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": "3:5",
"budget": 2401
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006089",
"normalized_example_id": "ex_006089",
"source_idx": 6089,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3001,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3:5",
"budget": 3001
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006089",
"normalized_example_id": "ex_006089",
"source_idx": 6089,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3001,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2037",
"budget": 1247
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006090",
"normalized_example_id": "ex_006090",
"source_idx": 6090,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2078,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2037",
"budget": 1662
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006090",
"normalized_example_id": "ex_006090",
"source_idx": 6090,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2078,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2037",
"budget": 2078
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006090",
"normalized_example_id": "ex_006090",
"source_idx": 6090,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2078,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(-\\infty,0)\\cup(1,+\\infty)",
"budget": 5028
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006091",
"normalized_example_id": "ex_006091",
"source_idx": 6091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8380,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(-\\infty,0)\\cup(1,+\\infty)",
"budget": 6704
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006091",
"normalized_example_id": "ex_006091",
"source_idx": 6091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8380,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(-\\infty,0)\\cup(1,+\\infty)",
"budget": 8380
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006091",
"normalized_example_id": "ex_006091",
"source_idx": 6091,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8380,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "F_{n+1}",
"budget": 2540
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006092",
"normalized_example_id": "ex_006092",
"source_idx": 6092,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4233,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "F_{n+1}",
"budget": 3386
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006092",
"normalized_example_id": "ex_006092",
"source_idx": 6092,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4233,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "F_{n+1}",
"budget": 4233
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006092",
"normalized_example_id": "ex_006092",
"source_idx": 6092,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4233,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{1}+P_{2}+P_{3}",
"budget": 4272
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006094",
"normalized_example_id": "ex_006094",
"source_idx": 6094,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7120,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{1}+P_{2}+P_{3}",
"budget": 5696
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006094",
"normalized_example_id": "ex_006094",
"source_idx": 6094,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7120,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{1}+P_{2}+P_{3}",
"budget": 7120
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006094",
"normalized_example_id": "ex_006094",
"source_idx": 6094,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7120,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "b^{2}",
"budget": 6091
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006095",
"normalized_example_id": "ex_006095",
"source_idx": 6095,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10152,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "b^{2}",
"budget": 8122
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006095",
"normalized_example_id": "ex_006095",
"source_idx": 6095,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10152,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "b^{2}",
"budget": 10152
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006095",
"normalized_example_id": "ex_006095",
"source_idx": 6095,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10152,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006096",
"normalized_example_id": "ex_006096",
"source_idx": 6096,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7616,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6093
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006096",
"normalized_example_id": "ex_006096",
"source_idx": 6096,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7616,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7616
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006096",
"normalized_example_id": "ex_006096",
"source_idx": 6096,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7616,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\left(x-\\frac{1}{2}\\right)^{2}+\\left(y+\\frac{7}{2}\\right)^{2}=\\frac{178}{4}",
"budget": 1958
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006097",
"normalized_example_id": "ex_006097",
"source_idx": 6097,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3264,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\left(x-\\frac{1}{2}\\right)^{2}+\\left(y+\\frac{7}{2}\\right)^{2}=\\frac{178}{4}",
"budget": 2611
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006097",
"normalized_example_id": "ex_006097",
"source_idx": 6097,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3264,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\left(x-\\frac{1}{2}\\right)^{2}+\\left(y+\\frac{7}{2}\\right)^{2}=\\frac{178}{4}",
"budget": 3264
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006097",
"normalized_example_id": "ex_006097",
"source_idx": 6097,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3264,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 1483
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006098",
"normalized_example_id": "ex_006098",
"source_idx": 6098,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2472,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 1978
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006098",
"normalized_example_id": "ex_006098",
"source_idx": 6098,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2472,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "40",
"budget": 2472
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006098",
"normalized_example_id": "ex_006098",
"source_idx": 6098,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2472,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "S_{ABC}=12,S_{CNK}=\\sqrt{2}",
"budget": 3626
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006099",
"normalized_example_id": "ex_006099",
"source_idx": 6099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6043,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "S_{ABC}=12,S_{CNK}=\\sqrt{2}",
"budget": 4834
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006099",
"normalized_example_id": "ex_006099",
"source_idx": 6099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6043,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "S_{ABC}=12,S_{CNK}=\\sqrt{2}",
"budget": 6043
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006099",
"normalized_example_id": "ex_006099",
"source_idx": 6099,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6043,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "13",
"budget": 1193
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006103",
"normalized_example_id": "ex_006103",
"source_idx": 6103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1989,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "13",
"budget": 1591
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006103",
"normalized_example_id": "ex_006103",
"source_idx": 6103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1989,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "13",
"budget": 1989
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006103",
"normalized_example_id": "ex_006103",
"source_idx": 6103,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1989,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3:1",
"budget": 3886
