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": 5051
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007045",
"normalized_example_id": "ex_007045",
"source_idx": 7045,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6314,
... |
[
{
"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": 6314
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007045",
"normalized_example_id": "ex_007045",
"source_idx": 7045,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6314,
... |
[
{
"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": "11",
"budget": 2935
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007047",
"normalized_example_id": "ex_007047",
"source_idx": 7047,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4892,
... |
[
{
"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": "11",
"budget": 3914
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007047",
"normalized_example_id": "ex_007047",
"source_idx": 7047,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4892,
... |
[
{
"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": "11",
"budget": 4892
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007047",
"normalized_example_id": "ex_007047",
"source_idx": 7047,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4892,
... |
[
{
"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": "B",
"budget": 3881
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007050",
"normalized_example_id": "ex_007050",
"source_idx": 7050,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6469,
... |
[
{
"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": "B",
"budget": 5175
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007050",
"normalized_example_id": "ex_007050",
"source_idx": 7050,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6469,
... |
[
{
"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": "B",
"budget": 6469
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007050",
"normalized_example_id": "ex_007050",
"source_idx": 7050,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6469,
... |
[
{
"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{(2n)!}{n!(n+1)!}",
"budget": 2113
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007051",
"normalized_example_id": "ex_007051",
"source_idx": 7051,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3522,
... |
[
{
"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{(2n)!}{n!(n+1)!}",
"budget": 2818
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007051",
"normalized_example_id": "ex_007051",
"source_idx": 7051,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3522,
... |
[
{
"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{(2n)!}{n!(n+1)!}",
"budget": 3522
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007051",
"normalized_example_id": "ex_007051",
"source_idx": 7051,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3522,
... |
[
{
"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": "37",
"budget": 2867
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007053",
"normalized_example_id": "ex_007053",
"source_idx": 7053,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4779,
... |
[
{
"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": "37",
"budget": 3823
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007053",
"normalized_example_id": "ex_007053",
"source_idx": 7053,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4779,
... |
[
{
"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": "37",
"budget": 4779
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007053",
"normalized_example_id": "ex_007053",
"source_idx": 7053,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4779,
... |
[
{
"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": "(4x-1)\\cos4x-\\sin4x+C",
"budget": 1517
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007054",
"normalized_example_id": "ex_007054",
"source_idx": 7054,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2528,
... |
[
{
"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": "(4x-1)\\cos4x-\\sin4x+C",
"budget": 2022
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007054",
"normalized_example_id": "ex_007054",
"source_idx": 7054,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2528,
... |
[
{
"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": "(4x-1)\\cos4x-\\sin4x+C",
"budget": 2528
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007054",
"normalized_example_id": "ex_007054",
"source_idx": 7054,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2528,
... |
[
{
"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": "(-2,2,-3,1)",
"budget": 1829
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007057",
"normalized_example_id": "ex_007057",
"source_idx": 7057,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3048,
... |
[
{
"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": "(-2,2,-3,1)",
"budget": 2438
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007057",
"normalized_example_id": "ex_007057",
"source_idx": 7057,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3048,
... |
[
{
"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": "(-2,2,-3,1)",
"budget": 3048
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007057",
"normalized_example_id": "ex_007057",
"source_idx": 7057,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3048,
... |
[
{
"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": "1,2",
"budget": 3807
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007061",
"normalized_example_id": "ex_007061",
"source_idx": 7061,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6345,
... |
[
{
"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": "1,2",
"budget": 5076
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007061",
"normalized_example_id": "ex_007061",
"source_idx": 7061,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6345,
... |
[
{
"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": "1,2",
"budget": 6345
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007061",
"normalized_example_id": "ex_007061",
"source_idx": 7061,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6345,
... |
[
{
"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": "2870",
"budget": 3864
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007062",
"normalized_example_id": "ex_007062",
"source_idx": 7062,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6440,
... |
[
{
"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": "2870",
"budget": 5152
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007062",
"normalized_example_id": "ex_007062",
"source_idx": 7062,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6440,
... |
[
{
