id stringlengths 16 16 | text stringlengths 13.5k 135k | thinking stringclasses 0
values | template stringclasses 67
values | family stringclasses 13
values | pool stringclasses 9
values | system stringclasses 15
values | temperature float32 0.6 0.95 | top_p float32 0.85 0.98 | max_tokens int32 16.4k 16.4k | seed int64 3.32M 2.15B | enable_thinking bool 2
classes | response dict |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
aug-0:0000000017 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
You are generating ReAct-style agent training traces. Given a document and an action API of three functions:
Search[query] -> returns a passage
Lookup[term] -> returns the next occurre... | null | agentic_doc_to_react | agentic | agentic | You write concise, content-grounded conversations that stay close to the source. | 0.933 | 0.96 | 16,384 | 1,996,601,397 | false | {
"id": "chatcmpl-bdcfa340868c46c9",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Question: What is the total number of families residing in Shebertu according to the 2006 census?\nThought 1: I need to find the census population d... |
aug-0:0000000002 | <|im_start|>user
Produce a structured function-calling conversation in Hermes style grounded in the document.
1. In the system turn, declare 2-4 tools as JSON inside <tools>...</tools>, each {"name","description","parameters":{json-schema}}.
2. Write a realistic user request answerable only by calling tool(s).
3. The ... | null | agentic_hermes_toolcall | agentic | agentic | 0.869 | 0.937 | 16,384 | 621,298,943 | false | {
"id": "chatcmpl-9070e191c0f9fe75",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "<tool_call>{\"name\": \"extract_concepts\", \"arguments\": {\"document_text\": \"Analysis of algorithms\\nFor example, if the sorted list to which w... | |
aug-0:0000000047 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
For the following paragraph give me a paraphrase of the same using a very small vocabulary and extremely simple sentences that... | null | wrap_easy | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.755 | 0.968 | 16,384 | 1,826,627,621 | false | {
"id": "chatcmpl-be28d28032ebd66c",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Key Terms and Concepts:**\n\n1. **Justus Perthes**: A man who started a big shop.\n2. **Bernhardt Wilhelm Perthes**: Justus Perthes' grandson. H... |
aug-0:0000000029 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(que... | null | agentic_toolformer_qa | agentic | agentic | You write educational dialogues. Do not invent facts outside the provided context. | 0.622 | 0.888 | 16,384 | 1,238,943,212 | false | {
"id": "chatcmpl-9f9a8b937349cd8f",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To apply the learning strategy \"Create a Table of Contents\" to the given document, I will first extract the main sections and headings. \n\nThe ma... |
aug-0:0000000026 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
Rewrite the document below to remove personally-identifying and conversational noise while preserving all factual and technica... | null | transform_pseudonymize | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.703 | 0.928 | 16,384 | 1,315,336,105 | false | {
"id": "chatcmpl-81c301dbe35ecd14",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Teach someone else**\nTeaching someone else about the Gate Church of the Trinity can help retain information and solidify understanding. The infor... |
aug-0:0000000046 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
For the following paragraph give me a paraphrase of the same using very terse and abstruse language that only an erudite schol... | null | wrap_hard | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.913 | 0.885 | 16,384 | 889,563,720 | false | {
"id": "chatcmpl-8334e99254d85e76",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here is a summary of the document in my own words:\n\nThe document outlines the historical trajectory and current status of the renowned cartographi... |
aug-0:0000000039 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
Rewrite the document below to remove personally-identifying and conversational noise while preserving all factual and technica... | null | transform_pseudonymize | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.769 | 0.896 | 16,384 | 818,384,599 | false | {
"id": "chatcmpl-8bbccc6f1cf9a94d",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**HP Pavilion dv1000 Series**\n\n* **Target Audience:** Home and small business users.\n* **Design:** \"Thin and light\" widescreen laptops with... |
aug-0:0000000008 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Convert the following paragraph into a conversational format with multiple tags of "Question:" followed by "Answer:":
given a... | null | wrap_qa | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.881 | 0.913 | 16,384 | 1,733,008,472 | false | {
"id": "chatcmpl-8ed42da4116b5227",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Question:** What is the primary definition of the \"analysis of algorithms\" according to the provided text?\n**Answer:** In computer science, the... |
aug-0:0000000012 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Convert the following paragraph into a conversational format with multiple tags of "Question:" followed by "Answer:":
tion or ... | null | wrap_qa | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.839 | 0.891 | 16,384 | 2,099,980,028 | false | {
"id": "chatcmpl-824d56c899389553",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Question: What is the main subject of the provided document?\nAnswer: The document is about a village named Shebertu, located in Iran.\n\nQuestion: ... |
aug-0:0000000032 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Rewrite the document as a comprehensive FAQ (Frequently Asked Questions). Extract or infer the key questions a reader would have about this topic, then provide clear, direct answers. Order qu... | null | synthpile_faq | synthpile | general_rewrite | You write educational dialogues. Do not invent facts outside the provided context. | 0.844 | 0.972 | 16,384 | 461,176,380 | false | {
"id": "chatcmpl-af4bf87411a76437",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "# Biography of Don M. Carlson: Sectional Breakdown\n\n## 1. Early Life and Education\n* **Full Name:** Don M. Carlson\n* **Birth Date:** Decembe... |
aug-0:0000000015 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.687 | 0.852 | 16,384 | 2,122,854,924 | false | {
"id": "chatcmpl-ad8e6fe0b405636d",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The user has provided a series of examples showing how to break down, analyze, and extract information from a document about the village of Shebertu... |
aug-0:0000000037 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
Rewrite the document below to remove personally-identifying and conversational noise while preserving all factual and technica... | null | transform_pseudonymize | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.629 | 0.974 | 16,384 | 1,536,626,756 | false | {
"id": "chatcmpl-81732863e82c7c3e",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "HP Pavilion dv1000 series\n| developer = [[Hewlett-Packard]] ([[HP Inc.]])\n| cpu = \n| graphics = \n| website = \n... |
aug-0:0000000024 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.612 | 0.903 | 16,384 | 824,995,566 | false | {
"id": "chatcmpl-8f772d30aac8e538",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Data-Cleaning and Templatization Report\n\n**Status:** No physical quantities requiring unit conversion were found in the provided text.\n**Reas... |
aug-0:0000000094 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Rewrite the document as a structured table that organizes the key information, then generate one question-answer pair based on the table. First extract the main data points and organize them in... | null | synthpile_table | synthpile | general_rewrite | You write concise, content-grounded conversations that stay close to the source. | 0.898 | 0.889 | 16,384 | 974,299,746 | false | {
"id": "chatcmpl-8711b1fb62e5f8d1",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "| Category | Key Information |\n| :--- | :--- |\n| **Location** | County Durham, England; 6 miles southeast of Consett; 7 miles southwest of Durham.... |
aug-0:0000000006 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Read the document below. Imagine three different kinds of users who might encounter this content (e.g. a student, an expert practitioner, and a casual reader). For each user, generate ONE pla... | null | persona_instruction_fewshot | persona | general_dialogue | You write educational dialogues. Do not invent facts outside the provided context. | 0.7 | 0.865 | 16,384 | 1,102,987,810 | false | {
"id": "chatcmpl-9b2d2177121386cd",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**User type:** Student\n**Instruction:** Can you explain what \"DTIME\" and \"iteration\" mean in the context of this document?\n**Answer:** Based o... |
aug-0:0000000028 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(question)]" where "question"... | null | agentic_toolformer_qa | agentic | agentic | You are a careful, faithful synthetic-dialogue generator. | 0.8 | 0.861 | 16,384 | 1,454,145,055 | false | {
"id": "chatcmpl-80884f792811df1e",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Based on the provided document, here is a timeline of Don Carlson's life and political career, organized chronologically:\n\n* **1938**\n * *... |
aug-0:0000000000 | <|im_start|>system
You are an expert at writing instruction-following tasks whose compliance can be checked by a deterministic program. Your verifier must be correct.<|im_end|>
<|im_start|>user
You generate hard instruction-following data with MULTIPLE simultaneous, verifiable constraints (IFEval / AutoIF style).
