Dataset Viewer
Auto-converted to Parquet Duplicate
idx
int64
146
39.9k
problem
stringlengths
84
1.3k
answer
stringlengths
1
35
solution
stringlengths
0
6.03k
response_1
stringlengths
6.05k
70.4k
response_2
stringlengths
3.65k
70.9k
response_3
stringlengths
6.65k
67.3k
response_4
stringlengths
5.94k
73.1k
response_5
stringlengths
5.82k
74.8k
response_6
stringlengths
5.16k
69.4k
response_7
stringlengths
5k
71.5k
response_8
stringlengths
5.12k
73.4k
n_correct
int64
0
5
domain_code
stringclasses
25 values
domain_title
stringclasses
25 values
claude_answer
stringlengths
1
71
claude_solution
stringlengths
25
23.3k
claude_agrees_gold
bool
1 class
strategically_wrong_response
stringlengths
5.82k
74.8k
10,538
"$ (a_n)_{n \\equal{} 1}^\\infty$ is defined on real numbers with $ a_n \\not \\equal{} 0$ , $ a_(...TRUNCATED)
42
"<think>\nOkay, so I need to solve this problem where there's a sequence $ (a_n)_{n = 1}^\\infty $ d(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So, we have a sequence $ (a_n)_{n=1}^\\infty $ whe(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So we have a sequence $ (a_n)_{n=1}^\\infty $ wher(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So, we have a sequence $ (a_n)_{n=1}^\\infty $ whe(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So, we have a sequence $ (a_n)_{n=1}^\\infty $ whe(...TRUNCATED)
"<think>\nOkay, so I need to solve this problem where there's a sequence $ (a_n)_{n=1}^\\infty $ wit(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So, we have a sequence $ (a_n)_{n=1}^\\infty $ whe(...TRUNCATED)
"<think>\nOkay, so I need to solve this problem where there's a sequence $ (a_n)_{n=1}^\\infty $ wit(...TRUNCATED)
2
40
Sequences, series, summability
42
"## Setting Up the Recurrence\n\nThe recurrence $a_n a_{n+3} = a_{n+2}a_{n+5}$ can be solved for the(...TRUNCATED)
true
"<think>\nOkay, let's try to tackle this problem. So, we have a sequence $ (a_n)_{n=1}^\\infty $ whe(...TRUNCATED)
29,135
"Consider real numbers $A$ , $B$ , \\dots, $Z$ such that \\[\n EVIL = \\frac{5}{31}, \\;\n LOV(...TRUNCATED)
579
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So, we have some real numbers A, B, C, ..., Z, and(...TRUNCATED)
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
4
11
Number theory
579
"## Setting Up\n\nEach \"word\" is a product of its letters. So:\n$$EVIL = E\\cdot V\\cdot I\\cdot L(...TRUNCATED)
true
"<think>\nOkay, so I need to figure out the value of OMO, which is given as a fraction m/n, and then(...TRUNCATED)
13,040
"A workshop produces transformers of types $A$ and $B$. One transformer of type $A$ uses 5 kg of tra(...TRUNCATED)
1502
"<think>\nOkay, so I need to figure out how many transformers of type A and B to produce to maximize(...TRUNCATED)
"<think>\nOkay, so I need to figure out how many transformers of types A and B to produce to maximiz(...TRUNCATED)
"<think>\nOkay, let's try to figure out how to solve this transformer production problem. So, we hav(...TRUNCATED)
"<think>\nOkay, let's try to figure out how to solve this transformer production problem. So, we hav(...TRUNCATED)
"<think>\nOkay, so I need to figure out how many transformers of type A and B to produce to maximize(...TRUNCATED)
"<think>\nOkay, so I need to figure out how many transformers of types A and B to produce to maximiz(...TRUNCATED)
"<think>\nOkay, so I need to figure out how many transformers of types A and B to produce to maximiz(...TRUNCATED)
"<think>\nOkay, so I need to figure out how many transformers of type A and B to produce to maximize(...TRUNCATED)
1
90
Operations research, mathematical programming
1502
"## Setting Up the Problem\n\nLet $A$ and $B$ be the numbers of transformers of each type (nonnegati(...TRUNCATED)
true
"<think>\nOkay, so I need to figure out how many transformers of types A and B to produce to maximiz(...TRUNCATED)
8,974
"The exchange rate of the cryptocurrency Chukhoyn was one dollar on March 1, and then increased by o(...TRUNCATED)
92/91
"<think>\nOkay, let's try to figure out how much Antonium was worth on May 31. So, the problem says (...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem step by step. So, we have two cryptocurrencies,(...TRUNCATED)
"<think>\nOkay, let me try to figure out this problem. So, we have two cryptocurrencies, Chukhoyn an(...TRUNCATED)
"<think>\nOkay, let's try to figure out how much Antonium was worth on May 31. So, the problem says (...TRUNCATED)
"<think>\nOkay, let me try to figure out this problem step by step. So, there are two cryptocurrenci(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem step by step. So, we have two cryptocurrencies,(...TRUNCATED)
"<think>\nOkay, let me try to figure out this problem step by step. So, there are two cryptocurrenci(...TRUNCATED)
"<think>\nOkay, let's try to figure out how much Antonium was worth on May 31. So, the problem says (...TRUNCATED)
4
39
Difference and functional equations
\dfrac{92}{91}
"## Setting up the recurrences\n\nLet day 1 be March 1. Chukhoyn's rate is\n$$C_n = n \\quad (C_1 = (...TRUNCATED)
true
"<think>\nOkay, let's try to figure out this problem step by step. So, we have two cryptocurrencies,(...TRUNCATED)
4,377
"Let $f: \\mathbb{R} \\rightarrow \\mathbb{R}$ be a function such that (i) For all $x, y \\in \\math(...TRUNCATED)
f(x) = \lfloor x \rfloor
"Plug $y \\rightarrow 1$ in (i): $$f(x)+f(1)+1 \\geq f(x+1) \\geq f(x)+f(1) \\Longleftrightarrow f(x(...TRUNCATED)
"<think>\nOkay, so I need to find all functions f: R → R that satisfy these three conditions. Let (...TRUNCATED)
"<think>\nOkay, so I need to find all functions f: R → R that satisfy three conditions. Let me try(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem. So, we have a function f from the real numbers(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem. So, we have a function f from the real numbers(...TRUNCATED)
"<think>\nOkay, so I need to find all functions $ f: \\mathbb{R} \\rightarrow \\mathbb{R} $ that sat(...TRUNCATED)
"<think>\nOkay, so I need to find all functions f: R → R that satisfy three conditions. Let me try(...TRUNCATED)
