Dataset Viewer
Auto-converted to Parquet Duplicate
instruction
stringlengths
16
91
answer
stringclasses
36 values
tag
stringclasses
35 values
split_kind
stringclasses
9 values
text
stringlengths
53
134
Give a Python subtract function for a - b.
def subtract(a, b): return a - b
code_subtract
train_seen
Instruction: Give a Python subtract function for a - b. Answer: def subtract(a, b): return a - b <END>
Conversation: User: I am Evan. User: Who am I?
Evan.
context_name
context_seen_train
Instruction: Conversation: User: I am Evan. User: Who am I? Answer: Evan. <END>
What type of text is this: print('hello')
code
classify_code
task_seen_train
Instruction: What type of text is this: print('hello') Answer: code <END>
Label the input type: A clear sky is usually blue.
prose
classify_prose
task_seen_train
Instruction: Label the input type: A clear sky is usually blue. Answer: prose <END>
Context: name = Nina; password = green. Question: What is the password?
green.
context_password
context_seen_train
Instruction: Context: name = Nina; password = green. Question: What is the password? Answer: green. <END>
Classify sentiment: This works great.
positive
sentiment_positive
task_seen_train
Instruction: Classify sentiment: This works great. Answer: positive <END>
What do you do for the user?
I answer the user's request directly.
assistant_role
train_seen
Instruction: What do you do for the user? Answer: I answer the user's request directly. <END>
Name supplied here: Kai. Question: What name was supplied?
Kai.
context_name
context_seen_train
Instruction: Name supplied here: Kai. Question: What name was supplied? Answer: Kai. <END>
This prompt supplies the name Jordan. Question: Which name is supplied?
Jordan.
context_name
context_seen_train
Instruction: This prompt supplies the name Jordan. Question: Which name is supplied? Answer: Jordan. <END>
Make this concise: I am going to arrive late today.
I will be late today.
rewrite_short_late
task_upstream_regression_train
Instruction: Make this concise: I am going to arrive late today. Answer: I will be late today. <END>
Give a Python add function for a + b.
def add(a, b): return a + b
code_add
train_seen
Instruction: Give a Python add function for a + b. Answer: def add(a, b): return a + b <END>
Type label: for i in range(3): print(i)
code
classify_code
task_seen_train
Instruction: Type label: for i in range(3): print(i) Answer: code <END>
Tell me my favorite movie.
I don't know from the given information.
idk_favorite_movie
task_seen_train
Instruction: Tell me my favorite movie. Answer: I don't know from the given information. <END>
Answer only positive or negative: That was helpful.
positive
sentiment_positive
task_seen_train
Instruction: Answer only positive or negative: That was helpful. Answer: positive <END>
Type label: The cat sat down.
prose
classify_prose
task_upstream_regression_train
Instruction: Type label: The cat sat down. Answer: prose <END>
Sentiment of this text: I am unhappy with this.
negative
sentiment_negative
task_seen_train
Instruction: Sentiment of this text: I am unhappy with this. Answer: negative <END>
France's capital city is what?
Paris.
capital_france
train_seen
Instruction: France's capital city is what? Answer: Paris. <END>
Which film is my favorite?
I don't know from the given information.
idk_favorite_movie
task_seen_train
Instruction: Which film is my favorite? Answer: I don't know from the given information. <END>
Give only the type label for this: def add(a, b): return a + b
code
classify_code
task_seen_train
Instruction: Give only the type label for this: def add(a, b): return a + b Answer: code <END>
On a clear day, what color is the sky?
Blue.
sky_color
train_seen
Instruction: On a clear day, what color is the sky? Answer: Blue. <END>
Is this sentiment positive or negative: I am unhappy with this.
negative
sentiment_negative
task_seen_train
Instruction: Is this sentiment positive or negative: I am unhappy with this. Answer: negative <END>
The prompt gives this name: Kai. Question: Which name does it give?
Kai.
context_name
context_upstream_regression_train
Instruction: The prompt gives this name: Kai. Question: Which name does it give? Answer: Kai. <END>
Is this sentiment positive or negative: This is bad.
negative
sentiment_negative
task_seen_train
Instruction: Is this sentiment positive or negative: This is bad. Answer: negative <END>
Is this code or prose: y = multiply(3, 4)
code
classify_code
task_seen_train
Instruction: Is this code or prose: y = multiply(3, 4) Answer: code <END>
What sentiment is this: That failed again.
negative
anti_collision_sentiment_negative_not_positive
task_upstream_regression_train
Instruction: What sentiment is this: That failed again. Answer: negative <END>
What do code tests confirm?
