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": "2014",
"budget": 1319
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003651",
"normalized_example_id": "ex_003651",
"source_idx": 3651,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2199,
... |
[
{
"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": "2014",
"budget": 1759
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003651",
"normalized_example_id": "ex_003651",
"source_idx": 3651,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2199,
... |
[
{
"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": "2014",
"budget": 2199
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003651",
"normalized_example_id": "ex_003651",
"source_idx": 3651,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2199,
... |
[
{
"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": "\\arccos\\frac{1}{3}",
"budget": 2589
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003653",
"normalized_example_id": "ex_003653",
"source_idx": 3653,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4315,
... |
[
{
"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": "\\arccos\\frac{1}{3}",
"budget": 3452
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003653",
"normalized_example_id": "ex_003653",
"source_idx": 3653,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4315,
... |
[
{
"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": "\\arccos\\frac{1}{3}",
"budget": 4315
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003653",
"normalized_example_id": "ex_003653",
"source_idx": 3653,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4315,
... |
[
{
"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": "21!-1",
"budget": 760
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003656",
"normalized_example_id": "ex_003656",
"source_idx": 3656,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1266,
... |
[
{
"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": "21!-1",
"budget": 1013
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003656",
"normalized_example_id": "ex_003656",
"source_idx": 3656,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1266,
... |
[
{
"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": "21!-1",
"budget": 1266
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003656",
"normalized_example_id": "ex_003656",
"source_idx": 3656,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1266,
... |
[
{
"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": "17",
"budget": 2003
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003657",
"normalized_example_id": "ex_003657",
"source_idx": 3657,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3338,
... |
[
{
"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": "17",
"budget": 2670
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003657",
"normalized_example_id": "ex_003657",
"source_idx": 3657,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3338,
... |
[
{
"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": "17",
"budget": 3338
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003657",
"normalized_example_id": "ex_003657",
"source_idx": 3657,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3338,
... |
[
{
"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": "104",
"budget": 2721
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003658",
"normalized_example_id": "ex_003658",
"source_idx": 3658,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4535,
... |
[
{
"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": "104",
"budget": 3628
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003658",
"normalized_example_id": "ex_003658",
"source_idx": 3658,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4535,
... |
[
{
"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": "104",
"budget": 4535
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003658",
"normalized_example_id": "ex_003658",
"source_idx": 3658,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4535,
... |
[
{
"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{\\sqrt{12a^2+9b^2}}{6}",
"budget": 1927
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003660",
"normalized_example_id": "ex_003660",
"source_idx": 3660,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3211,
... |
[
{
"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{\\sqrt{12a^2+9b^2}}{6}",
"budget": 2569
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003660",
"normalized_example_id": "ex_003660",
"source_idx": 3660,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3211,
... |
[
{
"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{\\sqrt{12a^2+9b^2}}{6}",
"budget": 3211
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003660",
"normalized_example_id": "ex_003660",
"source_idx": 3660,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3211,
... |
[
{
"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": "71",
"budget": 5586
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003664",
"normalized_example_id": "ex_003664",
"source_idx": 3664,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9310,
... |
[
{
"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": "71",
"budget": 7448
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003664",
"normalized_example_id": "ex_003664",
"source_idx": 3664,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9310,
... |
[
{
"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": "71",
"budget": 9310
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003664",
"normalized_example_id": "ex_003664",
"source_idx": 3664,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9310,
... |
[
{
"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": "(48,3),(18,4),(12,5),(8,7),(6,10),(4,25)",
"budget": 1419
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003665",
"normalized_example_id": "ex_003665",
"source_idx": 3665,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2365,
... |
[
{
"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": "(48,3),(18,4),(12,5),(8,7),(6,10),(4,25)",
"budget": 1892
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003665",
"normalized_example_id": "ex_003665",
"source_idx": 3665,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2365,
... |
[
{
"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": "(48,3),(18,4),(12,5),(8,7),(6,10),(4,25)",
"budget": 2365
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003665",
"normalized_example_id": "ex_003665",
"source_idx": 3665,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2365,
... |
[
{
"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}",
"budget": 2119
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003666",
"normalized_example_id": "ex_003666",
