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": "-\\frac{7}{3}",
"budget": 2386
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002856",
"normalized_example_id": "ex_002856",
"source_idx": 2856,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3977,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{7}{3}",
"budget": 3182
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002856",
"normalized_example_id": "ex_002856",
"source_idx": 2856,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3977,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{7}{3}",
"budget": 3977
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002856",
"normalized_example_id": "ex_002856",
"source_idx": 2856,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3977,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 2319
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002858",
"normalized_example_id": "ex_002858",
"source_idx": 2858,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3865,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 3092
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002858",
"normalized_example_id": "ex_002858",
"source_idx": 2858,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3865,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 3865
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002858",
"normalized_example_id": "ex_002858",
"source_idx": 2858,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3865,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}",
"budget": 2392
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002859",
"normalized_example_id": "ex_002859",
"source_idx": 2859,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3987,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}",
"budget": 3190
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002859",
"normalized_example_id": "ex_002859",
"source_idx": 2859,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3987,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}",
"budget": 3987
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002859",
"normalized_example_id": "ex_002859",
"source_idx": 2859,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3987,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 4318
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002860",
"normalized_example_id": "ex_002860",
"source_idx": 2860,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7196,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 5757
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002860",
"normalized_example_id": "ex_002860",
"source_idx": 2860,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7196,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 7196
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002860",
"normalized_example_id": "ex_002860",
"source_idx": 2860,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7196,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "52",
"budget": 3802
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002863",
"normalized_example_id": "ex_002863",
"source_idx": 2863,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6336,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "52",
"budget": 5069
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002863",
"normalized_example_id": "ex_002863",
"source_idx": 2863,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6336,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "52",
"budget": 6336
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002863",
"normalized_example_id": "ex_002863",
"source_idx": 2863,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6336,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\varphi(x)=xe^{x}-e^{x}+1",
"budget": 1639
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002866",
"normalized_example_id": "ex_002866",
"source_idx": 2866,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2731,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\varphi(x)=xe^{x}-e^{x}+1",
"budget": 2185
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002866",
"normalized_example_id": "ex_002866",
"source_idx": 2866,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2731,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\varphi(x)=xe^{x}-e^{x}+1",
"budget": 2731
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002866",
"normalized_example_id": "ex_002866",
"source_idx": 2866,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2731,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{175}{256}",
"budget": 2262
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002867",
"normalized_example_id": "ex_002867",
"source_idx": 2867,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3770,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{175}{256}",
"budget": 3016
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002867",
"normalized_example_id": "ex_002867",
"source_idx": 2867,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3770,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{175}{256}",
"budget": 3770
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002867",
"normalized_example_id": "ex_002867",
"source_idx": 2867,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3770,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,3,6),(2,4,4),(3,3,3)",
"budget": 5538
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002868",
"normalized_example_id": "ex_002868",
"source_idx": 2868,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9230,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,3,6),(2,4,4),(3,3,3)",
"budget": 7384
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002868",
"normalized_example_id": "ex_002868",
"source_idx": 2868,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9230,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(2,3,6),(2,4,4),(3,3,3)",
"budget": 9230
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002868",
"normalized_example_id": "ex_002868",
"source_idx": 2868,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9230,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "D",
"budget": 5309
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002869",
"normalized_example_id": "ex_002869",
"source_idx": 2869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8849,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "D",
"budget": 7079
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002869",
"normalized_example_id": "ex_002869",
"source_idx": 2869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8849,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "D",
"budget": 8849
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002869",
"normalized_example_id": "ex_002869",
"source_idx": 2869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8849,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 748
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002872",
"normalized_example_id": "ex_002872",
"source_idx": 2872,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 997
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002872",
"normalized_example_id": "ex_002872",
"source_idx": 2872,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 1246
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002872",
"normalized_example_id": "ex_002872",
"source_idx": 2872,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{4}\\cdot(10-7x)\\cos4x+\\frac{7}{16}\\cdot\\sin4x+C",
"budget": 1541
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002873",
"normalized_example_id": "ex_002873",
"source_idx": 2873,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2568,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}\\cdot(10-7x)\\cos4x+\\frac{7}{16}\\cdot\\sin4x+C",
"budget": 2054
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002873",
"normalized_example_id": "ex_002873",
"source_idx": 2873,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2568,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}\\cdot(10-7x)\\cos4x+\\frac{7}{16}\\cdot\\sin4x+C",
"budget": 2568
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002873",
"normalized_example_id": "ex_002873",
