prompt listlengths 2 2 | label dict | metadata dict |
|---|---|---|
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 1376
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005074",
"normalized_example_id": "ex_005074",
"source_idx": 5074,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 1835
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005074",
"normalized_example_id": "ex_005074",
"source_idx": 5074,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 2294
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005074",
"normalized_example_id": "ex_005074",
"source_idx": 5074,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n\\in{2;3;6}",
"budget": 5178
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005079",
"normalized_example_id": "ex_005079",
"source_idx": 5079,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8630,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n\\in{2;3;6}",
"budget": 6904
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005079",
"normalized_example_id": "ex_005079",
"source_idx": 5079,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8630,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n\\in{2;3;6}",
"budget": 8630
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005079",
"normalized_example_id": "ex_005079",
"source_idx": 5079,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8630,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "180",
"budget": 1005
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005081",
"normalized_example_id": "ex_005081",
"source_idx": 5081,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1675,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "180",
"budget": 1340
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005081",
"normalized_example_id": "ex_005081",
"source_idx": 5081,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1675,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "180",
"budget": 1675
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005081",
"normalized_example_id": "ex_005081",
"source_idx": 5081,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1675,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 2921
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005085",
"normalized_example_id": "ex_005085",
"source_idx": 5085,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4868,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 3894
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005085",
"normalized_example_id": "ex_005085",
"source_idx": 5085,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4868,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 4868
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005085",
"normalized_example_id": "ex_005085",
"source_idx": 5085,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4868,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "45^\\circ, 45^\\circ, 90^\\circ",
"budget": 1720
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005086",
"normalized_example_id": "ex_005086",
"source_idx": 5086,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2866,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "45^\\circ, 45^\\circ, 90^\\circ",
"budget": 2293
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005086",
"normalized_example_id": "ex_005086",
"source_idx": 5086,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2866,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "45^\\circ, 45^\\circ, 90^\\circ",
"budget": 2866
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005086",
"normalized_example_id": "ex_005086",
"source_idx": 5086,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2866,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0,1,2",
"budget": 2938
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005087",
"normalized_example_id": "ex_005087",
"source_idx": 5087,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4897,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0,1,2",
"budget": 3918
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005087",
"normalized_example_id": "ex_005087",
"source_idx": 5087,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4897,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0,1,2",
"budget": 4897
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005087",
"normalized_example_id": "ex_005087",
"source_idx": 5087,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4897,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "42",
"budget": 1339
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005091",
"normalized_example_id": "ex_005091",
"source_idx": 5091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2231,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "42",
"budget": 1785
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005091",
"normalized_example_id": "ex_005091",
"source_idx": 5091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2231,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "42",
"budget": 2231
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005091",
"normalized_example_id": "ex_005091",
"source_idx": 5091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2231,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2000",
"budget": 4018
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005093",
"normalized_example_id": "ex_005093",
"source_idx": 5093,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6696,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2000",
"budget": 5357
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005093",
"normalized_example_id": "ex_005093",
"source_idx": 5093,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6696,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2000",
"budget": 6696
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005093",
"normalized_example_id": "ex_005093",
"source_idx": 5093,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6696,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{3}+\\log _{9} 2",
"budget": 3922
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005098",
"normalized_example_id": "ex_005098",
"source_idx": 5098,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6536,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{3}+\\log _{9} 2",
"budget": 5229
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005098",
"normalized_example_id": "ex_005098",
"source_idx": 5098,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6536,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{3}+\\log _{9} 2",
"budget": 6536
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005098",
"normalized_example_id": "ex_005098",
"source_idx": 5098,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6536,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1+\\sqrt{5}}{2}",
"budget": 1569
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005099",
"normalized_example_id": "ex_005099",
"source_idx": 5099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2615,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1+\\sqrt{5}}{2}",
"budget": 2092
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005099",
"normalized_example_id": "ex_005099",
"source_idx": 5099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2615,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1+\\sqrt{5}}{2}",
"budget": 2615
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005099",
"normalized_example_id": "ex_005099",
"source_idx": 5099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2615,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-\\sqrt{2};\\sqrt{2}+2\\sin2]",
"budget": 5074
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005100",
"normalized_example_id": "ex_005100",
"source_idx": 5100,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8457,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-\\sqrt{2};\\sqrt{2}+2\\sin2]",
"budget": 6766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005100",
"normalized_example_id": "ex_005100",
"source_idx": 5100,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8457,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[-\\sqrt{2};\\sqrt{2}+2\\sin2]",
"budget": 8457
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005100",
"normalized_example_id": "ex_005100",
"source_idx": 5100,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8457,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,3,24) \\text{ and } (2,4,8)",
