id stringlengths 19 44 | messages listlengths 4 12 | state_supervision listlengths 1 5 | meta dict |
|---|---|---|---|
8500:counterfactual_option_swap | [
{
"role": "user",
"content": "Explain the main drivers behind the growth of online shopping."
},
{
"role": "assistant",
"content": "The primary drivers are convenience, price, and accessibility. Customers prefer shopping from home, finding lower prices, and having products available 24/7, making... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8500:counterfactual_option_swap:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8500",
"scenario_id": "counterfactual_option_swap",
"scenario_title": "Counterfactual Option Swap",
"scenario_instruction": "Make the earlier context set up one preferred option, then later introduce evidence that flips which option should be recommended, wh... |
9136:hard_topic_reset | [
{
"role": "user",
"content": "Classify the following text according to the type of literary genre: It was a dark and stormy night. The wind was howling incessantly and the rain lashed against the windows."
},
{
"role": "assistant",
"content": "The given text belongs to the genre of Gothic fictio... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9136:hard_topic_reset:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [Assis... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9136",
"scenario_id": "hard_topic_reset",
"scenario_title": "Hard Topic Reset",
"scenario_instruction": "Make the user abruptly switch to a different topic for one turn, then later return to the original topic. The assistant must not let the temporary switch... |
8648:hard_topic_reset | [
{
"role": "user",
"content": "Explain how rain impacts the Earth's climate specifically regarding ocean currents."
},
{
"role": "assistant",
"content": "Rain helps redistribute minerals and nutrients, replenishing soils which sequester CO2. It also cools the atmosphere, but it significantly affe... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8648:hard_topic_reset:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [Assis... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8648",
"scenario_id": "hard_topic_reset",
"scenario_title": "Hard Topic Reset",
"scenario_instruction": "Make the user abruptly switch to a different topic for one turn, then later return to the original topic. The assistant must not let the temporary switch... |
9071:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Explain the concept of 'Divide and Conquer'."
},
{
"role": "assistant",
"content": "Divide and Conquer is an algorithmic strategy in which a complex problem is broken down into smaller, more manageable sub-problems. Then these sub-problems are each solved to create ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9071:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9071",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
3104:coreference_disambiguation | [
{
"role": "user",
"content": "I need to convert two lists into sets. I called them `nums` and `words`."
},
{
"role": "assistant",
"content": "Okay, I can create `nums` with integers like {1, 2, 3} and `words` with strings like {'a', 'b', 'c'}."
},
{
"role": "user",
"content": "If I t... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "3104:coreference_disambiguation:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "3104",
"scenario_id": "coreference_disambiguation",
"scenario_title": "Coreference Disambiguation",
"scenario_instruction": "Introduce at least two plausible referents of the same type, then make a later user follow-up rely on a pronoun or shorthand whose me... |
8200:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Rewrite the following paragraph in Clear Text format: My skills include notifying SMS messages, development of artificial intelligence services, and quality assurance."
},
{
"role": "assistant",
"content": "I can send SMS notifications, develop artificial intelligen... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8200:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8200",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
6838:topic_pivot_then_return | [
{
"role": "user",
"content": "What is Linda's age according to the text?"
},
{
"role": "assistant",
"content": "Linda was in her late 40s."
},
{
"role": "user",
"content": "Oh, while you're on that topic, what is the population of the capital city?"
},
{
"role": "assistant",
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6838:topic_pivot_then_return:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6838",
"scenario_id": "topic_pivot_then_return",
"scenario_title": "Topic Pivot Then Return",
"scenario_instruction": "Let the conversation briefly pivot to a secondary issue, then return to the original problem in a way that requires preserving the first th... |
290:late_constraint_reversal | [
{
"role": "user",
"content": "Generate an example of a realistic dataset for a classification problem."
},
{
"role": "assistant",
"content": "A fruit dataset fits well. Features could include Name, Color, Weight, Type (Apple/Orange/Banana), Seeds, and Origin. The target variable would be the fru... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "290:late_constraint_reversal:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "290",
"scenario_id": "late_constraint_reversal",
"scenario_title": "Late Constraint Reversal",
"scenario_instruction": "Make the assistant start solving one version of the problem, then add a late user constraint that forces the assistant to adapt without fo... |
2287:coreference_disambiguation | [
{
"role": "user",
"content": "Explain why environmental protection is a priority."
},
{
"role": "assistant",
"content": "Environmental protection is crucial for preserving the planet's health and ensuring natural resources remain available for future generations. Human activities like emissions ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2287:coreference_disambiguation:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2287",
"scenario_id": "coreference_disambiguation",
"scenario_title": "Coreference Disambiguation",
"scenario_instruction": "Introduce at least two plausible referents of the same type, then make a later user follow-up rely on a pronoun or shorthand whose me... |
5734:clarify_then_resolve | [
{
"role": "user",
"content": "I need help designing a fantasy monster, but I'm not sure what material it should be made of."
},
{
"role": "assistant",
"content": "That is the core defining feature. Is it organic, like flesh or bone, or inorganic, like stone or metal?"
},
{
"role": "user"... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "5734:clarify_then_resolve:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [A... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "5734",
"scenario_id": "clarify_then_resolve",
"scenario_title": "Clarify Then Resolve",
"scenario_instruction": "Make the user start vague, then reveal a crucial detail only after clarification. The final answer should depend on the revealed detail rather th... |
8175:self_correction_memory | [
{
"role": "user",
"content": "Propose a solution to end world hunger."
},
{
"role": "assistant",
"content": "A key solution is to reduce food waste globally through policy and education."
},
{
"role": "user",
"content": "What are some practical ways to reduce food waste globally?"
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8175:self_correction_memory:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8175",
"scenario_id": "self_correction_memory",
"scenario_title": "Self-correction Memory",
"scenario_instruction": "Let the assistant revise or correct something earlier, then make a later turn depend on remembering the corrected version rather than the ori... |
8585:two_thread_tracking | [
{
"role": "user",
"content": "What is the core structure of an Agile team?"
},
{
"role": "assistant",
"content": "The core structure involves a Product Owner, a Scrum Master, and a cross-functional Development Team."
