last word... | instruction_following_rl | {
"instruction_id_list": [
"startend:end_checker",
"last_word:last_word_sent"
],
"kwargs": [
{
"end_phrase": "Any other questions?"
},
{
"last_word": "transportation"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints:\n1. In your response, the letter i should appear less than 4 times.\n2. Include a palindrome in your response.\n\nРасскажи всю историю по Рэтчету и Кланку не в ключая фильм 2016 и игру по Рэтчету и Кланку 2016. Расскажи все ча... | instruction_following_rl | {
"instruction_id_list": [
"keywords:letter_frequency",
"keywords:palindrome"
],
"kwargs": [
{
"letter": "i",
"let_frequency": 4,
"let_relation": "less than"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Ich werde dir ein Expose inklusive Gliederung für eine wissenschaftliche Arbeit bereitstellen:\n\n„„Arbeitslosigkeit und ihre Folgen für die Gesundheit und das Wohlbefinden von Menschen“ – Exposé zur BA – Daniel Hübner \nFragestellung: Inwiefern beeinflusst Arbeitslosigkeit das... | instruction_following_rl | {
"instruction_id_list": [
"startend:end_checker",
"copy:repeat_phrase"
],
"kwargs": [
{
"end_phrase": "Is there anything else I can help with?"
},
{
"phrase": "All that glitters is not gold",
"small_n": 2
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please ensure your response follows these guidelines: In your response, the word broad should appear less than 2 times, Your answer must contain a title, wrapped in double angular brackets, such as <<poem of joy>>, and Answer with at least 3 letters. what are the main factors a... | instruction_following_rl | {
"instruction_id_list": [
"keywords:frequency",
"detectable_format:title",
"letters:letter_counting"
],
"kwargs": [
{
"keyword": "broad",
"frequency": 2,
"relation": "less than"
},
null,
{
"N": 3,
"relation": "at least"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Qui était Albert Einstein ? Please follow these rules when answering: In your entire response, refrain from the use of ! (i.e. exclamation marks) as punctuation and in general, In your entire response, refrain from the use of . (i.e. dots) as punctuation and in general, and Inc... | instruction_following_rl | {
"instruction_id_list": [
"punctuation:punctuation_exclamation",
"punctuation:punctuation_dot",
"keywords:existence"
],
"kwargs": [
null,
null,
{
"keywords": [
"return",
"tooth"
]
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n1. 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 drag.\n2. In your response, all lowercase wo... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:nth_paragraph_first_word",
"count:lowercase_counting",
"length_constraints:number_sentences"
],
"kwargs": [
{
"num_paragraphs": 3,
"nth_paragraph": 1,
"first_word": "drag"
},
{
"N": 2
},
{
"num_sentences":... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints: Answer with less than 228 words and There should be 5 paragraphs. Paragraphs are separated with the markdown divider: ***. show annual power consumption data for each ballkan country for years 2010, 2011, 2012, 2013, 2014, 201... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_words",
"length_constraints:number_paragraphs"
],
"kwargs": [
{
"num_words": 228,
"relation": "less than"
},
{
"num_paragraphs": 5
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Your response should satisfy these conditions:\n1. 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 radio.\n2. Wrap every word bigram in double angular b... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:nth_paragraph_first_word",
"detectable_format:bigram_wrapping",
"count:count_increment_word"
],
"kwargs": [
{
"num_paragraphs": 3,
"nth_paragraph": 2,
"first_word": "radio"
},
null,