"budget": 4232
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005101",
"normalized_example_id": "ex_005101",
"source_idx": 5101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7053,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,3,24) \\text{ and } (2,4,8)",
"budget": 5642
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005101",
"normalized_example_id": "ex_005101",
"source_idx": 5101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7053,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,3,24) \\text{ and } (2,4,8)",
"budget": 7053
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005101",
"normalized_example_id": "ex_005101",
"source_idx": 5101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7053,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 748
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005103",
"normalized_example_id": "ex_005103",
"source_idx": 5103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 997
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005103",
"normalized_example_id": "ex_005103",
"source_idx": 5103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 1246
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005103",
"normalized_example_id": "ex_005103",
"source_idx": 5103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = n",
"budget": 4179
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005105",
"normalized_example_id": "ex_005105",
"source_idx": 5105,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6965,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = n",
"budget": 5572
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005105",
"normalized_example_id": "ex_005105",
"source_idx": 5105,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6965,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = n",
"budget": 6965
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005105",
"normalized_example_id": "ex_005105",
"source_idx": 5105,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6965,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 3383
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005107",
"normalized_example_id": "ex_005107",
"source_idx": 5107,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5638,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 4510
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005107",
"normalized_example_id": "ex_005107",
"source_idx": 5107,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5638,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "12",
"budget": 5638
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005107",
"normalized_example_id": "ex_005107",
"source_idx": 5107,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5638,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8",
"budget": 4618
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005108",
"normalized_example_id": "ex_005108",
"source_idx": 5108,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7696,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8",
"budget": 6157
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005108",
"normalized_example_id": "ex_005108",
"source_idx": 5108,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7696,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8",
"budget": 7696
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005108",
"normalized_example_id": "ex_005108",
"source_idx": 5108,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7696,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4",
"budget": 3923
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005111",
"normalized_example_id": "ex_005111",
"source_idx": 5111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6539,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4",
"budget": 5231
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005111",
"normalized_example_id": "ex_005111",
"source_idx": 5111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6539,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4",
"budget": 6539
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005111",
"normalized_example_id": "ex_005111",
"source_idx": 5111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6539,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = 2^n + 1",
"budget": 886
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005113",
"normalized_example_id": "ex_005113",
"source_idx": 5113,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1476,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = 2^n + 1",
"budget": 1181
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005113",
"normalized_example_id": "ex_005113",
"source_idx": 5113,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1476,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_n = 2^n + 1",
"budget": 1476
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005113",
"normalized_example_id": "ex_005113",
"source_idx": 5113,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1476,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{5\\sqrt{7}}{14}",
"budget": 2346
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005116",
"normalized_example_id": "ex_005116",
"source_idx": 5116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3910,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{5\\sqrt{7}}{14}",
"budget": 3128
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005116",
"normalized_example_id": "ex_005116",
"source_idx": 5116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3910,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{5\\sqrt{7}}{14}",
"budget": 3910
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005116",
"normalized_example_id": "ex_005116",
"source_idx": 5116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3910,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "171",
"budget": 1828
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005120",
"normalized_example_id": "ex_005120",
"source_idx": 5120,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3046,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "171",
"budget": 2437
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005120",
"normalized_example_id": "ex_005120",
"source_idx": 5120,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3046,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "171",
"budget": 3046
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005120",
"normalized_example_id": "ex_005120",
"source_idx": 5120,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3046,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 5422
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005122",
"normalized_example_id": "ex_005122",
"source_idx": 5122,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9036,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 7229
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005122",
"normalized_example_id": "ex_005122",
"source_idx": 5122,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9036,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 9036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005122",
"normalized_example_id": "ex_005122",
"source_idx": 5122,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9036,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2013",
"budget": 4667
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005123",
"normalized_example_id": "ex_005123",
"source_idx": 5123,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7779,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2013",
"budget": 6223
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005123",
"normalized_example_id": "ex_005123",
"source_idx": 5123,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7779,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2013",
"budget": 7779
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005123",
"normalized_example_id": "ex_005123",
"source_idx": 5123,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7779,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "72",
"budget": 5806
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005125",
"normalized_example_id": "ex_005125",
"source_idx": 5125,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9677,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "72",
"budget": 7742
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005125",