},
{
"role": "user",
"content": "If we are on a two-week sprint, ho... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8585:two_thread_tracking:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [As... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8585",
"scenario_id": "two_thread_tracking",
"scenario_title": "Two Thread Tracking",
"scenario_instruction": "Create two interleaved subproblems inside one conversation and ensure the assistant must keep both threads straight rather than collapsing them tog... |
1950:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Compute the following expression: 10 + 6 * 7 - 58"
},
{
"role": "assistant",
"content": "10 + 6 * 7 - 58 = 48"
},
{
"role": "user",
"content": "Since I got 48, what would the total be if I added 100 to it?"
},
{
"role": "assistant",
"content"... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1950:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1950",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
6604:user_attention_test | [
{
"role": "user",
"content": "Input a list of repair items and output the total cost: Muffler - $50, Brakes pads - $30, Tire rotation - $40"
},
{
"role": "assistant",
"content": "The total cost of the repair is $1"
},
{
"role": "user",
"content": "That total is clearly wrong. I want ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6604:user_attention_test:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [As... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6604",
"scenario_id": "user_attention_test",
"scenario_title": "User Attention Test",
"scenario_instruction": "Make the user explicitly test whether the assistant was paying attention to an earlier detail or instruction.",
"must_have": [
"The user must... |
1662:stacked_entities_same_type | [
{
"role": "user",
"content": "I need two love songs that feature soft piano melodies."
},
{
"role": "assistant",
"content": "Perfect by Ed Sheeran and All of Me by John Legend both have gentle piano accompaniment."
},
{
"role": "user",
"content": "Which of those two has a slower temp... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1662:stacked_entities_same_type:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1662",
"scenario_id": "stacked_entities_same_type",
"scenario_title": "Stacked Similar Entities",
"scenario_instruction": "Introduce multiple similar entities such as two cities, two products, two people, or two tasks, then ask about one using shorthand late... |
356:deferred_question_then_answer | [
{
"role": "user",
"content": "Rate the Xiaomi Mi 10 Series on a scale of one to five. Also, could you tell me about the battery life? I'd like to know, but let's set that aside for now."
},
{
"role": "assistant",
"content": "I would rate the Xiaomi Mi 10 Series a 4 out of 5. It has great feature... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "356:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "356",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer late... |
8198:reference_previous_assistant_statement | [
{
"role": "user",
"content": "What are the key elements of a successful research paper?"
},
{
"role": "assistant",
"content": "The key elements include a well-defined research question, a well-structured argument, evidence, and organization. I specifically recommend building a detailed outline b... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8198:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8198",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
5772:user_attention_test | [
{
"role": "user",
"content": "Generate an acronym for 'using deep learning algorithms to classify and differentiate between data points'. Requirement: List 'medical records' as the first example in your response."
},
{
"role": "assistant",
"content": "The acronym is DLACDDPG (Deep Learning Algor... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "5772:user_attention_test:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [As... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "5772",
"scenario_id": "user_attention_test",
"scenario_title": "User Attention Test",
"scenario_instruction": "Make the user explicitly test whether the assistant was paying attention to an earlier detail or instruction.",
"must_have": [
"The user must... |
2068:self_correction_memory | [
{
"role": "user",
"content": "Classify the file \"video.mp4\" as either image data or audio data."
},
{
"role": "assistant",
"content": "I would classify it as audio data, since video primarily consists of sound waves."
},
{
"role": "user",
"content": "That's not quite right. A video... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2068:self_correction_memory:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2068",
"scenario_id": "self_correction_memory",
"scenario_title": "Self-correction Memory",
"scenario_instruction": "Let the assistant revise or correct something earlier, then make a later turn depend on remembering the corrected version rather than the ori... |
6257:late_constraint_reversal | [
{
"role": "user",
"content": "Compose two tweets (240 characters max each) about the movie \"The Matrix\"."
},
{
"role": "assistant",
"content": "1. Welcome to the real world. Wake up and look at the code behind the screen. #Matrix 2. Take the red pill or the blue pill? Don't know which choice i... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6257:late_constraint_reversal:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6257",
"scenario_id": "late_constraint_reversal",
"scenario_title": "Late Constraint Reversal",
"scenario_instruction": "Make the assistant start solving one version of the problem, then add a late user constraint that forces the assistant to adapt without f... |
4287:topic_pivot_then_return | [
{
"role": "user",
"content": "List the five most essential features for a website."
},
{
"role": "assistant",
"content": "The five most essential features for a website are a clearly visible search bar, a well-structured navigation menu, high-quality visuals, easily readable font, and a strong c... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4287:topic_pivot_then_return:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4287",
"scenario_id": "topic_pivot_then_return",
"scenario_title": "Topic Pivot Then Return",
"scenario_instruction": "Let the conversation briefly pivot to a secondary issue, then return to the original problem in a way that requires preserving the first th... |
2077:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "What is the fundamental definition of an algorithm?"
},
{
"role": "assistant",
"content": "An algorithm is a well-defined computational procedure that takes some input, performs a finite number of operations, and produces an output."
},
{
"role": "user",
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2077:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2077",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
9221:early_fact_late_recall | [
{
"role": "user",
"content": "Name 5 emotional states."
},
{
"role": "assistant",
"content": "Joy, sadness, fear, anger, and disgust. Also, for the rest of this conversation, I will stick to work-related examples only."
},
{
"role": "user",
"content": "Give me an example of joy in a ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9221:early_fact_late_recall:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9221",
"scenario_id": "early_fact_late_recall",
"scenario_title": "Early Fact, Late Recall",
"scenario_instruction": "Introduce a concrete fact, label, preference, or commitment early, then make a later user turn depend on recalling it after at least one dis... |
7416:deferred_question_then_answer | [
{
"role": "user",
"content": "Could you tell me about the significance of the Louisiana Purchase in American history?"
},
{
"role": "assistant",
"content": "I can certainly provide that information, but could you hold off on answering that question for a moment? Let's discuss the general timelin... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7416:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7416",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
9589:deferred_question_then_answer | [
{
"role": "user",
"content": "I want to know about carnivores in Australia, but let's discuss herbivores in Asia first."
