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": "987",
"budget": 3529
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000771",
"normalized_example_id": "ex_000771",
"source_idx": 771,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5881,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "987",
"budget": 4705
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000771",
"normalized_example_id": "ex_000771",
"source_idx": 771,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5881,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "987",
"budget": 5881
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000771",
"normalized_example_id": "ex_000771",
"source_idx": 771,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5881,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[6,\\infty)",
"budget": 2944
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000774",
"normalized_example_id": "ex_000774",
"source_idx": 774,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4907,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[6,\\infty)",
"budget": 3926
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000774",
"normalized_example_id": "ex_000774",
"source_idx": 774,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4907,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "[6,\\infty)",
"budget": 4907
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000774",
"normalized_example_id": "ex_000774",
"source_idx": 774,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4907,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{3}",
"budget": 1214
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000776",
"normalized_example_id": "ex_000776",
"source_idx": 776,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2024,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{3}",
"budget": 1619
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000776",
"normalized_example_id": "ex_000776",
"source_idx": 776,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2024,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{3}",
"budget": 2024
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000776",
"normalized_example_id": "ex_000776",
"source_idx": 776,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2024,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\geqslant 2",
"budget": 2099
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000782",
"normalized_example_id": "ex_000782",
"source_idx": 782,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3499,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\geqslant 2",
"budget": 2799
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000782",
"normalized_example_id": "ex_000782",
"source_idx": 782,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3499,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 \\geqslant 2",
"budget": 3499
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000782",
"normalized_example_id": "ex_000782",
"source_idx": 782,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3499,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1004",
"budget": 5614
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000783",
"normalized_example_id": "ex_000783",
"source_idx": 783,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9356,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1004",
"budget": 7485
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000783",
"normalized_example_id": "ex_000783",
"source_idx": 783,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9356,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1004",
"budget": 9356
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000783",
"normalized_example_id": "ex_000783",
"source_idx": 783,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9356,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 = 5",
"budget": 1013
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000784",
"normalized_example_id": "ex_000784",
"source_idx": 784,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1688,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 = 5",
"budget": 1350
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000784",
"normalized_example_id": "ex_000784",
"source_idx": 784,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1688,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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 = 5",
"budget": 1688
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000784",
"normalized_example_id": "ex_000784",
"source_idx": 784,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1688,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "39",
"budget": 5701
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000790",
"normalized_example_id": "ex_000790",
"source_idx": 790,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9502,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "39",
"budget": 7602
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000790",
"normalized_example_id": "ex_000790",
"source_idx": 790,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9502,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "39",
"budget": 9502
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000790",
"normalized_example_id": "ex_000790",
"source_idx": 790,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9502,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 5099
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000792",
"normalized_example_id": "ex_000792",
"source_idx": 792,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8498,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 6798
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000792",
"normalized_example_id": "ex_000792",
"source_idx": 792,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8498,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 8498
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000792",
"normalized_example_id": "ex_000792",
"source_idx": 792,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8498,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2486
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000795",
"normalized_example_id": "ex_000795",
"source_idx": 795,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4143,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3314
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000795",
"normalized_example_id": "ex_000795",
"source_idx": 795,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4143,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4143
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000795",
"normalized_example_id": "ex_000795",
"source_idx": 795,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4143,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "26",
"budget": 4671
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000801",
"normalized_example_id": "ex_000801",
"source_idx": 801,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7785,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "26",
"budget": 6228
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000801",
"normalized_example_id": "ex_000801",
"source_idx": 801,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7785,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "26",
"budget": 7785
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000801",
"normalized_example_id": "ex_000801",
"source_idx": 801,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7785,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "91",
"budget": 847
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000802",
"normalized_example_id": "ex_000802",
"source_idx": 802,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1411,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "91",
"budget": 1129
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000802",
"normalized_example_id": "ex_000802",
"source_idx": 802,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1411,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "91",
"budget": 1411
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000802",
"normalized_example_id": "ex_000802",
"source_idx": 802,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1411,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "119",
"budget": 2434
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000803",
"normalized_example_id": "ex_000803",
"source_idx": 803,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4056,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "119",