"normalized_example_id": "ex_005125",
"source_idx": 5125,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9677,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "72",
"budget": 9677
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005125",
"normalized_example_id": "ex_005125",
"source_idx": 5125,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9677,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2, 1), (1, 2), (-1, -20), (-20, -1)",
"budget": 5576
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005127",
"normalized_example_id": "ex_005127",
"source_idx": 5127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2, 1), (1, 2), (-1, -20), (-20, -1)",
"budget": 7435
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005127",
"normalized_example_id": "ex_005127",
"source_idx": 5127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2, 1), (1, 2), (-1, -20), (-20, -1)",
"budget": 9294
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005127",
"normalized_example_id": "ex_005127",
"source_idx": 5127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 3635
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005129",
"normalized_example_id": "ex_005129",
"source_idx": 5129,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6059,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 4847
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005129",
"normalized_example_id": "ex_005129",
"source_idx": 5129,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6059,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 6059
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005129",
"normalized_example_id": "ex_005129",
"source_idx": 5129,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6059,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{\\sqrt{2}}",
"budget": 6114
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005130",
"normalized_example_id": "ex_005130",
"source_idx": 5130,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10190,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{\\sqrt{2}}",
"budget": 8152
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005130",
"normalized_example_id": "ex_005130",
"source_idx": 5130,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10190,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{\\sqrt{2}}",
"budget": 10190
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005130",
"normalized_example_id": "ex_005130",
"source_idx": 5130,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10190,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "259",
"budget": 4389
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005131",
"normalized_example_id": "ex_005131",
"source_idx": 5131,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7315,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "259",
"budget": 5852
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005131",
"normalized_example_id": "ex_005131",
"source_idx": 5131,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7315,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "259",
"budget": 7315
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005131",
"normalized_example_id": "ex_005131",
"source_idx": 5131,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7315,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4044120",
"budget": 1111
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005132",
"normalized_example_id": "ex_005132",
"source_idx": 5132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1851,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4044120",
"budget": 1481
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005132",
"normalized_example_id": "ex_005132",
"source_idx": 5132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1851,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4044120",
"budget": 1851
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005132",
"normalized_example_id": "ex_005132",
"source_idx": 5132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1851,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_{1}=\\frac{4}{9},\\quadb_{1}=-\\frac{20}{9},\\quadc_{1}=\\frac{100}{9},\\quada_{2}=4,\\quadb_{2}=12,\\quadc_{2}=36",
"budget": 4186
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005133",
"normalized_example_id": "ex_005133",
"source_idx": 5133,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6977,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_{1}=\\frac{4}{9},\\quadb_{1}=-\\frac{20}{9},\\quadc_{1}=\\frac{100}{9},\\quada_{2}=4,\\quadb_{2}=12,\\quadc_{2}=36",
"budget": 5582
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005133",
"normalized_example_id": "ex_005133",
"source_idx": 5133,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6977,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_{1}=\\frac{4}{9},\\quadb_{1}=-\\frac{20}{9},\\quadc_{1}=\\frac{100}{9},\\quada_{2}=4,\\quadb_{2}=12,\\quadc_{2}=36",
"budget": 6977
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005133",
"normalized_example_id": "ex_005133",
"source_idx": 5133,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6977,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "280",
"budget": 3400
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005134",
"normalized_example_id": "ex_005134",
"source_idx": 5134,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5667,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "280",
"budget": 4534
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005134",
"normalized_example_id": "ex_005134",
"source_idx": 5134,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5667,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "280",
"budget": 5667
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005134",
"normalized_example_id": "ex_005134",
"source_idx": 5134,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5667,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 3047
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005135",
"normalized_example_id": "ex_005135",
"source_idx": 5135,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5078,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 4062
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005135",
"normalized_example_id": "ex_005135",
"source_idx": 5135,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5078,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "10",
"budget": 5078
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005135",
"normalized_example_id": "ex_005135",
"source_idx": 5135,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5078,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1}{27}",
"budget": 4922
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005137",
"normalized_example_id": "ex_005137",
"source_idx": 5137,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8204,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1}{27}",
"budget": 6563
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005137",
"normalized_example_id": "ex_005137",
"source_idx": 5137,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8204,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{1}{27}",
"budget": 8204
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005137",
"normalized_example_id": "ex_005137",
"source_idx": 5137,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8204,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-6",
"budget": 2197
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005138",
"normalized_example_id": "ex_005138",
"source_idx": 5138,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3661,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-6",
"budget": 2929
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005138",
"normalized_example_id": "ex_005138",
"source_idx": 5138,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3661,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-6",
"budget": 3661
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005138",
"normalized_example_id": "ex_005138",
"source_idx": 5138,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3661,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1:1",
"budget": 2520
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_005140",
"normalized_example_id": "ex_005140",
"source_idx": 5140,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4200,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.