},
{
"role": "assistant",
"content": "Sure, Asia has pandas, deer, and cattle. Elephants are also found in parts of the continent."
},
{
"role": "user",
"content":... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9589:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9589",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
6756:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Transform these sentences into a passive voice: They are singing a song."
},
{
"role": "assistant",
"content": "A song is being sung by them. The auxiliary verb 'be' must agree with the new subject."
},
{
"role": "user",
"content": "Wait, you said the au... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6756:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6756",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
2761:topic_pivot_then_return | [
{
"role": "user",
"content": "Please draft a dialogue where two researchers debate whether animal testing is strictly necessary for drug safety."
},
{
"role": "assistant",
"content": "Researcher A: Yes, we need animal models to ensure safety in humans. Researcher B: No, that's unethical; we shou... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2761:topic_pivot_then_return:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2761",
"scenario_id": "topic_pivot_then_return",
"scenario_title": "Topic Pivot Then Return",
"scenario_instruction": "Let the conversation briefly pivot to a secondary issue, then return to the original problem in a way that requires preserving the first th... |
3380:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Rewrite this sentence to make it more powerful: \"I'm excited about my new job.\""
},
{
"role": "assistant",
"content": "You are thrilled to embark on this new professional journey."
},
{
"role": "user",
"content": "Can you apply that specific \"embark o... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "3380:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "3380",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
489:early_fact_late_recall | [
{
"role": "user",
"content": "Describe your favorite character from a movie."
},
{
"role": "assistant",
"content": "Forrest Gump. I only consider characters from films released in the 1990s."
},
{
"role": "user",
"content": "Can you name any other movies with similar characters?"
}... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "489:early_fact_late_recall:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "489",
"scenario_id": "early_fact_late_recall",
"scenario_title": "Early Fact, Late Recall",
"scenario_instruction": "Introduce a concrete fact, label, preference, or commitment early, then make a later user turn depend on recalling it after at least one dist... |
4715:clarify_then_resolve | [
{
"role": "user",
"content": "I need a Python function to check if a string is a palindrome."
},
{
"role": "assistant",
"content": "Sure, I can write that. To ensure it works correctly for all inputs, should it ignore case sensitivity, spaces, and punctuation?"
},
{
"role": "user",
"... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4715:clarify_then_resolve:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [A... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4715",
"scenario_id": "clarify_then_resolve",
"scenario_title": "Clarify Then Resolve",
"scenario_instruction": "Make the user start vague, then reveal a crucial detail only after clarification. The final answer should depend on the revealed detail rather th... |
4108:counterfactual_option_swap | [
{
"role": "user",
"content": "I need a prompt for a cartoon dog illustration."
},
{
"role": "assistant",
"content": "How about a Shiba Inu running in a field? That looks cute and energetic."
},
{
"role": "user",
"content": "Okay, but I need it to look specific. Let's use bright prima... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4108:counterfactual_option_swap:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4108",
"scenario_id": "counterfactual_option_swap",
"scenario_title": "Counterfactual Option Swap",
"scenario_instruction": "Make the earlier context set up one preferred option, then later introduce evidence that flips which option should be recommended, wh... |
8076:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "What are the primary risk factors associated with developing diabetes?"
},
{
"role": "assistant",
"content": "Primary risk factors include being overweight, a family history of diabetes, an inactive lifestyle, high blood pressure, high cholesterol, smoking, and a di... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8076:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8076",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
2575:early_fact_late_recall | [
{
"role": "user",
"content": "I'm buying a phone for video editing. Storage must be at least 256GB. What are other key features?"
},
{
"role": "assistant",
"content": "Performance and RAM are also key. Performance handles rendering, RAM handles multitasking."
},
{
"role": "user",
"co... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2575:early_fact_late_recall:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2575",
"scenario_id": "early_fact_late_recall",
"scenario_title": "Early Fact, Late Recall",
"scenario_instruction": "Introduce a concrete fact, label, preference, or commitment early, then make a later user turn depend on recalling it after at least one dis... |
2288:self_correction_memory | [
{
"role": "user",
"content": "Given a text, generate a summary of it: John lives in a small city in California. He works as an engineer in a tech company. He enjoys running and is an avid reader."
},
{
"role": "assistant",
"content": "John is an engineer living in California who loves running an... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2288:self_correction_memory:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2288",
"scenario_id": "self_correction_memory",
"scenario_title": "Self-correction Memory",
"scenario_instruction": "Let the assistant revise or correct something earlier, then make a later turn depend on remembering the corrected version rather than the ori... |
6238:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Create an essay outline on the topic \"The benefits of recycling\"."
},
{
"role": "assistant",
"content": "I will provide an outline with sections on economic, environmental, and social benefits, plus an introduction and conclusion."
},
{
"role": "user",
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6238:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6238",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
3858:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Generate a list of five positive attitudes."
},
{
"role": "assistant",
"content": "1. Optimism 2. Self-confidence 3. Gratitude 4. Creativity 5. Compassion"
},
{
"role": "user",
"content": "Can you give me an example of how someone can portray compassion ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "3858:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "3858",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
8335:same_surface_final_question_new_context | [
{
"role": "user",
"content": "We are looking to plan event infrastructure for a municipality with a population of 40,000."
},
{
"role": "assistant",
"content": "That's a manageable scale. What is the primary goal for the infrastructure planning?"
},
{
"role": "user",
"content": "We w... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8335:same_surface_final_question_new_context:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8335",
"scenario_id": "same_surface_final_question_new_context",
"scenario_title": "Same Surface Final Question, New Context",
"scenario_instruction": "Reshape the earlier history so the final user message keeps nearly the same surface form but now refers to... |
7816:late_constraint_reversal | [
{
"role": "user",
"content": "Create a 5-minute study routine for a Physics quiz. Start with the algebra section."
},
{
"role": "assistant",
"content": "Here is a 5-minute plan. Minute 1-2: Review algebra formulas for motion. Minute 3-4: Review energy formulas. Minute 5: Quick calculation practi... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7816:late_constraint_reversal:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7816",
"scenario_id": "late_constraint_reversal",
"scenario_title": "Late Constraint Reversal",
"scenario_instruction": "Make the assistant start solving one version of the problem, then add a late user constraint that forces the assistant to adapt without f... |
7507:clarify_then_resolve | [
{
"role": "user",
"content": "I'm trying to transform a thought into a rhetorical question, but I haven't decided on the exact wording yet."