"source_idx": 3666,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3532,
... |
[
{
"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}",
"budget": 2826
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003666",
"normalized_example_id": "ex_003666",
"source_idx": 3666,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3532,
... |
[
{
"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}",
"budget": 3532
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003666",
"normalized_example_id": "ex_003666",
"source_idx": 3666,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3532,
... |
[
{
"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": 4906
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003667",
"normalized_example_id": "ex_003667",
"source_idx": 3667,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8177,
... |
[
{
"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": 6542
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003667",
"normalized_example_id": "ex_003667",
"source_idx": 3667,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8177,
... |
[
{
"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": 8177
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003667",
"normalized_example_id": "ex_003667",
"source_idx": 3667,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8177,
... |
[
{
"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": "\\textbf{(E)}15",
"budget": 5967
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003669",
"normalized_example_id": "ex_003669",
"source_idx": 3669,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9945,
... |
[
{
"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": "\\textbf{(E)}15",
"budget": 7956
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003669",
"normalized_example_id": "ex_003669",
"source_idx": 3669,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9945,
... |
[
{
"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": "\\textbf{(E)}15",
"budget": 9945
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003669",
"normalized_example_id": "ex_003669",
"source_idx": 3669,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9945,
... |
[
{
"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\\sqrt{5}",
"budget": 3118
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003670",
"normalized_example_id": "ex_003670",
"source_idx": 3670,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5197,
... |
[
{
"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\\sqrt{5}",
"budget": 4158
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003670",
"normalized_example_id": "ex_003670",
"source_idx": 3670,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5197,
... |
[
{
"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\\sqrt{5}",
"budget": 5197
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003670",
"normalized_example_id": "ex_003670",
"source_idx": 3670,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5197,
... |
[
{
"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": "7.5",
"budget": 4222
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003672",
"normalized_example_id": "ex_003672",
"source_idx": 3672,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7036,
... |
[
{
"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": "7.5",
"budget": 5629
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003672",
"normalized_example_id": "ex_003672",
"source_idx": 3672,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7036,
... |
[
{
"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": "7.5",
"budget": 7036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003672",
"normalized_example_id": "ex_003672",
"source_idx": 3672,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7036,
... |
[
{
"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": "61",
"budget": 6016
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003678",
"normalized_example_id": "ex_003678",
"source_idx": 3678,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10026,
... |
[
{
"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": "61",
"budget": 8021
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003678",
"normalized_example_id": "ex_003678",
"source_idx": 3678,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10026,
... |
[
{
"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": "61",
"budget": 10026
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003678",
"normalized_example_id": "ex_003678",
"source_idx": 3678,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10026,
... |
[
{
"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{8}{3}",
"budget": 6093
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003679",
"normalized_example_id": "ex_003679",
"source_idx": 3679,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10155,
... |
[
{
"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{8}{3}",
"budget": 8124
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003679",
"normalized_example_id": "ex_003679",
"source_idx": 3679,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10155,
... |
[
{
"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{8}{3}",
"budget": 10155
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003679",
"normalized_example_id": "ex_003679",
"source_idx": 3679,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10155,
... |
[
{
"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": "P(x) = (x-1)^n",
"budget": 2797
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003681",
"normalized_example_id": "ex_003681",
"source_idx": 3681,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4662,
... |
[
{
"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": "P(x) = (x-1)^n",
"budget": 3730
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003681",
"normalized_example_id": "ex_003681",
"source_idx": 3681,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4662,
... |
[
{
"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": "P(x) = (x-1)^n",
"budget": 4662
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003681",
"normalized_example_id": "ex_003681",
"source_idx": 3681,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4662,
... |
[
{
"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": "24",
"budget": 4722
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003685",
"normalized_example_id": "ex_003685",
"source_idx": 3685,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7870,
... |
[
{
"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": "24",
"budget": 6296
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003685",
"normalized_example_id": "ex_003685",
"source_idx": 3685,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7870,