"budget": 3245
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000803",
"normalized_example_id": "ex_000803",
"source_idx": 803,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4056,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "119",
"budget": 4056
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000803",
"normalized_example_id": "ex_000803",
"source_idx": 803,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4056,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2095
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000805",
"normalized_example_id": "ex_000805",
"source_idx": 805,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3491,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2793
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000805",
"normalized_example_id": "ex_000805",
"source_idx": 805,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3491,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3491
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000805",
"normalized_example_id": "ex_000805",
"source_idx": 805,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3491,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\sqrt{5}",
"budget": 5770
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000806",
"normalized_example_id": "ex_000806",
"source_idx": 806,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9617,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\sqrt{5}",
"budget": 7694
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000806",
"normalized_example_id": "ex_000806",
"source_idx": 806,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9617,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\sqrt{5}",
"budget": 9617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000806",
"normalized_example_id": "ex_000806",
"source_idx": 806,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9617,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "60",
"budget": 2353
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000808",
"normalized_example_id": "ex_000808",
"source_idx": 808,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3921,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "60",
"budget": 3137
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000808",
"normalized_example_id": "ex_000808",
"source_idx": 808,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3921,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "60",
"budget": 3921
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000808",
"normalized_example_id": "ex_000808",
"source_idx": 808,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3921,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 2770
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000810",
"normalized_example_id": "ex_000810",
"source_idx": 810,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4617,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 3694
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000810",
"normalized_example_id": "ex_000810",
"source_idx": 810,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4617,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "5",
"budget": 4617
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000810",
"normalized_example_id": "ex_000810",
"source_idx": 810,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4617,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "60\\mathrm{~}/\\mathrm{}",
"budget": 700
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000811",
"normalized_example_id": "ex_000811",
"source_idx": 811,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1166,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "60\\mathrm{~}/\\mathrm{}",
"budget": 933
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000811",
"normalized_example_id": "ex_000811",
"source_idx": 811,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1166,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "60\\mathrm{~}/\\mathrm{}",
"budget": 1166
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000811",
"normalized_example_id": "ex_000811",
"source_idx": 811,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1166,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{1-x}",
"budget": 1336
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000813",
"normalized_example_id": "ex_000813",
"source_idx": 813,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2227,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{1-x}",
"budget": 1782
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000813",
"normalized_example_id": "ex_000813",
"source_idx": 813,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2227,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{1-x}",
"budget": 2227
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000813",
"normalized_example_id": "ex_000813",
"source_idx": 813,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2227,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "51",
"budget": 2426
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000814",
"normalized_example_id": "ex_000814",
"source_idx": 814,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4044,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "51",
"budget": 3235
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000814",
"normalized_example_id": "ex_000814",
"source_idx": 814,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4044,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "51",
"budget": 4044
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000814",
"normalized_example_id": "ex_000814",
"source_idx": 814,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4044,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "873451215or873451233",
"budget": 4360
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000815",
"normalized_example_id": "ex_000815",
"source_idx": 815,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7266,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "873451215or873451233",
"budget": 5813
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000815",
"normalized_example_id": "ex_000815",
"source_idx": 815,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7266,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "873451215or873451233",
"budget": 7266
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000815",
"normalized_example_id": "ex_000815",
"source_idx": 815,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7266,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{333}{1997}",
"budget": 5020
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000817",
"normalized_example_id": "ex_000817",
"source_idx": 817,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8366,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{333}{1997}",
"budget": 6693
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000817",
"normalized_example_id": "ex_000817",
"source_idx": 817,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8366,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{333}{1997}",
"budget": 8366
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000817",
"normalized_example_id": "ex_000817",
"source_idx": 817,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8366,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{x^{4}-1}",
"budget": 1788
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000818",
"normalized_example_id": "ex_000818",
"source_idx": 818,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2980,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{x^{4}-1}",
"budget": 2384
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000818",
"normalized_example_id": "ex_000818",
"source_idx": 818,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2980,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{x^{4}-1}",
"budget": 2980
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000818",
"normalized_example_id": "ex_000818",
"source_idx": 818,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2980,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "20",
"budget": 875
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000819",
"normalized_example_id": "ex_000819",
"source_idx": 819,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1458,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "20",
"budget": 1166
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000819",
"normalized_example_id": "ex_000819",