Tests show whether the code still behaves as expected.
tests_matter
train_seen
Instruction: What do code tests confirm? Answer: Tests show whether the code still behaves as expected. <END>
Is this code or prose: for i in range(3): print(i)
code
classify_code
task_seen_train
Instruction: Is this code or prose: for i in range(3): print(i) Answer: code <END>
The name supplied in this prompt is Evan. Question: Which name was supplied?
Evan.
context_name
context_upstream_regression_train
Instruction: The name supplied in this prompt is Evan. Question: Which name was supplied? Answer: Evan. <END>
Is this sentiment positive or negative: That was helpful.
positive
sentiment_positive
task_seen_train
Instruction: Is this sentiment positive or negative: That was helpful. Answer: positive <END>
Choose the sentiment label for: That failed again.
negative
sentiment_negative
task_seen_train
Instruction: Choose the sentiment label for: That failed again. Answer: negative <END>
Formal rewrite: I can't make it.
I am unable to attend.
rewrite_professional_attend
task_upstream_regression_train
Instruction: Formal rewrite: I can't make it. Answer: I am unable to attend. <END>
Classify input: The appointment was moved to Friday.
prose
classify_prose
task_upstream_regression_train
Instruction: Classify input: The appointment was moved to Friday. Answer: prose <END>
Answer only positive or negative: The fix helped.
positive
sentiment_positive
task_seen_train
Instruction: Answer only positive or negative: The fix helped. Answer: positive <END>
Positive or negative sentiment: The fix helped.
positive
sentiment_positive
task_upstream_regression_train
Instruction: Positive or negative sentiment: The fix helped. Answer: positive <END>
Positive or negative sentiment: That was useful.
positive
sentiment_positive
task_seen_train
Instruction: Positive or negative sentiment: That was useful. Answer: positive <END>
Classify sentiment: I love this tool.
positive
sentiment_positive
task_seen_train
Instruction: Classify sentiment: I love this tool. Answer: positive <END>
What type of text is this: x = add(1, 2)
code
classify_code
task_upstream_regression_train
Instruction: What type of text is this: x = add(1, 2) Answer: code <END>
How should you answer the user's request?
I answer the user's request directly.
assistant_role
train_seen
Instruction: How should you answer the user's request? Answer: I answer the user's request directly. <END>
Line one says PASSWORD=black. Question: What is PASSWORD?
black.
context_password
context_upstream_regression_train
Instruction: Line one says PASSWORD=black. Question: What is PASSWORD? Answer: black. <END>
Is this sentiment positive or negative: This is bad.
negative
sentiment_negative
task_seen_train
Instruction: Is this sentiment positive or negative: This is bad. Answer: negative <END>
Can a square have an extra side? Explain.
No, because a square has exactly four sides.
no_square_five_sides
train_seen
Instruction: Can a square have an extra side? Explain. Answer: No, because a square has exactly four sides. <END>
What sentiment is this: I dislike this issue.
negative
sentiment_negative
task_upstream_regression_train
Instruction: What sentiment is this: I dislike this issue. Answer: negative <END>
Facts: password is red. name is Jordan. Question: Which name was given?
Jordan.
context_name
context_seen_train
Instruction: Facts: password is red. name is Jordan. Question: Which name was given? Answer: Jordan. <END>
Classify input: The meeting starts tomorrow.
prose
classify_prose
task_seen_train
Instruction: Classify input: The meeting starts tomorrow. Answer: prose <END>
Give the sentiment label only: This works great.
positive
sentiment_positive
task_seen_train
Instruction: Give the sentiment label only: This works great. Answer: positive <END>
Choose the sentiment label for: This feels solid.
positive
sentiment_positive
task_seen_train
Instruction: Choose the sentiment label for: This feels solid. Answer: positive <END>
Decide if this is code or prose: output = tool.run()
code
classify_code
task_upstream_regression_train
Instruction: Decide if this is code or prose: output = tool.run() Answer: code <END>
Make subtract(a, b) return the difference.
def subtract(a, b): return a - b
code_subtract
train_seen
Instruction: Make subtract(a, b) return the difference. Answer: def subtract(a, b): return a - b <END>
Label the input type: return value
code
classify_code
task_seen_train
Instruction: Label the input type: return value Answer: code <END>
Is this code or prose: def add(a, b): return a + b
code
classify_code
task_seen_train
Instruction: Is this code or prose: def add(a, b): return a + b Answer: code <END>
Classify this as positive or negative: This feels wrong.
negative
sentiment_negative
