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": "I_{2}",
"budget": 2789
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004000",
"normalized_example_id": "ex_004000",
"source_idx": 4000,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2789,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 2763
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004001",
"normalized_example_id": "ex_004001",
"source_idx": 4001,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4605,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 3684
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004001",
"normalized_example_id": "ex_004001",
"source_idx": 4001,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4605,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 4605
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004001",
"normalized_example_id": "ex_004001",
"source_idx": 4001,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4605,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8",
"budget": 4895
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004003",
"normalized_example_id": "ex_004003",
"source_idx": 4003,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8158,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8",
"budget": 6526
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004003",
"normalized_example_id": "ex_004003",
"source_idx": 4003,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8158,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8",
"budget": 8158
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004003",
"normalized_example_id": "ex_004003",
"source_idx": 4003,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8158,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4000",
"budget": 2779
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004006",
"normalized_example_id": "ex_004006",
"source_idx": 4006,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4631,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4000",
"budget": 3705
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004006",
"normalized_example_id": "ex_004006",
"source_idx": 4006,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4631,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "4000",
"budget": 4631
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004006",
"normalized_example_id": "ex_004006",
"source_idx": 4006,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4631,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{ac}{b}",
"budget": 4690
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004007",
"normalized_example_id": "ex_004007",
"source_idx": 4007,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7816,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{ac}{b}",
"budget": 6253
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004007",
"normalized_example_id": "ex_004007",
"source_idx": 4007,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7816,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{ac}{b}",
"budget": 7816
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004007",
"normalized_example_id": "ex_004007",
"source_idx": 4007,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7816,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\sqrt{6}",
"budget": 2044
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004009",
"normalized_example_id": "ex_004009",
"source_idx": 4009,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3406,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\sqrt{6}",
"budget": 2725
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004009",
"normalized_example_id": "ex_004009",
"source_idx": 4009,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3406,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\sqrt{6}",
"budget": 3406
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004009",
"normalized_example_id": "ex_004009",
"source_idx": 4009,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3406,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "605000",
"budget": 5786
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004011",
"normalized_example_id": "ex_004011",
"source_idx": 4011,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9643,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "605000",
"budget": 7714
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004011",
"normalized_example_id": "ex_004011",
"source_idx": 4011,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9643,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "605000",
"budget": 9643
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004011",
"normalized_example_id": "ex_004011",
"source_idx": 4011,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9643,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "63360,73260,83160,93060,93960",
"budget": 2947
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004012",
"normalized_example_id": "ex_004012",
"source_idx": 4012,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4912,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "63360,73260,83160,93060,93960",
"budget": 3930
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004012",
"normalized_example_id": "ex_004012",
"source_idx": 4012,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4912,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "63360,73260,83160,93060,93960",
"budget": 4912
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004012",
"normalized_example_id": "ex_004012",
"source_idx": 4012,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4912,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{4}{3(3x^{2}-2x+1)^{2}}",
"budget": 2150
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004013",
"normalized_example_id": "ex_004013",
"source_idx": 4013,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3583,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{4}{3(3x^{2}-2x+1)^{2}}",
"budget": 2866
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004013",
"normalized_example_id": "ex_004013",
"source_idx": 4013,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3583,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{4}{3(3x^{2}-2x+1)^{2}}",
"budget": 3583
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004013",
"normalized_example_id": "ex_004013",
"source_idx": 4013,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3583,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\mathrm{~}^{3}",
"budget": 2924
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004015",
"normalized_example_id": "ex_004015",
"source_idx": 4015,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4873,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\mathrm{~}^{3}",
"budget": 3898
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004015",
"normalized_example_id": "ex_004015",
"source_idx": 4015,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4873,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\mathrm{~}^{3}",
"budget": 4873
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004015",
"normalized_example_id": "ex_004015",
"source_idx": 4015,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4873,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 1983
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004016",
"normalized_example_id": "ex_004016",
"source_idx": 4016,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3305,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 2644
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004016",
"normalized_example_id": "ex_004016",
"source_idx": 4016,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3305,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 3305
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004016",
"normalized_example_id": "ex_004016",
"source_idx": 4016,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3305,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;0)",
"budget": 2491
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004018",
"normalized_example_id": "ex_004018",
"source_idx": 4018,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4151,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;0)",
"budget": 3321
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004018",
"normalized_example_id": "ex_004018",
"source_idx": 4018,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4151,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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;0)",