"source_idx": 2873,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2568,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "90",
"budget": 5567
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002874",
"normalized_example_id": "ex_002874",
"source_idx": 2874,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9278,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "90",
"budget": 7422
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002874",
"normalized_example_id": "ex_002874",
"source_idx": 2874,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9278,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "90",
"budget": 9278
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002874",
"normalized_example_id": "ex_002874",
"source_idx": 2874,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9278,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{5}+1}{4}",
"budget": 3937
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002876",
"normalized_example_id": "ex_002876",
"source_idx": 2876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6561,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{5}+1}{4}",
"budget": 5249
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002876",
"normalized_example_id": "ex_002876",
"source_idx": 2876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6561,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{5}+1}{4}",
"budget": 6561
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002876",
"normalized_example_id": "ex_002876",
"source_idx": 2876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6561,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 3757
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002877",
"normalized_example_id": "ex_002877",
"source_idx": 2877,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6262,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 5010
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002877",
"normalized_example_id": "ex_002877",
"source_idx": 2877,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6262,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 6262
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002877",
"normalized_example_id": "ex_002877",
"source_idx": 2877,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6262,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 5905
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002879",
"normalized_example_id": "ex_002879",
"source_idx": 2879,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9842,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 7874
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002879",
"normalized_example_id": "ex_002879",
"source_idx": 2879,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9842,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 9842
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002879",
"normalized_example_id": "ex_002879",
"source_idx": 2879,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9842,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "65",
"budget": 1725
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002880",
"normalized_example_id": "ex_002880",
"source_idx": 2880,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2875,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "65",
"budget": 2300
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002880",
"normalized_example_id": "ex_002880",
"source_idx": 2880,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2875,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "65",
"budget": 2875
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002880",
"normalized_example_id": "ex_002880",
"source_idx": 2880,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2875,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 4787
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002881",
"normalized_example_id": "ex_002881",
"source_idx": 2881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7979,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 6383
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002881",
"normalized_example_id": "ex_002881",
"source_idx": 2881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7979,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 7979
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002881",
"normalized_example_id": "ex_002881",
"source_idx": 2881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7979,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "31",
"budget": 1060
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002884",
"normalized_example_id": "ex_002884",
"source_idx": 2884,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "31",
"budget": 1413
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002884",
"normalized_example_id": "ex_002884",
"source_idx": 2884,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "31",
"budget": 1766
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002884",
"normalized_example_id": "ex_002884",
"source_idx": 2884,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1766,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^2",
"budget": 4518
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002890",
"normalized_example_id": "ex_002890",
"source_idx": 2890,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7530,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^2",
"budget": 6024
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002890",
"normalized_example_id": "ex_002890",
"source_idx": 2890,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7530,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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^2",
"budget": 7530
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002890",
"normalized_example_id": "ex_002890",
"source_idx": 2890,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7530,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, \\sqrt{2})",
"budget": 3451
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002891",
"normalized_example_id": "ex_002891",
"source_idx": 2891,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5752,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, \\sqrt{2})",
"budget": 4602
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002891",
"normalized_example_id": "ex_002891",
"source_idx": 2891,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5752,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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, \\sqrt{2})",
"budget": 5752
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002891",
"normalized_example_id": "ex_002891",
"source_idx": 2891,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5752,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_{1}\\in(0,1)",
"budget": 3672
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002892",
"normalized_example_id": "ex_002892",
"source_idx": 2892,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6120,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_{1}\\in(0,1)",
"budget": 4896
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002892",
"normalized_example_id": "ex_002892",
"source_idx": 2892,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6120,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "a_{1}\\in(0,1)",
"budget": 6120
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002892",
"normalized_example_id": "ex_002892",
"source_idx": 2892,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6120,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "83.25",
"budget": 4612
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002897",
"normalized_example_id": "ex_002897",
"source_idx": 2897,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7686,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "83.25",
"budget": 6149
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002897",
"normalized_example_id": "ex_002897",
"source_idx": 2897,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7686,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "83.25",
"budget": 7686
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002897",
"normalized_example_id": "ex_002897",