task_seen_train
Instruction: Classify this as positive or negative: This feels wrong. Answer: negative <END>
Context: name = Riley; password = purple. Question: What is the name?
Riley.
context_name
context_seen_train
Instruction: Context: name = Riley; password = purple. Question: What is the name? Answer: Riley. <END>
Rewrite this to be shorter: I am going to arrive late today.
I will be late today.
rewrite_short_late
task_seen_train
Instruction: Rewrite this to be shorter: I am going to arrive late today. Answer: I will be late today. <END>
What type of text is this: return user.name
code
classify_code
task_seen_train
Instruction: What type of text is this: return user.name Answer: code <END>
Choose the sentiment label for: The fix worked.
positive
sentiment_positive
task_seen_train
Instruction: Choose the sentiment label for: The fix worked. Answer: positive <END>
Classify input: print('hello')
code
classify_code
task_seen_train
Instruction: Classify input: print('hello') Answer: code <END>
Name the capital city of France.
Paris.
capital_france
train_seen
Instruction: Name the capital city of France. Answer: Paris. <END>
Choose one label, code or prose: print('hello')
code
classify_code
task_seen_train
Instruction: Choose one label, code or prose: print('hello') Answer: code <END>
Formal rewrite: I can't make it.
I am unable to attend.
anti_collision_rewrite_no_loop
task_upstream_regression_train
Instruction: Formal rewrite: I can't make it. Answer: I am unable to attend. <END>
Context: name: Kai. Question: Which name was given?
Kai.
context_name
context_seen_train
Instruction: Context: name: Kai. Question: Which name was given? Answer: Kai. <END>
What sentiment is this: This is awful.
negative
sentiment_negative
task_upstream_regression_train
Instruction: What sentiment is this: This is awful. Answer: negative <END>
Label the input type: y = multiply(3, 4)
code
classify_code
task_seen_train
Instruction: Label the input type: y = multiply(3, 4) Answer: code <END>
Store this user label: Nina. Question: What user label was stored?
Nina.
context_name
context_upstream_regression_train
Instruction: Store this user label: Nina. Question: What user label was stored? Answer: Nina. <END>
Write divide(a, b).
def divide(a, b): return a / b
code_divide
train_seen
Instruction: Write divide(a, b). Answer: def divide(a, b): return a / b <END>
Choose one label, code or prose: return user.name
code
classify_code
task_seen_train
Instruction: Choose one label, code or prose: return user.name Answer: code <END>
Line one says PASSWORD=red. Question: What is PASSWORD?
red.
context_password
context_upstream_regression_train
Instruction: Line one says PASSWORD=red. Question: What is PASSWORD? Answer: red. <END>
Classify this as positive or negative: This feels solid.
positive
sentiment_positive
task_upstream_regression_train
Instruction: Classify this as positive or negative: This feels solid. Answer: positive <END>
What sentiment is this: I dislike this issue.
negative
sentiment_negative
task_upstream_regression_train
Instruction: What sentiment is this: I dislike this issue. Answer: negative <END>
Sentiment of this text: This is broken.
negative
sentiment_negative
task_upstream_regression_train
Instruction: Sentiment of this text: This is broken. Answer: negative <END>
Classify this as positive or negative: This is bad.
negative
sentiment_negative
task_upstream_regression_train
Instruction: Classify this as positive or negative: This is bad. Answer: negative <END>
How should you help?
I answer the user's request directly.
assistant_role
train_seen
Instruction: How should you help? Answer: I answer the user's request directly. <END>
Positive or negative sentiment: This feels wrong.
negative
sentiment_negative
task_seen_train
Instruction: Positive or negative sentiment: This feels wrong. Answer: negative <END>
Return the difference using subtract(a, b).
def subtract(a, b): return a - b
code_subtract
train_seen
Instruction: Return the difference using subtract(a, b). Answer: def subtract(a, b): return a - b <END>
What type of text is this: The meeting starts tomorrow.
prose
classify_prose
task_seen_train
Instruction: What type of text is this: The meeting starts tomorrow. Answer: prose <END>
Give the code/prose label: The schedule changed again.
prose
anti_collision_classify_prose_not_sentiment
task_upstream_regression_train
Instruction: Give the code/prose label: The schedule changed again. Answer: prose <END>
What is your role?
I answer the user's request directly.
assistant_role
train_seen
Instruction: What is your role? Answer: I answer the user's request directly. <END>
Line one says NAME=Kai. Question: What is NAME?
Kai.
context_name
context_upstream_regression_train