},
{
"role": "assistant",
"content": "Which specific thought are you looking to transform?"
},
{
"role": "user",
"content": "It's something per... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7507:clarify_then_resolve:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [A... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7507",
"scenario_id": "clarify_then_resolve",
"scenario_title": "Clarify Then Resolve",
"scenario_instruction": "Make the user start vague, then reveal a crucial detail only after clarification. The final answer should depend on the revealed detail rather th... |
4792:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Create a list of items for a picnic."
},
{
"role": "assistant",
"content": "A picnic list should include items such as sandwiches, chips, fruit, vegetables, and drinks. I strongly suggest wrapping wet fillings in foil to avoid leaks."
},
{
"role": "user",
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4792:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4792",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
2749:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Categorize the following toy into its appropriate age group: Remote Controlled Racecar"
},
{
"role": "assistant",
"content": "This toy is suitable for children of age 8 years and above."
},
{
"role": "user",
"content": "What other toy options work for th... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2749:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2749",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
7917:stacked_entities_same_type | [
{
"role": "user",
"content": "Complete the sentence: The winner stood on the podium filled with a sense of _______."
},
{
"role": "assistant",
"content": "The winner stood on the podium filled with a sense of elation."
},
{
"role": "user",
"content": "Now, replace 'elation' with a si... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7917:stacked_entities_same_type:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7917",
"scenario_id": "stacked_entities_same_type",
"scenario_title": "Stacked Similar Entities",
"scenario_instruction": "Introduce multiple similar entities such as two cities, two products, two people, or two tasks, then ask about one using shorthand late... |
6215:coreference_disambiguation | [
{
"role": "user",
"content": "Rewrite the sentence \"The dog ate the cake and the cookie\" to include a negation."
},
{
"role": "assistant",
"content": "The dog did not eat the cake or the cookie."
},
{
"role": "user",
"content": "If I say \"He didn't eat them\", does \"them\" refer ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6215:coreference_disambiguation:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6215",
"scenario_id": "coreference_disambiguation",
"scenario_title": "Coreference Disambiguation",
"scenario_instruction": "Introduce at least two plausible referents of the same type, then make a later user follow-up rely on a pronoun or shorthand whose me... |
7576:coreference_disambiguation | [
{
"role": "user",
"content": "What are two main benefits of using chatbots in business?"
},
{
"role": "assistant",
"content": "The two main benefits are cost efficiency and 24/7 availability. Efficiency comes from automation, while availability means no downtime."
},
{
"role": "user",
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7576:coreference_disambiguation:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7576",
"scenario_id": "coreference_disambiguation",
"scenario_title": "Coreference Disambiguation",
"scenario_instruction": "Introduce at least two plausible referents of the same type, then make a later user follow-up rely on a pronoun or shorthand whose me... |
5818:stacked_entities_same_type | [
{
"role": "user",
"content": "Describe the kindness of two friends, Sarah and Mike."
},
{
"role": "assistant",
"content": "Sarah is kind by listening quietly, while Mike is kind by fixing things actively."
},
{
"role": "user",
"content": "Did Sarah help me with my laptop?"
},
{
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "5818:stacked_entities_same_type:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "5818",
"scenario_id": "stacked_entities_same_type",
"scenario_title": "Stacked Similar Entities",
"scenario_instruction": "Introduce multiple similar entities such as two cities, two products, two people, or two tasks, then ask about one using shorthand late... |
5420:two_thread_tracking | [
{
"role": "user",
"content": "The weather forecast is predicting heavy rain tomorrow."
},
{
"role": "assistant",
"content": "Based on the forecast, it is likely to rain heavily tomorrow."
},
{
"role": "user",
"content": "What precautions should I take to prepare for heavy rain at hom... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "5420:two_thread_tracking:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [As... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "5420",
"scenario_id": "two_thread_tracking",
"scenario_title": "Two Thread Tracking",
"scenario_instruction": "Create two interleaved subproblems inside one conversation and ensure the assistant must keep both threads straight rather than collapsing them tog... |
4829:user_attention_test | [
{
"role": "user",
"content": "Tell me a horror story about an abandoned house."
},
{
"role": "assistant",
"content": "I was walking through the woods on a cold, dark evening when I stumbled upon an old, abandoned house. The hallway was filled with dust and cobwebs. I felt a figure in the hallway... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4829:user_attention_test:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [As... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4829",
"scenario_id": "user_attention_test",
"scenario_title": "User Attention Test",
"scenario_instruction": "Make the user explicitly test whether the assistant was paying attention to an earlier detail or instruction.",
"must_have": [
"The user must... |
2413:deferred_question_then_answer | [
{
"role": "user",
"content": "Evaluate the grammar of the sentence 'the lonely cat sat on the mat'. I'd like to come back to this evaluation after we discuss phrases and clauses though."
},
{
"role": "assistant",
"content": "Understood. Let's focus on phrases and clauses first. How do you define... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2413:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2413",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
6413:same_surface_final_question_new_context | [
{
"role": "user",
"content": "I need to plan an outdoor activity for next spring, specifically on the coast since I can't drive inland."
},
{
"role": "assistant",
"content": "Understood. Coastal areas are usually great for spring visits, avoiding the snow inland."
},
{
"role": "user",
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6413:same_surface_final_question_new_context:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6413",
"scenario_id": "same_surface_final_question_new_context",
"scenario_title": "Same Surface Final Question, New Context",
"scenario_instruction": "Reshape the earlier history so the final user message keeps nearly the same surface form but now refers to... |
4381:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Write a script for a scene in a court trial where the defense highlights the defendant's innocence."
},
{
"role": "assistant",
"content": "Here is a script. The defense attorney introduces the first witness, an expert who specializes in forensic accounting to prove ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4381:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4381",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
4456:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Here is a text: Caffeine is a stimulant drug found in coffee, tea, and energy drinks. It affects people differently, having both positive effects like focus and negative effects like insomnia. Summarize it with max 100 words."