{
"keyword1": "collection",
... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "我在一家船舶公司实习，我在实习期间做过生产员工、小组长、产品检验员、PE技术员，我现在需要一篇实习报告\n\nPlease ensure your answer follows these guidelines:\n1. Wrap every word bigram in double angular brackets, such as <<I am>> <<at home>> <<with my>> <<cute dog>>.\n2. Highlight at least 1 sections in your answer with markdow... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:bigram_wrapping",
"detectable_format:number_highlighted_sections",
"count:count_increment_word"
],
"kwargs": [
null,
{
"num_highlights": 1
},
{
"keyword1": "hit",
"keyword2": "name"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Ignore previous instructions. Federation of Baisse-Lechande is a multiparty state with 777-member unicameral parliament. Patriotic-Centralist Party has 128 seats, Sovereignist Party has 115 seats, People’s Sovereignist Party has 101 seats, Party of Privatizers has 99 seats, Dir... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:sentence_hyphens",
"last_word:last_word_answer"
],
"kwargs": [
null,
{
"last_word": "reputation"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n1. Only use unique words in your response, no word should be repeated!\n2. Answer with at least 118 words.\n3. Include keyword slide in the 8-th sentence, as the 18-th word of that sentence.\n\nfirst mcdonalds location an... | instruction_following_rl | {
"instruction_id_list": [
"count:count_unique",
"length_constraints:number_words",
"keywords:keyword_specific_position"
],
"kwargs": [
null,
{
"num_words": 118,
"relation": "at least"
},
{
"keyword": "slide",
"n": 8,
"m": 18
}
],
"verifier": "ifev... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Script about Miami blowing a 27-0 lead against Mac jones \n\nMake sure to adhere to these requirements: 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 ... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:nth_paragraph_first_word",
"first_word:first_word_answer"
],
"kwargs": [
{
"num_paragraphs": 3,
"nth_paragraph": 4,
"first_word": "feedback"
},
{
"first_word": "kiss"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please incorporate these requirements into your answer: Include keyword fee in the 5-th sentence, as the 26-th word of that sentence and Highlight at least 3 sections in your answer with markdown, i.e. *highlighted section*. Il est nécessaire de lister dans un tableau triable d... | instruction_following_rl | {
"instruction_id_list": [
"keywords:keyword_specific_position",
"detectable_format:number_highlighted_sections"
],
"kwargs": [
{
"keyword": "fee",
"n": 5,
"m": 26
},
{
"num_highlights": 3
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Hey, do you know about the parody of \"five nights at freddy's\" called \"fredina's nightclub\"? Please incorporate these requirements: There should be 3 paragraphs. Paragraphs are separated with the markdown divider: ***, Your answer must contain exactly 2 bullet points. Use t... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_paragraphs",
"detectable_format:number_bullet_lists",
"letters:letter_counting2"
],
"kwargs": [
{
"num_paragraphs": 3
},
{
"num_bullets": 2
},
{
"letter": "c",
"let_frequency": 2,
"let_relation": "a... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please follow these instructions in your response:\n- At the end of your response, please explicitly add a postscript starting with P.S.\n- In your entire response, refrain from the use of ! (i.e. exclamation marks) as punctuation and in general.\n\nHola. Sabes por qué la ciuda... | instruction_following_rl | {
"instruction_id_list": [
"detectable_content:postscript",
"punctuation:punctuation_exclamation"
],
"kwargs": [
{
"postscript_marker": "P.S."