"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": "2870",
"budget": 6440
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007062",
"normalized_example_id": "ex_007062",
"source_idx": 7062,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6440,
... |
[
{
"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{\\mathrm{e}}{2}",
"budget": 1850
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007063",
"normalized_example_id": "ex_007063",
"source_idx": 7063,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3084,
... |
[
{
"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{\\mathrm{e}}{2}",
"budget": 2467
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007063",
"normalized_example_id": "ex_007063",
"source_idx": 7063,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3084,
... |
[
{
"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{\\mathrm{e}}{2}",
"budget": 3084
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007063",
"normalized_example_id": "ex_007063",
"source_idx": 7063,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3084,
... |
[
{
"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": "7",
"budget": 1063
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007064",
"normalized_example_id": "ex_007064",
"source_idx": 7064,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1771,
... |
[
{
"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": "7",
"budget": 1417
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007064",
"normalized_example_id": "ex_007064",
"source_idx": 7064,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1771,
... |
[
{
"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": "7",
"budget": 1771
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007064",
"normalized_example_id": "ex_007064",
"source_idx": 7064,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1771,
... |
[
{
"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{4\\sqrt{6}}{23}",
"budget": 1510
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007065",
"normalized_example_id": "ex_007065",
"source_idx": 7065,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2517,
... |
[
{
"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{4\\sqrt{6}}{23}",
"budget": 2014
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007065",
"normalized_example_id": "ex_007065",
"source_idx": 7065,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2517,
... |
[
{
"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{4\\sqrt{6}}{23}",
"budget": 2517
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007065",
"normalized_example_id": "ex_007065",
"source_idx": 7065,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2517,
... |
[
{
"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": "2a^4",
"budget": 3572
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007066",
"normalized_example_id": "ex_007066",
"source_idx": 7066,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5954,
... |
[
{
"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": "2a^4",
"budget": 4763
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007066",
"normalized_example_id": "ex_007066",
"source_idx": 7066,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5954,
... |
[
{
"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": "2a^4",
"budget": 5954
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007066",
"normalized_example_id": "ex_007066",
"source_idx": 7066,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5954,
... |
[
{
"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": "246",
"budget": 3633
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007067",
"normalized_example_id": "ex_007067",
"source_idx": 7067,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6055,
... |
[
{
"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": "246",
"budget": 4844
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007067",
"normalized_example_id": "ex_007067",
"source_idx": 7067,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6055,
... |
[
{
"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": "246",
"budget": 6055
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007067",
"normalized_example_id": "ex_007067",
"source_idx": 7067,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6055,
... |
[
{
"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": "K_{\\max}=\\frac{9}{16}",
"budget": 2359
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007070",
"normalized_example_id": "ex_007070",
"source_idx": 7070,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3931,
... |
[
{
"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": "K_{\\max}=\\frac{9}{16}",
"budget": 3145
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007070",
"normalized_example_id": "ex_007070",
"source_idx": 7070,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3931,
... |
[
{
"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": "K_{\\max}=\\frac{9}{16}",
"budget": 3931
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007070",
"normalized_example_id": "ex_007070",
"source_idx": 7070,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3931,
... |
[
{
"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": "252",
"budget": 1223
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007071",
"normalized_example_id": "ex_007071",
"source_idx": 7071,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2038,
... |
[
{
"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": "252",
"budget": 1630
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007071",
"normalized_example_id": "ex_007071",
"source_idx": 7071,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2038,
... |
[
{
"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": "252",
"budget": 2038
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007071",
"normalized_example_id": "ex_007071",
"source_idx": 7071,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2038,
... |
[
{
"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{2}{3}",
"budget": 5339
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007075",
"normalized_example_id": "ex_007075",
"source_idx": 7075,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8899,
... |
[
{
"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{2}{3}",
"budget": 7119
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007075",
"normalized_example_id": "ex_007075",
"source_idx": 7075,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8899,
... |
[
{
"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{2}{3}",
"budget": 8899
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007075",
"normalized_example_id": "ex_007075",
"source_idx": 7075,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8899,
... |
[
{