"source_idx": 2897,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7686,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6817",
"budget": 5717
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002901",
"normalized_example_id": "ex_002901",
"source_idx": 2901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9528,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6817",
"budget": 7622
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002901",
"normalized_example_id": "ex_002901",
"source_idx": 2901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9528,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "6817",
"budget": 9528
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002901",
"normalized_example_id": "ex_002901",
"source_idx": 2901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9528,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2R\\sqrt[3]{4}",
"budget": 1809
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002905",
"normalized_example_id": "ex_002905",
"source_idx": 2905,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3015,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2R\\sqrt[3]{4}",
"budget": 2412
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002905",
"normalized_example_id": "ex_002905",
"source_idx": 2905,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3015,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2R\\sqrt[3]{4}",
"budget": 3015
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002905",
"normalized_example_id": "ex_002905",
"source_idx": 2905,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3015,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1466
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002907",
"normalized_example_id": "ex_002907",
"source_idx": 2907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2444,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1955
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002907",
"normalized_example_id": "ex_002907",
"source_idx": 2907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2444,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2444
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002907",
"normalized_example_id": "ex_002907",
"source_idx": 2907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2444,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 944
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002908",
"normalized_example_id": "ex_002908",
"source_idx": 2908,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1573,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1258
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002908",
"normalized_example_id": "ex_002908",
"source_idx": 2908,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1573,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1573
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002908",
"normalized_example_id": "ex_002908",
"source_idx": 2908,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1573,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 4948
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002910",
"normalized_example_id": "ex_002910",
"source_idx": 2910,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 6597
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002910",
"normalized_example_id": "ex_002910",
"source_idx": 2910,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 8246
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002910",
"normalized_example_id": "ex_002910",
"source_idx": 2910,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8246,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3382
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002912",
"normalized_example_id": "ex_002912",
"source_idx": 2912,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5636,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4509
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002912",
"normalized_example_id": "ex_002912",
"source_idx": 2912,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5636,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5636
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002912",
"normalized_example_id": "ex_002912",
"source_idx": 2912,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5636,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 3122
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002914",
"normalized_example_id": "ex_002914",
"source_idx": 2914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5204,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 4163
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002914",
"normalized_example_id": "ex_002914",
"source_idx": 2914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5204,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 5204
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002914",
"normalized_example_id": "ex_002914",
"source_idx": 2914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5204,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{13}{3}",
"budget": 914
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002915",
"normalized_example_id": "ex_002915",
"source_idx": 2915,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1523,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{13}{3}",
"budget": 1218
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002915",
"normalized_example_id": "ex_002915",
"source_idx": 2915,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1523,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{13}{3}",
"budget": 1523
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002915",
"normalized_example_id": "ex_002915",
"source_idx": 2915,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1523,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 1624
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002919",
"normalized_example_id": "ex_002919",
"source_idx": 2919,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2707,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 2166
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002919",
"normalized_example_id": "ex_002919",
"source_idx": 2919,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2707,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}",
"budget": 2707
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002919",
"normalized_example_id": "ex_002919",
"source_idx": 2919,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2707,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2819
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002920",
"normalized_example_id": "ex_002920",
"source_idx": 2920,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4698,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002920",
"normalized_example_id": "ex_002920",
"source_idx": 2920,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4698,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4698
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002920",
"normalized_example_id": "ex_002920",
"source_idx": 2920,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4698,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n \\leq 12",
"budget": 2119
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002921",
"normalized_example_id": "ex_002921",
"source_idx": 2921,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3531,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n \\leq 12",
"budget": 2825
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002921",
"normalized_example_id": "ex_002921",
"source_idx": 2921,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3531,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "n \\leq 12",
"budget": 3531
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002921",
"normalized_example_id": "ex_002921",
"source_idx": 2921,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3531,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 2189
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_002928",
"normalized_example_id": "ex_002928",
"source_idx": 2928,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3649,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.