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006104",
"normalized_example_id": "ex_006104",
"source_idx": 6104,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6477,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3:1",
"budget": 5182
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006104",
"normalized_example_id": "ex_006104",
"source_idx": 6104,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6477,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3:1",
"budget": 6477
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006104",
"normalized_example_id": "ex_006104",
"source_idx": 6104,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6477,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 1894
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006105",
"normalized_example_id": "ex_006105",
"source_idx": 6105,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3157,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 2526
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006105",
"normalized_example_id": "ex_006105",
"source_idx": 6105,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3157,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{2}",
"budget": 3157
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006105",
"normalized_example_id": "ex_006105",
"source_idx": 6105,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3157,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "276",
"budget": 3119
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006106",
"normalized_example_id": "ex_006106",
"source_idx": 6106,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5199,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "276",
"budget": 4159
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006106",
"normalized_example_id": "ex_006106",
"source_idx": 6106,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5199,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "276",
"budget": 5199
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006106",
"normalized_example_id": "ex_006106",
"source_idx": 6106,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5199,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2107
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006109",
"normalized_example_id": "ex_006109",
"source_idx": 6109,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3511,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2809
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006109",
"normalized_example_id": "ex_006109",
"source_idx": 6109,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3511,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3511
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006109",
"normalized_example_id": "ex_006109",
"source_idx": 6109,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3511,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{2}{\\pi}",
"budget": 2123
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006111",
"normalized_example_id": "ex_006111",
"source_idx": 6111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3538,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{2}{\\pi}",
"budget": 2830
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006111",
"normalized_example_id": "ex_006111",
"source_idx": 6111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3538,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\frac{2}{\\pi}",
"budget": 3538
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006111",
"normalized_example_id": "ex_006111",
"source_idx": 6111,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3538,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8750",
"budget": 4247
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006112",
"normalized_example_id": "ex_006112",
"source_idx": 6112,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7079,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8750",
"budget": 5663
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006112",
"normalized_example_id": "ex_006112",
"source_idx": 6112,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7079,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8750",
"budget": 7079
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006112",
"normalized_example_id": "ex_006112",
"source_idx": 6112,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7079,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{39\\pi}{4}",
"budget": 4376
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006115",
"normalized_example_id": "ex_006115",
"source_idx": 6115,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{39\\pi}{4}",
"budget": 5835
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006115",
"normalized_example_id": "ex_006115",
"source_idx": 6115,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{39\\pi}{4}",
"budget": 7294
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006115",
"normalized_example_id": "ex_006115",
"source_idx": 6115,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7294,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "30",
"budget": 5453
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006116",
"normalized_example_id": "ex_006116",
"source_idx": 6116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9088,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "30",
"budget": 7270
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006116",
"normalized_example_id": "ex_006116",
"source_idx": 6116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9088,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "30",
"budget": 9088
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006116",
"normalized_example_id": "ex_006116",
"source_idx": 6116,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9088,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4080401",
"budget": 3836
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006118",
"normalized_example_id": "ex_006118",
"source_idx": 6118,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6393,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4080401",
"budget": 5114
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006118",
"normalized_example_id": "ex_006118",
"source_idx": 6118,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6393,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4080401",
"budget": 6393
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006118",
"normalized_example_id": "ex_006118",
"source_idx": 6118,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6393,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2c",
"budget": 3104
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006119",
"normalized_example_id": "ex_006119",
"source_idx": 6119,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5174,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2c",
"budget": 4139
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006119",
"normalized_example_id": "ex_006119",
"source_idx": 6119,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5174,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2c",
"budget": 5174
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006119",
"normalized_example_id": "ex_006119",
"source_idx": 6119,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5174,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x) = x^2 + ax + c",
"budget": 3446
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006120",
"normalized_example_id": "ex_006120",
"source_idx": 6120,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5744,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x) = x^2 + ax + c",
"budget": 4595
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006120",
"normalized_example_id": "ex_006120",
"source_idx": 6120,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5744,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(x) = x^2 + ax + c",
"budget": 5744
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006120",
"normalized_example_id": "ex_006120",
"source_idx": 6120,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5744,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3080
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006126",
"normalized_example_id": "ex_006126",
"source_idx": 6126,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5133,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4106
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006126",