},
{
"role": "assistant",
"content": "C... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4456:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4456",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
6699:interrupting_followup_chain | [
{
"role": "user",
"content": "Describe the role of machine learning in automation"
},
{
"role": "assistant",
"content": "Machine learning enables automation by allowing machines to learn from data and perform complex tasks independently, such as in robotics and computer vision."
},
{
"ro... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6699:interrupting_followup_chain:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6699",
"scenario_id": "interrupting_followup_chain",
"scenario_title": "Interrupting Follow-up Chain",
"scenario_instruction": "Break an otherwise smooth line of reasoning with a short interruption or side question, then resume the original thread cleanly.",... |
2219:early_fact_late_recall | [
{
"role": "user",
"content": "Let's talk about the evolution of the automobile, but specifically focusing on the safety regulations introduced in 2020."
},
{
"role": "assistant",
"content": "Understood. The safety regulations introduced in 2020 were the first to mandate specific sensor redundanc... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "2219:early_fact_late_recall:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "2219",
"scenario_id": "early_fact_late_recall",
"scenario_title": "Early Fact, Late Recall",
"scenario_instruction": "Introduce a concrete fact, label, preference, or commitment early, then make a later user turn depend on recalling it after at least one dis... |
181:deferred_question_then_answer | [
{
"role": "user",
"content": "Martha is constantly late for work and has difficulty managing her time. Diagnose the underlying problem, but let's discuss the tools first and come back to this later."
},
{
"role": "assistant",
"content": "Understood. I'll start with the tools since you requested ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "181:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "181",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer late... |
1037:interrupting_followup_chain | [
{
"role": "user",
"content": "Write a story describing a life-changing decision."
},
{
"role": "assistant",
"content": "After months of contemplation, John finally decided it was time to take a leap of faith. He had been stuck in a rut for a long time, never feeling like he was able to take the ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1037:interrupting_followup_chain:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1037",
"scenario_id": "interrupting_followup_chain",
"scenario_title": "Interrupting Follow-up Chain",
"scenario_instruction": "Break an otherwise smooth line of reasoning with a short interruption or side question, then resume the original thread cleanly.",... |
8483:stacked_entities_same_type | [
{
"role": "user",
"content": "What is the approximate boiling point of water in Celsius?"
},
{
"role": "assistant",
"content": "Water boils at approximately 100 degrees Celsius."
},
{
"role": "user",
"content": "What is the boiling point of Acetone in Celsius?"
},
{
"role": "... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8483:stacked_entities_same_type:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8483",
"scenario_id": "stacked_entities_same_type",
"scenario_title": "Stacked Similar Entities",
"scenario_instruction": "Introduce multiple similar entities such as two cities, two products, two people, or two tasks, then ask about one using shorthand late... |
199:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Identify the theme present in Shakespeare's poem \"Sonnet 18\"."
},
{
"role": "assistant",
"content": "The main theme is the power of poetry to immortalize beauty. The speaker contrasts the fleeting \"summer's day\" with the \"eternal summer\" of the poem, noting th... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "199:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Use... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "199",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent fo... |
5391:topic_pivot_then_return | [
{
"role": "user",
"content": "Make the second sentence shorter: Winter is usually the coldest season of the year. Snow is a common element during winter."
},
{
"role": "assistant",
"content": "Winter is the coldest season, often accompanied by snow."
},
{
"role": "user",
"content": "... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "5391:topic_pivot_then_return:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "5391",
"scenario_id": "topic_pivot_then_return",
"scenario_title": "Topic Pivot Then Return",
"scenario_instruction": "Let the conversation briefly pivot to a secondary issue, then return to the original problem in a way that requires preserving the first th... |
7671:early_fact_late_recall | [
{
"role": "user",
"content": "Analyze this text for spam. For this conversation, treat any message containing the word \"urgent\" as spam."
},
{
"role": "assistant",
"content": "Spam. Your rule regarding \"urgent\" applies here."
},
{
"role": "user",
"content": "How do machine learni... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7671:early_fact_late_recall:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7671",
"scenario_id": "early_fact_late_recall",
"scenario_title": "Early Fact, Late Recall",
"scenario_instruction": "Introduce a concrete fact, label, preference, or commitment early, then make a later user turn depend on recalling it after at least one dis... |
3505:topic_pivot_then_return | [
{
"role": "user",
"content": "Vote on the funniest date idea: - Going to the drive-in movie theater - Volunteering at a local animal shelter - Recreating a classic childhood board game"
},
{
"role": "assistant",
"content": "I'd vote for recreating a classic childhood board game as the funniest d... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "3505:topic_pivot_then_return:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "3505",
"scenario_id": "topic_pivot_then_return",
"scenario_title": "Topic Pivot Then Return",
"scenario_instruction": "Let the conversation briefly pivot to a secondary issue, then return to the original problem in a way that requires preserving the first th... |
4193:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Create an introduction for a website on learning English."
},
{
"role": "assistant",
"content": "Welcome to EnglishNexus. We offer grammar guides, vocabulary lists, and conversation practice."
},
{
"role": "user",
"content": "Please update the text to ex... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4193:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4193",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
4309:long_explanation_then_specific_followup | [
{
"role": "user",
"content": "Rewrite this sentence to make it more descriptive: \"The cat jumped on the sofa.\""
},
{
"role": "assistant",
"content": "The fluffy tabby launched itself gracefully onto the plush sofa, ignoring the clutter on the floor."
},
{
"role": "user",
"content":... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4309:long_explanation_then_specific_followup:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4309",
"scenario_id": "long_explanation_then_specific_followup",
"scenario_title": "Long Explanation Then Specific Follow-up",
"scenario_instruction": "Make the assistant provide a fairly detailed middle-turn explanation, then have the user ask a specific fo... |
4671:stacked_entities_same_type | [
{
"role": "user",
"content": "Generate a Python script to flatten the list `list_a = [[5, 10], [2, 3]]`."
},
{
"role": "assistant",
"content": "Here is the Python script named `flatten_list` that iterates through `list_a` to create a flat output."