Usin... | null | autoif_multi_constraint | autoif | autoif | You are an expert at writing instruction-following tasks whose compliance can be checked by a deterministic program. Your verifier must be correct. | 0.632 | 0.903 | 16,384 | 61,707,255 | false | {
"id": "chatcmpl-944514c27c8f11e9",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "[Instruction]\n**Topic:** Algorithm Analysis Categories\n**Constraints:**\n1. **Format:** The response must contain exactly **3** bullet points.\n2... |
aug-0:0000000081 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Task: Read the text, ask questions and answer them.
Follow these instructions:
1. Ask diverse questions that require different cognitive skills or cover different aspects of the text.
1. Ask qu... | null | synthpile_diverse_qa | synthpile | general_rewrite | You write concise, content-grounded conversations that stay close to the source. | 0.886 | 0.932 | 16,384 | 936,384,885 | false | {
"id": "chatcmpl-afae5c001c6e777d",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are the questions and answers based on the provided text:\n- Question: Are the pillboxes and trenches system located in the vicinity of Pok Wai... |
aug-0:0000000066 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
For the following paragraph give me a paraphrase of the same using very terse and abstruse language that only an erudite schol... | null | wrap_hard | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.617 | 0.92 | 16,384 | 1,953,206,530 | false | {
"id": "chatcmpl-8fde5b3c9d213e53",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Júlio César Redecker (1956–2007) was a Brazilian attorney and federal deputy representing Rio Grande do Sul, serving from 1995 until his untimely de... |
aug-0:0000000020 | <|im_start|>user
Here is an extract from a webpage: "hurch's frescoes were based on [[Bible|Biblical scenes]], and the exterior decor was based on Ukrainian folklore. Eighteenth-century compositions by F. Pavlovskyi, I. Maksimovych and A. Galik (with help from M. Yakubovych and I. Kadelskyi) have been preserved to this... | null | cosmopedia_blog | cosmopedia | education | 0.642 | 0.858 | 16,384 | 1,178,171,468 | false | {
"id": "chatcmpl-bf62c4bab660a6c7",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Stepping into the Gate Church of the Trinity within the sprawling grounds of the Kyiv Pechersk Lavra feels less like entering a standard place of wo... | |
aug-0:0000000108 | <|im_start|>system
You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any... | null | agentic_codeact | agentic | agentic | You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any standard library o... | 0.665 | 0.966 | 16,384 | 440,010,694 | false | {
"id": "chatcmpl-9438beb256e62c9a",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "<execute>\n# Attempting to parse the document text to extract ride locations for mapping\nimport re\n\ndocument_text = \"\"\"Joyland (Great Yarmouth... |
aug-0:0000000054 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.894 | 0.956 | 16,384 | 482,665,975 | true | {
"id": "chatcmpl-ae802aacd0ed0257",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The user is asking me to create flashcards with key terms and concepts based on the provided document. The document is about Unnamed Battery No. 1, ... |
aug-0:0000000048 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
You are an API designer. Given a document, hypothesize a realistic library of 4-8 functions whose combination could answer questions a knowledgeable user might ask about this domain. Function... | null | agentic_doc_to_tools | agentic | agentic | You write educational dialogues. Do not invent facts outside the provided context. | 0.932 | 0.965 | 16,384 | 2,135,074,721 | false | {
"id": "chatcmpl-99974ae0d8219369",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "[\n {\n \"name\": \"get_person_by_name\",\n \"description\": \"Retrieves biographical details for a person given their name or birth name.\",... |
aug-0:0000000077 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Read the document below. Imagine a curious user who has just read it. Generate ONE specific, well-formed instruction (i.e., a user prompt) that this person might naturally ask a helpful AI assi... | null | persona_instruction_zeroshot | persona | general_dialogue | You write concise, content-grounded conversations that stay close to the source. | 0.837 | 0.936 | 16,384 | 1,801,071,684 | false | {
"id": "chatcmpl-a7b7434c8f7efe76",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Instruction:**\nBased on the provided text about Pok Wai and the referenced picture of the shrine at No. 106, explain to a visitor why this specif... |
aug-0:0000000120 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Context:
<document>
ory:1985 in Missouri]]
[[Category:1985 United States mayoral elections|St. Louis]]
1985 St. Louis mayoral election
== General election =={{Election box begin no change
| p... | null | two_students | mind | general_dialogue | You write educational dialogues. Do not invent facts outside the provided context. | 0.698 | 0.904 | 16,384 | 328,998,319 | false | {
"id": "chatcmpl-9c166ef587eb7716",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Student A:** Hey, have you started looking at the 1985 St. Louis mayoral election data we need for our assignment? I'm trying to figure out who th... |
aug-0:0000000049 | <|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(question)]" where "question" is the question you want to ask. Here are some examples of API calls:
Input: Joe Biden... | null | agentic_toolformer_qa | agentic | agentic | 0.61 | 0.854 | 16,384 | 779,582,990 | false | {
"id": "chatcmpl-a4165a6ab72a1bdc",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To apply the \"Ask questions and seek clarification\" strategy to the provided document, I will identify specific points where information is implic... | |
aug-0:0000000001 | <|im_start|>user
Your task is to enrich the text below with agent function calls. First DECIDE which tool(s) would actually help complete or verify the text — invent them yourself (e.g. calculator, calendar, unit_convert, currency, search, lookup) and only use ones that are genuinely needed. Then interleave the calls i... | null | agentic_toolformer_freeform | agentic | agentic | 0.843 | 0.978 | 16,384 | 15,326,589 | false | {
"id": "chatcmpl-b717368e90f5f68b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To apply the learning strategy of breaking down the document into smaller sections, I need to organize the fragmented text into logical segments. Be... | |
aug-0:0000000063 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
For the following paragraph give me a paraphrase of the same using very terse and abstruse language that only an erudite schol... | null | wrap_hard | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.88 | 0.858 | 16,384 | 1,782,553,501 | false | {
"id": "chatcmpl-8fe371b217124926",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here is a tabular compendium organizing the titular subject's political affiliations and held offices, derived from the provided documentation:\n\n|... |
aug-0:0000000003 | <|im_start|>system
You are an expert at writing instruction-following tasks whose compliance can be checked by a deterministic program. Your verifier must be correct.<|im_end|>
<|im_start|>user
You generate hard instruction-following data with MULTIPLE simultaneous, verifiable constraints (IFEval / AutoIF style).
