question stringclasses 10
values | metadata stringclasses 10
values | task_source stringclasses 1
value | formatted_prompt listlengths 2 2 | response listlengths 1 1 | formatted_prompt__metadata listlengths 1 1 | response__metadata listlengths 1 1 | response__extractions listlengths 1 1 | response__extractions_metadata listlengths 1 1 | eval_correct listlengths 1 1 | eval_correct__metadata listlengths 1 1 | _judge_prompt__metadata listlengths 1 1 | _judge_response__metadata listlengths 1 1 | strategy_compliance int64 1 1 | compliance_reasoning stringclasses 10
values | compliance_elements_followed listlengths 0 0 | compliance_elements_missed listlengths 8 8 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
"Using ALL of the numbers 28, 98, 44, 14, 69, 49, 7, 46, create an expression that equals 303. You m(...TRUNCATED) | "{\"numbers\": [28, 98, 44, 14, 69, 49, 7, 46], \"target\": 303, \"operators_used\": [\"*\", \"*\", (...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["We need to find an expression using numbers 28,98,44,14,69,49,7,46 exactly once each, with operati(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"length","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Thi(...TRUNCATED) | [
[]
] | [
[]
] | [
[]
] | [
[]
] | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0052035,"finish_reason":"stop","input_tokens":33890,"model":"gpt-4o-mini-2024-07-18","out(...TRUNCATED) | 1 | "The model completely disregards the prescribed reasoning strategy and instead relies on intuition a(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 25, 69, 58, 14, 55, 24, 36, 60, create an expression that equals 176. You (...TRUNCATED) | "{\"numbers\": [25, 69, 58, 14, 55, 24, 36, 60], \"target\": 176, \"operators_used\": [\"-\", \"+\",(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["<think>\nWe need to find an expression using numbers 25,69,58,14,55,24,36,60 exactly once each, wi(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"stop","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Think(...TRUNCATED) | [["(69+55+60+14+36-58)*(25-24) = 176","10 + 5 * 2 = 20","(69+55+60+14+36-58)*(25-24) = 176","(69+55+(...TRUNCATED) | [[{"end_index":17753,"end_phrase":"</answer>","start_index":17703,"start_phrase":"<answer>"},{"end_i(...TRUNCATED) | [
[
true,
false,
true,
true,
true
]
] | [[{"answer_block":"(69+55+60+14+36-58)*(25-24) = 176","error":null,"final_answer":176.0,"is_correct"(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0012640499999999999,"finish_reason":"stop","input_tokens":7647,"model":"gpt-4o-mini-2024-(...TRUNCATED) | 1 | "The model completely ignores the prescribed strategy and does not follow the systematic tree search(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 55, 36, 6, 3, 14, 99, 17, 82, create an expression that equals 245. You mu(...TRUNCATED) | "{\"numbers\": [55, 36, 6, 3, 14, 99, 17, 82], \"target\": 245, \"operators_used\": [\"*\", \"-\", \(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["We need to solve a puzzle: Using numbers 55, 36, 6, 3, 14, 99, 17, 82, each exactly once, with ope(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"length","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Thi(...TRUNCATED) | [
[]
] | [
[]
] | [
[]
] | [
[]
] | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.00520725,"finish_reason":"stop","input_tokens":33919,"model":"gpt-4o-mini-2024-07-18","ou(...TRUNCATED) | 1 | "The model completely disregards the prescribed strategy and instead relies on intuition and unstruc(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 87, 69, 43, 4, 55, 34, 23, 2, create an expression that equals 874. You mu(...TRUNCATED) | "{\"numbers\": [87, 69, 43, 4, 55, 34, 23, 2], \"target\": 874, \"operators_used\": [\"*\", \"+\", \(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["<think>\nWe need to create an expression using all numbers 87, 69, 43, 4, 55, 34, 23, 2 exactly on(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"stop","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Think(...TRUNCATED) | [["23 * ((87 - 55) + (34 - 4) - (69 - 43 - 2)) = 874","23 * ((87 - 55) + (34 - 4) - (69 - 43 - 2)) =(...TRUNCATED) | [[{"end_index":61033,"end_phrase":"</answer>","start_index":60967,"start_phrase":"<answer>"},{"end_i(...TRUNCATED) | [