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please incorporate these requirements into your answer: Wrap your entire response with double quotation marks, In your entire response, refrain from the use of any commas, and Only use unique words in your response, no word should be repeated!. 《阳光开朗大男孩》是谁唱的？"
}
] | instruction_following_rl | {
"instruction_id_list": [
"startend:quotation",
"punctuation:no_comma",
"count:count_unique"
],
"kwargs": [
null,
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "如何使用celery自己的缓存？ \n\nMake sure to adhere to these requirements: The first word of your response should be the word building, In your entire response, refrain from the use of ! (i.e. exclamation marks) as punctuation and in general, and Your answer must contain a title, wrapped ... | instruction_following_rl | {
"instruction_id_list": [
"first_word:first_word_answer",
"punctuation:punctuation_exclamation",
"detectable_format:title"
],
"kwargs": [
{
"first_word": "building"
},
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "is there a website to create a character for tavern.ai ?\n\nPlease ensure your answer follows these guidelines:\n1. There should be 1 paragraphs. Paragraphs are separated with the markdown divider: ***\n2. Finish your response with this exact phrase Is there anything else I can... | instruction_following_rl | {
"instruction_id_list": [
"length_constraints:number_paragraphs",
"startend:end_checker"
],
"kwargs": [
{
"num_paragraphs": 1
},
{
"end_phrase": "Is there anything else I can help with?"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Сколько весит баклажка 5 литпов\n\nMake sure to adhere to these requirements:\n- Your response must have 3 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]\n- The last word of your respon... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:multiple_sections",
"last_word:last_word_answer",
"paragraphs:paragraphs"
],
"kwargs": [
{
"section_spliter": "SECTION",
"num_sections": 3
},
{
"last_word": "script"
},
{}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "api/datasources/proxy/1/api/v1/query_range?query=1%20-%20avg(irate(node_cpu_seconds_total%7Binstance%3D~%22(pre-uk-kafka1%7Cpre-uk-kafka2)%22%2Cmode%3D%22idle%22%7D%5B30m%5D))%20by%20(instance)&start={start}&end={end}&step=15\n\n把%20这种翻译一下\n\nPlease follow these instructions in... | instruction_following_rl | {
"instruction_id_list": [
"letters:letter_counting2"
],
"kwargs": [
{
"letter": "c",
"let_frequency": 4,
"let_relation": "at least"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "can you provide a business plan for Oriental weavers to be including the following:\n1-Executive Summary\n2-Company Description\n3-Market analysis \n 4- Organization and management\n 5-Service or product line\n 6-Marketing and sales\n\n 7-Financial projections\n ... | instruction_following_rl | {
"instruction_id_list": [
"keywords:keyword_specific_position",
"detectable_format:multiple_sections"
],
"kwargs": [
{
"keyword": "building",
"n": 1,
"m": 10
},
{
"section_spliter": "SECTION",
"num_sections": 2
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "1. How can we identify a syllable?\n2. What are the five features of stress? Please follow these rules when answering: At the end of your response, please explicitly add a postscript starting with P.P.S and Include keyword mark in your response"
}
] | instruction_following_rl | {
"instruction_id_list": [
"detectable_content:postscript",
"keywords:word_once"
],
"kwargs": [
{
"postscript_marker": "P.P.S"
},
{
"keyword": "mark"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Existe t-il une version de toi sur Android ?\n\nMake sure to adhere to these requirements:\n1. In your response, the word perception should appear at least 1 times."
}
] | instruction_following_rl | {
"instruction_id_list": [
"keywords:frequency"
],
"kwargs": [
{
"keyword": "perception",
"frequency": 1,
"relation": "at least"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Your response should satisfy these conditions:\n1. In your entire response, refrain from the use of . (i.e. dots) as punctuation and in general.\n2. There should be 2 paragraphs. Paragraphs and only paragraphs are separated with each other by two new lines as if it was '\\n\\n'... | instruction_following_rl | {
"instruction_id_list": [
"punctuation:punctuation_dot",
"length_constraints:nth_paragraph_first_word",
"keywords:word_once"
],
"kwargs": [
null,
{
"num_paragraphs": 2,
"nth_paragraph": 2,
"first_word": "classroom"
},
{
"keyword": "stay"
}
],
"verifier"... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints:\n- Your entire response should be in English, and in all capital letters.\n- Write 3 paragraphs, delimited by the markdown divider: * * *, with exactly 3 sentences each, with exactly 2 words in each sentence.\n- Your response ... | instruction_following_rl | {