Usin... | null | autoif_multi_constraint | autoif | autoif | You are an expert at writing instruction-following tasks whose compliance can be checked by a deterministic program. Your verifier must be correct. | 0.861 | 0.936 | 16,384 | 720,153,533 | false | {
"id": "chatcmpl-b91af3082cd4234a",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "[Instruction]\n**Constraint 1:** The response must contain **exactly 4** bullet points.\n**Constraint 2:** Each bullet point must be **≤ 15 words**.... |
aug-0:0000000067 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Rewrite the document as a structured table that organizes the key information, then generate one question-answer pair based on the table. First extract the main data points and organize them ... | null | synthpile_table | synthpile | general_rewrite | You write educational dialogues. Do not invent facts outside the provided context. | 0.646 | 0.918 | 16,384 | 1,811,691,082 | false | {
"id": "chatcmpl-b70f97b83bfba148",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "| Category | Details |\n| :--- | :--- |\n| **Birth** | 1956 (Taquara) |\n| **Death** | 2007 (TAM Airlines Flight 3054) |\n| **Political Parties** | ... |
aug-0:0000000131 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Task:
Read the text, ask questions and answer them.
Follow these instructions:
1. Ask diverse questions that require differen... | null | nemotron_cc_diverse_qa | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.781 | 0.885 | 16,384 | 1,865,864,570 | false | {
"id": "chatcmpl-b3d28f147fe60a20",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are the questions and answers based on the provided text:\n- Question: What is the primary focus of the book \"Creative Dreaming: Plan And Cont... |
aug-0:0000000061 | <|im_start|>user
Rewrite the document as a clear, step-by-step tutorial or instructional guide. Use numbered steps or bullet points where appropriate to enhance clarity. Preserve all essential information while ensuring the style feels didactic and easy to follow. Output only the tutorial, nothing else.
Document:
ruct... | null | synthpile_tutorial | synthpile | general_rewrite | 0.916 | 0.929 | 16,384 | 1,451,763,550 | false | {
"id": "chatcmpl-917c40104dc6a44b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "# Tutorial: How to Create a Comprehensive Study Guide for Unnamed Battery No. 1\n\nFollow these step-by-step instructions to transform the raw data ... | |
aug-0:0000000013 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.852 | 0.872 | 16,384 | 86,007,583 | false | {
"id": "chatcmpl-b2b208172fafbce9",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Concept Map: Shebertu**\n\n**Central Node: Shebertu (The Village)**\n* **Identity & Naming**\n * Official Name: Shebertu\n * Native Na... | |
aug-0:0000000069 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Task:
Read the text, ask questions and answer them.
Follow these instructions:
1. Ask diverse questions that require differen... | null | nemotron_cc_diverse_qa | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.805 | 0.914 | 16,384 | 1,103,914,784 | false | {
"id": "chatcmpl-a9c3e4f3cd51d477",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are the questions and answers based on the provided text:\n- Question: Was Júlio Redecker a member of the Brazilian Social Democracy Party (PSD... |
aug-0:0000000018 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
I need to critique a document. Consider the document below. FIRST, generate a conversational style that would be helpful, interesting and creative in conveying the general information in this d... | null | huu_conversation | huu | general_dialogue | You write concise, content-grounded conversations that stay close to the source. | 0.764 | 0.906 | 16,384 | 1,467,419,148 | false | {
"id": "chatcmpl-878fe84e0c7334d7",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Chosen Conversational Style: **The \"Field Agent & The Skeptic\"**\n* **Roles:** One is an enthusiastic field agent (\"Alex\") trying to map o... |
aug-0:0000000004 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.618 | 0.965 | 16,384 | 1,027,996,327 | false | {
"id": "chatcmpl-9d2319f08bcd4293",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef parameterize_algorithm_growth_analysis(\n t1, t2, \n n1, n2, \n power_rule_holds=True, \n algorithm_comparison_needed=Fal... | |
aug-0:0000000116 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Produce a structured function-calling conversation in Hermes style grounded in the document.