[
true,
true
]
] | [[{"answer_block":"23 * ((87 - 55) + (34 - 4) - (69 - 43 - 2)) = 874","error":null,"final_answer":87(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.004122300000000001,"finish_reason":"stop","input_tokens":26686,"model":"gpt-4o-mini-2024-(...TRUNCATED) | 1 | "The model completely disregards the prescribed strategy and instead relies on intuition and ad-hoc (...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 56, 63, 7, 50, 44, 1, 87, 52, create an expression that equals 132. You mu(...TRUNCATED) | "{\"numbers\": [56, 63, 7, 50, 44, 1, 87, 52], \"target\": 132, \"operators_used\": [\"-\", \"/\", \(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["We need to create an expression using all numbers exactly once: 56, 63, 7, 50, 44, 1, 87, 52, to e(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"length","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Thi(...TRUNCATED) | [
[]
] | [
[]
] | [
[]
] | [
[]
] | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0052086,"finish_reason":"stop","input_tokens":33908,"model":"gpt-4o-mini-2024-07-18","out(...TRUNCATED) | 1 | "The model completely disregards the prescribed strategy and instead relies on ad-hoc reasoning and (...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 32, 32, 36, 58, 2, 97, 60, 11, create an expression that equals 409. You m(...TRUNCATED) | "{\"numbers\": [32, 32, 36, 58, 2, 97, 60, 11], \"target\": 409, \"operators_used\": [\"/\", \"*\", (...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["<think>\nWe need to find an expression using numbers: 32, 32, 36, 58, 2, 97, 60, 11. Use each exac(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"stop","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Think(...TRUNCATED) | [["(97+2) * (36-32) * (32/32) + (60-58) + 11 = 409","(97+2)*(36-32)*(32/32)+(60-58)+11 = 409","... e(...TRUNCATED) | [[{"end_index":21229,"end_phrase":"</answer>","start_index":21165,"start_phrase":"<answer>"},{"end_i(...TRUNCATED) | [
[
false,
false,
false,
false,
false
]
] | [[{"answer_block":"(97+2) * (36-32) * (32/32) + (60-58) + 11 = 409","error":"Expression uses an unav(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0014529,"finish_reason":"stop","input_tokens":8874,"model":"gpt-4o-mini-2024-07-18","outp(...TRUNCATED) | 1 | "The model completely ignores the systematic tree search strategy and instead relies on ad-hoc reaso(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 38, 30, 47, 23, 39, 2, 91, 1, create an expression that equals 785. You mu(...TRUNCATED) | "{\"numbers\": [38, 30, 47, 23, 39, 2, 91, 1], \"target\": 785, \"operators_used\": [\"-\", \"*\", \(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["<think>\nWe need to find an expression using numbers: 38, 30, 47, 23, 39, 2, 91, 1, each exactly o(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"stop","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Think(...TRUNCATED) | [["tags, as a single arithmetic expression using the given numbers that equals the target. Use each (...TRUNCATED) | [[{"end_index":24084,"end_phrase":"</answer>","start_index":23647,"start_phrase":"<answer>"},{"end_i(...TRUNCATED) | [
[
false,
true,
true
]
] | [[{"answer_block":"tags, as a single arithmetic expression using the given numbers that equals the t(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0016912499999999998,"finish_reason":"stop","input_tokens":10459,"model":"gpt-4o-mini-2024(...TRUNCATED) | 1 | "The model completely disregards the prescribed strategy and instead relies on intuition and heurist(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 13, 1, 13, 68, 2, 2, 93, 62, create an expression that equals 102. You mus(...TRUNCATED) | "{\"numbers\": [13, 1, 13, 68, 2, 2, 93, 62], \"target\": 102, \"operators_used\": [\"/\", \"-\", \"(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["<think>\nWe need to find an expression using numbers: 13, 1, 13, 68, 2, 2, 93, 62. Use each exactl(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"stop","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Think(...TRUNCATED) | [["(93 * 68) / 62 * ((13 + 2) / (13 + 2)) * 1 = 102","tags.\" So we need to show reasoning then fina(...TRUNCATED) | [[{"end_index":11785,"end_phrase":"</answer>","start_index":11720,"start_phrase":"<answer>"},{"end_i(...TRUNCATED) | [