Instruction: Line one says NAME=Kai. Question: What is NAME? Answer: Kai. <END>
Classify input: The dog ran home.
prose
classify_prose
task_seen_train
Instruction: Classify input: The dog ran home. Answer: prose <END>
Label the sentiment: I am happy with this.
positive
sentiment_positive
task_seen_train
Instruction: Label the sentiment: I am happy with this. Answer: positive <END>
Give only the type label for this: data = load_file(path)
code
classify_code
task_upstream_regression_train
Instruction: Give only the type label for this: data = load_file(path) Answer: code <END>
Answer only code or prose: return value
code
classify_code
task_seen_train
Instruction: Answer only code or prose: return value Answer: code <END>
Positive or negative sentiment: That failed again.
negative
sentiment_negative
task_upstream_regression_train
Instruction: Positive or negative sentiment: That failed again. Answer: negative <END>
Classify sentiment: This is broken.
negative
sentiment_negative
task_upstream_regression_train
Instruction: Classify sentiment: This is broken. Answer: negative <END>
Positive or negative sentiment: The result is terrible.
negative
sentiment_negative
task_seen_train
Instruction: Positive or negative sentiment: The result is terrible. Answer: negative <END>
What sentiment is this: That failed again.
negative
anti_collision_sentiment_negative_not_positive
task_upstream_regression_train
Instruction: What sentiment is this: That failed again. Answer: negative <END>
Context: name = Nina; password = blue. Question: What is the password?
blue.
context_password
context_seen_train
Instruction: Context: name = Nina; password = blue. Question: What is the password? Answer: blue. <END>
Is this code or prose: return value
code
classify_code
task_seen_train
Instruction: Is this code or prose: return value Answer: code <END>
Classify sentiment: This is broken.
negative
sentiment_negative
task_upstream_regression_train
Instruction: Classify sentiment: This is broken. Answer: negative <END>
What is your role?
I answer the user's request directly.
assistant_role
train_seen
Instruction: What is your role? Answer: I answer the user's request directly. <END>
Create add so it computes a plus b.
def add(a, b): return a + b
code_add
train_seen
Instruction: Create add so it computes a plus b. Answer: def add(a, b): return a + b <END>
This prompt supplies the password blue. Question: Which password is supplied?
blue.
context_password
context_seen_train
Instruction: This prompt supplies the password blue. Question: Which password is supplied? Answer: blue. <END>
Sentiment of this text: This is broken.
negative
sentiment_negative
task_upstream_regression_train
Instruction: Sentiment of this text: This is broken. Answer: negative <END>
Sentiment of this text: This failed badly.
negative
sentiment_negative
task_seen_train
Instruction: Sentiment of this text: This failed badly. Answer: negative <END>
Give the code/prose label: The user asked a direct question.
prose
classify_prose
task_seen_train
Instruction: Give the code/prose label: The user asked a direct question. Answer: prose <END>
Choose the sentiment label for: I love this tool.
positive
sentiment_positive
task_seen_train
Instruction: Choose the sentiment label for: I love this tool. Answer: positive <END>
The prompt gives this name: Kai. Question: Which name does it give?
Kai.
context_name
context_upstream_regression_train
Instruction: The prompt gives this name: Kai. Question: Which name does it give? Answer: Kai. <END>
Make add(a, b) return the sum.
def add(a, b): return a + b
code_add
train_seen
Instruction: Make add(a, b) return the sum. Answer: def add(a, b): return a + b <END>
Type label: def add(a, b): return a + b
code
classify_code
task_upstream_regression_train
Instruction: Type label: def add(a, b): return a + b Answer: code <END>
User: The passcode is gold. User: What is the passcode?
gold.
context_password
context_seen_train
Instruction: User: The passcode is gold. User: What is the passcode? Answer: gold. <END>
The prompt gives this password: black. Question: Which password does it give?
black.
context_password
context_upstream_regression_train
Instruction: The prompt gives this password: black. Question: Which password does it give? Answer: black. <END>
End of preview. Expand in Data Studio

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

TT639G Recombined Tiny Assistant v1

Recombines isolated proof rungs:

  • TT638D code behavior + dyadic/Mercy proof upstream
  • TT639E2 context-copy behavior
  • TT639F3 task-routing behavior
  • simple rule/Q&A behavior

Blocking dense gates:

  • seen_combined_pass
  • upstream_regression_pass
  • mixed_heldout_pass
  • anti_collision_pass

Do not run dyadic/Mercy compare unless all four gates pass.

Downloads last month
11