1. In the system turn, declare 2-4 tools as JSON inside <tools>...</tools>, each {"name","description","parameters":{json-... | null | agentic_hermes_toolcall | agentic | agentic | You are a careful, faithful synthetic-dialogue generator. | 0.707 | 0.911 | 16,384 | 1,896,146,190 | true | {
"id": "chatcmpl-b2d1ad1127e6746b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The user is asking me to create a structured function-calling conversation in Hermes style based on the document provided. I need to:\n\n1. Create 2... |
aug-0:0000000090 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
Rewrite the document below to remove personally-identifying and conversational noise while preserving all factual and technica... | null | transform_pseudonymize | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.762 | 0.891 | 16,384 | 1,723,386,592 | false | {
"id": "chatcmpl-8f724969f81c3508",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "[author]\nTo apply the learning strategy of reviewing and summarizing the information to ensure understanding and retention of the key points, I wil... |
aug-0:0000000088 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Review the text and extract the key information. Follow these instructions:
- Carefully read the above text and provide a conc... | null | nemotron_cc_knowledge_list | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.823 | 0.883 | 16,384 | 839,556,592 | false | {
"id": "chatcmpl-94ff03da583922d3",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "* Paležnica Gornja (Палежница Горња) is a village located in the municipality of Doboj, Bosnia and Herzegovina.\n* Administratively, the village fal... |
aug-0:0000000040 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Context:
<document>
ons:**
* **Intel CPUs:**
* Pentium Dual-Core: Dual-core processing, Hyper-Threading
* Core Solo: Single-core processing
* Core Duo: Dual-core processing, Hyper-Threading
*... | null | problem_solving | mind | general_dialogue | You are a careful, faithful synthetic-dialogue generator. | 0.713 | 0.892 | 16,384 | 1,121,567,351 | false | {
"id": "chatcmpl-ab3125c0ec8c13db",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**User:** I've been studying the HP Pavilion dv1000 series documentation. I noticed the specs vary quite a bit between models. Can we break down how... |
aug-0:0000000022 | <|im_start|>system
You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any... | null | agentic_codeact | agentic | agentic | You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any standard library o... | 0.663 | 0.922 | 16,384 | 2,041,301,245 | false | {
"id": "chatcmpl-83555eeba5c12c5d",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "<execute>\n# Task: Extract and analyze bibliographic references from the document to create a structured summary.\n# I will parse the document to id... |
aug-0:0000000043 | <|im_start|>system
You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any... | null | agentic_codeact | agentic | agentic | You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any standard library o... | 0.886 | 0.91 | 16,384 | 775,526,445 | false | {
"id": "chatcmpl-902e5bbacb06c15f",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "<execute>\nimport re\nfrom datetime import datetime\n\n# Define the document text\ndocument_text = \"\"\"\nHP Pavilion dv1000 series\n| lifespan ... |
aug-0:0000000035 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.677 | 0.929 | 16,384 | 209,735,037 | false | {
"id": "chatcmpl-bc46b5ecffb028f5",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef don_carlson_profile(\n birth_date_year = 1938,\n birth_date_month = 12,\n birth_date_day = 30,\n birth_place_city = \"Abe... | |
aug-0:0000000072 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Your task is to rewrite knowledge from the provided text following these instructions.
- Rewrite the text as a passage or pass... | null | nemotron_cc_extract_knowledge | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.855 | 0.91 | 16,384 | 2,059,297,411 | false | {
"id": "chatcmpl-8847c33230206ac2",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Júlio César Redecker, born on July 12, 1956, in Taquari, Rio Grande do Sul, Brazil, was a Brazilian lawyer and politician who served as a member of ... |
aug-0:0000000060 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Read the document below and assume the role of a knowledgeable expert in its domain. Write a Quora-style article from this expert's perspective. Draw on your knowledge, skills, experience, or... | null | persona_quora_article | persona | education | You write educational dialogues. Do not invent facts outside the provided context. | 0.877 | 0.934 | 16,384 | 838,910,207 | false | {
"id": "chatcmpl-9aa628c8d10ae978",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Title: Visualizing the Fort: A Guide to Understanding Unnamed Battery No. 1**\n\n**By Dr. Elena Vance, Historic Preservation Specialist & Civil Wa... |
aug-0:0000000019 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Read the document below. Imagine three different kinds of users who might encounter this content (e.g. a student, an expert practitioner, and a casual reader). For each user, generate ONE plausible instruction they m... | null | persona_instruction_fewshot | persona | general_dialogue | You are a careful, faithful synthetic-dialogue generator. | 0.92 | 0.948 | 16,384 | 771,907,223 | false | {
"id": "chatcmpl-b0bc717c0914d245",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**User 1: Student**\n**Instruction:** \"I need to study for a geography exam and I'm struggling to remember the specific administrative hierarchy fo... |
aug-0:0000000057 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.835 | 0.95 | 16,384 | 701,102,559 | true | {
"id": "chatcmpl-86d5210f83d8cf90",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The user is asking me to extract quantities from the text and convert them to SI units. Let me carefully analyze the document and identify all physi... |
aug-0:0000000122 | <|im_start|>user
Context:
<document>
; />
[[Category:1985 United States mayoral elections|St. Louis]]
| party = Republican Party (United States)
| percentage= 16.24
{{reflist|30em}}{{1985 United States elections}}[[Category:Mayoral elections in St. Louis]]
1985 St. Louis mayoral election
| candidate = [[Vincent C. S... | null | two_students | mind | general_dialogue | 0.755 | 0.954 | 16,384 | 568,616,711 | false | {
"id": "chatcmpl-a776328b71ddabeb",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Student A:** Hey, I've been trying to organize the data from the 1985 St. Louis mayoral election document, but the formatting is a bit messy. Do y... | |
aug-0:0000000095 | <|im_start|>system
You are an artificial intelligence assistant. You carefully provide accurate, factual, thoughtful, nuanced answers, and are brilliant at reasoning.<|im_end|>
<|im_start|>user
Your task is to read and paraphrase the provided text following these instructions:
- Aim to create a condensed but accurate a... | null | nemotron_cc_distill | nemotron_cc | general_rewrite | You are an artificial intelligence assistant. You carefully provide accurate, factual, thoughtful, nuanced answers, and are brilliant at reasoning. | 0.684 | 0.944 | 16,384 | 1,131,643,609 | false | {
"id": "chatcmpl-977e3d76631c8ae4",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Paraphrased Text:\nCornsay is a hamlet and civil parish located in County Durham, England, within the North East region. According to the 2011 Unite... |