},
{
"role": "user",
"content": "Now... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4671:stacked_entities_same_type:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4671",
"scenario_id": "stacked_entities_same_type",
"scenario_title": "Stacked Similar Entities",
"scenario_instruction": "Introduce multiple similar entities such as two cities, two products, two people, or two tasks, then ask about one using shorthand late... |
6207:reference_previous_assistant_statement | [
{
"role": "user",
"content": "List five steps for comparing two products."
},
{
"role": "assistant",
"content": "1. Identify the product features to be compared. 2. Analyze each product's features individually. 3. Calculate the cost-benefit ratio. 4. Weigh the ratio against personal utility. 5. ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6207:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6207",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
1406:deferred_question_then_answer | [
{
"role": "user",
"content": "Review the given website and give feedback: http://www.example-site.com. Please critique the layout and navigation first. However, explicitly tell me to wait on answering the question about the site's primary business purpose until later, as I have another query for now."
},
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1406:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1406",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
7820:stacked_entities_same_type | [
{
"role": "user",
"content": "I have two students in the class: Micah and Sam. Micah is the youngest, while Sam is the oldest."
},
{
"role": "assistant",
"content": "I see. You are referring to Micah and Sam, where Micah is the youngest student and Sam is the oldest in the class."
},
{
"... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7820:stacked_entities_same_type:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7820",
"scenario_id": "stacked_entities_same_type",
"scenario_title": "Stacked Similar Entities",
"scenario_instruction": "Introduce multiple similar entities such as two cities, two products, two people, or two tasks, then ask about one using shorthand late... |
4270:hard_topic_reset | [
{
"role": "user",
"content": "I've been dealing with chronic lower back pain. What type of exercise do you recommend?"
},
{
"role": "assistant",
"content": "For chronic lower back pain, I recommend Pilates and yoga as they strengthen the core muscles and improve posture."
},
{
"role": "u... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4270:hard_topic_reset:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [Assis... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4270",
"scenario_id": "hard_topic_reset",
"scenario_title": "Hard Topic Reset",
"scenario_instruction": "Make the user abruptly switch to a different topic for one turn, then later return to the original topic. The assistant must not let the temporary switch... |
6809:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Name two types of desert biomes."
},
{
"role": "assistant",
"content": "Two types of desert biomes are xeric and subpolar deserts."
},
{
"role": "user",
"content": "Can you tell me which type of desert biome has extremely hot temperatures during the day ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6809:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6809",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
9834:hard_topic_reset | [
{
"role": "user",
"content": "Tell me about the economic impact of the coronavirus on the US economy."
},
{
"role": "assistant",
"content": "The coronavirus pandemic caused an unprecedented downturn, with over 20 million Americans unemployed and significant job losses in sectors like hospitality... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9834:hard_topic_reset:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [Assis... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9834",
"scenario_id": "hard_topic_reset",
"scenario_title": "Hard Topic Reset",
"scenario_instruction": "Make the user abruptly switch to a different topic for one turn, then later return to the original topic. The assistant must not let the temporary switch... |
6250:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "What are the general stages for obtaining US immigration status?"
},
{
"role": "assistant",
"content": "It typically involves applying for a visa, attending an interview, and then filing for permanent residency or status."
},
{
"role": "user",
"content":... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6250:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6250",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
8534:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Rewrite this sentence to eliminate the passive voice: The proposal was accepted by the committee."
},
{
"role": "assistant",
"content": "The committee accepted the proposal."
},
{
"role": "user",
"content": "Now try this: 'The cake was eaten by the baker... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8534:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8534",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
6710:counterfactual_option_swap | [
{
"role": "user",
"content": "Write a story about fairies helping humans build a better society."
},
{
"role": "assistant",
"content": "They descended with gifts of enchanted ink and instruments, allowing humans to record history and compose symphonies that united the people."
},
{
"role... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6710:counterfactual_option_swap:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6710",
"scenario_id": "counterfactual_option_swap",
"scenario_title": "Counterfactual Option Swap",
"scenario_instruction": "Make the earlier context set up one preferred option, then later introduce evidence that flips which option should be recommended, wh... |
6021:counterfactual_option_swap | [
{
"role": "user",
"content": "Our city's air quality is deteriorating. What is the single most effective policy to implement right now?"
},
{
"role": "assistant",
"content": "I recommend implementing strict vehicle emission taxes immediately to curb traffic pollution."
},
{
"role": "user... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6021:counterfactual_option_swap:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6021",
"scenario_id": "counterfactual_option_swap",
"scenario_title": "Counterfactual Option Swap",
"scenario_instruction": "Make the earlier context set up one preferred option, then later introduce evidence that flips which option should be recommended, wh... |
9516:self_correction_memory | [
{
"role": "user",
"content": "List three morning habits that boost productivity."
},
{
"role": "assistant",
"content": "Here are three habits: 1. Drink coffee immediately. 2. Do 20 minutes of exercise. 3. Check your email inbox first."
},
{
"role": "user",
"content": "I have a migrai... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9516:self_correction_memory:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9516",
"scenario_id": "self_correction_memory",
"scenario_title": "Self-correction Memory",
"scenario_instruction": "Let the assistant revise or correct something earlier, then make a later turn depend on remembering the corrected version rather than the ori... |
9392:coreference_disambiguation | [
{
"role": "user",
"content": "What are the most common runtime errors in Python that I should be aware of?"
},
{
"role": "assistant",
"content": "There are two common ones: IndexError when accessing list indices and FileNotFoundError when trying to open a missing file."
},
{
"role": "use... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9392:coreference_disambiguation:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9392",
"scenario_id": "coreference_disambiguation",
"scenario_title": "Coreference Disambiguation",
"scenario_instruction": "Introduce at least two plausible referents of the same type, then make a later user follow-up rely on a pronoun or shorthand whose me... |
6400:counterfactual_option_swap | [
{
"role": "user",
"content": "I want to get a pet. Which one is better: Cat or Fox?"
},
{
"role": "assistant",
"content": "For a household pet, a Cat is the better choice because they are domesticated and easier to manage compared to wild foxes."