... |
[
{
"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": "24",
"budget": 7870
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003685",
"normalized_example_id": "ex_003685",
"source_idx": 3685,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7870,
... |
[
{
"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{999}{1000}",
"budget": 1523
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003686",
"normalized_example_id": "ex_003686",
"source_idx": 3686,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2539,
... |
[
{
"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{999}{1000}",
"budget": 2031
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003686",
"normalized_example_id": "ex_003686",
"source_idx": 3686,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2539,
... |
[
{
"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{999}{1000}",
"budget": 2539
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003686",
"normalized_example_id": "ex_003686",
"source_idx": 3686,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2539,
... |
[
{
"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",
"budget": 3826
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003687",
"normalized_example_id": "ex_003687",
"source_idx": 3687,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6377,
... |
[
{
"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",
"budget": 5102
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003687",
"normalized_example_id": "ex_003687",
"source_idx": 3687,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6377,
... |
[
{
"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",
"budget": 6377
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003687",
"normalized_example_id": "ex_003687",
"source_idx": 3687,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6377,
... |
[
{
"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{6}{5}(\\sqrt[3]{\\frac{1+\\sqrt{x}}{\\sqrt{x}}})^{5}+C",
"budget": 2452
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003688",
"normalized_example_id": "ex_003688",
"source_idx": 3688,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4087,
... |
[
{
"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{6}{5}(\\sqrt[3]{\\frac{1+\\sqrt{x}}{\\sqrt{x}}})^{5}+C",
"budget": 3270
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003688",
"normalized_example_id": "ex_003688",
"source_idx": 3688,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4087,
... |
[
{
"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{6}{5}(\\sqrt[3]{\\frac{1+\\sqrt{x}}{\\sqrt{x}}})^{5}+C",
"budget": 4087
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003688",
"normalized_example_id": "ex_003688",
"source_idx": 3688,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4087,
... |
[
{
"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": "7560",
"budget": 1570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003689",
"normalized_example_id": "ex_003689",
"source_idx": 3689,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2617,
... |
[
{
"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": "7560",
"budget": 2094
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003689",
"normalized_example_id": "ex_003689",
"source_idx": 3689,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2617,
... |
[
{
"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": "7560",
"budget": 2617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003689",
"normalized_example_id": "ex_003689",
"source_idx": 3689,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2617,
... |
[
{
"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": "3",
"budget": 1133
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003690",
"normalized_example_id": "ex_003690",
"source_idx": 3690,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1888,
... |
[
{
"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": "3",
"budget": 1510
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003690",
"normalized_example_id": "ex_003690",
"source_idx": 3690,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1888,
... |
[
{
"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": "3",
"budget": 1888
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003690",
"normalized_example_id": "ex_003690",
"source_idx": 3690,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1888,
... |
[
{
"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 \\sqrt{2}",
"budget": 3391
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003691",
"normalized_example_id": "ex_003691",
"source_idx": 3691,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5651,
... |
[
{
"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 \\sqrt{2}",
"budget": 4521
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003691",
"normalized_example_id": "ex_003691",
"source_idx": 3691,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5651,
... |
[
{
"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 \\sqrt{2}",
"budget": 5651
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003691",
"normalized_example_id": "ex_003691",
"source_idx": 3691,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5651,
... |
[
{
"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": "32;64;96",
"budget": 1330
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003692",
"normalized_example_id": "ex_003692",
"source_idx": 3692,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2217,
... |
[
{
"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": "32;64;96",
"budget": 1774
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003692",
"normalized_example_id": "ex_003692",
"source_idx": 3692,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2217,
... |
[
{
"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": "32;64;96",
"budget": 2217
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003692",
"normalized_example_id": "ex_003692",
"source_idx": 3692,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2217,
... |
[
{
"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}{13}",
"budget": 1414
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003697",
"normalized_example_id": "ex_003697",
"source_idx": 3697,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2356,
... |
[
{
"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}{13}",
"budget": 1885
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003697",
"normalized_example_id": "ex_003697",
"source_idx": 3697,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2356,
... |
[
{
"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}{13}",
"budget": 2356
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003697",
"normalized_example_id": "ex_003697",