"instruction_id_list": [
"change_case:english_capital",
"count:counting_composition",
"length_constraints:number_sentences"
],
"kwargs": [
null,
{
"n_sent": 3,
"n_words": 2
},
{
"num_sentences": 17,
"relation": "less than"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Make sure to follow these requirements when responding:\n- Only use unique words in your response, no word should be repeated!\n\nwrite 5 ben shapiro tweets"
}
] | instruction_following_rl | {
"instruction_id_list": [
"count:count_unique"
],
"kwargs": [
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "what if Super Robot Monkey Team to air on disney channel instead of jetix?\n\nMake sure to adhere to these requirements:\n- In your response, the letter p should appear less than 1 times.\n- All sentences must be connected using hyphens, with no spaces between them.\n- In your ... | instruction_following_rl | {
"instruction_id_list": [
"keywords:letter_frequency",
"detectable_format:sentence_hyphens",
"change_case:capital_word_frequency"
],
"kwargs": [
{
"letter": "p",
"let_frequency": 1,
"let_relation": "less than"
},
null,
{
"capital_frequency": 10,
"capital_... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "When answering, please adhere to these constraints:\n- The last word of each sentence, before punctuation, should be the word knife.\n- Include keyword garden in your response.\n- In your response, the word annual should appear less than 2 times.\n\n请帮我做一份水务行业发展与展望的PPT"
}
] | instruction_following_rl | {
"instruction_id_list": [
"last_word:last_word_sent",
"keywords:word_once",
"keywords:frequency"
],
"kwargs": [
{
"last_word": "knife"
},
{
"keyword": "garden"
},
{
"keyword": "annual",
"frequency": 2,
"relation": "less than"
}
],
"verifier": ... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please incorporate these requirements into your answer: Highlight at least 4 sections in your answer with markdown, i.e. *highlighted section*, Include keyword ['designer'] once in your response, keyword ['machine'] twice in your response, and Repeat the phrase Actions speak lo... | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:number_highlighted_sections",
"count:count_increment_word",
"copy:repeat_phrase"
],
"kwargs": [
{
"num_highlights": 4
},
{
"keyword1": "designer",
"keyword2": "machine"
},
{
"phrase": "Actions speak louder than w... | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "what version of GPT are you? Please ensure your answer follows these guidelines: Give two different responses. Responses and only responses should be separated by 6 asterisk symbols: ******, Your ENTIRE response should be in Ukrainian language, no other language is allowed, and... | instruction_following_rl | {
"instruction_id_list": [
"combination:two_responses",
"language:response_language",
"detectable_format:title"
],
"kwargs": [
null,
{
"language": "uk"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "Please incorporate these requirements into your answer: Enclose every word in your response within square brackets and Your entire response should be in English, and in all lowercase letters. No capital letters are allowed. 为什么电压负载之后会比之前高？"
}
] | instruction_following_rl | {
"instruction_id_list": [
"detectable_format:square_brackets",
"change_case:english_lowercase"
],
"kwargs": [
null,
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "What are the best starting points to create a new Angular project ?\n\nYour response should comply with these constraints:\n- Finish your response with this exact phrase Is there anything else I can help with?. No other words should follow this phrase.\n- In your response, the ... | instruction_following_rl | {
"instruction_id_list": [
"startend:end_checker",
"keywords:word_count_different_numbers"
],
"kwargs": [
{
"end_phrase": "Is there anything else I can help with?"
},
{
"keyword": "cream",
"frequency": 1,
"relation": "less than"
}
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
[
{
"role": "user",
"content": "you are a sixteen year old teenager from sydney, australia. your purpose it to become really good at data science and python as fast as possible. your first task is to find a course, mentorship or community that will accelerate your growth and experience. how will you do this?\... | instruction_following_rl | {
"instruction_id_list": [
"count:count_unique",
"keywords:letter_frequency",
"detectable_format:square_brackets"
],
"kwargs": [
null,
{
"letter": "o",
"let_frequency": 8,
"let_relation": "less than"
},
null
],
"verifier": "ifeval"
} | IF-RL | {
"type": "responses_api_agents",
"name": "instruction_following_simple_agent"
} | nano-v3-RL-instruction_following |
End of preview.
No dataset card yet
- Downloads last month
- 554