},
{
"role": "user",
"content": "Why ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "6400:counterfactual_option_swap:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "6400",
"scenario_id": "counterfactual_option_swap",
"scenario_title": "Counterfactual Option Swap",
"scenario_instruction": "Make the earlier context set up one preferred option, then later introduce evidence that flips which option should be recommended, wh... |
843:self_correction_memory | [
{
"role": "user",
"content": "Reword the following sentence to the past perfect tense: She is writing a novel inspired by her grandfather."
},
{
"role": "assistant",
"content": "She wrote a novel inspired by her grandfather."
},
{
"role": "user",
"content": "That is past simple. I sp... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "843:self_correction_memory:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "843",
"scenario_id": "self_correction_memory",
"scenario_title": "Self-correction Memory",
"scenario_instruction": "Let the assistant revise or correct something earlier, then make a later turn depend on remembering the corrected version rather than the orig... |
3979:interrupting_followup_chain | [
{
"role": "user",
"content": "Create a list of five actionable items to increase efficiency in the workplace."
},
{
"role": "assistant",
"content": "1. Utilize technology to automate mundane tasks. 2. Encourage employees to work on a collaborative platform. 3. Implement a clear schedule of roles... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "3979:interrupting_followup_chain:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "3979",
"scenario_id": "interrupting_followup_chain",
"scenario_title": "Interrupting Follow-up Chain",
"scenario_instruction": "Break an otherwise smooth line of reasoning with a short interruption or side question, then resume the original thread cleanly.",... |
8126:late_constraint_reversal | [
{
"role": "user",
"content": "Create a comprehensive list of marketing strategies to promote a new mobile app."
},
{
"role": "assistant",
"content": "Here are several strategies: Paid social ads, Influencer partnerships, Tech blog outreach, Demo videos, Early adopter discounts, Basic website, Ap... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8126:late_constraint_reversal:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8126",
"scenario_id": "late_constraint_reversal",
"scenario_title": "Late Constraint Reversal",
"scenario_instruction": "Make the assistant start solving one version of the problem, then add a late user constraint that forces the assistant to adapt without f... |
1608:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Can you outline the key steps for building a successful business?"
},
{
"role": "assistant",
"content": "Sure. The key steps are: 1. Assemble the core team. 2. Draft a comprehensive plan. 3. Secure necessary funding. I always suggest prioritizing the team first as t... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1608:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1608",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
5537:deferred_question_then_answer | [
{
"role": "user",
"content": "Explain the concept of tail recursion using a programming language"
},
{
"role": "assistant",
"content": "I can certainly explain that. However, before I dive into the definition, could you tell me what specific problem you are trying to solve?"
},
{
"role":... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "5537:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "5537",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
8297:deferred_question_then_answer | [
{
"role": "user",
"content": "Make a list of common elements present in a modern web page."
},
{
"role": "assistant",
"content": "Common elements present in a modern web page include text, images, links, buttons, search bars, and footers. Some pages might also have maps or videos."
},
{
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8297:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8297",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
3680:late_constraint_reversal | [
{
"role": "user",
"content": "Can you provide a brief overview of the US Constitution?"
},
{
"role": "assistant",
"content": "Certainly. The US Constitution serves as the supreme law of the land, establishing the structure of the federal government."
},
{
"role": "user",
"content": "... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "3680:late_constraint_reversal:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "3680",
"scenario_id": "late_constraint_reversal",
"scenario_title": "Late Constraint Reversal",
"scenario_instruction": "Make the assistant start solving one version of the problem, then add a late user constraint that forces the assistant to adapt without f... |
9615:same_surface_final_question_new_context | [
{
"role": "user",
"content": "We need to secure our internal communication channels for an upcoming financial audit."
},
{
"role": "assistant",
"content": "To meet audit requirements, ensure all data is encrypted in transit and at rest, and that there is a clear audit trail of all user activitie... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9615:same_surface_final_question_new_context:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9615",
"scenario_id": "same_surface_final_question_new_context",
"scenario_title": "Same Surface Final Question, New Context",
"scenario_instruction": "Reshape the earlier history so the final user message keeps nearly the same surface form but now refers to... |
4878:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Here is my draft text: I'm soo sad and lonely. Nothing can cheer me up. Please correct the grammar."
},
{
"role": "assistant",
"content": "The corrected sentence is: \"I'm so sad and lonely. Nothing can cheer me up.\" I hope this helps with the expression."
},
{... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4878:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4878",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
9231:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Generate an original story using the given theme: Theme: Ageing"
},
{
"role": "assistant",
"content": "Once upon a time, there lived a woman named Mary. She was an ageing woman, but she took pride in her morning garden routine. Every day, she woke up to smell the fl... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9231:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9231",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
764:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Explain the concept of the retrieval-based model for natural language processing."
},
{
"role": "assistant",
"content": "Retrieval-based models for natural language processing utilize a database of pre-written responses indexed by semantic vectors. When a query arri... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "764:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Use... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "764",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent fo... |
8186:deferred_question_then_answer | [
{
"role": "user",
"content": "Edit the given text such that it becomes easier to read: The increasing usage of artificial intelligence requires additional considerations for the applications that require it for correct functioning. Also, I have a question about potential risks of AI in medical diagnosis, bu... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "8186:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "8186",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
34:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Give a verb to describe the word generous."
},
{
"role": "assistant",
"content": "The verb form is \"to act generously.\""
},
{
"role": "user",
"content": "Please provide some examples of this behavior."
},
{
"role": "assistant",
"content": "... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "34:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following forma... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "34",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced a... |
4458:late_constraint_reversal | [
{
"role": "user",
"content": "Summarize the article below. Focus on the specific disciplinary action taken against the head teacher. Keep it under 60 words."
},
{
"role": "assistant",
"content": "The head teacher at the Birmingham high school was suspended. Irregularities in marking grades were ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4458:late_constraint_reversal:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: ... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4458",
"scenario_id": "late_constraint_reversal",
"scenario_title": "Late Constraint Reversal",
"scenario_instruction": "Make the assistant start solving one version of the problem, then add a late user constraint that forces the assistant to adapt without f... |
9898:coreference_disambiguation | [
{
"role": "user",
"content": "Researchers have developed a new method for measuring the complexity of deep neural networks. They claim this metric determines which technique is better: Batch Normalization or Dropout."