"source_idx": 3697,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2356,
... |
[
{
"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": 1207
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003703",
"normalized_example_id": "ex_003703",
"source_idx": 3703,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2012,
... |
[
{
"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": 1610
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003703",
"normalized_example_id": "ex_003703",
"source_idx": 3703,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2012,
... |
[
{
"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": 2012
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003703",
"normalized_example_id": "ex_003703",
"source_idx": 3703,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2012,
... |
[
{
"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": 2479
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003704",
"normalized_example_id": "ex_003704",
"source_idx": 3704,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4131,
... |
[
{
"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": 3305
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003704",
"normalized_example_id": "ex_003704",
"source_idx": 3704,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4131,
... |
[
{
"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": 4131
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003704",
"normalized_example_id": "ex_003704",
"source_idx": 3704,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4131,
... |
[
{
"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": "x=y^{2}-4 y+3 \\text { or } x=-y^{2}-4 y+3",
"budget": 2600
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003705",
"normalized_example_id": "ex_003705",
"source_idx": 3705,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4334,
... |
[
{
"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": "x=y^{2}-4 y+3 \\text { or } x=-y^{2}-4 y+3",
"budget": 3467
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003705",
"normalized_example_id": "ex_003705",
"source_idx": 3705,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4334,
... |
[
{
"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": "x=y^{2}-4 y+3 \\text { or } x=-y^{2}-4 y+3",
"budget": 4334
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003705",
"normalized_example_id": "ex_003705",
"source_idx": 3705,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4334,
... |
[
{
"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_003706",
"normalized_example_id": "ex_003706",
"source_idx": 3706,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3662,
... |
[
{
"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": 2930
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003706",
"normalized_example_id": "ex_003706",
"source_idx": 3706,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3662,
... |
[
{
"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": 3662
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003706",
"normalized_example_id": "ex_003706",
"source_idx": 3706,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3662,
... |
[
{
"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": "1453789",
"budget": 4758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003707",
"normalized_example_id": "ex_003707",
"source_idx": 3707,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7930,
... |
[
{
"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": "1453789",
"budget": 6344
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003707",
"normalized_example_id": "ex_003707",
"source_idx": 3707,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7930,
... |
[
{
"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": "1453789",
"budget": 7930
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003707",
"normalized_example_id": "ex_003707",
"source_idx": 3707,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7930,
... |
[
{
"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,-1,1),(3,2,4)",
"budget": 1963
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003708",
"normalized_example_id": "ex_003708",
"source_idx": 3708,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3271,
... |
[
{
"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,-1,1),(3,2,4)",
"budget": 2617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003708",
"normalized_example_id": "ex_003708",
"source_idx": 3708,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3271,
... |
[
{
"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,-1,1),(3,2,4)",
"budget": 3271
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003708",
"normalized_example_id": "ex_003708",
"source_idx": 3708,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3271,
... |
[
{
"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": "x=y=z=1",
"budget": 4787
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003709",
"normalized_example_id": "ex_003709",
"source_idx": 3709,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7978,
... |
[
{
"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": "x=y=z=1",
"budget": 6382
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003709",
"normalized_example_id": "ex_003709",
"source_idx": 3709,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7978,
... |
[
{
"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": "x=y=z=1",
"budget": 7978
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003709",
"normalized_example_id": "ex_003709",
"source_idx": 3709,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7978,
... |
[
{
"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": "x = y = z = \\sqrt{2}",
"budget": 3570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003710",
"normalized_example_id": "ex_003710",
"source_idx": 3710,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5950,
... |
[
{
"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": "x = y = z = \\sqrt{2}",
"budget": 4760
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003710",
"normalized_example_id": "ex_003710",
"source_idx": 3710,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5950,
... |
[
{
"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": "x = y = z = \\sqrt{2}",
"budget": 5950
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003710",
"normalized_example_id": "ex_003710",
"source_idx": 3710,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5950,
... |
[
{
"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{7}",
"budget": 1435
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003712",
"normalized_example_id": "ex_003712",
"source_idx": 3712,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2392,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.