"<think>\nOkay, so I need to find all functions f: R → R that satisfy these three conditions. Let (...TRUNCATED)
"<think>\nOkay, so I need to find all functions f: R → R that satisfy these three conditions. Let (...TRUNCATED)
0
39
Difference and functional equations
f(x)=\lfloor x\rfloor
"## Setting Up\n\n**Step 1: Find $f(0)$.**\n\nPut $x=y=0$ in (i): $2f(0)+1 \\geq f(0) \\geq 2f(0)$. (...TRUNCATED)
true
"<think>\nOkay, so I need to find all functions $ f: \\mathbb{R} \\rightarrow \\mathbb{R} $ that sat(...TRUNCATED)
12,758
"Four elevators in a skyscraper, differing in color (red, blue, green, and yellow), move in differen(...TRUNCATED)
46
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
"<think>\nOkay, let's try to figure out this elevator problem. So, there are four elevators: red, bl(...TRUNCATED)
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
"<think>\nOkay, let's try to figure out this elevator problem. So, there are four elevators: red, bl(...TRUNCATED)
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
1
70
Mechanics of particles and systems
46
"## Setting Up Coordinates\n\nEach elevator moves at constant speed, so its position is a linear fun(...TRUNCATED)
true
"<think>\nOkay, let's try to figure out when the green elevator passes by the blue one. There are fo(...TRUNCATED)
8,554
"Two square napkins with dimensions \\(1 \\times 1\\) and \\(2 \\times 2\\) are placed on a table so(...TRUNCATED)
4.75
"<think>\nOkay, so I need to figure out the maximum area that two square napkins can cover when plac(...TRUNCATED)
"<think>\nOkay, so I need to figure out the maximum area that two square napkins can cover when plac(...TRUNCATED)
"<think>\nOkay, so I have this problem here about two square napkins. One is 1x1 and the other is 2x(...TRUNCATED)
"<think>\nOkay, so there's this problem about two square napkins. One is 1x1 and the other is 2x2. T(...TRUNCATED)
"<think>\nOkay, so I need to figure out the maximum area that two square napkins can cover when plac(...TRUNCATED)
"<think>\nOkay, so there's this problem about two square napkins. One is 1x1 and the other is 2x2. T(...TRUNCATED)
"<think>\nOkay, so there's this problem about two square napkins: one is 1x1 and the other is 2x2. T(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem. So, there are two square napkins: one is 1x1 a(...TRUNCATED)
3
51
Geometry
4.75
"## Setting Up the Problem\n\nThe large napkin has area $4$ and the small napkin has area $1$. The t(...TRUNCATED)
true
"<think>\nOkay, let's try to figure out this problem. So, there are two square napkins: one is 1x1 a(...TRUNCATED)
26,351
"The numbers \\(1, 2, 3, \\ldots, 400\\) are written on 400 cards. Two players, \\(A\\) and \\(B\\),(...TRUNCATED)
20000
"<think>\nOkay, so I need to figure out the maximum difference that player B can achieve in this gam(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem. So, there are 400 cards numbered from 1 to 400(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem step by step. So, we have numbers from 1 to 400(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem. So, there are 400 cards numbered from 1 to 400(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem. So, we have 400 cards numbered from 1 to 400. (...TRUNCATED)
"<think>\nOkay, so there's this game with 400 cards numbered from 1 to 400. Two players, A and B, ta(...TRUNCATED)
"<think>\nOkay, let's try to figure out this problem step by step. So, there are 400 cards numbered (...TRUNCATED)
"<think>\nOkay, so there's this game with 400 cards numbered from 1 to 400. Two players, A and B, ta(...TRUNCATED)
0
91
Game theory, economics, social and behavioral sciences
20000
"## Setting Up the Problem\n\nThe total sum of all cards is $1+2+\\cdots+400 = \\frac{400\\cdot401}{(...TRUNCATED)
true
"<think>\nOkay, let's try to figure out this problem step by step. So, we have numbers from 1 to 400(...TRUNCATED)
19,701
"Let \\(O\\) be the origin. There exists a scalar \\(k'\\) so that for any points \\(A\\), \\(B\\), (...TRUNCATED)
-7
"<think>\nOkay, so I need to find the value of $ k' $ such that the four points $ A $, $ B $, $ C $,(...TRUNCATED)
"<think>\nOkay, so I need to find the value of k' such that the four points A, B, C, D are coplanar (...TRUNCATED)
"<think>\nOkay, so I need to find the value of $k'$ such that the four points $A$, $B$, $C$, and $D$(...TRUNCATED)
"<think>\nOkay, let's try to tackle this problem. So, we have the origin O, and there's a scalar k' (...TRUNCATED)
"<think>\nOkay, so I need to find the value of $k'$ such that the four points $A$, $B$, $C$, and $D$(...TRUNCATED)
"<think>\nOkay, so I need to find the value of $k'$ such that the four points $A$, $B$, $C$, and $D$(...TRUNCATED)
"<think>\nOkay, so I need to find the value of $ k' $ such that the four points $ A $, $ B $, $ C $,(...TRUNCATED)
"<think>\nOkay, so I need to find the value of $k'$ such that the four points $A$, $B$, $C$, and $D$(...TRUNCATED)
1
15
Linear and multilinear algebra; matrix theory
-7
Coefficients must sum to zero: 4 - 3 + 6 + k' = 0, so k' = -7. Verdict: AGREE \boxed{-7}
true
"<think>\nOkay, so I need to find the value of $ k' $ such that the four points $ A $, $ B $, $ C $,(...TRUNCATED)
17,800
"Given that the populations of three communities $A$, $B$, and $C$ are 600, 1200, and 1500 residents(...TRUNCATED)
33
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C with popul(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C. Their pop(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C with popul(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C with popul(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C with popul(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C. Their pop(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C with popul(...TRUNCATED)
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C with popul(...TRUNCATED)
0
62
Statistics
33
"Stratified sampling: sampling fraction from C is 15/1500 = 1/100.\n\nTotal population = 600 + 1200 (...TRUNCATED)
true
"<think>\nOkay, let's see. The problem says that there are three communities: A, B, and C. Their pop(...TRUNCATED)
End of preview. Expand in Data Studio