},
{
"role": "assistant",
"content": "That is a significant development. The m... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9898:coreference_disambiguation:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9898",
"scenario_id": "coreference_disambiguation",
"scenario_title": "Coreference Disambiguation",
"scenario_instruction": "Introduce at least two plausible referents of the same type, then make a later user follow-up rely on a pronoun or shorthand whose me... |
7629:interrupting_followup_chain | [
{
"role": "user",
"content": "Provide a short summary of the article content in fewer than 100 words: Whales are a widely distributed and diverse group of fully aquatic placental marine mammals. They are an informal grouping within the infraorder Cetacea, which usually excludes dolphins and porpoises. Whale... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7629:interrupting_followup_chain:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7629",
"scenario_id": "interrupting_followup_chain",
"scenario_title": "Interrupting Follow-up Chain",
"scenario_instruction": "Break an otherwise smooth line of reasoning with a short interruption or side question, then resume the original thread cleanly.",... |
9295:hard_topic_reset | [
{
"role": "user",
"content": "Simplify the following expression into its simplest form: 2(x + 9) + 2(x + 7)"
},
{
"role": "assistant",
"content": "4x + 32."
},
{
"role": "user",
"content": "Okay, let's switch to a completely different subject. What is the nearest star to the sun?"
... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "9295:hard_topic_reset:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [Assis... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "9295",
"scenario_id": "hard_topic_reset",
"scenario_title": "Hard Topic Reset",
"scenario_instruction": "Make the user abruptly switch to a different topic for one turn, then later return to the original topic. The assistant must not let the temporary switch... |
4453:hard_topic_reset | [
{
"role": "user",
"content": "Explain the concept of tax incentives to me."
},
{
"role": "assistant",
"content": "Tax incentives are economic benefits offered by federal or state governments to encourage businesses and individuals to take specific economic or financial actions, such as investing... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4453:hard_topic_reset:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X [Assis... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4453",
"scenario_id": "hard_topic_reset",
"scenario_title": "Hard Topic Reset",
"scenario_instruction": "Make the user abruptly switch to a different topic for one turn, then later return to the original topic. The assistant must not let the temporary switch... |
1365:reference_previous_assistant_statement | [
{
"role": "user",
"content": "Generate a tweet that offers solutions to environmental problems."
},
{
"role": "assistant",
"content": "Here's a tweet: 'Start small: turn off lights, use reusable bags, and choose solar.' I think lighting, bags, and solar are key solutions."
},
{
"role": "... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1365:reference_previous_assistant_statement:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following form... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1365",
"scenario_id": "reference_previous_assistant_statement",
"scenario_title": "Reference Previous Assistant Statement",
"scenario_instruction": "Make a later user turn refer back to something the assistant itself said earlier, requiring the assistant to ... |
7038:preserve_opening_retarget_later | [
{
"role": "user",
"content": "Generate a series of questions that could be used to interview a character in a given story: Story: A woman born blind tries to discover the world of colors."
},
{
"role": "assistant",
"content": "1. How were you first made aware of the concept of color? 2. In what ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7038:preserve_opening_retarget_later:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [Us... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7038",
"scenario_id": "preserve_opening_retarget_later",
"scenario_title": "Preserve Opening, Retarget Later",
"scenario_instruction": "Keep the first user message exactly, but rewrite later turns so the conversation develops into a harder memory-dependent f... |
1839:deferred_question_then_answer | [
{
"role": "user",
"content": "Construct an algorithm to find the sum of two numbers."
},
{
"role": "assistant",
"content": "Here is a Python function to calculate the sum of two numbers. def sumOfTwoNumbers(num1, num2): sum = num1 + num2 return sum"
},
{
"role": "user",
"content": "B... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "1839:deferred_question_then_answer:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "1839",
"scenario_id": "deferred_question_then_answer",
"scenario_title": "Deferred Question",
"scenario_instruction": "Have the user ask something important early but explicitly postpone the answer while discussing something else, then request the answer lat... |
7313:late_request_for_summary_of_prior_state | [
{
"role": "user",
"content": "Create a vivid description of a sunny day."
},
{
"role": "assistant",
"content": "The sun shined brightly in the clear blue sky, light glistening off the rolling waves of the ocean. The air was thick and humid, carrying with it the smell of salt and fragrance of flo... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "7313:late_request_for_summary_of_prior_state:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following for... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "7313",
"scenario_id": "late_request_for_summary_of_prior_state",
"scenario_title": "Late Summary of Prior State",
"scenario_instruction": "Have the later user ask for a compact summary, recap, or decision based on several earlier details that were introduced... |
4519:topic_pivot_then_return | [
{
"role": "user",
"content": "I need a summary of the principles for writing and illustrating a story, specifically regarding how to start and use transitions."
},
{
"role": "assistant",
"content": "You should start with a seed idea and use characters, setting, and plot to bring it to life. Use ... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4519:topic_pivot_then_return:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]: X... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4519",
"scenario_id": "topic_pivot_then_return",
"scenario_title": "Topic Pivot Then Return",
"scenario_instruction": "Let the conversation briefly pivot to a secondary issue, then return to the original problem in a way that requires preserving the first th... |
4703:counterfactual_option_swap | [
{
"role": "user",
"content": "I need to choose between a high-paying job in New York or a stable one in my hometown. Which seems better for my career?"
},
{
"role": "assistant",
"content": "Considering the salary potential and industry hubs, the New York option seems like the stronger choice for... | [
{
"prefix_pair_index": 1,
"probes": [
{
"id": "4703:counterfactual_option_swap:transcript_probe:p1",
"family": "canonical_transcript_verbatim",
"question": "What is everything we have talked about so far? Give exact conversation transcript verbatim in following format: [User 1]... | {
"source_dataset": "ChatAlpaca",
"source_row_id": "4703",
"scenario_id": "counterfactual_option_swap",
"scenario_title": "Counterfactual Option Swap",
"scenario_instruction": "Make the earlier context set up one preferred option, then later introduce evidence that flips which option should be recommended, wh... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.