"normalized_example_id": "ex_006126",
"source_idx": 6126,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5133,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5133
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006126",
"normalized_example_id": "ex_006126",
"source_idx": 6126,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5133,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "250\\mathrm{}",
"budget": 1385
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006127",
"normalized_example_id": "ex_006127",
"source_idx": 6127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2309,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "250\\mathrm{}",
"budget": 1847
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006127",
"normalized_example_id": "ex_006127",
"source_idx": 6127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2309,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "250\\mathrm{}",
"budget": 2309
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006127",
"normalized_example_id": "ex_006127",
"source_idx": 6127,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2309,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "E",
"budget": 956
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006128",
"normalized_example_id": "ex_006128",
"source_idx": 6128,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1593,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "E",
"budget": 1274
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006128",
"normalized_example_id": "ex_006128",
"source_idx": 6128,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1593,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "E",
"budget": 1593
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006128",
"normalized_example_id": "ex_006128",
"source_idx": 6128,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1593,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "C",
"budget": 2604
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006130",
"normalized_example_id": "ex_006130",
"source_idx": 6130,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4340,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "C",
"budget": 3472
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006130",
"normalized_example_id": "ex_006130",
"source_idx": 6130,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4340,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "C",
"budget": 4340
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006130",
"normalized_example_id": "ex_006130",
"source_idx": 6130,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4340,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 4130
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006132",
"normalized_example_id": "ex_006132",
"source_idx": 6132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6883,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 5506
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006132",
"normalized_example_id": "ex_006132",
"source_idx": 6132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6883,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "B",
"budget": 6883
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006132",
"normalized_example_id": "ex_006132",
"source_idx": 6132,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6883,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4703
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006133",
"normalized_example_id": "ex_006133",
"source_idx": 6133,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7838,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6270
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006133",
"normalized_example_id": "ex_006133",
"source_idx": 6133,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7838,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7838
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006133",
"normalized_example_id": "ex_006133",
"source_idx": 6133,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7838,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{35}",
"budget": 1730
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006134",
"normalized_example_id": "ex_006134",
"source_idx": 6134,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2884,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{35}",
"budget": 2307
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006134",
"normalized_example_id": "ex_006134",
"source_idx": 6134,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2884,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{35}",
"budget": 2884
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006134",
"normalized_example_id": "ex_006134",
"source_idx": 6134,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2884,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "19",
"budget": 3620
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006135",
"normalized_example_id": "ex_006135",
"source_idx": 6135,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6033,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "19",
"budget": 4826
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006135",
"normalized_example_id": "ex_006135",
"source_idx": 6135,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6033,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "19",
"budget": 6033
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006135",
"normalized_example_id": "ex_006135",
"source_idx": 6135,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6033,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2}{3^{999}}",
"budget": 1280
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006136",
"normalized_example_id": "ex_006136",
"source_idx": 6136,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2134,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2}{3^{999}}",
"budget": 1707
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006136",
"normalized_example_id": "ex_006136",
"source_idx": 6136,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2134,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2}{3^{999}}",
"budget": 2134
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006136",
"normalized_example_id": "ex_006136",
"source_idx": 6136,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2134,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 2043
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006138",
"normalized_example_id": "ex_006138",
"source_idx": 6138,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3405,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 2724
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006138",
"normalized_example_id": "ex_006138",
"source_idx": 6138,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3405,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 3405
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006138",
"normalized_example_id": "ex_006138",
"source_idx": 6138,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3405,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{p}(1-(1-p)^{n})",
"budget": 1663
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006141",
"normalized_example_id": "ex_006141",
"source_idx": 6141,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2772,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{p}(1-(1-p)^{n})",
"budget": 2218
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006141",
"normalized_example_id": "ex_006141",
"source_idx": 6141,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2772,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{p}(1-(1-p)^{n})",
"budget": 2772
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006141",
"normalized_example_id": "ex_006141",
"source_idx": 6141,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2772,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5or-3or-7",
"budget": 3164
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006144",
"normalized_example_id": "ex_006144",
"source_idx": 6144,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5273,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5or-3or-7",
"budget": 4218
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_006144",
"normalized_example_id": "ex_006144",
"source_idx": 6144,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5273,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.