"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": "8281",
"budget": 2062
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007076",
"normalized_example_id": "ex_007076",
"source_idx": 7076,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3436,
... |
[
{
"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": "8281",
"budget": 2749
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007076",
"normalized_example_id": "ex_007076",
"source_idx": 7076,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3436,
... |
[
{
"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": "8281",
"budget": 3436
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007076",
"normalized_example_id": "ex_007076",
"source_idx": 7076,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3436,
... |
[
{
"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": "0.375",
"budget": 3515
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007077",
"normalized_example_id": "ex_007077",
"source_idx": 7077,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5859,
... |
[
{
"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": "0.375",
"budget": 4687
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007077",
"normalized_example_id": "ex_007077",
"source_idx": 7077,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5859,
... |
[
{
"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": "0.375",
"budget": 5859
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007077",
"normalized_example_id": "ex_007077",
"source_idx": 7077,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5859,
... |
[
{
"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": "K = 8\\sqrt{95}",
"budget": 3668
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007079",
"normalized_example_id": "ex_007079",
"source_idx": 7079,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6113,
... |
[
{
"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": "K = 8\\sqrt{95}",
"budget": 4890
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007079",
"normalized_example_id": "ex_007079",
"source_idx": 7079,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6113,
... |
[
{
"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": "K = 8\\sqrt{95}",
"budget": 6113
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007079",
"normalized_example_id": "ex_007079",
"source_idx": 7079,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6113,
... |
[
{
"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": "2011",
"budget": 3354
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007083",
"normalized_example_id": "ex_007083",
"source_idx": 7083,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5590,
... |
[
{
"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": "2011",
"budget": 4472
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007083",
"normalized_example_id": "ex_007083",
"source_idx": 7083,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5590,
... |
[
{
"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": "2011",
"budget": 5590
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007083",
"normalized_example_id": "ex_007083",
"source_idx": 7083,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5590,
... |
[
{
"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": 940
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007084",
"normalized_example_id": "ex_007084",
"source_idx": 7084,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1566,
... |
[
{
"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": 1253
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007084",
"normalized_example_id": "ex_007084",
"source_idx": 7084,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1566,
... |
[
{
"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": 1566
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007084",
"normalized_example_id": "ex_007084",
"source_idx": 7084,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1566,
... |
[
{
"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": "401802",
"budget": 3477
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007086",
"normalized_example_id": "ex_007086",
"source_idx": 7086,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5795,
... |
[
{
"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": "401802",
"budget": 4636
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007086",
"normalized_example_id": "ex_007086",
"source_idx": 7086,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5795,
... |
[
{
"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": "401802",
"budget": 5795
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007086",
"normalized_example_id": "ex_007086",
"source_idx": 7086,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5795,
... |
[
{
"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 \\leq a \\leq 6",
"budget": 856
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007087",
"normalized_example_id": "ex_007087",
"source_idx": 7087,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1426,
... |
[
{
"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 \\leq a \\leq 6",
"budget": 1141
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007087",
"normalized_example_id": "ex_007087",
"source_idx": 7087,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1426,
... |
[
{
"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 \\leq a \\leq 6",
"budget": 1426
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007087",
"normalized_example_id": "ex_007087",
"source_idx": 7087,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1426,
... |
[
{
"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": "C",
"budget": 898
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007089",
"normalized_example_id": "ex_007089",
"source_idx": 7089,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1497,
... |
[
{
"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": "C",
"budget": 1198
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007089",
"normalized_example_id": "ex_007089",
"source_idx": 7089,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1497,
... |
[
{
"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": "C",
"budget": 1497
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007089",
"normalized_example_id": "ex_007089",
"source_idx": 7089,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1497,
... |
[
{
"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": 4441
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007091",
"normalized_example_id": "ex_007091",
"source_idx": 7091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7401,
... |
[
{
"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": 5921
