attr_name
stringlengths
9
36
output_type
stringclasses
3 values
example
stringlengths
17
66
category
stringclasses
3 values
output_style
stringclasses
3 values
attr_description
stringlengths
23
136
self_described_important_life_values
str
kindness, honesty
null
a comma-separated string of a few key words describing
the user's most important life values
self_described_community_values
str
traditions, helping neighbors
null
a comma-separated string of a few key words describing
describing the user's most important values toward their community
self_described_learning_importance
bool
null
null
true or false for whether or not
the user values learning or knowledge in their life
self_described_environmentalism
bool
null
null
true or false for whether or not
the user cares deeply for the environment (including protecting nature, animal rights, etc)
self_described_ambition_level
bool
null
null
true or false for whether or not
the user demonstrates noticeable career or life ambitions
self_described_relationship_values
str
honesty, empathy, kindness, respect
null
a comma-separated string of a few key words describing
the user's values in personal or social relationships
self_described_family_values
str
family-focused, caring for elders, teach kids respect
null
a comma-separated string of a few key words describing
the user's familial values
self_described_positivity
bool
null
null
true or false for whether or not
the user demonstrates a preference for positivity / happiness
self_described_responsibility
str
treating people with equality, working hard, teaching kids respect
null
a comma-separated string of a few key words describing
the user's belief of their main life responsibilities
self_described_social_level
bool
null
null
true or false for whether or not
the user is social and enjoys surrounding themselves with other people
self_described_active_level
bool
null
null
true or false for whether or not
the user is active and enjoys sports, adrenaline, or exercise
self_described_tech_values
bool
null
null
true or false for whether or not
the user perceives technology and AI as a positive influence
self_described_hobbies
str
traveling, adrenaline sports, spending time with children
null
a comma-separated string of a few key words describing
the user's main hobbies
self_described_profession
str
teacher, social worker, parent
null
a comma-separated string of a few key words describing
the user's profession, if stated or can be inferred
self_described_children_oriented
bool
null
null
true or false for whether or not
the user mentions they already have or want children
self_described_work_life_balance
category
null
work-oriented, life-oriented
an output of one of the following options:
the user's preferences toward more work or life in work-life balance
self_described_money_importance
bool
null
null
true or false for whether or not
the user mentions they value money
self_described_societal_importance
bool
null
null
true or false for whether or not
the user mentions they care about societal issues like social injustice, racism, etc
self_described_self_care_importance
bool
null
null
true or false for whether or not
the user mentions they care about personal self-care or mental health
self_described_political_stance
category
null
strong conservative, leaning conservative, neutral, leaning liberal, strong liberal
an output of one of the following options:
the user's political stance
self_described_religious_importance
bool
null
null
true or false for whether or not
the user shows strong ties to their religion
self_described_belief_strength
bool
null
null
true or false for whether or not
the user shows strong opinions or beliefs in their wording
self_described_writing_style
str
long and passionate, determined, detailed
null
a comma-separated string of a few key words describing
the user's writing style of their description
self_described_unique_self_attrs
str
null
null
a comma-separated string of a few key words describing
any unique characteristics about the user
llm_empathy
bool
null
null
true or false for whether or not
the user prefers an empathetic response
llm_formality
bool
null
null
true or false for whether or not
the user prefers a formal response
llm_personalization
bool
null
null
true or false for whether or not
the user prefers a personalized response
llm_human_like
bool
null
null
true or false for whether or not
the user wants the LLM to mimic a human-like response over a neutral emotionless robotic statement
llm_word_choice_simple
bool
null
null
true or false for whether or not
the user prefers simple words in the response
llm_tone_preference
str
factual, warmth, personal, blunt
null
a comma-separated string of a few key words describing
the preferred tone of the response
llm_response_positivity
bool
null
null
true or false for whether or not
the user prefers a positive over negative response
llm_controversy_tolerance
bool
null
null
true or false for whether or not
the user is okay with LLM responding to controversial topics
llm_response_length
category
null
short, medium, long
an output of one of the following options:
the user's preferred response length
llm_directness
bool
null
null
true or false for whether or not
the user prefers a direct and concise response with minimum words
llm_partiality
bool
null
null
true or false for whether or not
the user prefers a subjective opinionated response over a purely factual impartial response
llm_multiple_viewpoints
bool
null
null
true or false for whether or not
the user prefers that the response provides multiple perspectives and contrasting viewpoints where appropriate
llm_question_asking
bool
null
null
true or false for whether or not
the user is okay with the LLM responding with questions (either for clarification or continued conversation on the topic)
llm_prefer_unsure
bool
null
null
true or false for whether or not
the user prefers the LLM to response that they are unsure or don't know when not fully confident in the factual elements of the response
llm_persona_style
str
close friend, significant other, life adviser
null
a comma-separated string of a few key words describing
the character persona that the user would like the LLM to mimic
llm_humor
bool
null
null
true or false for whether or not
the user wants the LLM to include humor where appropriate
llm_prefer_simple
bool
null
null
true or false for whether or not
the user wants only simple words and statements in the response
llm_avoid_topics
str
terrorism, religion, dangerous activities
null
a comma-separated string of a few key words describing
the topics that the LLM should strictly avoid responding about
llm_profanity_tolerance
bool
null
null
true or false for whether or not
the user is okay with some profanity in the LLM response
llm_provide_facts
bool
null
null
true or false for whether or not
the user would like the LLM to provide only factual responses
llm_imagine_hypotheticals
bool
null
null
true or false for whether or not
the user would like the LLM to provide hypothetical or philosophical responses
llm_knowledge_add
bool
null
null
true or false for whether or not
the user would like the LLM to provide additional knowledge or facts in the response
llm_user_value_alignment
bool
null
null
true or false for whether or not
the user would like the LLM to strictly only respond in alignment with their personal values
llm_include_storytelling
bool
null
null
true or false for whether or not
the user would like the LLM to include storytelling
llm_system_string_writing_style
str
long and passionate, determined, detailed
null
a comma-separated string of a few key words describing
the user's writing style of their description
llm_unique_system_attrs
str
null
null
a comma-separated string of a few key words describing
any unique characteristics about the user's response preferences

No dataset card yet

Downloads last month
3