"budget": 4151
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004018",
"normalized_example_id": "ex_004018",
"source_idx": 4018,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4151,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4259
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004021",
"normalized_example_id": "ex_004021",
"source_idx": 4021,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7098,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5678
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004021",
"normalized_example_id": "ex_004021",
"source_idx": 4021,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7098,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 7098
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004021",
"normalized_example_id": "ex_004021",
"source_idx": 4021,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7098,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\ln2",
"budget": 4763
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004025",
"normalized_example_id": "ex_004025",
"source_idx": 4025,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7938,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\ln2",
"budget": 6350
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004025",
"normalized_example_id": "ex_004025",
"source_idx": 4025,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7938,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\ln2",
"budget": 7938
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004025",
"normalized_example_id": "ex_004025",
"source_idx": 4025,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7938,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 3238
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004027",
"normalized_example_id": "ex_004027",
"source_idx": 4027,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5396,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 4317
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004027",
"normalized_example_id": "ex_004027",
"source_idx": 4027,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5396,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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",
"budget": 5396
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004027",
"normalized_example_id": "ex_004027",
"source_idx": 4027,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5396,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004028",
"normalized_example_id": "ex_004028",
"source_idx": 4028,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5950,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 4760
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004028",
"normalized_example_id": "ex_004028",
"source_idx": 4028,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5950,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 5950
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004028",
"normalized_example_id": "ex_004028",
"source_idx": 4028,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5950,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "27",
"budget": 5315
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004030",
"normalized_example_id": "ex_004030",
"source_idx": 4030,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8859,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "27",
"budget": 7087
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004030",
"normalized_example_id": "ex_004030",
"source_idx": 4030,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8859,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "27",
"budget": 8859
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004030",
"normalized_example_id": "ex_004030",
"source_idx": 4030,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8859,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2475",
"budget": 777
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004031",
"normalized_example_id": "ex_004031",
"source_idx": 4031,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1295,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2475",
"budget": 1036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004031",
"normalized_example_id": "ex_004031",
"source_idx": 4031,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1295,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2475",
"budget": 1295
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004031",
"normalized_example_id": "ex_004031",
"source_idx": 4031,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1295,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{35}{12}",
"budget": 1333
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004032",
"normalized_example_id": "ex_004032",
"source_idx": 4032,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2222,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{35}{12}",
"budget": 1778
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004032",
"normalized_example_id": "ex_004032",
"source_idx": 4032,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2222,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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{35}{12}",
"budget": 2222
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004032",
"normalized_example_id": "ex_004032",
"source_idx": 4032,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2222,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "570",
"budget": 2883
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004033",
"normalized_example_id": "ex_004033",
"source_idx": 4033,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4805,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "570",
"budget": 3844
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004033",
"normalized_example_id": "ex_004033",
"source_idx": 4033,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4805,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "570",
"budget": 4805
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004033",
"normalized_example_id": "ex_004033",
"source_idx": 4033,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4805,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3.75\\pi",
"budget": 5698
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004035",
"normalized_example_id": "ex_004035",
"source_idx": 4035,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9497,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3.75\\pi",
"budget": 7598
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004035",
"normalized_example_id": "ex_004035",
"source_idx": 4035,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9497,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3.75\\pi",
"budget": 9497
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004035",
"normalized_example_id": "ex_004035",
"source_idx": 4035,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9497,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35,46,74",
"budget": 5874
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004042",
"normalized_example_id": "ex_004042",
"source_idx": 4042,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9790,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35,46,74",
"budget": 7832
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004042",
"normalized_example_id": "ex_004042",
"source_idx": 4042,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9790,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35,46,74",
"budget": 9790
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004042",
"normalized_example_id": "ex_004042",
"source_idx": 4042,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9790,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "32767",
"budget": 4292
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004045",
"normalized_example_id": "ex_004045",
"source_idx": 4045,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7154,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "32767",
"budget": 5723
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004045",
"normalized_example_id": "ex_004045",
"source_idx": 4045,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7154,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "32767",
"budget": 7154
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004045",
"normalized_example_id": "ex_004045",