"source_idx": 819,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1458,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "20",
"budget": 1458
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000819",
"normalized_example_id": "ex_000819",
"source_idx": 819,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1458,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3310
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000821",
"normalized_example_id": "ex_000821",
"source_idx": 821,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5516,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4413
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000821",
"normalized_example_id": "ex_000821",
"source_idx": 821,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5516,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5516
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000821",
"normalized_example_id": "ex_000821",
"source_idx": 821,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5516,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1096
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000825",
"normalized_example_id": "ex_000825",
"source_idx": 825,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1827,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1462
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000825",
"normalized_example_id": "ex_000825",
"source_idx": 825,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1827,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1827
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000825",
"normalized_example_id": "ex_000825",
"source_idx": 825,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1827,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(x,y)\\in{(0,0),(\\frac{1}{2}(1+\\sqrt{5}),\\frac{1}{2}(1+\\sqrt{5})),(\\frac{1}{2}(1-\\sqrt{5}),\\frac{1}{2}(1-\\sqrt{5}))}",
"budget": 2140
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000831",
"normalized_example_id": "ex_000831",
"source_idx": 831,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3567,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(x,y)\\in{(0,0),(\\frac{1}{2}(1+\\sqrt{5}),\\frac{1}{2}(1+\\sqrt{5})),(\\frac{1}{2}(1-\\sqrt{5}),\\frac{1}{2}(1-\\sqrt{5}))}",
"budget": 2854
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000831",
"normalized_example_id": "ex_000831",
"source_idx": 831,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3567,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(x,y)\\in{(0,0),(\\frac{1}{2}(1+\\sqrt{5}),\\frac{1}{2}(1+\\sqrt{5})),(\\frac{1}{2}(1-\\sqrt{5}),\\frac{1}{2}(1-\\sqrt{5}))}",
"budget": 3567
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000831",
"normalized_example_id": "ex_000831",
"source_idx": 831,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3567,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "101",
"budget": 2634
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000832",
"normalized_example_id": "ex_000832",
"source_idx": 832,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4390,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "101",
"budget": 3512
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000832",
"normalized_example_id": "ex_000832",
"source_idx": 832,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4390,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "101",
"budget": 4390
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000832",
"normalized_example_id": "ex_000832",
"source_idx": 832,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4390,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2}}{2}",
"budget": 1484
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000833",
"normalized_example_id": "ex_000833",
"source_idx": 833,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2474,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2}}{2}",
"budget": 1979
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000833",
"normalized_example_id": "ex_000833",
"source_idx": 833,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2474,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{2}}{2}",
"budget": 2474
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000833",
"normalized_example_id": "ex_000833",
"source_idx": 833,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2474,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14",
"budget": 2708
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000834",
"normalized_example_id": "ex_000834",
"source_idx": 834,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4513,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14",
"budget": 3610
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000834",
"normalized_example_id": "ex_000834",
"source_idx": 834,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4513,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14",
"budget": 4513
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000834",
"normalized_example_id": "ex_000834",
"source_idx": 834,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4513,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 4527
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000835",
"normalized_example_id": "ex_000835",
"source_idx": 835,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7545,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 6036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000835",
"normalized_example_id": "ex_000835",
"source_idx": 835,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7545,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 7545
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000835",
"normalized_example_id": "ex_000835",
"source_idx": 835,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7545,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{3\\left(a^{2}+b^{2}+c^{2}+d^{2}\\right)}}{6}",
"budget": 1741
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000837",
"normalized_example_id": "ex_000837",
"source_idx": 837,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2902,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{3\\left(a^{2}+b^{2}+c^{2}+d^{2}\\right)}}{6}",
"budget": 2322
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000837",
"normalized_example_id": "ex_000837",
"source_idx": 837,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2902,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{3\\left(a^{2}+b^{2}+c^{2}+d^{2}\\right)}}{6}",
"budget": 2902
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000837",
"normalized_example_id": "ex_000837",
"source_idx": 837,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2902,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 2954
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000838",
"normalized_example_id": "ex_000838",
"source_idx": 838,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4924,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 3939
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000838",
"normalized_example_id": "ex_000838",
"source_idx": 838,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4924,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "16",
"budget": 4924
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000838",
"normalized_example_id": "ex_000838",
"source_idx": 838,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4924,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3\\sqrt{3}}{2}",
"budget": 5963
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000840",
"normalized_example_id": "ex_000840",
"source_idx": 840,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9938,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3\\sqrt{3}}{2}",
"budget": 7950
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000840",
"normalized_example_id": "ex_000840",
"source_idx": 840,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9938,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{3\\sqrt{3}}{2}",
"budget": 9938
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000840",
"normalized_example_id": "ex_000840",
"source_idx": 840,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9938,
"... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(m, n) = (0, 2), (3, 5), (4, 7)",
"budget": 4690
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_000846",
"normalized_example_id": "ex_000846",
"source_idx": 846,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7817,
"... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.