aug-0:0000000103 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.834 | 0.925 | 16,384 | 1,480,668,088 | false | {
"id": "chatcmpl-a203274acd9b58f3",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The provided text contains no physical quantities with units (such as mass, length, time, temperature, etc.) that require conversion. It consists of... |
aug-0:0000000149 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
For the following paragraph give me a paraphrase of the same using a very small vocabulary and extremely simple sentences that... | null | wrap_easy | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.927 | 0.927 | 16,384 | 1,310,591,655 | false | {
"id": "chatcmpl-af34d3e30da21f6c",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here is a simple story about the artists, like a bedtime story for a small child:\n\nBig man had a brush.\nHe drew a big sun.\nOne man drew a big ca... |
aug-0:0000000074 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.755 | 0.851 | 16,384 | 1,591,952,330 | false | {
"id": "chatcmpl-aeecb925489ef445",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "There are no physical quantities (values with units such as mass, length, time, temperature, etc.) present in the provided text. The text contains:\... |
aug-0:0000000055 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.911 | 0.967 | 16,384 | 813,000,095 | false | {
"id": "chatcmpl-b0fdc1eb3d7efac6",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "I have analyzed the provided text and the instructions. However, there is a critical mismatch between your **system role definition** and the **cont... |
aug-0:0000000062 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.934 | 0.962 | 16,384 | 127,631,935 | false | {
"id": "chatcmpl-b9571808121bc7d0",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef unnamed_battery_no_1_structurer(\n name_str,\n added_date_str,\n ref_num_str,\n built_year_int,\n location_str,\n r... | |
aug-0:0000000154 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.855 | 0.886 | 16,384 | 70,906,911 | false | {
"id": "chatcmpl-a3fe1b10c26ea430",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "There are no physical quantities (such as mass, volume, time, temperature, or percentage) present in the provided text. The text contains a list of ... |
aug-0:0000000051 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.804 | 0.946 | 16,384 | 2,115,275,823 | false | {
"id": "chatcmpl-812c85bf41622438",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Based on the input document provided, here is the text parameterization system output. The prose describing the succession of the Justus Perthes bus... | |
aug-0:0000000044 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.807 | 0.98 | 16,384 | 1,649,005,502 | false | {
"id": "chatcmpl-b9b37ffb3c0809b6",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef hp_pavilion_dv1000_parameterization(\n # --- Categorical Structure ---\n product_name = 'HP Pavilion dv1000 series',\n produ... | |
aug-0:0000000052 | <|im_start|>system
You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any... | null | agentic_codeact | agentic | agentic | You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any standard library o... | 0.871 | 0.931 | 16,384 | 303,960,565 | true | {
"id": "chatcmpl-81c625a4e310cab9",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The user is asking me to apply a learning strategy called \"Visualize the Setting\" to a document about Justus Perthes. This is a continuation of a ... |
aug-0:0000000123 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
You are an API designer. Given a document, hypothesize a realistic library of 4-8 functions whose combination could answer questions a knowledgeable user might ask about this domain. Functions ... | null | agentic_doc_to_tools | agentic | agentic | You write concise, content-grounded conversations that stay close to the source. | 0.747 | 0.916 | 16,384 | 466,215,224 | false | {
"id": "chatcmpl-ae4a2fddc6da3770",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "{\"name\": \"extract_election_timeline\", \"description\": \"Extracts chronological events and dates from the document to create a timeline of the e... |
aug-0:0000000010 | <|im_start|>user
You are a smart software engineer. Please evaluate the following code on a scale of 1 to 10 based on the following criteria:
1. Are variable names descriptive and consistent with naming conventions?
2. Are comments and doc-strings appropriately written to explain the purpose and functionality of the c... | null | swallow_code_sgcr | swallow | code | 0.612 | 0.923 | 16,384 | 894,361,291 | false | {
"id": "chatcmpl-ab9936dc08fdbbb0",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Evaluation: 3\n\n### Suggestions:\nThe provided \"code\" is actually a series of text responses analyzing a Wikipedia-style Infobox for a villag... | |
aug-0:0000000092 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Read the document below. Imagine three different kinds of users who might encounter this content (e.g. a student, an expert practitioner, and a casual reader). For each user, generate ONE pla... | null | persona_instruction_fewshot | persona | general_dialogue | You write educational dialogues. Do not invent facts outside the provided context. | 0.949 | 0.921 | 16,384 | 896,974,915 | false | {
"id": "chatcmpl-bd4e739e30ff3a75",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are three labeled triples representing different user types interacting with the document provided:\n\n**User type: History Student**\n**Instru... |
aug-0:0000000041 | <|im_start|>user
I need to teach a document. Consider the document below. FIRST, generate a conversational style that would be helpful, interesting and creative in conveying the general information in this document. SECOND, apply the conversational style to the document. THIRD, when generating the conversation, add at ... | null | huu_conversation | huu | general_dialogue | 0.83 | 0.962 | 16,384 | 1,713,677,810 | false | {
"id": "chatcmpl-a47bdcb6169da713",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Part 1: Conversational Style Generation\n\n**Selected Style:** **(6) Two-Students (The \"Riveting History\" & \"The Skeptic\")**\n\n**Why this s... | |
aug-0:0000000082 | <|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(question)]" where "question" is the question you want to ask. Here are some examples of API calls:
Input: Joe Biden... | null | agentic_toolformer_qa | agentic | agentic | 0.647 | 0.862 | 16,384 | 1,887,260,493 | false | {
"id": "chatcmpl-b3b9e25a842aa40b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To reinforce my learning and memory, I will review the document regularly, ideally every week or every two weeks. This will help me to:\n\n1. Recal... | |
aug-0:0000000083 | <|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(question)]" where "question" is the question you want to ask. Here are some examples of API calls:
Input: Joe Biden... | null | agentic_toolformer_qa | agentic | agentic | 0.799 | 0.963 | 16,384 | 1,680,256,238 | false | {
"id": "chatcmpl-a02d645de8f3d117",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Let's apply the \"Break down the document into smaller chunks\" strategy to the provided document.\n\n**1. Metadata**\n* **Title:** Paležnica Gorn... | |
aug-0:0000000059 | <|im_start|>system
You are an expert at writing instruction-following tasks whose compliance can be checked by a deterministic program. Your verifier must be correct.<|im_end|>
<|im_start|>user
You generate instruction-following training data with checkable constraints (IFEval / AutoIF style).