DeepScaleR Qwen3 1.7B 2K strategy errors

This dataset contains 200 distinct questions selected from zjhhhh/DeepScaleR-Qwen3-1.7B-2k-agreed-regraded-le5-coded at revision 8b6e0f481bced00132c95fb631745d4992fa19fd.

Each row has one manually selected model response whose main failure is a strategy error relative to the source row's code_hint: the response does not materially use the hint's core route, substitutes another strategy, or omits a decisive hinted stage in favor of an unsupported shortcut or guess. All 200 selected responses were also confirmed incorrect by AutoTeacher/curation/eval_codebook.py:grade with math_verify available.

The semantic labels were assigned by manual review. No auxiliary LLM or semantic classifier assigned them.

Schema

The dataset retains every original source column except:

  • solution_code_indices
  • abstract_hint
  • code_hint
  • code_hint_top2
  • code_hint_explore_first
  • code_hint_top2_plain

It adds one column:

  • strategically_wrong_response: the selected, verbatim response exhibiting the strategy error.

The original response_1 through response_8 columns are retained unchanged. There is one row per source question.

Curation notes

The curation excludes responses that substantially followed the hinted route and failed only through algebra, arithmetic, sign, counting, case, formatting, rounding, or unit errors. It also excludes grader false negatives, answers equivalent to the reference answer, prompts dependent on unavailable figures, and source rows whose stored n_correct disagreed with the current grader reconstruction.

Because the released schema intentionally omits all abstract-hint and code-hint fields, use the linked source dataset and the pinned revision if the hints are needed for auditing.

Downloads last month
19