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007091",
"normalized_example_id": "ex_007091",
"source_idx": 7091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7401,
... |
[
{
"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": 7401
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007091",
"normalized_example_id": "ex_007091",
"source_idx": 7091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7401,
... |
[
{
"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}{2^{1984}}",
"budget": 5445
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007093",
"normalized_example_id": "ex_007093",
"source_idx": 7093,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9075,
... |
[
{
"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}{2^{1984}}",
"budget": 7260
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007093",
"normalized_example_id": "ex_007093",
"source_idx": 7093,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9075,
... |
[
{
"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}{2^{1984}}",
"budget": 9075
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007093",
"normalized_example_id": "ex_007093",
"source_idx": 7093,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9075,
... |
[
{
"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": "1210",
"budget": 1387
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007094",
"normalized_example_id": "ex_007094",
"source_idx": 7094,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2311,
... |
[
{
"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": "1210",
"budget": 1849
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007094",
"normalized_example_id": "ex_007094",
"source_idx": 7094,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2311,
... |
[
{
"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": "1210",
"budget": 2311
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007094",
"normalized_example_id": "ex_007094",
"source_idx": 7094,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2311,
... |
[
{
"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": "2022",
"budget": 1817
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007095",
"normalized_example_id": "ex_007095",
"source_idx": 7095,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3029,
... |
[
{
"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": "2022",
"budget": 2423
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007095",
"normalized_example_id": "ex_007095",
"source_idx": 7095,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3029,
... |
[
{
"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": "2022",
"budget": 3029
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007095",
"normalized_example_id": "ex_007095",
"source_idx": 7095,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3029,
... |
[
{
"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": "107",
"budget": 5489
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007096",
"normalized_example_id": "ex_007096",
"source_idx": 7096,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9148,
... |
[
{
"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": "107",
"budget": 7318
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007096",
"normalized_example_id": "ex_007096",
"source_idx": 7096,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9148,
... |
[
{
"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": "107",
"budget": 9148
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007096",
"normalized_example_id": "ex_007096",
"source_idx": 7096,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9148,
... |
[
{
"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{3}{16}",
"budget": 1249
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007099",
"normalized_example_id": "ex_007099",
"source_idx": 7099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2082,
... |
[
{
"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{3}{16}",
"budget": 1666
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007099",
"normalized_example_id": "ex_007099",
"source_idx": 7099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2082,
... |
[
{
"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{3}{16}",
"budget": 2082
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007099",
"normalized_example_id": "ex_007099",
"source_idx": 7099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2082,
... |
[
{
"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-(\\frac{9}{10})^{n}",
"budget": 1231
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007100",
"normalized_example_id": "ex_007100",
"source_idx": 7100,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2052,
... |
[
{
"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-(\\frac{9}{10})^{n}",
"budget": 1642
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007100",
"normalized_example_id": "ex_007100",
"source_idx": 7100,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2052,
... |
[
{
"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-(\\frac{9}{10})^{n}",
"budget": 2052
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007100",
"normalized_example_id": "ex_007100",
"source_idx": 7100,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2052,
... |
[
{
"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{4 \\sqrt{11}}{11}",
"budget": 3790
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007101",
"normalized_example_id": "ex_007101",
"source_idx": 7101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6316,
... |
[
{
"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{4 \\sqrt{11}}{11}",
"budget": 5053
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007101",
"normalized_example_id": "ex_007101",
"source_idx": 7101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6316,
... |
[
{
"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{4 \\sqrt{11}}{11}",
"budget": 6316
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007101",
"normalized_example_id": "ex_007101",
"source_idx": 7101,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6316,
... |
[
{
"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": "(999;998),(999;-998),(-999;-998),(-999;998)",
"budget": 1423
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007106",
"normalized_example_id": "ex_007106",
"source_idx": 7106,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2371,
... |
[
{
"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": "(999;998),(999;-998),(-999;-998),(-999;998)",
"budget": 1897
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_007106",
"normalized_example_id": "ex_007106",
"source_idx": 7106,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2371,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.