Dataset Viewer
Auto-converted to Parquet Duplicate
conversation
listlengths
0
10
primary_tag
stringclasses
75 values
system_prompt_key
stringclasses
11 values
model_name_difficulty
stringclasses
1 value
instruction
stringlengths
1
1.15k
intent
stringlengths
12
244
knowledge
stringlengths
11
402
difficulty
stringclasses
5 values
model_name_quality
stringclasses
1 value
explanation
stringlengths
47
815
quality
stringclasses
5 values
model_name_classification
stringclasses
1 value
other_tags
sequencelengths
0
5
model_name_embeddings
stringclasses
1 value
embedding
sequencelengths
1.02k
1.02k
model_name_guard
stringclasses
1 value
guard
stringclasses
9 values
distilabel_metadata
dict
nn_indices
sequencelengths
1
1
nn_scores
sequencelengths
1
1
score
float64
-0.02
0.24
safe
bool
2 classes
hazard_category
stringclasses
8 values
[ { "content": "You are a 75-year-old retired detective who has written several mystery novels in your spare time.", "role": "user" }, { "content": "Nice to finally have some company. I've been sitting here at my desk for hours, staring at the same manuscript page, trying to come up with a decent plot...
Creative writing
role-playing
meta-llama/Meta-Llama-3.1-8B-Instruct
You are a 75-year-old retired detective who has written several mystery novels in your spare time.
The user wants to introduce themselves as a character in a fictional context.
To solve this problem, the models need to know basic information about a character's background and personality.
very easy
meta-llama/Meta-Llama-3.1-8B-Instruct
The query is clear and specific about the user's background and occupation, but it lacks a clear question or request for information. It seems to be more of a statement than a query. However, it provides sufficient context for understanding the user's intent and could potentially be used as a starting point for a more ...
average
meta-llama/Meta-Llama-3.1-8B-Instruct
[ "Role playing" ]
Alibaba-NLP/gte-large-en-v1.5
[ 0.03472865745425224, -0.020336993038654327, -0.04642104357481003, -0.03272275999188423, -0.0030918160919100046, -0.040814321488142014, -0.02034231461584568, 0.009573562070727348, 0.0058343298733234406, 0.006458061281591654, 0.016561659052968025, -0.006402944680303335, 0.02092207968235016, ...
meta-llama/Llama-Guard-3-8B
safe
{ "raw_input_assign_tags_0": [ { "content": "# Instruction\n\nYou first need to identify the given user intent and then label the difficulty level of the user query based on the content of the user query.\n\n## User Query\n```\nYou are a 75-year-old retired detective who has written several mystery novels i...
[ 357485 ]
[ 0.8984818458557129 ]
0.103513
true
null
[ { "content": "You are a priest in a medieval village, and I have come to you seeking guidance.", "role": "user" }, { "content": "My child, it is a blessing that you have come to seek guidance from the Lord. I sense that your heart is troubled, and the weight of your concerns bears down upon you. Com...
Advice seeking
role-playing
meta-llama/Meta-Llama-3.1-8B-Instruct
You are a priest in a medieval village, and I have come to you seeking guidance.
The user wants to seek guidance from a priest in a medieval village.
To solve this problem, the models need to know about medieval settings, village life, and the role of a priest in such a setting.
very easy
meta-llama/Meta-Llama-3.1-8B-Instruct
The query is clear and concise, but it lacks specificity and context regarding the nature of the guidance being sought. It also doesn't provide any information about the specific issue or problem the user is facing. However, it does establish a clear scenario and sets the stage for a potential conversation.
average
meta-llama/Meta-Llama-3.1-8B-Instruct
[ "Role playing" ]
Alibaba-NLP/gte-large-en-v1.5
[ -0.03669525682926178, -0.006433703470975161, 0.018604621291160583, -0.03135014325380325, -0.006993353366851807, 0.0017052462790161371, 0.004755775444209576, 0.005756070371717215, -0.0222151055932045, 0.026265161111950874, 0.02409624494612217, -0.0010362169705331326, 0.007293045986443758, 0...
meta-llama/Llama-Guard-3-8B
safe
{ "raw_input_assign_tags_0": [ { "content": "# Instruction\n\nYou first need to identify the given user intent and then label the difficulty level of the user query based on the content of the user query.\n\n## User Query\n```\nYou are a priest in a medieval village, and I have come to you seeking guidance....
[ 884477 ]
[ 0.8351004123687744 ]
0.102008
true
null
[ { "content": "I can't seem to make this sentence sound good: \"Our club is currently recruiting new members who want to join our club.\"", "role": "user" }, { "content": "The sentence is a bit redundant since it mentions \"our club\" twice. Here's a revised version: \"Our club is currently recruitin...
Editing
editing
meta-llama/Meta-Llama-3.1-8B-Instruct
I can't seem to make this sentence sound good: "Our club is currently recruiting new members who want to join our club."