Using the document below... | null | autoif_constraint | autoif | autoif | You are an expert at writing instruction-following tasks whose compliance can be checked by a deterministic program. Your verifier must be correct. | 0.637 | 0.857 | 16,384 | 176,611,997 | true | {
"id": "chatcmpl-bbd5c49ca9937e29",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The user wants me to create an instruction-following training data example based on the provided document about \"Unnamed Battery No. 1\". I need to... |
aug-0:0000000097 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
Rewrite the document below to remove personally-identifying and conversational noise while preserving all factual and technica... | null | transform_pseudonymize | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.819 | 0.888 | 16,384 | 708,254,693 | false | {
"id": "chatcmpl-b755f56101c14b69",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "[Expert]\n**Section 1:**\nCornsay is a hamlet and civil parish in County Durham, England. The population of the civil parish taken at the 2011 censu... |
aug-0:0000000115 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
For the following paragraph give me a diverse paraphrase of the same in high quality English language as in sentences on Wikip... | null | wrap_medium | wrap | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.83 | 0.96 | 16,384 | 210,748,600 | false | {
"id": "chatcmpl-9c36a159d1081dd1",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The supercell subsequently generated a tornado in the **Bucca** and **Kolan** region. Meteorological reports indicated that the storm's intensity wa... |
aug-0:0000000085 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Your task is to complete a given piece of text by inserting Wikipedia search API calls where helpful. You can do so by writing "[WikiSearch(term)]" where "term" is the search term you want to l... | null | agentic_toolformer_wiki | agentic | agentic | You write concise, content-grounded conversations that stay close to the source. | 0.746 | 0.959 | 16,384 | 1,046,668,603 | false | {
"id": "chatcmpl-b53c8fb164155dc8",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Based on the provided document, here is a concept question to test your understanding:\n\n**Concept Question:**\nConsidering the administrative stru... |
aug-0:0000000114 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Context:
<document>
1992 Queensland storms and the 1992 tornado outbreaks were significant events that occurred in Australia in 1992. The exact details of the storms and the tornado outbreak... | null | debate | mind | general_dialogue | You write educational dialogues. Do not invent facts outside the provided context. | 0.84 | 0.923 | 16,384 | 334,369,071 | false | {
"id": "chatcmpl-a0424f182e3bc3de",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Participant A:** Based on the document, the 1992 Queensland storms represent a significant natural disaster event in Australia, specifically invol... |
aug-0:0000000112 | <|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(question)]" where "question" is the question you want to ask. Here are some examples of API calls:
Input: Joe Biden... | null | agentic_toolformer_qa | agentic | agentic | 0.648 | 0.878 | 16,384 | 1,109,746,476 | false | {
"id": "chatcmpl-82417a62c0d4125b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To apply the strategy of breaking down the document into smaller chunks, I will divide the text into logical sections based on the events and detail... | |
aug-0:0000000033 | <|im_start|>user
Here is an extract from a webpage: "hington State Legislature by Districts Since 1889 |pages=144–145 |url=http://leg.wa.gov/History/Legislative/Documents/MembersOfLeg2019.pdf |publisher=[[Washington State Legislature]] |accessdate=April 23, 2019}}</ref> Carlson, a Republican, lost his seat to Dem... | null | cosmopedia_textbook_narrative | cosmopedia | education | 0.897 | 0.907 | 16,384 | 131,540,138 | false | {
"id": "chatcmpl-8c57278a5b94305a",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The story of Don Carlson is not merely a chronicle of a man who sat in a government chamber for a dozen years; it is a case study in the precarious ... | |
aug-0:0000000109 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Your task is to rewrite knowledge from the provided text following these instructions.
- Rewrite the text as a passage or pass... | null | nemotron_cc_extract_knowledge | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.732 | 0.875 | 16,384 | 241,694,066 | false | {
"id": "chatcmpl-a1fb6c3870c54b87",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Joyland, located in Great Yarmouth, is a theme park renowned for its heritage rides that have remained unchanged through generations, allowing paren... |
aug-0:0000000126 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
You are generating ReAct-style agent training traces. Given a document and an action API of three functions:
Search[query] -> returns a passage
Lookup[term] -> returns the next occur... | null | agentic_doc_to_react | agentic | agentic | You write educational dialogues. Do not invent facts outside the provided context. | 0.677 | 0.853 | 16,384 | 1,374,306,994 | false | {
"id": "chatcmpl-985a0214809bcd65",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Question: What was the total number of votes received by the incumbent mayor in the general election compared to his vote count in the Democratic pr... |
aug-0:0000000014 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Here is an extract from a webpage: "ebsite =
|subdivision_name3 =[[Chang Almas District|Chang Almas]]
|utc_offset_DST = +4:30
|settlement_type = village
}}
|popul... | null | cosmopedia_textbook_academic | cosmopedia | education | You write concise, content-grounded conversations that stay close to the source. | 0.816 | 0.868 | 16,384 | 1,364,825,044 | false | {
"id": "chatcmpl-b96b84c483399ea9",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The study of rural demographics in the Kurdistan Province of Iran offers a compelling case study for understanding the complex interplay between adm... |
aug-0:0000000071 | <|im_start|>user
Context:
<document>
an (1956–2007)}}
| birth_name = Júlio César Redecker
| death_cause = [[TAM Airlines Flight 3054|Plane crash]]
* [[National Renewal Alliance|ARENA]] (1974–80)
| term_end1 = July 17, 2007
| term_start1 = February 6, 2007
* [[Brazilian Social Democracy Party|PSDB... | null | two_professors | mind | general_dialogue | 0.914 | 0.944 | 16,384 | 757,600,439 | false | {
"id": "chatcmpl-9efa804006e94177",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Professor A:** Good morning, colleagues. Today we are reviewing the biographical data on Júlio César Redecker to ensure our students understand th... | |
aug-0:0000000025 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Here is an extract from a webpage: " was built in 1106-1108, as part of the [[Pechersk Lavra fortification]], atop the main entrance to the monastery. The church was founded by the grandson of the [[Rulers of Kievan ... | null | cosmopedia_wikihow | cosmopedia | education | You are a careful, faithful synthetic-dialogue generator. | 0.873 | 0.948 | 16,384 | 112,622,051 | false | {
"id": "chatcmpl-9a731b5248d5995c",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "# How to Create an Effective Historical Study Timeline from Primary Source Texts\n\n**Article by:** WikiHow Editorial Team \n**Category:** Educatio... |
aug-0:0000000065 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
I need to teach a document. Consider the document below. FIRST, generate a list of one question that covers every piece of technical information in this document. SECOND, after generating all t... | null | huu_questions | huu | general_dialogue | You write concise, content-grounded conversations that stay close to the source. | 0.928 | 0.956 | 16,384 | 713,882,674 | false | {
"id": "chatcmpl-a82bf8080cc7edde",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here is the comprehensive breakdown of the document regarding Júlio Redecker, following your four-step instruction.\n\n### Step 1: Comprehensive Tec... |