"source_idx": 4045,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7154,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8.2",
"budget": 1454
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004046",
"normalized_example_id": "ex_004046",
"source_idx": 4046,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2423,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8.2",
"budget": 1938
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004046",
"normalized_example_id": "ex_004046",
"source_idx": 4046,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2423,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "8.2",
"budget": 2423
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004046",
"normalized_example_id": "ex_004046",
"source_idx": 4046,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2423,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7",
"budget": 4081
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004048",
"normalized_example_id": "ex_004048",
"source_idx": 4048,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6801,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7",
"budget": 5441
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004048",
"normalized_example_id": "ex_004048",
"source_idx": 4048,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6801,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "7",
"budget": 6801
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004048",
"normalized_example_id": "ex_004048",
"source_idx": 4048,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6801,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3368
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004050",
"normalized_example_id": "ex_004050",
"source_idx": 4050,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4491
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004050",
"normalized_example_id": "ex_004050",
"source_idx": 4050,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5614
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004050",
"normalized_example_id": "ex_004050",
"source_idx": 4050,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 5614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 1807
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004052",
"normalized_example_id": "ex_004052",
"source_idx": 4052,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3011,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 2409
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004052",
"normalized_example_id": "ex_004052",
"source_idx": 4052,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3011,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 3011
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004052",
"normalized_example_id": "ex_004052",
"source_idx": 4052,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3011,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}",
"budget": 3814
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004053",
"normalized_example_id": "ex_004053",
"source_idx": 4053,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6356,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}",
"budget": 5085
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004053",
"normalized_example_id": "ex_004053",
"source_idx": 4053,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6356,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{4}",
"budget": 6356
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004053",
"normalized_example_id": "ex_004053",
"source_idx": 4053,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6356,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 4094
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004054",
"normalized_example_id": "ex_004054",
"source_idx": 4054,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6823,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 5458
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004054",
"normalized_example_id": "ex_004054",
"source_idx": 4054,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6823,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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": 6823
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004054",
"normalized_example_id": "ex_004054",
"source_idx": 4054,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6823,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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}{10}",
"budget": 1541
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004055",
"normalized_example_id": "ex_004055",
"source_idx": 4055,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2568,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{10}",
"budget": 2054
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004055",
"normalized_example_id": "ex_004055",
"source_idx": 4055,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2568,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{10}",
"budget": 2568
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004055",
"normalized_example_id": "ex_004055",
"source_idx": 4055,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2568,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2018",
"budget": 4568
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004057",
"normalized_example_id": "ex_004057",
"source_idx": 4057,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2018",
"budget": 6091
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004057",
"normalized_example_id": "ex_004057",
"source_idx": 4057,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2018",
"budget": 7614
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004057",
"normalized_example_id": "ex_004057",
"source_idx": 4057,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7614,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{n}=1+\\frac{3}{2}n(n-1),S_{n}=\\frac{1}{2}n(n^{2}+1)",
"budget": 1361
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004058",
"normalized_example_id": "ex_004058",
"source_idx": 4058,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2269,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{n}=1+\\frac{3}{2}n(n-1),S_{n}=\\frac{1}{2}n(n^{2}+1)",
"budget": 1815
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004058",
"normalized_example_id": "ex_004058",
"source_idx": 4058,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2269,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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_{n}=1+\\frac{3}{2}n(n-1),S_{n}=\\frac{1}{2}n(n^{2}+1)",
"budget": 2269
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004058",
"normalized_example_id": "ex_004058",
"source_idx": 4058,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2269,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1214",
"budget": 4771
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004059",
"normalized_example_id": "ex_004059",
"source_idx": 4059,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7951,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1214",
"budget": 6361
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004059",
"normalized_example_id": "ex_004059",
"source_idx": 4059,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7951,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1214",
"budget": 7951
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004059",
"normalized_example_id": "ex_004059",
"source_idx": 4059,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7951,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\binom{n}{4}",
"budget": 4760
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004060",
"normalized_example_id": "ex_004060",
"source_idx": 4060,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7933,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\binom{n}{4}",
"budget": 6346
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004060",
"normalized_example_id": "ex_004060",
"source_idx": 4060,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7933,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step 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\\binom{n}{4}",
"budget": 7933
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_004060",
"normalized_example_id": "ex_004060",
"source_idx": 4060,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7933,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.