[
true,
false,
true,
true
]
] | [[{"answer_block":"(93 * 68) / 62 * ((13 + 2) / (13 + 2)) * 1 = 102","error":null,"final_answer":102(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0010083,"finish_reason":"stop","input_tokens":5874,"model":"gpt-4o-mini-2024-07-18","outp(...TRUNCATED) | 1 | "The model completely disregards the prescribed strategy and instead relies on intuition and ad-hoc (...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 87, 1, 21, 9, 57, 45, 6, 17, create an expression that equals 113. You mus(...TRUNCATED) | "{\"numbers\": [87, 1, 21, 9, 57, 45, 6, 17], \"target\": 113, \"operators_used\": [\"/\", \"*\", \"(...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["<think>\nWe need to use numbers: 87, 1, 21, 9, 57, 45, 6, 17 to form expression equals 113. Use ea(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"stop","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Think(...TRUNCATED) | [["57 * ((87 - 45) / 21) - (17 - (9 + 6 + 1)) = 113","tags.\n\nThus, final answer: <answer>57 * ((87(...TRUNCATED) | [[{"end_index":78196,"end_phrase":"</answer>","start_index":78131,"start_phrase":"<answer>"},{"end_i(...TRUNCATED) | [
[
true,
false,
true
]
] | [[{"answer_block":"57 * ((87 - 45) / 21) - (17 - (9 + 6 + 1)) = 113","error":null,"final_answer":113(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0048725999999999995,"finish_reason":"stop","input_tokens":31632,"model":"gpt-4o-mini-2024(...TRUNCATED) | 1 | "The model completely disregards the prescribed strategy and instead relies on an unstructured, intu(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
"Using ALL of the numbers 60, 90, 39, 30, 53, 15, 18, 6, create an expression that equals 239. You m(...TRUNCATED) | "{\"numbers\": [60, 90, 39, 30, 53, 15, 18, 6], \"target\": 239, \"operators_used\": [\"+\", \"*\", (...TRUNCATED) | countdown | [{"content":"You are a math problem solver. Solve the given arithmetic problem step by step. Show yo(...TRUNCATED) | ["We need to create an expression using all numbers exactly once: 60, 90, 39, 30, 53, 15, 18, 6. Use(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/moonshotai/Kimi-K2-Th(...TRUNCATED) | [{"cost":0.0,"finish_reason":"length","input_tokens":255,"model":"together_ai/moonshotai/Kimi-K2-Thi(...TRUNCATED) | [
[]
] | [
[]
] | [
[]
] | [
[]
] | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"gpt-4o-mini-2024-07-18","raw_prom(...TRUNCATED) | [{"cost":0.0052116,"finish_reason":"stop","input_tokens":33900,"model":"gpt-4o-mini-2024-07-18","out(...TRUNCATED) | 1 | "The model completely ignores the prescribed strategy and instead relies on intuition and ad-hoc rea(...TRUNCATED) | [] | ["Lists all available numbers and target at the start (INVENTORY step)","Picks pairs of numbers and (...TRUNCATED) |
t1-strategy-countdown-anti-kimi-k2-thinking-kimi-think
Strategy compliance evaluation on countdown arithmetic — anti-strategy variant.
Model was explicitly told NOT to use tree search (anti-strategy control). Judge still scores against tree search rubric. Compliance is scored by an LLM judge (1-5 Likert).
Results
| Metric | Value |
|---|---|
| pass@1 | 0.5000 |
| Strategy compliance (mean) | 1.00 |
| Strategy compliance (min) | 1 |
| Strategy compliance (max) | 1 |
| Total problems | 10 |
Compliance Distribution
| Score | Count | Meaning |
|---|---|---|
| 1 | 10 | Non-compliant |
| 2 | 0 | Minimal |
| 3 | 0 | Partial |
| 4 | 0 | Mostly compliant |
| 5 | 0 | Fully compliant |
Details
| Parameter | Value |
|---|---|
| Model | together_ai/moonshotai/Kimi-K2-Thinking |
| Thinking | True |
| Temperature | 0.7 |
| Max tokens | 32768 |
| Judge model | openai/gpt-4o-mini |
| Num examples | 10 |
| Num args | 8 |
| Variant | anti-strategy |
Usage
from datasets import load_dataset
ds = load_dataset("reasoning-degeneration-dev/t1-strategy-countdown-anti-kimi-k2-thinking-kimi-think", split="train")
scores = ds["strategy_compliance"]
print(f"Mean compliance: {sum(scores) / len(scores):.2f}")
Tracked in reasoning-degeneration-dev/PROJECT-MANIFEST
- Downloads last month
- 7