aug-0:0000000089 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Your task is to add calls to a Calculator API to a piece of text. The calculations should help you get the information required to complete the text. You can call the API by writing "[Calcula... | null | agentic_toolformer_calc | agentic | agentic | You write educational dialogues. Do not invent facts outside the provided context. | 0.82 | 0.877 | 16,384 | 2,103,663,655 | false | {
"id": "chatcmpl-83352e462d709f84",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Response:\nHere are the flashcards based on the document provided, focusing on key terms, coordinates, and demographic data.\n\n**Flashcard 1:**... |
aug-0:0000000163 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Rewrite the document as a structured table that organizes the key information, then generate one question-answer pair based on the table. First extract the main data points and organize them in... | null | synthpile_table | synthpile | general_rewrite | You write concise, content-grounded conversations that stay close to the source. | 0.606 | 0.903 | 16,384 | 2,054,106,975 | false | {
"id": "chatcmpl-a9f596e6dd2cc468",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "| Category | Details |\n| :--- | :--- |\n| **Full Name** | Thomas Karl Ring |\n| **Birth** | November 28, 1892, at 5:00 PM in Muggenhof (near Nuremb... |
aug-0:0000000087 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.697 | 0.858 | 16,384 | 262,377,052 | false | {
"id": "chatcmpl-bf5760e29e9e10a7",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Based on the provided document, here is the timeline of information regarding **Paležnica Gornja**:\n\n### Timeline of Events and Data for Paležnica... |
aug-0:0000000101 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.884 | 0.902 | 16,384 | 1,649,142,121 | false | {
"id": "chatcmpl-ab4b88f286717b98",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To apply the strategy of **using mnemonics** to the provided document, I will create associative phrases, acronyms, and memory hooks for the key dat... | |
aug-0:0000000084 | <|im_start|>system
You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input.<|im_end|>
<|im_start|>user
You are a unit-conversion reasoning system. The text below describes a procedure or dataset with physical quantities. Each qua... | null | transform_unit_conversion | transform | transform | You are a careful data-cleaning and templatization engine. You transform text precisely as instructed and never add facts not present in the input. | 0.938 | 0.927 | 16,384 | 79,358,612 | false | {
"id": "chatcmpl-8ce6de4d66515f98",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Step-by-Step Reasoning\n\n**1. Extraction of Quantities**\nI will scan the provided text for any explicit (value, unit) pairs representing physi... |
aug-0:0000000031 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.709 | 0.937 | 16,384 | 34,609,041 | false | {
"id": "chatcmpl-bad4725939c5b3c6",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef don_carlson_profile(\n # --- Core Identity ---\n name = \"Don M. Carlson\",\n birth_name = None, # Explicitly None or empty ... | |
aug-0:0000000129 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Your task is to add calls to a Calculator API to a piece of text. The calculations should help you get the information required to complete the text. You can call the API by writing "[Calcula... | null | agentic_toolformer_calc | agentic | agentic | You write educational dialogues. Do not invent facts outside the provided context. | 0.82 | 0.939 | 16,384 | 745,432,028 | false | {
"id": "chatcmpl-8265d49e31feb57b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "To apply the **7. Review and Repeat** strategy, let's break down the information into key points and associations:\n\n1. **Name:** Patricia Garfiel... |
aug-0:0000000056 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
I need to critique a document. Consider the document below. FIRST, generate a list of three questions that covers between them every piece of factual information in this document. SECOND, after generating all the que... | null | huu_questions | huu | general_dialogue | You are a careful, faithful synthetic-dialogue generator. | 0.658 | 0.912 | 16,384 | 851,905,404 | false | {
"id": "chatcmpl-ac3c15bd974e2384",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "### Part 1: Three Questions Covering All Factual Information\n\n1. **What are the specific geographical coordinates, location name, area size, and ... |
aug-0:0000000142 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Your task is to rewrite knowledge from the provided text following these instructions.
- Rewrite the text as a passage or pass... | null | nemotron_cc_extract_knowledge | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.94 | 0.978 | 16,384 | 2,137,253,781 | false | {
"id": "chatcmpl-a83682b530a25c7b",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The Office of the Commissioner for Public Appointments in Scotland was established by the Public Appointments and Public Bodies etc. (Scotland) Act ... |
aug-0:0000000100 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Your task is to add calls to a Question Answering API to a piece of text. The questions should help you get information required to complete the text. You can call the API by writing "[QA(quest... | null | agentic_toolformer_qa | agentic | agentic | You write concise, content-grounded conversations that stay close to the source. | 0.891 | 0.891 | 16,384 | 1,074,257,405 | false | {
"id": "chatcmpl-83013d02d5c20076",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Cornsay is a hamlet and civil parish located in County Durham, England. It is situated approximately 6 miles southeast of Consett and 7 miles southw... |
aug-0:0000000099 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.884 | 0.898 | 16,384 | 1,599,770,477 | false | {
"id": "chatcmpl-9dd584f69310e7ae",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef cornsay_document_parameters(\n location_entity='Cornsay',\n colliery_name='Cornsay Colliery',\n village_green_entity='villag... | |
aug-0:0000000132 | <|im_start|>system
You are an artificial intelligence assistant. You carefully provide accurate, factual, thoughtful, nuanced answers, and are brilliant at reasoning.<|im_end|>
<|im_start|>user
Your task is to read and paraphrase the provided text following these instructions:
- Aim to create a condensed but accurate a... | null | nemotron_cc_distill | nemotron_cc | general_rewrite | You are an artificial intelligence assistant. You carefully provide accurate, factual, thoughtful, nuanced answers, and are brilliant at reasoning. | 0.89 | 0.948 | 16,384 | 1,322,429,997 | false | {
"id": "chatcmpl-b508652870b11dc7",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Paraphrased Text:\nPatricia L. Garfield was a distinguished American academic and writer whose primary specialization was the study of dreams, with ... |
aug-0:0000000138 | <|im_start|>system
You write educational dialogues. Do not invent facts outside the provided context.<|im_end|>
<|im_start|>user
Task: Read the text, ask questions and answer them.
Follow these instructions:
1. Ask diverse questions that require different cognitive skills or cover different aspects of the text.
1. Ask ... | null | synthpile_diverse_qa | synthpile | general_rewrite | You write educational dialogues. Do not invent facts outside the provided context. | 0.703 | 0.966 | 16,384 | 508,175,522 | false | {
"id": "chatcmpl-b7cd63dbea335771",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are the questions and answers based on the provided text:\n- Question: Did the Socialist Propaganda League of America merge with other parties ... |
aug-0:0000000078 | <|im_start|>user
Consider the following document. First list three learning strategies specific to this document that I can use to help me learn and remember all of the common sense information contained. Then for EACH strategy, apply the strategy to the document in detail.