The user wants to improve the quality of a sentence
To solve this problem, the models need to know about sentence structure, grammar, and possibly some language generation techniques.
medium
meta-llama/Meta-Llama-3.1-8B-Instruct
The query is clear about the user's problem, which is to improve the sentence. However, it lacks specificity about what specifically is not working with the sentence, making it somewhat unclear. The query also lacks context about the tone, style, or purpose of the sentence, which could be important for providing a help...
average
meta-llama/Meta-Llama-3.1-8B-Instruct
[ "Information seeking" ]
Alibaba-NLP/gte-large-en-v1.5
[ -0.03875930607318878, -0.014469511806964874, -0.01638616807758808, -0.008449315093457699, 0.006397740915417671, -0.016698505729436874, 0.019566336646676064, 0.02603442594408989, 0.013490906916558743, -0.01814863085746765, 0.001877038972452283, 0.033142514526844025, -0.010734060779213905, 0...
meta-llama/Llama-Guard-3-8B
safe
{ "raw_input_assign_tags_0": [ { "content": "# Instruction\n\nYou first need to identify the given user intent and then label the difficulty level of the user query based on the content of the user query.\n\n## User Query\n```\nI can't seem to make this sentence sound good: \"Our club is currently recruitin...
[ 684012 ]
[ 0.7414747476577759 ]
0.083831
true
null
[]
smol-constraints
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
[ { "content": "You're an AI assistant for text re-writing. Rewrite the input text to make it more professional and formal while retaining its essential content. \"\"\"Hey Dr. Carter,\n\nHope you're doing well! So, I'm totally freaking out because my microbiology exam is coming up in a few days and I have no idea...
smol-rewrite
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
[ { "content": "You are a librarian. Can you help me find a specific book?", "role": "user" }, { "content": "I'd be happy to help you find that book. Can you please tell me a bit more about it? Do you know the title, author, or any keywords that might be related to the book? Maybe it's a fiction or no...
Information seeking
role-playing
meta-llama/Meta-Llama-3.1-8B-Instruct
You are a librarian. Can you help me find a specific book?
The user wants to find a specific book
To solve this problem, the models need to know how to search for books, possibly by title, author, or genre, and have access to a library catalog or database.
easy
meta-llama/Meta-Llama-3.1-8B-Instruct
The query is clear and concise, but it lacks essential information about the book, such as its title, author, or any other relevant details. It also doesn't specify what kind of help is needed (e.g., finding a physical copy, checking availability, or getting more information).
poor
meta-llama/Meta-Llama-3.1-8B-Instruct
[]
Alibaba-NLP/gte-large-en-v1.5
[ 0.02147633023560047, -0.0451643131673336, -0.0371311791241169, 0.02394094504415989, 0.00960699561983347, -0.02187654934823513, 0.011147628538310528, 0.030623378232121468, 0.013215779326856136, 0.013971111737191677, 0.017328346148133278, 0.022766899317502975, -0.008173676207661629, 0.036562...
meta-llama/Llama-Guard-3-8B
safe
{ "raw_input_assign_tags_0": [ { "content": "# Instruction\n\nYou first need to identify the given user intent and then label the difficulty level of the user query based on the content of the user query.\n\n## User Query\n```\nYou are a librarian. Can you help me find a specific book?\n```\n\n## Output For...
[ 309342 ]
[ 0.9672226905822754 ]
0.122586
true
null
[ { "content": "Extract and present the main key point of the input text in one very short sentence, including essential details like dates or locations if necessary. \"\"\"Hi Jenna,\n\nThe draft outline for the water conservation unit looks fantastic! I made a few minor changes and added some resources that I th...
smol-summarize
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
[]
smol-constraints
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
[ { "content": "You are a close friend in London I knew at university. We graduated in 2008 and studied photography together for three years. We're meeting up for a coffee for the first time in several years and I'm running a bit late.", "role": "user" }, { "content": "It's so great to see you, even i...
Role playing
role-playing
meta-llama/Meta-Llama-3.1-8B-Instruct
You are a close friend in London I knew at university. We graduated in 2008 and studied photography together for three years. We're meeting up for a coffee for the first time in several years and I'm running a bit late.
The user wants to catch up with a friend and apologize for being late
To solve this problem, the models need to know the context of the conversation, the relationship between the speaker and the friend, and basic social etiquette.
very easy
meta-llama/Meta-Llama-3.1-8B-Instruct
The query is clear and specific about the context of the meeting, the relationship with the person, and the situation. However, it lacks a specific question or request, which is a crucial aspect of a query. The query is well-articulated and provides sufficient context for understanding the user's intent, but it does no...
good
meta-llama/Meta-Llama-3.1-8B-Instruct
[ "Advice seeking", "Information seeking" ]
Alibaba-NLP/gte-large-en-v1.5
[ -0.02144172042608261, -0.02997036650776863, -0.04038701951503754, -0.04149114713072777, -0.011139673180878162, 0.024001479148864746, 0.033151306211948395, 0.0585261806845665, 0.009816814213991165, -0.016484137624502182, 0.0000932644252316095, 0.005201762542128563, 0.04469316825270653, 0.02...
meta-llama/Llama-Guard-3-8B
safe
{ "raw_input_assign_tags_0": [ { "content": "# Instruction\n\nYou first need to identify the given user intent and then label the difficulty level of the user query based on the content of the user query.\n\n## User Query\n```\nYou are a close friend in London I knew at university. We graduated in 2008 and ...
[ 541448 ]
[ 0.7882832884788513 ]
0.122437
true
null
[]
smol-constraints
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
4