<document>
({{zh|t=壆圍}}) is a [[Walled villa... | null | huu_learning | huu | general_dialogue | 0.621 | 0.972 | 16,384 | 108,186,226 | false | {
"id": "chatcmpl-91389a6b0725f55d",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Based on the document provided about **Pok Wai**, here are three specific learning strategies tailored to its content, followed by a detailed applic... | |
aug-0:0000000141 | <|im_start|>system
You write concise, content-grounded conversations that stay close to the source.<|im_end|>
<|im_start|>user
Create a logical reasoning problem inspired by the relationships, classifications, or arguments in the document below. The problem must be reasonable and solvable, with a clear answer derivable... | null | persona_logic_problem | persona | reasoning | You write concise, content-grounded conversations that stay close to the source. | 0.711 | 0.903 | 16,384 | 1,374,933,307 | false | {
"id": "chatcmpl-8ae9b7217d3dc3a5",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Based on the document provided, here are the questions and their answers to help engage with the material:\n\n1. **What is the primary purpose of t... |
aug-0:0000000064 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Here is an extract from a webpage: "ate = {{birth date|1956|7|12}}
{{Infobox officeholder
| party = {{Plainlist|
| profession = Lawyer, politician
| office = [[Chamber of Deputies (Br... | null | cosmopedia_textbook_narrative | cosmopedia | education | You are a careful, faithful synthetic-dialogue generator. | 0.701 | 0.888 | 16,384 | 466,202,980 | false | {
"id": "chatcmpl-8a08ea04a53782bf",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The story of Júlio Redecker is not merely a biographical entry in a database; it is a stark, almost cinematic vignette of the volatility inherent in... |
aug-0:0000000145 | <|im_start|>user
Here is an extract from a webpage: "as first appointed as the Acting Commissioner in April 2021 and is now appointed as the Ethical Standards Commissioner from 1 March 2023. Parliament approved the appointment of Ian Bruce to take up the full-time post of Commissioner for a fixed term of 6 years. Ian B... | null | cosmopedia_blog | cosmopedia | education | 0.678 | 0.937 | 16,384 | 1,077,118,459 | false | {
"id": "chatcmpl-9f20a008d0156982",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The landscape of public integrity in Scotland has shifted significantly with the transition of the Commissioner for Ethical Standards in Public Life... | |
aug-0:0000000075 | <|im_start|>user
You are a text parameterization system.
Your task is to convert prose into a structured symbolic representation that preserves:
- quantities,
- ranges,
- relationships,
- dependencies,
- implied calculations,
- categorical structure,
- and normalized parameters.
DO NOT summarize.
DO NOT simplify.
DO ... | null | transform_parameterize_scientific | transform | transform | 0.694 | 0.948 | 16,384 | 1,875,424,679 | false | {
"id": "chatcmpl-aba1ff948beda1c6",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "```python\ndef document_structure_analyzer(\n location_name = 'Pok Wai',\n war_period = 'post World War II',\n structure_type = 'pillboxes ... | |
aug-0:0000000148 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
You are generating ReAct-style agent training traces. Given a document and an action API of three functions:
Search[query] -> returns a passage
Lookup[term] -> returns the next occurrence of `term` in the la... | null | agentic_doc_to_react | agentic | agentic | You are a careful, faithful synthetic-dialogue generator. | 0.617 | 0.896 | 16,384 | 1,714,539,223 | false | {
"id": "chatcmpl-9df92994d3424425",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Question: Which painter from the list was born in 1945 and shares a surname with a painter born in 1857?\n\nThought 1: I need to find a painter born... |
aug-0:0000000091 | <|im_start|>system
You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any... | null | agentic_codeact | agentic | agentic | You are a helpful assistant that solves tasks by writing and executing Python code. Emit a code action by wrapping Python in <execute> ... </execute>; after each one you receive the interpreter result as 'Observation:'. Code persists across turns (variables and imports carry over) and you may use any standard library o... | 0.842 | 0.857 | 16,384 | 1,135,838,315 | false | {
"id": "chatcmpl-b7e34ea2bd97d839",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "<execute>\n# Extract and validate the coordinates from the document to visualize the location\nimport re\n\ndocument_text = \"\"\"\nPaležnica Gornja... |
aug-0:0000000155 | <|im_start|>user
Task: Read the text, ask questions and answer them.
Follow these instructions:
1. Ask diverse questions that require different cognitive skills or cover different aspects of the text.
1. Ask questions in various forms such as:
- Yes/No questions that require determining whether a statement is true ... | null | synthpile_diverse_qa | synthpile | general_rewrite | 0.913 | 0.886 | 16,384 | 173,632,774 | false | {
"id": "chatcmpl-9f0680db98cbd815",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are the questions and answers based on the provided text:\n- Question: Is Haydar Hatemi a painter from Turkey? Answer: Yes\n- Question: What is... | |
aug-0:0000000106 | <|im_start|>system
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions.<|im_end|>
<|im_start|>user
Task:
Read the text, ask questions and answer them.
Follow these instructions:
1. Ask diverse questions that require differen... | null | nemotron_cc_diverse_qa | nemotron_cc | general_rewrite | A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the questions. | 0.783 | 0.924 | 16,384 | 1,953,385,123 | false | {
"id": "chatcmpl-96b8bb292bf99892",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Here are the questions and answers based on the provided text:\n- Question: Who designed both the Tyrolean Tubs and the Snails rides? Answer: Horace... |
aug-0:0000000157 | <|im_start|>system
You are a careful, faithful synthetic-dialogue generator.<|im_end|>
<|im_start|>user
Context:
<document>
rak Arıkan was influenced by contemporary art movements.
* Haluk Akakçe was influenced by modern and contemporary art.
* Taner Ceylan was influenced by contemporary art movements.
**Notable Works... | null | two_professors | mind | general_dialogue | You are a careful, faithful synthetic-dialogue generator. | 0.63 | 0.962 | 16,384 | 1,025,143,247 | false | {
"id": "chatcmpl-bea61445f81b9bf5",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "**Professor A:** I have been reviewing the list of painters from Turkey we compiled. It seems we have successfully organized them alphabetically, cr... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.